FOR RAY WRITTER

I attached Iist of 7 documents to help you do the desirtation. For chapter 2, I need about 20 pages, chapter 1 about 12 pages and chapter 3 about 20 pages. I will let you know how many pages I will need for the chapter 4 and 5 later. The first paper attached is my prospectus. A prospectus a brief paper approved by the program director to allow me to work on a proposal. The title of my proposal topic I am working on is called “ Factors Associated with Marternal Mortality in Accra” That is the first attchments. The second attachment is the checklist you will have to follow to write the dessirtaton. The third attachment is the literature review important information to help you. The fourth and fith attachment is a similar paper some of the student did to graduate. Use that as a guid or example to comple mine since thiers is similar to my topic. The sixth and the seventh attachments is to help you with chapter 3 and 4. So when you open the first attachment which is my prospectus there are articles I listed in the paper that i will use for the chapter 2 literature review. Use them and also use the articles I have in the reference pages. Also in the similar paper attached, you will see that both students used a lot of articles. please use any of thier articles that will be relevant to my paper.I sent you those similar papers to use as a example to write mine. Those student used relevant article in theirs paper and you can use their articles used also to write mine. Follow how they wrote their paper. I need an outline and Chater 2 first wich is the 20 page. For the prospectus attached is the chapter 1. You can see that ifyou follow the checklist, you will see that I needed to add few missing things to complete the chapter 1. That is why I said write about 12 pages to add it to what I have already for the chapter 1 titled “Factors Associated With Marternal Mortality in Accra”. So copy and past the prospectus I have written and just add again the 12 pages to it. Again By following the similar paper attached will help you write the paper. Again don’t forget I have attached a checklist to help you write all the chapters. This paper is APA formart, No plegrirism and check grammer thouroughly. Thanks

Prospectus

Factors Associated with Maternal Mortality in Greater Accra Ghana 2016;

Case-Control Study

Prospectus: Factors associated with Maternal Mortality in greater Accra Ghana 2016;

Case-control Study

Problem Statement

Maternal death is defined as “the death of a woman while pregnant or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to or aggravated by the pregnancy or its management but not from accidental or incidental causes’’ (Menendez et al., 2008, p. 2). Though the causes and risk factors for maternal death are known and preventable, it is a major health problem concentrated in resource-poor regions of the world (Menendez et al., 2008) including Ghana. The reduction of maternal deaths is a key international development goal, therefore health policy and interventions targeted at significantly reducing it should be evidence-based (Khan et al., 2006).

Ghana documented a Maternal Mortality Ratio (MMR) of 350 per 100,000 live births for the year 2012 (Mahama, 2013), however some districts for example Osu Klottey sub metro of the Accra Metropolitan area in its report recorded 428 maternal deaths per 100,000 live births at the end of 2012. This is a 39 percent increment on the 309 per 100,000 live births recorded in 2011. According to Addo and Gudu (2017), the Accra Metropolitan Area, an urban and commercial metropolis in the Greater Accra region has seen collaborative implementation of health policies and programs geared towards reducing maternal mortality over the past three years. Urban populations are mostly assumed to have access to more quality health care systems than their rural counterparts (Addo & Gudu, 2017). However, urban health systems in many low income countries (LIC) and lower middle income countries (MIC) have weak to non-existent public health structures (Coast et al., 2012). They also lack uniform implementation strategies and inadequate infrastructure to improve population health (Coast et al., 2012). Even though Ghana in collaboration with its development partners has implemented interventions to reduce maternal mortality to achieve the United Nations’ Millenium Development Goal (MDG 5) targets, institutional maternal mortality was very high in Osu Klottey Sub Metro for 2016 with the majority (80%) of maternal deaths being among individuals who were antenatal clinic non attendants (Mahama, 2013). Studies have shown lack of access to obstetrics care due to the lack of health facilities, poor transportation system and greater distances between client home and health facilities (Kaye et al., 2003).

Although the causes of maternal deaths are well established, knowledge on effective management of conditions has not translated into significantly improved outcomes (Coast et al., 2012). Observations at health institutions in the Accra Metro area show that service delivery factors such as prenatal care coverage and the presence of a skilled attendant at delivery may play a significant role in the mortalities and therefore needs to be investigated to inform policy decisions if the Sustainable Development Goal (SDG) goal 5 is to be met. The causes of maternal deaths in Ghana follow the trends of the developing country with haemorrhage, hypertensive disorders, abortion related complications, and septicemia leading in that order (Mensah et al., 2011) . In this research study, I will examine the association between sociodemographic and service delivery factors and maternal mortality.

Purpose

The purpose of this study is to examine the socio-demographic and service delivery factors associated with maternal mortality in the Accra Metropolitan Area of Ghana. The maternal mortality rates are not the same in every region and therefore there is a need to investigate whether there are factors that are exclusive to some geographic areas.

Significance

Since Ghana did not achieve its Millennium Development Goal 5 (MDG 5) target, there has been renewed effort to achieve the Sustainable Development Goals on maternal death reduction, yet very little research has been done on the factors for which intervention would yield the most impact. As Accra Metro is a high urbanized setting characterized by rural urban migration, with so much pressure on relatively few health facilities (Report 2016), policy makers need more information on the major risk factors in this setting to guide decision making and resource allocation. The information gathered could inform the Metro and Regional Health Directorate on other policy interventions to help reduce maternal death in the Metro area. The positive social change implication that could result is to improve the quality of institutional antenatal, intra-partum, and post-partum service delivery in the Metropolis and add to the body of knowledge to reduce maternal death in Ghana.

Background

Selected papers and works relating to factors associated with maternal mortality and how to improve outcomes in West Africa particularly Ghana are described below:

1. Menendez et al. (2008) presented the causes and risk factors of maternal deaths as a major health problem in resource-poor regions of the world.

2. Addo and Gudu (2017) set out the various factors that are associated with the utilization of skilled service delivery among women that live in the rural part of northern Ghana.

3. Khan et al. (2006) demonstrated that reduction of maternal deaths is a key international development goal, therefore health policy and interventions targeted at significantly reducing it should be evidence-based.

4. Mahama (2013) reported that Ghana documented Maternal Mortality Ratio (MMR) of 350 per 100,000 live births for the year 2012.

5. The annual report at Accra Metropolitan area, Ghana (2016) documented 428 maternal deaths per 100,000 live births. This is a 39 percent increment on the 309 per 100,000 live births recorded in 2015.

6. Coast and McDaid (2012) reported that urban populations are mostly assumed to have improved access to health care as compared to their rural counterparts, however, urban health systems in many Low Income Countries and Lower Middle Income Countries have weak to non-existent public health structures and lack uniform implementation of strategies and necessary infrastructure.

7. Based on information from UNICEF and WHO, Blencowe (2012) indicated that the sub-Saharan African region had a Maternal Mortality Ratio of 500 deaths per 100,000 live births which is the highest in the world. This has made the region a dangerous place to give birth.

8. Osotimehin (2012) reported maternal death preventable interventions. These include improving access to voluntary family planning, investing in health workers with midwifery skills, and ensuring access to emergency obstetrics care when complications arise.

9. Mensah et al. (2011) described the causes of maternal deaths in Ghana. It followed the trends of causes in developing country, with haemorrhage, hypertensive disorders, abortion related complications, and septicemia leading in that order.

Framework

The theory of social capital is very broad and has found a place in public policy, public health, and more specifically in epidemiology. This theory is founded on several assertions. Chief among these assertions are that social relationships are a determinant of health (Begum, Aziz-un-Nisa, & Begum, 2003). The external environment and the daily social interactions and support systems play a pivotal role in an individual’s overall health. This theory is also founded on the premise that poor social capital is one of the leading causes of physical and mental distress (Krieger, 2011). To elaborate, a strong social connection has been shown to lead to improved all-cause mortality rates. Lack of social connection can have an adverse impact on health outcomes. Social epidemiologists are tasked with identifying the social aspects that affect the pattern of disease distribution and its mechanisms in a populace. Social relationships, social inequalities, and social capital are some of the most important concepts of social epidemiology. Krieger takes the position that social epidemiologists exploit indicators of ‘life chances’ such as occupation, skills and income which inform on social inequality. The underlying factors linked to social equality are the most important determinants of health. The knowledge, skills, and resources possessed by individuals are factors contributing to the social stratification and consequentially the health outcomes of a given population (Krieger, 2011). Research indicates a social gradient of health whereby most of the individuals with a lower socioeconomic position have been shown to have poor health (Krieger, 2011).

Social capital occurs at different levels. These include the macro-level (social, economic and political aspects of society), mesolevel (organizations and the neighborhood) and the individual context through social interactions. Many ecological studies conducted indicate a positive association between social capital and health outcomes (Mensah, Bentil, Adjepong, & Dolo, 2011). The concept of social inequality is of fundamental importance to epidemiology and health research since it is evident that social factors such as level of education and income levels impact access to health and the quality of health care services in a particular region. Epidemiologists can capitalize on this premise to establish health patterns in a given population informed by the socio-economic status of the persons residing there. Social support structures influence help-seeking behavior, adherence to medical treatment and use of health care services (Pearce, 1996). The theoretical aspects of social capital theory and factors behind them such as social inequalities, social capital, and social relationships will form an integral part of my research. This theory will serve as the framework to determine how and to what extent sociodemographic and service delivery factors can affect health care outcomes such as maternal mortality.

Research Questions

RQ1:

· Is there a significant association between sociodemographic factors (marital status, education, income, and health insurance) and maternal mortality in Greater Accra Ghana?

RQ2;

· Is there a significant association between service delivery factors (prenatal care, delivery location, and presence of a skilled attendant at delivery) and maternal mortality in Greater Accra Ghana?

RQ 3:

· Is there a significant difference in maternal survival rates between women in Greater Accra Ghana in terms of health insurance coverage, annual median income above GH¢30.00, education above the high school level, marital status, and location of infant birth?

Nature of the Study

The nature of this study will be a quantitative unmatched case control study. This choice of design is appropriate in determining the strength and direction of association between risk factors and maternal death. For the study population/inclusion criteria, my population will be women of child bearing age living in Accra metropolitan area in Ghana. Cases would be maternal deaths that occurred in hospitals in the Osu Klottey Sub Metro of Accra metropolitan area in 2016. Controls would be mothers who delivered at the Ridge Regional and 37 military hospitals in 2016 that were alive at the end of the puerperal periods. Cases are defined as obstetric patients that died or were declared dead upon arrival, or after admission (including those who died before the fetus was delivered). Controls would be randomly selected from mothers who delivered at Ridge and 37 military hospitals in 2016 and are alive at the end of puerperal period using simple random sampling. For sampling Frame, the second post-natal attendants register in both hospitals would be used for control selections. The study will analyze and utilize secondary data for the year 2016.

Possible Types and Source of Information or Data

· Review medical records (folders) of cases and controls for the period of one year in a hospital in the Accra metropolitan area.

· Review records from the Registry of Births and Deaths in the Accra metropolitan area.

References

Addo, B., & Gudu, W. (2017). Factors associated with utilization of skilled service delivery among women in rural Northern Ghana: A cross sectional study. BMC Pregnancy and Childbirth, 17(1), 159.

Accra Metropolitan, District Analytical Report: Population and Housing Census. (2016). Retrieved from http://www.statsghana.gov.gh/docfiles/2010_District_Report/Greater%20Accra/AMA

Begum, S., Aziz-un, N., & Begum, I. (2003). Analysis of maternal mortality in a tertiary care hospital to determine causes and preventable factors. Journal of Ayub Medical College, Abbottabad, 15(2), 49-52. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/14552250

Blencowe, H., Cousens, S., Oestergaard, M. Z., Chou, D., Moller, A. B., Narwal, R.,… Lawn, J. E. (2012). National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: A systematic analysis and implications. The Lancet, 379(9832), 2162-2172.

Coast, E., McDaid, D., Leone, T., Pitchforth, E., Matthews, Z., Iemmi, V., … Jones, E. (2012). What are the effects of different models of delivery for improving maternal and infant health outcomes for poor people in urban areas in low and lower middle income countries? London, United Kindom: EPPI Center.

Kaye, D., Mirembe, F., Aziga, F., & Namulema, B. (2003). Maternal mortality and associated near-misses among emergency intrapartum obstetric referrals in Mulago Hospital, Kampala, Uganda. East African Medical Journal, 80(3), 144–149. https://doi.org/10.4314/eamj.v80i3.8684.

Krieger, N. (2011). Epidemiology and the people’s health: Theory and context. New York, NY: Oxford University Press.

Mahama, J. (2013). Sessional Address to Parliament. P. a. g. public. Accra. Retrieved from http://www.ohcs.gov.gh/sites/default/files/2013%20ANNUAL%20PERFORMANCE%20REPORT%20-%20Final

Menéndez, C., Romagosa, C., Ismail, M. R., Carrilho, C., Saute, F., Osman, N., … Ordi, J. (2008). An autopsy study of maternal mortality in Mozambique: The contribution of infectious diseases. PLoS Medicine, 5(2), 0220–0226. https://doi.org/10.1371/journal.pmed.0050044

Mensah, F. O., Bentil, E., Adjepong, M., & Dolo, O. (2011). Causes of maternal mortality in Ghana- A case study at the Koforidua Regional Hospital., 1–17. Retrieved from http://hdl.handle.net/123456789/548

Osotimehin, B. (2012). Maternal deaths halved in 20 years but faster progress needed. New York: Social Media Outreach.

Quantitative Dissertation Checklist

· The following provides guidance for reporting on quantitative studies.

· All items may not be relevant to your particular study; please consult with your chair for guidance.

· The checklist items may not necessarily be in the order that works best for your dissertation. Please consult with your committee; however, the checklist should work well in the absence of other considerations.

· Instructions for Students:

· Indicate on the checklist the page number (use the actual document page number, not the MS Word pagination) where the appropriate indicator is located.

· Respond to comments from the chair and/or URR comments in the comment history box. Do not delete previous comments(just add your response and use some means to clearly identify your remarks (different font/bold/italics/color).

· Instructions for the chair and/or URR

· Provide specific feedback in the comment history column. Do not delete previous comments(just add your response and use some means to clearly identify your remarks (different font/bold/italics/color).

· If you made detailed comments on the draft (using track changes and comments), you can make reference to the draft rather than restate everything in the checklist comment history section.

Date: (click here and type today’s date ()

Student’s Name:

Student ID (for office use only) —

School: (click here and pull down to select school name () FORMDROPDOWN

Committee Members’ Names:

Chairperson

Member

University Research Reviewer

Front Matter

Checklist Items

Comment History

Title [Insert Dissertation Title]

Type of relationship between variables.

Key variables (independent and dependent).

Applicable population.

Abstract

Describe the research problem and why it is important.

Identify the purpose of the study.

State the theoretical foundations and/or conceptual frameworks, as appropriate.

Summarize the key research question(s).

Describe, concisely, the overall research design, methods, and data analysis procedures.

Identify key results, conclusions, and recommendations that capture the heart of the research (for the final study only).

Conclude with a statement on the implications for positive social change.

Checklist Items

Comment History

CHAPTER 1
				Pg/NA
				Introduction
Describe the topic of the study, why the study needs to be conducted, and the potential positive social change implications of the study
Preview major sections of the chapter

Background

Briefly summarize research literature related to the scope of the study topic

Describe a gap in knowledge in the discipline the study will address

End the section on why the study is needed

Problem Statement

State the research problem

Summarize evidence of consensus that the problem is current, relevant, and significant to the discipline

Frame the problem in a way that builds upon or counters previous research findings focusing primarily on research conducted in the last five years

Address a meaningful gap in the current research literature

Purpose of the Study

Provide a concise statement that serves as the connection between the problem being addressed and the focus of the study. The purpose contains:

Indication that this is a quantitative study.

The study intent (such as describe, compare, correlate, explore, and develop).

The independent, dependent, and covariate variables.

Research Question(s) and Hypotheses

State the research questions.

State the null and alternative hypotheses that identify the independent and dependent variables being studied, the association being tested, and how the variables are being measured.

Theoretical and/or Conceptual Framework for the Study Studies must include either a theoretical foundation or a conceptual framework section or both.
	Theoretical Foundation
Identify the theory or theories and provide the origin or source.
State concisely the major theoretical propositions and/or major hypotheses with a reference to more detailed explanation in chapter 2.
Explain how the theory relates to the study approach and research questions.
Conceptual Framework This applies to some epidemiological studies (as well as to some other quantitative studies).
Identify and define the concept and/or phenomenon that grounds the study.
Concisely describe the conceptual framework (a description of the body of research that supports the need for the study) as derived from the literature with more detailed analysis in chapter 2.
State the logical connections among key elements of the framework with a reference to a more thorough explanation in chapter 2.
State how the framework relates to the study approach and key research questions, as well as to instrument development and data analysis, where appropriate.
Nature of the Study
Provide a concise rationale for selection of the design and/ or tradition.
Briefly describe the key study variables (independent, dependent, and covariates).
Briefly summarize the methodology (from whom and how data are collected and how data will be analyzed).

Definitions

Provide concise definitions of the independent variable, dependent variable(s), and any covariates (with more detailed analysis of coding, etc. described in chapter 3).

Define terms used in the study that have multiple meanings (e.g., socioeconomic status, educator, health service professional, etc.). Do not include common terms or terms that can easily be looked up in a dictionary.

Include citations that identify support in the professional literature for the definition or operational definition.

Assumptions

Clarify aspects of the study that are believed but cannot be demonstrated to be true. Only include assumptions critical to the meaningfulness of the study.

Describe the reasons why the assumption(s) was (were) necessary in the context of the study.

Scope and Delimitations

Describe specific aspects of the research problem that are addressed in the study and why the specific focus was chosen (issue of internal validity).

Define the boundaries of the study by identifying populations included and excluded and theories and/or conceptual frameworks most related to the area of study that were not investigated (this is an issue of external validity).

Address potential generalizability.

Limitations

Describe limitations of the study related to design and/or methodological weaknesses (including issues related to limitations of internal and external validity, construct validity, and confounder variables).

Describe any biases that could influence study outcomes and how they are addressed.

Describe reasonable measures to address limitations.

Significance
Identify potential contributions of the study that advance knowledge in the discipline. This is an elaboration of what the problem addresses.
Identify potential contributions of the study that advance practice and/or policy, as applicable.
Describe potential implications for positive social change that are consistent with and bounded by the scope of the study.
	Summary
Summarize main points of the chapter.
Provide transition to chapter 2.

Checklist Items

Pg/NA

Comment History

Introduction

CHAPTER 2
Restate the problem and the purpose.
Provide a concise synopsis of the current literature that establishes the relevance of the problem.
	Preview major sections of the chapter.

Literature Search Strategy

List accessed library databases and search engines used.

List key search terms and combinations of search terms (with more detailed search terms located in an appendix, if appropriate).

Describe scope of literature review in terms of years searched as well as types of literature and sources searched, including seminal literature as well as current peer-reviewed literature.

In cases where there is little current research, and few (if any) dissertations and/or conference proceedings, describe how this was handled.

Theoretical Foundation

Name the theory or theories.

Provide origin or source of the theory.

Describe major theoretical propositions and/or major hypotheses, including delineation of any assumptions appropriate to the application of the theory.

Provide a literature- and research- based analysis of how the theory has been applied previously in ways similar to the current study.

Provide the rationale for the choice of this theory.

Describe how and why the selected theory relates to the present study and how the research questions relate to, challenge, or build upon existing theory.

Conceptual Framework (As appropriate)

Identify and define the concept and/or phenomenon.

Synthesize primary writings by key theorists, philosophers, and/or seminal researchers related to the concept or phenomenon.

Provide key statements and definitions inherent in the framework.

Describe how the concept or phenomenon has been applied and articulated in previous research and how the current study benefits from this framework.

Literature Review Related to Key Variables and/or Concepts

Provide an exhaustive review of the current literature that includes the following information:

Describe studies related to the constructs of interest and chosen methodology and methods that are consistent with the scope of the study.

Describe ways researchers in the discipline have approached the problem and the strengths and weakness inherent in their approaches.

Justify from the literature the rationale for selection of the variables or concepts.

Review and synthesize studies related to the key independent, dependent, and covariate variables to produce a description and explanation of what is known about the variables, what is controversial (i.e., mixed findings by researchers), and what remains to be studied.

Review and synthesize studies related to the research questions.

Summary and Conclusions

Concisely summarize major themes in the literature.

Summarize what is known as well as what is not known in the discipline related to the topic of study.

Describe how the present study fills at least one of the gaps in the literature and will extend knowledge in the discipline.

Provide transitional material to connect the gap in the literature to the methods described in chapter 3.

Checklist Items

Pg/NA

Comment History

Introduction

Preview major sections of the chapter.

CHAPTER 3

Restate the study purpose as described in chapter 1.

Research Design and Rationale

Concisely state the study variables (independent, dependent, covariate, mediating, and/or moderating variables, as appropriate.

Identify the research design and its connection to the research questions.

Explain any time and resource constraints consistent with the design choice.

Describe how design choice is consistent with research designs needed to advance knowledge in the discipline.

If conducting an intervention study, defend the choice of intervention.

Methodology

(needs to be described in sufficient depth so that other researchers can replicate the study)

Population

Define the target population.

State target population size (if known) or approximate/estimated size.

Sampling and Sampling Procedures

Identify and justify the type of sampling strategy.

Explain specific procedures for how the sample will be drawn.

Describe the sampling frame (inclusion and exclusion criteria).

Use a power analysis to determine sample size and include:

· Justification for the effect size, alpha level, and power level chosen.

· Cite the source for calculating or the tool used to calculate the sample size.

Procedures For Recruitment, Participation, and Data Collection (for students collecting their own data)

Thoroughly describe recruiting procedures and particular demographic information that will be collected.

Describe how participants will be provided informed consent.

Describe how data are collected.

Explain how participants exit the study (for example, debriefing procedures, etc.).

Describe any follow-up procedures (such as requirements to return for follow-up interviews, treatments, etc.).

Additional Information if Conducting a Pilot Study:

Describe the relationship of the pilot study to the main study (for example, what is the purpose of the pilot study?).

Additional Information if Conducting an Intervention

Describe clearly and thoroughly the nature of the treatment, intervention, or experimental manipulation, how it will be designed and administered, and by whom and to whom it will be administered.

For Students Using Archival Data

Include all procedures for recruitment, participation, and data collection associated with the main study.

Describe the procedure for gaining access to the data set.

Describe necessary permissions to gain access to the data (with permission letters located in an appendix).

If historical or legal documents are used as sources of data, demonstrate the reputability of the sources and justify why they represent the best sources of data.

Instrumentation and Operationalization of Constructs

For published instruments provide:

· Name of developer(s) and year of publication.

· Appropriateness to the current study.

· Permission from developer to use the instrument (permission letter should be included in an appendix).

· Published reliability and validity values relevant to their use in the study.

· Where and/or with what populations the instrument was previously used and how validity/reliability are/were established in the study sample.

For all researcher instruments provide:

· Basis for development (literature sources or other bases for development, such as a pilot study).

· Plan to provide evidence for reliability (for example, internal consistency and test/retest).

· Plan to provide evidence for validity (for example, predictive and construct validity).

· Establish sufficiency of instrumentation to answer research questions.

For intervention studies or those involving manipulation of an independent variable:

Identify materials/programs applied as treatment or manipulation.

Provide information on the developer of the materials and/or programs.

If published, state where, how, and with which populations the instrument was previously used.

If researcher-developed materials, state the basis for development and how the materials were developed.

Provide evidence that another agency will sponsor intervention studies (such as clinical interventions).

Operationalization

For each variable describe:

Its operational definition.

How each variable is measured or manipulated.

How the variable/scale score is calculated, what the scores represent, and an example item.

Data Analysis Plan

Identify software used for analyses.

Provide explanation of data cleaning and screening procedures as appropriate to the study.

Restate the research questions and hypotheses here as written in chapter 1.

Describe in detail the analysis plan including the elements below including:

· Statistical tests that will be used to test the hypothesis(es).

· Procedures used to account for multiple statistical tests, as appropriate.

· Rationale for inclusion of potential covariates and/or confounding variables.

· How results will be interpreted (key parameter estimates, confidence intervals and/or probability values, odds ratios, etc.).

Threats to Validity

Describe threats to external validity (for example, testing reactivity, interaction effects of selection and experimental variables, specificity of variables, reactive effects of experimental arrangements, and multiple-treatment interference, as appropriate to the study) and how they will be and/or were addressed.

Describe threats to internal validity (for example, history, maturation, testing, instrumentation, statistical regression, experimental mortality, and selection-maturation interaction, as appropriate to the study) and how they will be and/or were addressed.

Describe any threats to construct or statistical conclusion validity.

Summary

Ethical Procedures

Agreements to gain access to participants or data (include actual documents in the IRB application).

Describe the treatment of human participants including the following (include actual documents in the Institutional Review Board [IRB] application):

· Institutional permissions, including IRB approvals that are needed (proposal) or were obtained (completed dissertation);include relevant IRB approval numbers in the final dissertation.

· Ethical concerns related to recruitment materials and processes and a plan to address them.

· Ethical concerns related to data collection and/or intervention activities (these could include participants refusing participation or early withdrawal from the study and response to any predicable adverse events) and a plan to address them.

Describe treatment of data (including archival data), including issues of:

· Whether data are anonymous or confidential and any concerns related to each.

· Protections for confidential data (data storage procedures, data dissemination, who will have access to the data, and when data will be destroyed).

Other ethical issues as applicable (these issues could include doing a study within one’s own work environment, conflict of interest or power differentials, and justification for use of incentives).

Summary of design and methodology of the method of inquiry.

Transition to chapter 4.

Checklist Items

Pg/NA

Comment History

Introduction

CHAPTER 4

Review briefly the purpose, research questions, and hypotheses.

Preview the organization of chapter 4.

Pilot Study (if applicable)

Concisely report the results of the pilot study.

Report any impact of the pilot study on the main study (for example, changes in instrumentation, data analysis strategies, etc.).

Data Collection

Describe the time frame for data collection as well as actual recruitment and response rates.

Present any discrepancies in data collection from the plan presented in chapter 3.

Report baseline descriptive and demographic characteristics of the sample

Describe how representative the sample is of the population of interest or how proportional it is to the larger population if non-probability sampling is used (external validity).

Provide results of basic univariate analyses that justify inclusion of covariates in the model, if applicable.

Checklist Items

Pg/NA

Comment History

Introduction

Treatment and/or Intervention Fidelity (as appropriate)

Describe whether the treatment was administered as planned and any challenges that prevented planned implementation as described in chapter 3.

Describe any adverse events (those with serious consequences) related to the intervention.

Results

Report descriptive statistics that appropriately characterize the sample.

Evaluate statistical assumptions as appropriate to the study.

Report statistical analysis findings, organized by research questions and/or hypotheses, including:

· Exact statistics and associated probability values.

· Confidence intervals around the statistics, as appropriate.

· Effect sizes, as appropriate.

Report results of post-hoc analyses of statistical tests, if applicable.

Report any additional statistical tests of hypotheses that emerged from the analysis of main hypotheses, as appropriate for the study.

Include tables and figures to illustrate results, as appropriate, and per the current edition of the Publication Manual of the American Psychological Association

Summary

Summarize answers to research questions.

Provide transitional material from the findings and introduce the reader to the prescriptive material in chapter 5.

CHAPTER 5

Concisely reiterate the purpose and nature of the study and why it was conducted.

Concisely summarize key findings.

Interpretation of the Findings

Describe in what ways findings confirm, disconfirm, or extend knowledge in the discipline by comparing them with what has been found in the peer-reviewed literature described in chapter 2.

Analyze and interpret the findings in the context of the theoretical and/or conceptual framework, as appropriate.

· Ensure interpretations do not exceed the data, findings, and scope.

Limitations of the Study

Describe the limitations to generalizability and/or trustworthiness, validity, and reliability that arose from execution of the study. These should be used to revise what was written in chapter 1 for the proposal.

Recommendations

Describe recommendations for further research that are grounded in the strengths and limitations of the current study as well as the literature reviewed in chapter 2.

· Ensure recommendations do not exceed the study boundaries.

Implications

Positive Social Change

· Describe the potential impact for positive social change at the appropriate level (individual, family, organizational, and societal/policy).

· Ensure implications for social change do not exceed the study boundaries.

Describe methodological, theoretical, and/or empirical implications, as appropriate.

Describe recommendations for practice, as appropriate.

Conclusion

Provide a strong “take home” message that captures the key essence of the study.

Checklist Items

Comment History

APA
Form and Style Check

Citations and Referencing

All citations have been crosschecked to ensure that there are corresponding references (and that there are no references that do not have associated citations).

All sources are cited correctly per
APA formatting requirements
(for example, studies listed in alphabetical order by first author; no first names of authors).

Grammar, Spelling, and Syntax

The paper has been thoroughly checked for
grammar
, spelling, and syntax errors.

For the final dissertation, the dissertation has been checked for correct
verb tense
representing a completed study.

Headings

Headings are used, consistent with the
Walden Dissertation Template
, to make sections of thought distinct.

Use of the Writing Center Template

The
Writing Center Dissertation Template
(APA, 6th edition) was used to construct the proposal and/or dissertation so that all formatting is correct.

Therole of the literature review

Your literature review gives your readers an understanding of the evolution of scholarly research on your topic.

In your literature review you will:

•survey the scholarly landscape

•provide a synthesis of the issues, trends, and concepts

•possibly provide some historical background

Throughout the literature review, your emphasis should fall on the current scholarly conversation. This is why the rubric often specifies that you need resources from peer-reviewed journals, published within the last five years of your anticipated graduation date. It’s in these recent, peer-reviewed journals that the scholarly debate is being carried out!

The literature review also shows the “gap” in the conversation — and how your own doctoral study will fill that gap and contribute to the scholarly knowledge. This is where you make the case for the importance and usefulness for your own work.

Searching comprehensively

Your literature review should be as comprehensive as possible — you want to include all of the relevant resources dealing with your topic. Missing important articles or researchers will significantly weaken your scholarship! So, searching comprehensively becomes important.

To ensuring comprehensiveness:

•Identify the databases that will cover your topic

◦Spend some time reading the descriptions of the databases in your subject area

◦Contact the Library to get advice from a librarian on appropriate databases

◦Some topics cross over subject/theoretical boundaries, and librarians can suggest databases that you may not have considered

•Search in more than one database

◦Some of our databases are huge, containing thousands of journals, but no single database covers every journal relevant to a topic

◦Searching in each relevant database, one at a time, gives you a better sense of control over your search, as well as a more accurate idea of the journals/databases that you’ve covered

Using a multi-database search (such as Thoreau) is not necessarily recommended; in doing so, you lose the ability to use subject terms and search limits that may be unique to each database.

•Explore resources outside of the databases:

◦Government websites

◦Professional organizations

◦Research groups

◦Think tanks

These can all be important sources of statistics and reliable information. These will not be peer-reviewed resources (i.e. since they are not journals, they do not employ the same sort of editorial process that results in peer-review). Evaluating for reliability is important!

Beyond the Library: Google Scholar

Google Scholar provides a good way to take your search beyond the databases; it searches very broadly and will pull in resources you may not have discovered before.

Google’s definition of scholarly includes government sites, think tanks, research organizations, journal websites, and of course colleges and universities.

Unfortunately, there is no way to limit your Google Scholar search to only peer-reviewed resources — so, you will need to invest time and skill in evaluating the resource, before deciding if it’s something that can be included.

Learn more with our Google Scholar guide.

Scholarly/peer-reviewed resources

The standards for scholarly rigor set in the rubric require you to focus your research in the scholarly (i.e. peer-reviewed) journals, most of which are found in our research databases. Discover databases in your field of study on our Articles by Topic page.

Being able to verify that a resource is peer-reviewed is very important as it helps to show you are meeting scholarly rigor.

The most reliable way is to verify peer-review is to look up the journal in Ulrich’s Periodical’s Directory.

Ulrich’s can tell you, authoritatively, if the journal follows a peer-reviewed (or “refereed”) editorial process.

Finding other dissertations/doctoral studies

Finding dissertations and doctoral studies can be useful. While these are not peer-reviewed resources, they are valuable sources of information and citations that can inform your own work.

Literature review process

The literature review process is an iterative one. In this process you will:

1.Choose a topic.

2.Search for current literature on the topic.

3.Evaluate the literature you find.

4.Search for more literature, using what you learned in the evaluation step to inform your search.

5.Once you have exhausted the literature, synthesize and summarize it.

For more information

See these guides for more on these topics:

• Find Full Text

• Google Scholar

• Keyword Searching: Finding Articles on Your Topic

• Peer Review

Basics of Literature Reviews

A literature review is a written approach to examining published information on a particular topic or field. Authors use this review of literature to create a foundation and justification for their research or to demonstrate knowledge on the current state of a field. This review can take the form of a course assignment or a section of a longer capstone project. Read on for more information about writing a strong literature review!

Students often misinterpret the term literature review to mean merely a collection of source summaries, similar to annotations or article abstracts. Although summarizing is an element of a literature review, you will want to approach this assignment as a comprehensive representation of your understanding of a topic or field, such as what has already been done or what has been found. Then, also using these sources, you can demonstrate the need for future research, specifically, your future research.

There is usually no required format or template for a literature review. However, there are some actions to keep in mind when constructing your review:

1. Include an

introduction

and

conclusion

. Even if the literature review will be part of a longer document, these paragraphs can act as bookends to your material. Provide background information for your reader, such as including references to the pioneers in the field in the beginning and offering closure in the end by discussing the implications of future research to the field.

2. Avoid direct quotations. Just like in an annotated bibliography, you will want to

paraphrase

all of the material you present in a literature review. This assignment is a chance for you to demonstrate your knowledge on a topic, and putting ideas into your own words will ensure that you are interpreting the found material for your reader. Paraphrasing will also ensure your review of literature is in your authorial voice.

3. Organize by topic or theme rather than by author. When compiling multiple sources, our tendency as writers can be to summarize each source and then compare and contrast the sources at the end. Instead, organize your sources by your identified themes and patterns. This organization helps demonstrate your synthesis of the material and inhibits you from creating a series of book reports.

4. Use

headings

. APA encourages the use of headings within longer pieces of text to display a shift in topic and create a visual break for the reader. Headings in a literature review can also help you as the writer organize your material by theme and note any layers, or subtopics, within the field.

5. Use comparative terms. A literature review can be lengthy and dense, so you will want to make your text appealing to your reader.

Transitions

and comparison terms will allow you to demonstrate where authors agree or disagree on a topic and highlight your interpretation of the literature.

Synthesizing your sources

In order to demonstrate your knowledge on a field through a review of literature, the key component is synthesis. To synthesize is to combine independent elements and form a cohesive whole; in essence, your literature review should integrate your sources and

· Identify patterns

· Critically discuss strengths and weaknesses of sources or the field

· Compare and contrast methods, approaches, and findings of authors

· Evaluate and interpret what is known in your field and what, if anything, is missing

A Metaphor for Synthesis

Imagine you are at a dinner party with other researchers and theorists from your field. Everyone is sitting around the table and discussing the state of your field of research. The beginning portion of your literature review would be similar to those dinner party guests who started the conversation by discussing foundational research and theories. The body of your literature review could take many forms: What guests are agreeing, and which are arguing? What are the debatable issues, and are there any subtopics of those key topics? Does one particular guest keep interrupting the table’s conversation? The final portion of your literature review would be similar to the host of the dinner party ending the debate with a comprehensive speech that touches on all opinions yet provides closure for the conversation.

Commentary v. opinion

In order to synthesize your sources, you must first analyze them to help provide rationale for why they are a part of your literature review and what role they play within your field. It can be difficult, however, to demonstrate analysis without inserting one’s opinions or beliefs.

Consider the following, analysis-free excerpt. This approach is typical of a student looking to avoid opinion in the paper:

….is to deny the student (Sigree, 1999).
As Harper (2001) noted, instructors cannot identify every one of their students’ emotional intelligences (EI). Faculty members do not have the time, and students simply are not that forthcoming with their learning preferences (Harper, 2001). Furthermore, as Harper warned, if instructors decide to attempt a complete analysis of every student’s EI, they will inevitably hold the entire class back. After all, taking time to adequately diagnose a student’s EI means less time for helping students meet the expectations set forth by the No Child Left Behind Act (Harper, 2001).
Finkelstein and Kramer’s (2002) findings…

There is no analysis or critique in this excerpt. There is strong paraphrasing, and this passage provides a decent overview of Harper, but it addresses only Harper’s ideas and does not explain why this information is important and how it relates to the author’s overall purpose for the paper. The reader needs to know the answer to “So what, and who cares?”

What is missing from this summary is context and analysis. Consider the following revision:

….is to deny the student (Sigree, 1999).
    Harper (2001), however, disagreed with Sigree’s (1999) assertion. Harper noted that despite the obvious benefits of diagnosing a student’s emotional intelligence (EI; Jones & Hammer, 1998; Mooney, 1998; Sigree, 1999), instructors cannot identify every one of their students’ EIs (Harper, 2001). Faculty members do not have the time, and students are not that forthcoming with their learning preferences (Harper, 2001).
     For Harper (2001), though, the real issue was not with instructors’ belief in EI, but rather in how this belief affected classroom logistics. Instructors who follow Earnhart’s (1996) advice to “Take the time to understand how each of your students learn” (p. 33) are being impractical, Harper argued. Taking time to adequately diagnose a student’s EI means less time for helping students meet the expectations set forth by the No Child Left Behind Act (Harper, 2001). Although Earnhart’s (1996) vision is ideal, Harper takes a more practical stance.
    With Harper’s (2001) concerns in mind, I cannot endorse Finkelstein and Kramer’s (2002) findings…

Here, the author is synthesizing the literature. We know, based on the author’s direction, how Harper interacts with the other literature on the topic. We know that Harper is probably in the minority, and we know what the author’s take on Harper is. Finally, we know how and why the author is using Harper: Harper will be used to refute Finkelstein and Kramer, which presumably is the author’s intent or thesis. Notice how the author demonstrated analysis and synthesis in just a few additional sentences. Note, also, that the author includes an “I” statement in the passage but did not insert his or her opinion.

LIterature review matrix

As you read and evaluate your literature there are several different ways to organize your research. Courtesy of Dr. Gary Burkholder in the School of Psychology, these sample matrices are one option to help organize your articles. These documents allow you to compile details about your sources, such as the foundational theories, methodologies, and conclusions; begin to note similarities among the authors; and retrieve citation information for easy insertion within a document.

You can review the sample matrixes to see a completed form or download the blank matrix for your own use.

Walden

University

ScholarWorks

Walden Dissertations and Doctoral Studies
Walden Dissertations and Doctoral Studies

Collection

2018

Maternal Socioeconomic Status and Human
Papilloma Virus Vaccine Uptake
Shawn Lockett

Walden University

Follow this and additional works at: http://scholarworks.waldenu.edu/dissertations

Part of the Public Health Education and Promotion Commons

This Dissertation is brought to you for free and open access by the Walden Dissertations and Doctoral Studies Collection at ScholarWorks. It has been
accepted for inclusion in Walden Dissertations and Doctoral Studies by an authorized administrator of ScholarWorks. For more information, please
contact ScholarWorks@waldenu.edu.

http://www.waldenu.edu/?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F4650&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F4650&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu/dissertations?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F4650&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu/dissanddoc?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F4650&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu/dissertations?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F4650&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/743?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F4650&utm_medium=PDF&utm_campaign=PDFCoverPages

mailto:ScholarWorks@waldenu.edu

Walden University

College of Health Sciences

This is to certify that the doctoral study

Shawn Lockett

has been found to be complete and satisfactory in all respects,

and that any and all revisions required by
the review committee have been made.

Review Committee
Dr. Peter Anderson, Committee Chairperson, Public Health Faculty

Dr. Hope King, Committee Member, Public Health Faculty
Dr. Ronald Hudak, University Reviewer, Public Health Faculty

Chief Academic Officer
Eric Riedel, Ph.D.

Walden University

201

Abstract

Maternal Socioeconomic Status and Human Papilloma Virus Vaccine

Uptake

Shawn Terrence Lockett

MPH, University of Oklahoma, 199

BS, University of Oklahoma, 199

Doctoral Study Submitted in Partial Fulfillment

of the Requirements for the Degree of

Doctor of Public Health

Walden University

February 2018

Abstract

There are more than 79 million people

in the U.S.

currently infected with human

papillomavirus (HPV), with an estimated 14 million new infections annually. There is a

lack of knowledge about the maternal socioeconomic influences and uptake of the HPV

vaccine series. Infection with HPV can cause cervical cancer in women, and there are

over 11,000 cervical cancer diagnoses in the U.S. responsible for 4000 deaths annually.

Vaccination coverage to prevent HPV infection does not meet the Healthy People 20

goals of an 80% vaccination rate in the U.S. In this study, associations were tested

between maternal SES variables and uptake of the HPV vaccine in male and female

adolescents ages 13-17 fro

1,125 participants who lived within the estimation areas of

New York City, New York and Houston, Texas in 2014. The health belief model was

used as the theoretical framework for the study. This was a cross-sectional quantitative

study using multiple logistic regression analysis of 4 maternal predictor variables. It was

found that 3 of the variables (income, p > .05, education β = -.026, p > .05, and age β = –

.096, p > .05) were not significantly related to uptake of the HPV vaccine series, whereas

ethnicity was found to be significant (Non-Hispanic White β = .429, p = .029, Non-

Hispanic Black β = .587, p = .002, and Non-Hispanic Other β = .586, p =.011). Hispanics

were nearly 2 times more likely to be vaccinated than other groups. The potential social

change implications of this research are that public health workers can use the findings to

develop targeted interventions to increase HPV vaccination uptake and reduce the

incidence of cervical cancer.

Maternal Socioeconomic Status and Human Papilloma Virus Vaccine Uptake

by
Shawn Terrence Lockett

MPH, University of Oklahoma, 19

BS, University of Oklahoma, 19

Doctoral Study Submitted in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Public Health

Walden University
February 201

Dedication

This research is dedicated to my son, Donovan Edward Lockett (December 7,

2005 – 9 August 2013). Donovan’s death at age 7 was my inspiration for returning to

school to finish my long-term goal of completing a doctorate. I will always remember my

son and his positive effect on my life from the day of his birth. I was proud to be his

father.

To my other children, Kaitlynn, Veronica, and Alexander for their unconditional

love and support through years of moving to different countries, meeting new friends and

leaving old friends related to our transient lifestyle within the Department of State.

To my late grandmother, Henrietta Lockett who helped form my ethical compass

and has been guiding me these last 17 years from Heaven.

Acknowledgments

I would like to profusely thank my advisor and committee chair, Dr. Peter

Anderson, for his frank coaching and providing the support that I needed to finish my

Dr.PH study and research; for his considerable patience, sincerity, positive motivation,

and knowledge. His guidance fully supported me through all the research and writing of

my final study. The research process has humbled me, and I thank Dr. Anderson for his

steady mentorship as I navigated the arduous research process.

In addition to my advisor, I would like to thank my former spouse, Stephanie

Laxa Lockett for all her dedication, support, and space that she provided me while I

completed the Dr.PH program. She was indeed the foundation that I needed to complete

the program. I would also like to thank my parents: Warren and Pauline Langdon for

providing me the environment and guidance that allowed me to reach my goals.

My close friends also played a significant role in the completion of my Dr.PH

program. I would like to thank my friends, Les Landry, Joe Santos, Dale Rush and

Michael Voorhies for the emotional support and motivation they provided me.

Lastly, I would like to thank my Department of State medical colleagues, Dr.

Jennifer Tseng, Dr. Jason Coe, Dr. Barry Fisher, and Dr. Edward Miron who provided the

professional mentorship that kept me focused on achieving my goal.

Table of Contents

List of Tables ………………………………………………………………………………………………………..

Section 1: Foundation of the Study and Literature Review ………………………………………….

Introduction ……………………………………………………………………………………………………..1

Problem Statement ……………………………………………………………………………………………

Purpose of the Study …………………………………………………………………………………………

Research Question(s) and Hypotheses …………………………………………………………………5

Theoretical Foundation for the Study ………………………………………………………………….7

Nature of the Study …………………………………………………………………………………………..9

Literature Review Search Strategy ……………………………………………………………………

Literature Review of Key Concepts …………………………………………………………………..1

Maternal Income ………………………………………………………………………………………

Maternal Education …………………………………………………………………………………..

Maternal Age …………………………………………………………………………………………..

Ethnicity ………………………………………………………………………………………………….

Decisional Influences ………………………………………………………………………………………

Critics and Differing Opinions………………………………………………………………………….

Definitions……………………………………………………………………………………………………..

Assumptions …………………………………………………………………………………………………..

Scope and Delimitations ………………………………………………………………………………….

Scope and Delimitations …………………………………………………………………………… 44

Significance and Potential for Social Change ……………………………………………………..45

Significance of Study ………………………………………………………………………………..

Social Change …………………………………………………………………………………………. 45

Summary ……………………………………………………………………………………………………….4

Conclusion …………………………………………………………………………………………………….

Section 2: Research Design and Data Collection ……………………………………………………..

Introduction ……………………………………………………………………………………………………48

Research Design and Rationale ………………………………………………………………………..48

Methodology ………………………………………………………………………………………………….

Study Population ……………………………………………………………………………………… 50

Sampling and Sampling Procedures ………………………………………………………………….

Access to Secondary Data ………………………………………………………………………….

Instrumentation and Operationalization of Constructs …………………………………………

Instrumentation ……………………………………………………………………………………….. 54

Operationalization …………………………………………………………………………………….

Data Analysis Plan ………………………………………………………………………………………….

Research Question(s) and Hypotheses ……………………………………………………………….56

Threats to Validity ………………………………………………………………………………………….

Ethical Considerations …………………………………………………………………………………….

Human Subjects ………………………………………………………………………………………. 60

Ethical Issues ………………………………………………………………………………………….. 60

Summary ……………………………………………………………………………………………………….

Section 3: Presentation of the Results and Findings ………………………………………………….63

iii

Introduction ……………………………………………………………………………………………………

Data Collection of Secondary Data Set ……………………………………………………………..

Discrepancies …………………………………………………………………………………………..

Univariate Analysis …………………………………………………………………………………………

Descriptive Characteristics of the Sample Population …………………………………… 70

Bivariate Analysis …………………………………………………………………………………………..

Logistic Regression Analysis ……………………………………………………………………………

Results

Research Question 1 …………………………………………………………………………………

Research Question 2 ………………………………………………………………………………… 81

Research Question 3 …………………………………………………………………………………

Research Question 4 ………………………………………………………………………………… 82

Summary ……………………………………………………………………………………………………….

Section 4: Application to Professional Practice and Implications for Social

Change …………………………………………………………………………………………………….

Introduction ……………………………………………………………………………………………………85

Concise Summary of Findings ………………………………………………………………………….85

Interpretation of the Findings……………………………………………………………………………

Ethnicity …………………………………………………………………………………………………. 86

Maternal Age ………………………………………………………………………………………….. 86

Maternal Income ………………………………………………………………………………………

Maternal Education ………………………………………………………………………………….. 87

Conceptual Framework ……………………………………………………………………………………

Limitations of the Study…………………………………………………………………………………..

Recommendations …………………………………………………………………………………………..

Implications for Professional Practice and Social Change ……………………………………

Professional Practice ………………………………………………………………………………… 91

Implications for Research …………………………………………………………………………. 91

Positive Social Change ……………………………………………………………………………..

Conclusion …………………………………………………………………………………………………….93

References …………………………………………………………………………………………………………..

List of Tables

Table 1. Health Belief Model…………………………………………………………………………………..9

Table 2. Maternal Age ………………………………………………………………………………………….

Table 3. Maternal Income ……………………………………………………………………………………..71

Table 4. Maternal Education ………………………………………………………………………………….

Table 5. Ethnicity …………………………………………………………………………………………………72

Table 6. HPV Vaccine Series Uptake ……………………………………………………………………..

Table 7. Estimation Area of Residence ……………………………………………………………………73

Table 8. Gender of Child ………………………………………………………………………………………73

Table 9. Crosstabulation Ethnicity and HPV Vaccine Uptake …………………………………..

Table 10. Crosstabulation Maternal Age and HPV Vaccine Uptake ……………………………75

Table 11. Crosstabulation Maternal Education and HPV Vaccine Uptake …………………..

Table 12. Crosstabulation Maternal Income and HPV Vaccine Uptake ………………………

Table 13. Logistic Regression Results for Maternal Education, Maternal Age,

Maternal Race/Ethnicity, and Maternal Income as Predictors of Teens’

HPV Vaccine Series Uptake ……………………………………………………………………….80

Section 1: Foundation of the Study and Literature Review

Introduction

The genital human papillomavirus (HPV) is the most common sexually

transmitted disease in the United States (Centers for Disease Control and Prevention

[CDC], 2013). There are more than 79 million people in the United States currently

infected with HPV, and an estimated 14 million new infections occur every year (CDC,

2014a). Infection with HPV can cause cervical cancer in women and is the second

leading cause of cancer deaths in women worldwide (CDC,

2013).

The first HPV vaccine, Gardasil, was a significant step forward in the fight

against cervical cancer. Gardasil, introduced in June 2006, immunized against HPV

serotypes 6 and 11, plus the oncogenic HPV serotypes 16 and 18 (CDC, 2015a). A

second vaccine, Cervarix, introduced in 2007, also immunized against the oncogenic

HPV serotypes 16 and 18 (CDC, 2013). Collectively, Gardasil and Cervarix, both of

which are a three-shot vaccination series, immunized adolescent females to the serotypes

that account for 70% of cervical cancers (van Keulen et al., 2013). As previously stated,

Gardasil also immunized against non-oncogenic HPV serotypes 6 and 11, which cause

genital warts and which can affect men as well as women. Because of its efficacy against

genital warts and oncogenic strains of HPV, Gardasil has been the only vaccine approved

for use in both adolescent males and adolescent females since 2009 (U.S. Food and Drug

Administration, 2013). Cervarix is also approved for use in HPV infection and cervical

cancer prevention, but its use was restricted for use in females only (CDC, 2013).

In 2013, there were more than 6.2 million new HPV cases reported in the United

States, and HPV was responsible for over 11,000 new cases of cervical cancer

contributing to 4,000 deaths (Nettleman & Garcia-Chen, 2013). In 2014, there were an

estimated 14 million new HPV infections (CDC, 2014b). Furthermore, despite the release

of a safe, efficacious HPV vaccine in 2006, cervical cancer remains the second largest

killer of women worldwide (Crowcroft, Hamid, Deeks, & Frank, 2012; Union for

International Cancer Control (UICC), 2015). There has been some success against

cervical cancer as the incidences in the United States have significantly decreased since

the introduction of the Papanicolaou (Pap) test, or “Pap smear,” in 1941 (Techakehakij &

Feldman, 2008).

The Pap smear test enabled clinicians to screen for early-stage cervical cancer and

earlier detection of cervical tissue changes related to HPV infection. The result of the

screening test was earlier identification, intervention, and improved overall outcomes as

the mortality rate of cervical cancer in the United States decreased by 70% after the

introduction of the screening test (Techakehakij & Feldman, 2008). However, even

though there has been a significant improvement in the identification and treatment of

cervical cancer, it is still a significant burden to those affected by the disease (National

Institutes of Health, 2013). Both men and women can carry HPV, and together they

equally contribute to an epidemic that accounts for the most prevalent sexually

transmitted disease in the United States (Malkowski, 2014; Vanderpool, Van Meter

Dressler, Stradtman, & Crosby, 2015).

There are significant health disparities associated with race, ethnicity, and

SES(SES) regarding HPV vaccine uptake, which puts vulnerable groups at increased risk

of contracting cervical cancer (Btoush, Brown, Fogarty, & Carmody, 2015; Daniel-Ulloa,

Gilbert, & Parker, 2016). Researchers and public health officials have taken significant

steps over the years to study these disparities and improve intervention programs, but the

barriers remain, resulting in only a modest increase in the uptake of the HPV vaccine

(Schmidt & Parsons, 2014).

The potential for significant social change related to this study is based on the

evidence of an association between measurable maternal SES influences (maternal

income, maternal education) and uptake of the HPV vaccine series. Public health officials

and researchers could potentially use the results of the research to reduce the burden of

cervical cancer in women through the enhancement of vaccination programs contributing

to the decreased incidence of a significant health disparity for women.

Section 1 is an introduction to the subject of cervical cancer and its impact on

morbidity and mortality. I introduce the HPV vaccine series and its role in decreasing the

incidence of HPV infection, as HPV infection is a precursor to the development of

cervical cancer. I then describe the study, beginning with a discussion of the problem

addressed by the research, followed by a statement of the purpose of the study. I also

introduce the research questions and demonstrate how this research filled the gap in the

existing literature.

Problem Statement

There is a lack of knowledge about the maternal socioeconomic (SES) influences

and the voluntary uptake of the HPV vaccine series. Due to multifactorial issues, there

has been resistance to the uptake of the HPV vaccine (Navarro-Illana, Aznar, & Díez-

Domingo, 2014). Apte, Pierre-Joseph, Vercruysse, and Perkins (2015) reported that in the

United States, 57% of female adolescents and 34% of male adolescents initiated the HPV

vaccine series in 2013. The percentage of vaccinated male and female participants has

grown in recent years but does not meet the Healthy People 2020 goals of 80% (Savoy,

2014). Additionally, according to one study, over 30% of females and more than 50% of

males did not complete the series of three vaccinations (Apte et al., 2015). Lastly, there

was evidence of family influences decreasing the uptake of the HPV vaccine series for

their adolescent children. Cullen, Stokley, and Markowitz (2014) concluded that

increasing parent education could increase uptake of the vaccine. Attanasio and

McAlpine (2014) reported that the mother’s education level influenced the accuracy of

recall of HPV vaccinations given to their children. The lack of acceptance and subsequent

completion of the HPV vaccination series poses a continued public health threat, and

more research was needed to improve intervention programs, remove barriers, and

increase confidence in the safety and efficacy of the HPV vaccine (Savoy, 2014). Other

than research performed by Musto et al. (2013), which analyzed school-based service

delivery models based on the SES status of schools, there are limited studies in recent

literature addressing the associations between maternal SES influences and the voluntary

uptake of the HPV vaccine series. In my research, I tested for such associations to gain

knowledge that researchers and public health officials could use to potentially enhance

educational programs designed to improve vaccination rates, increase prevention, and

reduce the overall incidence of cervical cancer. The problem statement for this study is as

follows: There is a lack of knowledge about the maternal socioeconomic influences and

the voluntary uptake of the HPV vaccine series. In this study, I sought to identify an

association between measurable maternal SES influences and uptake of the HPV Vaccine

based on the responses by participants of the 2014 National Immunization Survey-Teen

(NIS-Teen).

Purpose of the Study

In this cross-sectional quantitative research study, I investigated the association

between maternal SES variables and uptake of the HPV vaccine in female and male

adolescents, aged 13-17, based on postal codes within the city of Columbus, Ohio. I

included additional variables to observe for associations with maternal age and ethnicity.

Research Question(s) and Hypotheses

The objective of this research study was to explore the association between

maternal SES and uptake of the HPV vaccine series. The research questions were as

follows:

Q1: What is the association between maternal income and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in communities with

postal codes in

the

Columbus, Ohio metropolis?

H01: No association exists between maternal income and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 after controlling for

ethnicity

and maternal age based on postal codes in Columbus, Ohio.

Ha1: An association exists between maternal income and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 after controlling for

ethnicity and maternal age based on postal codes in Columbus, Ohio.

Q2: What is the association between maternal education and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in the postal codes

within the

Columbus,

Ohio metropolis?

H02: No association exists between maternal education and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in the postal codes

within

the Columbus, Ohio

metropolis.

Ha2: An association exists between maternal education and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in the postal codes

within

the

Columbus,

Ohio metropolis.

RQ3: What is the association between maternal age and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in communities with
postal codes in the Columbus, Ohio metropolis?

H03: There is no association between maternal age and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in communities

with postal codes in the Columbus, Ohio metropolis.

Ha3: There is an association between maternal age and uptake of the HPV

vaccine series in male and female adolescents aged 13-17 in communities
with postal codes in the Columbus, Ohio metropolis.

Q4: What is the association between ethnicity and uptake of the HPV vaccine

series in male and female adolescents aged 13-17 in the postal codes within the

Columbus, Ohio metropolis?

H04: There is no association between ethnicity and uptake of the HPV vaccine

series in male and female adolescents aged 13-17 in the postal codes within

the Columbus, Ohio metropolis.

Ha4: There is an association between ethnicity and uptake of the HPV vaccine

series in male and female adolescents aged 13-17 in the postal codes within
the

Columbus, Ohio metropolis.

Theoretical Foundation for the Study

The theoretical framework for this study was the health belief model (HBM). The

HBM was developed in the 1950s by psychologists in the U.S. Public Health Service to

determine the rationale of people to not participate in programs that prevent and detect

disease (Skinner, Tiro, & Champion, 2015). In the 1950s, there was a widespread failure

of people to participate in screening and preventative programs for the early detection of

asymptomatic disease (Rosenstock, 1974). The HBM consists of five core constructs

proposed in order to influence an individual to perform a particular healthy behavior:

perceived susceptibility, perceived severity, perceived benefits, perceived barriers, cues

to action, and self-efficacy. The HBM has some limitations, most notably the low

predictive capacity (R2 < 0.21 on average) of existing HBM variables coupled with the

small effect size of individual variables (Orji, Vassileva, & Mandryk, 2012). The second

limitation of the HBM was the lack of clear guidance on its usage in combination and

relationship between the individual variables being studied (Orji et al., 2012). In my

study, I focused on the modifying factors that influence individual beliefs as defined

within the constructs of the HBM. As applied to this research, under the HBM, I would

evaluate my independent variables of maternal income, maternal education, and maternal

age to see if they significantly influence the dependent variable of uptake of the HPV

vaccine series through the constructs of the HBM. My rationale for using the HBM was

based on the hypothesis that differences in maternal SES and maternal age may have an

association with the uptake of the HPV vaccine series in communities defined by postal

codes in the Columbus, Ohio metropolis. The HBM was used in this study to look for an

association with measurable maternal modifying factors that may influence the uptake of

the HPV vaccine series by way of the HBM. Low SES has been associated with many

different disease processes (Goldberg, 2014). For example, research by Nicolai et al.

(2013) showed that higher rates of the precursors of cervical cancer, cervical

intraepithelial neoplasia grades 2, 2/3 and 3 (CIN2+) and adenocarcinoma in situ (AIS)

were associated with higher levels of poverty and occurred disproportionately among

Black residents. In this study, I focused on maternal SES factors by exploring potential

associations based on maternal income, maternal education, and maternal age, as well as

the ethnicity of the participants, through the analysis of secondary data derived from the

2014 NIS-Teen survey.

Table 1

Health Belief Model

Modifying Factors Individual Beliefs Action
Maternal Income Perceived

susceptibility and
severity

Mother’s belief that
her child can get
HPV and HPV can
lead to cervical
cancer

Uptake of HPV
vaccine

Maternal

Education

Perceived benefits Mother’s belief that
vaccination of her
child with the HPV
vaccine series will
prevent HPV
infection and
cervical cancer

Non uptake of the
HPV vaccine series

Maternal Age Perceived Barriers Mother’s personal
barriers to vaccinate
her children (i.e.,
insurance coverage,
cost, knowledge
about disease or
vaccine)

Ethnicity Cues to Action Strategies to activate
mother’s readiness to
vaccinate

Self-Efficacy Non-applicable after
uptake of vaccine

Nature of the Study

This was a cross-sectional quantitative study in which I investigated the

association between maternal income, maternal education, maternal age and ethnicity,

and the outcome of the uptake of the HPV vaccine series. I analyzed the categorical

independent variables (maternal income, maternal education, maternal age and ethnicity)

with the categorical dependent variable (uptake of the HPV vaccine series). The most

appropriate method of statistical analysis for these variables was a multiple logistical

regression. My rationale for using multiple logistic regression was its appropriateness to

explore for a functional association between the independent variables and the dependent

variable. This statistical plan can be used to predict probabilities of an effect of multiple

independent variables on a categorical dichotomous dependent variable, and in some

circumstances multiple logistic regression can be used to make inferences about which

independent variables have a larger effect or stronger association with the dependent

variable (McDonald, 2014).

My study analyzed a secondary dataset that contains detailed information about

the uptake of the HPV vaccine series, maternal income, maternal education, maternal age,

and ethnicity by postal code. I defined the dependent variable, HPV vaccination uptake as

a dichotomous (yes/no) response on whether the male of female adolescents received at

least one dose of the vaccine. The independent variables were (a) maternal income

defined as the income reported by the respondents living within certain postal codes

within the Columbus, Ohio metropolitan area, (b) maternal education, (c) maternal age,

and (d) ethnicity. Maternal income was defined by total reported combined family

income, separated into the following groups: less than $20,000, $20,000 to $39,999,

$40,000 to $59,999, $60,000 to $75,000, and more than $75,000. I defined maternal

education by the highest level of education attained, separated into the following

categories: no high school diploma; high school graduate or GED; completed a

vocational, trade, or business school program; some college credit but no degree;

associate degree (AA, AS); bachelor’s degree (BA, BS, AB); master’s degree (MA, MS,

MSW, MBA); and doctorate (PhD, EDD) or professional degree (MD, DDS, DVM, JD).

Maternal age was measured by dividing the mothers’ age into the following groups: less

than 25 years old, 25-34 years old, 35-44 years, and 45+ years old. Ethnicity was

measured by the following six categories: White, Black/African-American, Native

American, Asian, Native Hawaiian, and Pacific Islander. The secondary dataset I used for

the study was the 2014 NIS-Teen. The rationale for the dataset was that it surveys HPV

vaccination, maternal education, and annual income. The 2013 NIS-Teen was an

instrument that researchers used to record the responses of over 18,000 households across

the United States (CDC, 2015c). For the purposes of my research, the 2014 NIS-Teen for

participants in Columbus, Ohio needed to be at least 1,188 participants based on

calculations using G*Power 3.1 Statistical Power Analysis for a logistic regression two-

tailed analysis with 0.95 power (1-β err prob; Faul, Erdfelder, Lang, & Buchner, 2007).

Since the NIS-Teen is an annual survey, there should be enough current information and,

therefore, less risk of the research being duplicated. Researchers used the NIS-Teen to

show at-risk groups for vaccine-preventable diseases (CDC, 2013).

Literature Review Search Strategy

I conducted a systematic literature search for pertinent research articles on the

factors affecting HPV vaccine uptake in different populations. Searches of several

databases, including Medline, Google Scholar, PubMed, Cinahl, and EBSCO resulted in

80 published articles relevant to the research. The keywords used in the search were

human papillomavirus vaccine, human papillomavirus vaccine and maternal income,

human papillomavirus vaccine and communities, human papillomavirus vaccine and

maternal education, and human papillomavirus vaccine and health belief model. I used

the search terms in various order to gather as many relevant articles to satisfy an

exhaustive search of literature that was less than five-years-old. The search for relevant

books and journal articles ranged from January 2011 to January 2016 unless otherwise

identified as an essential source for the development of the study.

Literature Review of Key Concepts

Key concepts noted during the exhaustive review of the literature concerning the

uptake of the HPV vaccine series were centered on maternal income, maternal education,

maternal age, and ethnicity. Below, I will highlight findings from previous research

regarding these key concepts that show tendencies relevant to my research.

Maternal Income

Higher income has been associated with increased uptake of the HPV vaccine

series. According to Link and Phelan (1995), money was a significant component of SES,

and the more money a person has, the better their health, with some exceptions. Musto et

al. (2013) investigated possible differences in HPV vaccine uptake in Calgary between

in-school and community delivery models and also whether SES contributed to the

phenomenon. Using 35,592 vaccination records the Calgary Zone Public Health

immunization database for all grades 5th through 9th-grade females for school years

2008–2011, logistic regression methods were used to examine the delivery system and

SES status on being vaccinated (Musto et al., 2013). The authors concluded that HPV

vaccination completion rates were 75% (95% confidence interval = 74.7%, 75.8%) for

females who received vaccination in school compared to 36% (95% confidence interval =

35.3%, 37.2%) for females who received the vaccination in the community (Musto et al.,

2013). Additionally, the researchers found that the participant’s neighborhood SES was

related to the likelihood of being HPV vaccinated depended on the delivery model

available (Musto et al., 2013). Limitations of this study were that the authors used an

area-based material deprivation index as an alternative measure for individual SES as

individual SES reporting was not available (Musto et al., 2013). Based on the authors’

admission, the usage of this index may have potentially misclassified SES, and there may

be some misclassification bias (Musto et al., 2013). A strength of the research was the

linkage of postal code with the SES data was over 99% reducing the risk of selection bias

(Musto et al., 2013). Millen, Ginde, Anderson, Fang, & Camargo (2009) examined

knowledge and attitudes about HPV vaccine among emergency department patients if the

vaccine was mandatory. The researchers hypothesized that most women would be aware

of HPV, but few would know its association with cervical cancer or support mandatory

vaccinations (Millen et al., 2009). The researchers reported that one-third of those

surveyed had no knowledge of HPV, which correlated with recent U.S. survey data

concerning knowledge of HPV by women (Millen et al., 2009). Additionally, one-half of

patients surveyed supported state-administered mandatory HPV vaccination programs.

Participants were three times more likely to support mandatory programs based on the

knowledge of HPV being a sexually transmitted disease whereas cervical cancer

knowledge did not increase support for mandatory vaccinations (Millen et al., 2009).

However, the limitation of the study was that it had been conducted in a higher status

SES area within the Boston metropolis that was less ethnically diverse and more educated

than the more urban emergency departments (Millen et al., 2009). The authors asserted

that that lower SES areas would have less knowledge of HPV but never tested the

assertion (Millen et al., 2009). Another limitation was selection bias of the participants by

the investigators and participants, as people with certain medical conditions, such as

mental status changes were excluded from the study, as were non-respondents, which

could have affected the results of the survey (Millen et al., 2009). Cowburn et al. (2014)

tested for an association between insurance continuity and HPV vaccine use in a network

of federally qualified health facilities. Using retrospective electronic health record (EHR)

entries for females aged 9 to 26 from 2008 to 2010; the researchers categorized the

participants’ length of insurance in an ordinal fashion and studied HPV vaccine initiation

prevalence across the range of insurance coverages. They found that participants were

less likely to start the HPV vaccination series if they were insured less than 66% of the

time of the study, were 13 years or older, and belonged to an ethnic minority (Cowburn et

al., 2014). The authors concluded that disparities existed in the health facilities

researched in the study despite the fact that HPV vaccines are available to many of the

patients regardless of their ability to pay (Cowburn et al., 2014). Limitations of the

research included the potential for incomplete vaccination records if patients had received

immunizations outside of the network of facilities studied, which could have caused

underreported vaccination status especially in older children (Cowburn et al., 2014).

Btoush, Brown, Fogarty, and Carmody (2015) examined the prevalence and correlates of

HPV vaccination initiation among adolescents in low-income urban areas using

electronic health records from multisite community health centers in 2011. Their research

indicated that 27.4% of the adolescents and the study initiated HPV vaccination (Btoush

et al., 2015). Of those vaccinated, initiation was higher among males and higher among

Blacks than Hispanics (Btoush et al., 2015). HPV vaccination initiation was lower in

older adolescents, non-English speakers, and those who had received care from non-

pediatricians (Btoush et al., 2015). The limitations of the study are related to the 2009

inclusion of males in the U.S. Advisory Committee on Immunization Practices (ACIP)

recommendations and the 2011 data analysis published by the researchers in 2015. An

important finding was evidence of the lack of vaccination among patients of non-

pediatricians. Smith et al. (2011) reviewed the usage of the HPV vaccine in Ontario

where the government spent over 100 million dollars to offer free quadrivalent HPV

vaccinations to young females. The researchers using administrative and immunization

databases conducted a population-based retrospective study cohort study of females

eligible to receive the vaccination in selected cities in the Ontario province of Canada. Of

the females eligible for vaccination and living within the study boundaries, 1,425 (56.6%)

received at least one dose of the HPV vaccination, and less than half (48.2%) completed

the vaccination series (Smith et al., 2011). The researchers found no differences in health

utilization between vaccinated and unvaccinated females, except that females in the

lowest income quintile were less likely to receive HPV vaccine than the quintile above

(Smith et al., 2011). Additionally, HPV-vaccinated females were more likely to have

received other childhood vaccinations than their unvaccinated counterparts showing an

association of parents’ attitudes and vaccinations (Smith et al., 2011). HPV vaccine series

initiation and completion were not associated with age, health services utilization, or

medical history, although there was an association with low-income neighborhoods,

which were less likely to complete the vaccine series than the females living in middle-

income neighborhoods (Smith et al., 2011). Females residing in rural areas were more

likely to complete their series than females living in urban areas (Smith et al., 2011). A

limitation of a study was that school grade was not always available for review in the

databases, so the researchers used the birth year to identify eligible females. Another

limitation was that investigators did not have vaccination information after December 31,

2009, meaning that females received their vaccinations after this date may have been

misclassified (Smith et al., 2011). There was also the potential for misclassification of

health systems usage because the database used in the research did not capture care

received in clinics that did not update the database (Smith et al., 2011). Lastly, the

validity of using neighborhood income as a proxy for household income has not been

assessed (Smith et al., 2011). The researchers concluded that the females in the lower

SES groups were least likely to complete the HPV vaccination series suggesting that

future intervention programs be modified to enhance the delivery of the vaccine to this

vulnerable population (Smith et al., 2011).

The following research provides evidence that some programs like the Vaccines

for Children Program (VFC) may be having an impact in compensating for families

lacking resources regarding the HPV vaccine. Bednarczyk, Curran, Orenstein, and Omer

(2014) conducted a study of the atypical demographic patterns of HPV vaccine initiation

phenomenon. The researchers used the NIS-Teen data from 2008-2011 and used

regression analysis to calculate the average annual increase by sociodemographic

characteristics. The researchers found that HPV series initiation increased overall 16%

during the time evaluated (Bednarczyk et al., 2014). The researchers also found that since

2008, adolescents living below the poverty level had higher HPV vaccination initiation

than adolescents above the poverty level (Bednarczyk et al., 2014). There were also some

variations in HPV initiation by ethnicity as Hispanic adolescents were consistently higher

in initiation followed by Black and White adolescents (Bednarczyk et al., 2014). There

were also consistent findings when you compared ethnicity and poverty status and HPV

vaccination. All ethnic groups had higher initiation of the HPV vaccine series if they

were below the poverty level than groups above the poverty line (Bednarczyk et al.,

2014). The research was limited by the small samples of some ethnic groups (Bednarczyk

et al., 2014). Additionally, the researchers used a 4-level race/ethnicity which may have

overlooked some racial differences and poverty status (Bednarczyk et al., 2014). The

researchers concluded that more research was needed to explore provider

recommendation and sociodemographic factors. The researchers also found that the VFC

program may have had a larger impact than previously thought as all adolescents below

the poverty level were consistently higher initiators for the HPV vaccine series

(Bednarczyk et al., 2014).

My study overcame the limits of previous studies by analyzing maternal income

defined by postal code. Unlike the study by Musto et al. (2013) which used an area-based

material deprivation index, this study established maternal SES by reported income

within the boundaries of the postal codes of the participants. This method may provide

evidence of a maternal influence on the uptake of the HPV vaccine series based on

maternal income, a limitation found in the study by Musto et al. (2013). Additionally, this

study evaluated all postal codes within the Columbus, Ohio metropolis and provided a

broader, temporal, and more complete evidence of an association between maternal

income and uptake of the HPV vaccine series. My research examined maternal income in

an ordinal fashion based on the responses on the 2014 NIS-Teen survey. By analyzing a

broader range of maternal income and HPV uptake, this study overcomes the limitations

of Millen et al. (2009) and Btoush et al. (2015) which examined communities of higher or

lower income only. By examining more than only the polar extremes of income across

communities in the same region, there may be some valuable associations with maternal

income found that could enhance future HPV vaccine series intervention programs.

Lastly, this was a secondary data research conducted by using responses from the 2014

NIS-Teen. There were no limitations associated with the reviewing medical records of a

particular facility. As part of the NIS-Teen Survey, permission was obtained from parents

to survey the primary provider’s vaccination records, and NIS-Teen researchers had

verified the vaccine administration, thus minimizing the risk of recall bias from parents

on when and if their children received the HPV vaccine.

Maternal Education

Knowledge is a modifying factor of the HBM as knowledge can moderate

individual beliefs as well as health literacy (Phelan & Link, 2013; Skinner, Tiro, &

Champion, 2015). Health literacy can decrease the asymmetry of information given

through access to care thus influencing the avoidance of disease (Phelan & Link, 2013).

The following publications show some contradictory evidence of maternal education and

uptake of the HPV vaccine series. My study analyzed maternal knowledge as the variable

maternal education as reported on the 2014 NIS-Teen to explore for an association with

HPV vaccine series uptake.

Dorell et al. (2014) conducted research using the NIS-Teen data to correlate

parents who refuse or delay the HPV vaccine series for their children with other

vaccinations compared to parents who do not refuse or delay HPV vaccination. The

researchers used the parental attitudes module of the 2010 NIS-Teen survey which

included 1808 completed household interviews (Dorell et al., 2014). The researchers

separated the respondents into four groups based on answers about delaying or refusing

the HPV vaccination. The results showed that 10% delayed only 16.6% refused only, and

3.4% both delayed and refused (Dorell et al., 2014). The females in the delayed only

group tended to be White, from higher income homes and have mothers with college

degrees (Dorell et al., 2014). The major rationale for delaying or refusing the HPV

vaccination found through the analysis of the responses was related to knowledge or

vaccine necessity, vaccine safety, and access (Dorell et al., 2014). The research had some

limitations related to the random digit dialing aspect of the survey, and it was limited to

households with landlines. The study also had the risk of non-response bias, parental

recall, and incomplete vaccination records may also have affected the outcome (Dorell et

al., 2014). The researchers concluded that parental education about HPV might help

improve acceptance of the vaccine.

Additionally, Feiring et al. (2015) studied the parental influence on HPV vaccine

in Norway. In this research, the researchers examined parental education and income as a

factor in the uptake of the HPV vaccine. The researchers used a national immunization

registry to study the uptake of the HPV vaccine and the income and education of parents

of adolescent females. Norway offered the HPV vaccine to the public free of charge to all

12-year-old females since 2009 (Feiring et al., 2015). The researchers found an

association with high maternal education with a lower probability of initiation of the

vaccine series whereas lower education was associated was associated with the higher

likelihood of the initiation of the vaccine series (Feiring et al., 2015). Conversely, high

maternal income was found to be significantly associated with a higher probability of

initiating the HPV vaccine than lower maternal income (Feiring et al., 2015). Paternal

income and education were found to have the same associations with HPV vaccine

initiation, but weaker association than of the mothers (Feiring et al., 2015). The

researchers concluded that more research was needed to determine the factors responsible

for the socioeconomic differences so that interventions could target these differences

(Feiring et al., 2015). The limitations of this study included incomplete information on

vaccinations, income and immigrant education obtained abroad may not be as precisely

accounted for in the survey and may have affected results (Feiring et al., 2015). The

strength of the study was that it was a national registry covering the entire population of

Norway limiting selection bias (Feiring et al., 2015).

The previous studies showed an inverse association with education and

acceptance of the vaccine, but the research by Feiring et al. (2015) reinforced the

association between increased income and the uptake of the HPV vaccine while the study

by Dorell et al. (2014) showed the opposite. The following studies showed a different

association in regards to education and the uptake of the HPV vaccine. Yu et al. (2016)

examined the awareness, knowledge, and acceptability of the HPV vaccine in mothers of

teenage daughters in Shandong, China. The goal to study was to examine the variables of

awareness, knowledge, and acceptability of the vaccine in preparation for the

introduction of the HPV vaccine rollout in China. The researchers wanted to gather

information on attitudes regarding the HPV vaccine to provide evidence to inform health

educators and improve programs targeting this population. Researchers used a cross-

sectional approach using self-administered surveys on a population of 1850 mothers who

had daughters aged 9 to 17 attending schools in the region of Weihai, Shandong, China

(Yu et al., 2016). Researchers used 12 public schools grouped by school level and

location for the research. The mothers were asked to complete the survey of which

researchers collected 1592 surveys. Researchers excluded 14 questionnaires related to

logic errors for a total of 1578 mothers, 85.3% of those initially invited finished the study

(Yu et al., 2016). The findings of the research showed that 19.33% were aware of HPV

before the investigation n=305 (Yu et al., 2016). For the mothers who were aware of

HPV, 14.75% had no knowledge 58.69 % had low knowledge and 26.56% at higher

knowledge (Yu et al., 2016). Additionally, 26.49% mothers voiced a willingness to

accept the vaccination for their daughters n= 418. The authors used Chi-square tests for

analysis and identified five variables that were significantly associated with the

acceptance of the HPV vaccine. These areas found to be associated with acceptance of

the vaccine were daughter’s age, maternal education, maternal occupation, household

income, and knowledge level (Yu et al., 2016). There was increased vaccine acceptability

associated with older daughters, higher income, and knowledge score (Yu et al., 2016).

The prevalent reasons for the refusal of the HPV vaccination were that mothers felt their

daughters were too young to have the risk of cervical cancer (30.95%); not sure about the

use of a new vaccine on their daughters (24.91%); and worried about the safety of the

vaccine (22.85%) (Yu et al., 2016). The limitations of the research were that the research

was conducted and in an economically developed city in China, and not a multi-center

study (Yu et al., 2016). Additionally, there was potential for response bias as the survey

was completed by the mothers who may have been influenced to give socially desirable

responses (Yu et al., 2016). The researchers concluded with the recommendation for the

prioritization of education to raise awareness and knowledge about HPV and the HPV

vaccine. As evidenced by the research, decreased knowledge was associated with poor

acceptance of HPV vaccination which was contrary to the research in Norway which

higher education was associated with decreased acceptance. However, the HPV vaccine

in Norway was available at the time of the study whereas the HPV vaccine was

unavailable in China and the research may not reflect the intentions of mothers once it is

available in China.

Researchers in Botswana conducted a cross-sectional survey on the HPV vaccine

with adults recruited from general medicine and HIV clinics in the capital of Botswana.

The goal of the researchers was to study the intentions of parents and adults to get the

HPV vaccine for their adolescent daughters. There were 376 participants in the study and

the researchers reported that 77% of the respondents were female, and their median age

was 37 years old (DiAngi, Panozzo, Ramogola-Masire, Steenhoff, & Brewer, 2011). The

participants had varying levels of education. 31% of participants completed up to primary

school level (6th grade), 41% of the participants have a secondary school education (high

school), and 28% of the participants had a tertiary education or above (DiAngi et al.,

2011). The income of the respondents showed that many of the participants were poor.

Many of the interviewees had no regular income (48%) or made less than 360 U.S.

dollars a month (29%). Geographically, over two-thirds (65%) of the participants

reported living within 30 kilometers of the capital. 83% of the respondents had children,

and 77% of the children of the participants had one or more daughters (DiAngi et al.,

2011). The results of the survey showed that in the population surveyed only 9% had ever

heard of the HPV vaccine before the study (DiAngi et al., 2011). Additionally, 88% of

the respondents said that they would definitely vaccinate their daughter, and they were

more likely to vaccinate if they had a lower education level or if they lived more than

kilometers outside the capital (DiAngi et al., 2011). The researchers concluded that

providing more information about HPV and a widely available HPV vaccine while

minimizing barriers would improve uptake in Botswana (DiAngi et al., 2011). The

limitations of the study were centered on the convenience sample of participants from

adults with health care access, and the oversampling of HIV-positive patients in the

sample may have confounded the results (DiAngi et al., 2011).

In South Africa, researchers studied the acceptability of the HPV vaccine in

educated participants attending a masters-level program in KwaZulu-Natal South Africa.

A cross-sectional self-administered anonymous survey was conducted on 146 participants

to test their knowledge of HPV and cervical cancer and whether they would accept the

HPV vaccination for their daughters. The researchers found that in this group that 74%

had heard of cervical cancer, but only 26.2% had ever heard of HPV (Hoque & Van Hal,

2014). The participants, after reading the information sheet on HPV and cervical cancer,

the intention to vaccinate their daughters increased from 88% to 97.2% (Hoque & Van

Hal, 2014). The majority of those surveyed (75.4%) believed that the vaccination should

be given before their daughters were aware of sexual activity (Hoque & Van Hal, 2014).

The group that declined to vaccinate tended to want more information on the safety of the

vaccination. The limitations of the study are that it only surveyed one university and

because of the education level of the participants, the results cannot be generalized to a

larger population (Hoque & Van Hal, 2014). A strength of the study was that the

participants, masters-level candidates and future leaders in their perspective fields, could

initiate societal changes through their leadership and knowledge of HPV and cervical

cancer (Hoque & Van Hal, 2014).

Markovitz, Song, Paustian, and El Reda (2014) also found that higher household

education was positively associated with both initiation and completion of the HPV

vaccination series but that higher household income was only positively associated with

completion. This discovery was noted during research of an association between maternal

preventive care utilization and HPV vaccine uptake by their adolescent daughters

(Markovitz et al., 2014).

My analysis of maternal education could contribute to previous research and help

deconflict some of the findings found in the past. This study could overcome some limits

of previous studies by showing maternal income across the population defined by postal

code. Additionally, because this research was on populations in areas within the same

metropolis, there may be some association with HPV vaccination uptake and maternal

education based on the community surveyed because secondary education i.e. beyond

high school education was not based on postal code, and the overlap of education and

community could show a tendency with HPV uptake. Unlike the study by Dorell et al.

2014), my research attempted to show an association with maternal education and uptake

across a defined area. My use of this method might provide some clearer evidence of a

correlation between maternal education and community influence based on postal code

and uptake of the HPV vaccine series. Feiring et al. (2015) researched a nationwide

database for a correlation between education, income, and uptake of the HPV vaccine

series. A major limitation in this research was incomplete vaccination information. My

research uses the 2014 NIS-Teen which allows researchers to query participants’

vaccination prescribers to verify vaccination information. Additionally, my research uses

the HBM methodology and examines maternal tendencies delineated by maternal income

delineated by postal code to search for an association between these variables.

Maternal

Age

Maternal age could be a significant factor in the mother’s decision-making

process for the acceptance of HPV vaccine series in adolescent males and females. There

were no studies found that directly examined this phenomenon, but the variable was

discussed in a study conducted in Tanzania. Watson-Jones et al. (2012) conducted a case-

control study of the characteristics of the receivers and non-receivers of the HPV

vaccination in Tanzania as well as their rationale for not taking the vaccination.

Researchers utilized a randomized trial of HPV vaccinations in 134 primary schools.

Researchers randomized 67 of the schools to an age-based strategy and the other 67 to a

school-based strategy. A sample of 250 females who did not take the vaccine (cases) was

compared to a sample of 250 females who did receive the vaccine (controls). An analysis

of the responses the researchers determined that 53% did not get a dose of the vaccine

because they were absent from school on vaccination day, 40% because a parent refused,

and 1% because the girl refused (Watson-Jones et al., 2012). For the parent group that

received the vaccination, the common reasons for accepting the vaccination was

protection from cervical cancer (89%), health benefits (22%), and knowing someone who

had cancer (13%) (Watson-Jones et al., 2012). For the pupil group that received the

vaccination, the common reasons for accepting the vaccination was protection from

cervical cancer (91%), health benefits (24%), and parental wishes (21%) (Watson-Jones

et al., 2012). For the parent group that did not agree to the vaccination of their daughters,

the common reasons for not accepting the vaccination was concern over side effects

(40%), infertility (23%), or insufficient knowledge about the vaccine (22%) (Watson-

Jones et al., 2012). For pupil group that had not received the vaccination, the common

reasons for not accepting the vaccination was absent from school on vaccination day

(33%), both parents refused (24%), and concerns about side effects (22%) (Watson-Jones

et al., 2012). Further analysis of parents who refused vaccination of their daughters

showed a tendency to be older household members with less education (Watson-Jones et

al., 2012). In conclusion, Watson-Jones et al. (2012) recommended that sensitization

messages targeted at older and poor parents are crucial for vaccine acceptance in

Tanzania. The limitations of the study were related to potential selection bias as 60%

cases responded to the survey compared to more than 80% of the controls. This

deficiency could confound results and might not be representative of the non-receivers of

the HPV vaccine series.

Maternal age has not been adequately researched in recent literature. My study

provided new information about tendencies of the acceptance of the HPV vaccine and

mothers based on their age. This new information could provide insight for future

researchers to tailor interventions that target mothers of adolescent children to increase

HPV vaccination acceptance. Since this research was from secondary data obtained from

the 2014 NIS-Teen Survey, the limitation of selection bias seen in the study by Watson-

Jones et al. (2012) would not be a constraint because of the random digit dialing method

used to recruit participants of the survey.

Ethnicity

Blackman et al. (2013) compared and contrasted the knowledge and attitudes

toward the HPV and the vaccine within different cultures of African descent. The

researchers conducted a cross-sectional survey of African-Americans and Afro-

Caribbean’s living in the US and the Bahamas. The evidence indicated that there was a

significant difference between the two countries in knowledge about HPV and the HPV

vaccine. People from the Bahamas were significantly less knowledgeable about HPV and

the vaccine than African-Americans residing in the United States (Blackman et al., 2013).

Attitudes related to the vaccine were similar although Bahamians tended not to support

vaccination without parental consent versus African-Americans (80% to 57%) (Blackman

et al., 2013). Limitations of this study included a low response rate from Bahamian

parents on the rationale for the unwillingness to vaccinate their children compared to the

replies from African-Americans (Blackman et al., 2013). Luque, Raychowdhury, and

Weaver (2012) examined the provider’s perspectives of the VFC program for Hispanics

in rural Southern Georgia. The researchers performed structured interviews with

providers and focus groups with parents of Hispanic immigrant’s parents to understand

from the provider’s perspective the barriers to access and compliance of the HPV

vaccine. There were two focus groups of parents of females aged 9 to 18 years with

mothers and fathers in separate panels. Predominate barriers gathered from VFC

providers were related to: (1) low English proficiency of the parents; (2) Medicaid

reimbursement shortfalls; (3) mobile population creating difficulty completing a 3-dose

series over a 6-month span of time; (4) lack of transportation access; and (5) lack of

knowledge of the HPV vaccine (Luque et al., 2012). The limitations of the study include

a small sample size of parents for recruitment related to immigration status and Georgia’s

immigration law, and a media-driven controversy surrounding the HPV vaccine (Luque

et al., 2012). There were also limitations in the recruitment of VFC providers related to

contractual limitations on research participation (Luque et al., 2012). Ultimately, the

researchers concluded that inadequate insurance coverage by the VFC program was a

major barrier for not vaccinating adolescents with the additional reluctance to discuss

sexuality and lack of education about HPV and the vaccine (Luque et al., 2012). Kumar

and Whynes (2011) researched for an association between uptake, deprivation and ethnic

background that had been established in pilot research. Based on national immunization

programs in England, the HPV vaccine rates across the country were inconsistent and

varied by location, and the researchers sought to identify the factors explaining the

variation (Kumar & Whynes, 2011). The researchers analyzed published data of HPV

vaccination uptake, material deprivation, ethnic compositions of the different localities,

primary care access, and quality, and preventative services such as usage of cervical

screening and childhood immunization services. The analysis showed that ethnicity was

associated with attitudes towards cervical screening and other childhood vaccinations

while material deprivation and access to quality care were not significant (Kumar &

Whynes, 2011). The researchers found that ethnicity, childhood immunizations, and

usage of preventive and primary care and cervical screening were predictive of the uptake

of the HPV vaccine (Kumar & Whynes, 2011). The researchers also found an association

with increased material deprivation independent of race and lower access to quality care

with the decreased uptake of the HPV vaccine. The limitations of the study were related

to the data only tracked the first two doses of the HPV vaccine and the unavailability of

the data to the boundaries within the localities researched (Kumar & Whynes, 2011).

Lechuga, Swain, and Weinhardt (2011) performed a generalizability study to investigate

the strongest predictors of the mother’s intentions to vaccinate their daughters across

three cultural groups: Hispanic, non-Hispanic White and African American. The

researchers recruited a convenience sample of 150 mothers, 50 from each cultural group

from public health clinics in Milwaukee, Wisconsin and assessed their personal and

normative predictors of intentions to vaccinate their daughters (Lechuga et al., 2011). The

convenience sample of 150 mothers was drawn from Women Infant and Children (WIC)

federal program clinics at one of the four clinics in the Milwaukee, Wisconsin

metropolis. The research results indicated that the predictors of HPV vaccine intentions

varied by cultural group and that culture moderated the influence of norms on intentions

(Lechuga et al., 2011). Additionally, researchers discovered that in their attempt to

control for demographic differences through the recruitment of mothers enrolled in the

WIC program, that there was a significant amount of variability in insurance status

(Lechuga et al., 2011). Hispanic mothers in the study were more likely to be uninsured

and only have a high school education compared to both White and African American

mothers (Lechuga et al., 2011). The perceptions about the vaccine varied based on each

cultural group. For example, the non-Hispanic White mothers had the perception at the

vaccine would lead to increased sexual risk-taking, African American mothers believed

that the vaccine would cause a decrease in protective behaviors such as screening

(Lechuga et al., 2011). Hispanic mothers were more influenced by social norms as it was

a significant contributor to health decision-making (Lechuga et al., 2011). The limitations

of the research stem from the use of a small convenience sample, thus limiting the

generalizability of the results. Additionally, the varying levels of health insurance and

education within the small sample of mothers may have confounded the results. Lastly,

the intention to vaccinate was researched, not the initiation and completion of the HPV

vaccination series (Lechuga et al., 2011).

As noted above, there was much research on racial differences and HPV vaccine series

uptake. My research overcame some limitations of previous research because it was

examining a cross-section of populations defined not by race or ethnicity, but

socioeconomic status. Previous studies such as Blackman et al. (2013) and Luque,

Raychowdhury, and Weaver, (2012) examined specific ethnic groups for an association

with HPV vaccine uptake. My study does not specifically seek racial or ethnic groups,

but all social demographics confined to individual postal codes defined by maternal

income in Columbus, Ohio. My research might add new information to the field

concerning maternal influence via SES, regardless of race or ethnicity, and the uptake of

the HPV vaccine series.

Decisional Influences

The overall premise of the HBM was that people are likely to adopt a health

protective behavior if they believe: that they are susceptible to disease or condition; the

condition could have serious consequences; the remedy for the problem could eliminate

or reduce the susceptibility or severity of the problem; there are benefits to taking action;

and that the perceived costs are outweighed by the benefit of the action (Skinner et al.,

2015). These beliefs, shaped by modifying factors such as age, gender, ethnicity,

personality, socioeconomics, and knowledge may moderate an individual’s beliefs and

subsequent actions (Skinner et al., 2015). The constructs of the HBM collectively affect

behaviors, but precise relationships, weighting, or the combination of variables cannot be

delineated into action in the individual (Skinner et al., 2015). In adult women under 26

years of age, the decision to initiate the HPV vaccination series depends on their personal

choice to get the immunization. The forces that influence initiation or rejection of the

vaccination series vary person to person but have been studied by several researchers.

Harper et al. (2014) studied decisional satisfaction associated with HPV vaccination.

Researchers performed a prospective survey of urban college women aged 18 to 26 years

old about their HPV vaccination experience. The result of the study showed personal

satisfaction was very high regardless of the participant’s decision to accept or reject the

vaccination (Harper et al., 2014). There was variance in the initiation of the vaccine

based on perceived value of the vaccination by the participants. Participants who saw the

value of the vaccination as a method to prevent cervical cancer were more significantly

associated uptake than those who perceived the vaccination as a preventive measure for

genital warts (Harper et al., 2014). Additionally, the authors concluded that based the

participant’s responses targeting those who are neutral to HPV vaccination are a more

effective group to engage than those with high satisfaction to reject vaccination (Harper

et al., 2014). The limitations of the research are related to half the population had already

made a choice to receive at least one dose of the HPV vaccine series (Harper et al., 2014).

Additionally, within the decisional framework of the study, the researchers did not offer

the choice of no vaccination to the participants, so it was unknown how not having that

option might affect the results (Harper et al., 2014).

Knowledge and awareness of a disease process can influence parental choices to

vaccinate their children. Trim et al. (2011) conducted a systematic review of critical

surveys about HPV to understand how knowledge, attitudes, and behaviors were

influenced before and after the FDA release of HPV vaccination. The authors compared

the findings of previous research which studied parental knowledge attitudes and

behaviors towards the HPV vaccine. Additionally, the authors studied the factors that

influenced the decision to vaccinate their children. The authors used published articles

printed between the years 2001 and 2011. The findings from the research showed some

knowledge trends that changed throughout the study. Researchers found that parental

awareness of HPV increased in 2008 and 2009. Parental awareness of the HPV vaccine

increased in 2007 from 14% aware of the HPV vaccine in 2006 to 59% aware of the HPV

vaccine in 2007 and awareness continued to rise into 2008, but dropped slightly by 2010

(Trim et al., 2011). Behavior trends also fluctuated during the study as parents began

vaccinating their children after the 2006 release of the quadrivalent HPV vaccine,

reaching its peak in 2009 and 2010. For attitude trends, the highest percentage of parents

who intended to vaccinate their children peaked at 86% in 2005, the year before the HPV

vaccine release (Trim et al., 2011). Parental intent to vaccinate their children gradually

rose from 67% in 2007, and 80% 2008, but declined slightly over the last three years of

the study (Trim et al., 2011). There were also barriers for parents to accept the HPV

vaccine for their children. Parental knowledge of the HPV vaccine was a significant

factor in the acceptance of the vaccine for their children. In the review of the research, the

authors found evidence that in 37% of the studies reviewed, concerns about the safety of

the HPV vaccine were the parent’s primary barrier with additional concerns about the

potential for side effects (Trim et al., 2011). Parents wanted more information to make an

informed decision was cited in 25% of the studies from the analysis. Conversely, parents

who were concerned about the risk of cancer in their child were more likely to accept the

HPV vaccination for their child (Trim et al., 2011). Parents differed in their attitudes on

when the vaccine should be given. In 19% of the studies reviewed by the researchers

showed a trend not to vaccinate if the parent believed that their child was too young for

the HPV vaccination (Trim et al., 2011). Earlier initiation to sex in adolescents related to

the HPV vaccine was also a barrier for parents. In 25% of the studies, researchers

examined parental concerns about increased risky sexual behaviors in adolescents after

vaccination (Trim et al., 2011). Research performed by Smith, Kaufman, Strumpf, and

Lévesque (2015) and Zimet et al. (2013) included in the review showed no evidence of

increased sexuality post-HPV vaccination. The strengths of the research were that it

included the knowledge, attitudes, and behaviors from a large number of parents from

several countries (Trim et al., 2011). The limitation of the research was the lack of the

ability to validate parental responses (Trim et al., 2011). The authors concluded based on

their analysis that parents wanted more information and reassurance from their providers

that the HPV vaccine was safe to give to their children (Trim et al., 2011). The parental

decisional findings by Trim et al. (2011) were similar to the findings of the research

conducted by Hofman et al. (2013). Hofman et al. (2013) studied parent’s decision-

making strategies through focus groups in the Netherlands. The researchers used four

focus groups of primarily Dutch parents (one urban and two rural) and one group of

Turkish parents, who represented the largest ethnic minority in the Netherlands. All of

the parents in each group had at least one daughter between the ages of 8 and 15. The

researchers concluded post analysis of the parental responses that many parents felt

uneasy about HPV vaccination. The concern was related to the safety and the

effectiveness of the HPV vaccine (Hofman et al., 2013). The common theme from the

analysis was child protection motivation, and with some of the parents, the motivation

was to vaccinate whereas there were also some parents who were motivated to protect

their daughter by not vaccinating (Hofman et al., 2013). The strength of the research was

that it provided information about parental attitudes and decisional strategies about HPV

uptake before the vaccine was discussed in the media. The limitations of the research

were that most of the participants in the focus groups were mothers. The other limitation

was related to the sample size of Turkish parents group as there were too few Turkish

parents studied to compare to Dutch parents (Hofman et al., 2013).

My research provided new information concerning decisional influences for HPV

vaccination based on the responses of parents in communities defined by postal code. The

variables of maternal income, maternal education, and maternal age could provide

additional insight of the decisional influences of parents in communities of various SES

by way of the health belief model. Maternal SES and maternal age could be contributing

influences for parent’s decisions to vaccinate adolescent males and females with the HPV

vaccine based on the core constructs of the health belief model.

Critics and Differing Opinions

Bresse, Goergen, Prager, and Joura (2014) researched the cost effectiveness and health

impact of universal vaccination against HPV in Austria. The focus of the study was to

note the cost savings of preventing cancers caused by HPV 16/18 in a cohort of 9-year-

old males and females (Bresse et al., 2014). The authors concluded that with vaccination,

the HPV-related cancer burden would decrease by 71% over 100 years (Bresse et al.,

2014). This total includes not only cervical cancers but also anal, penile and

oropharyngeal cancers (Bresse et al., 2014). Additionally, Crowcroft et al. (2012)

concluded that high vaccine coverage improves communities, reduces absolute risk, and

increase equity. Their research computed the comparative risks for invasive cervical

cancer in a population or subgroup before and after the implementation of a vaccination

program. A simple static multi-sensitivity analysis was completed to compare the relative

risk of HPV infections that would lead to invasive cervical cancers if they were not

prevented or detected (Crowcroft et al., 2012). The researchers evaluated 3,793,902

scenarios and in 63.9% of the considered scenarios; HPV vaccination would lead to a

better population outcome regardless of the effectiveness of the vaccine (Crowcroft et al.,

2012). A limitation in the research by Crowcroft et al. (2012) was not estimating the prior

probability distribution for their parameters as Bayesian methods require. A limitation of

both these studies was that according to Ruiz et al. (2012) there are other prevalent strains

of oncogenic HPV other than HPV 16/18 and would not be covered by the present

vaccine. Currently, there are 12 known oncogenic strains of HPV

(16/18/31/33/35/39/45/51/52/56/58/59) (Ruiz et al., 2012).

Usage of the HPV vaccine in males has been a recommendation in the United

States since 2009, but usage in males had only recently achieved approval in Canada in

2015. The National Advisory Committee on Immunizations in Canada recommended that

males aged 9-26 receive the HVP vaccination series (Smith et al., 2015). In 2009, the

ACIP recommended that males receive the HPV vaccine aged 9-26 but subsequently

modified the initiation age in males to ages 11-26 in 2011 (CDC, 2011). The current

recommendation by ACIP for HPV vaccination is routine vaccination at age 11 or 12

years with HPV4 or HPV2 for females and with HPV4 for males; the vaccination series

can be started beginning at age nine years (CDC, 2014a). There were additional studies

using the NIS-Teen survey data for analysis,

A different approach for the prediction of HPV vaccine uptake was done by (Hechter et

al., 2013), who studied the maternal use of preventative care and history of sexually

transmitted disease as a predictor of uptake of HPV vaccine in adolescent males. This

innovative study linked maternal information with electronic medical records of males

aged 9-17 enrolled in a health maintenance organization (HMO) in Southern California.

Based on the various criteria conducted during the study, the researchers found some

interesting results useful for future research. For example, there was an association

between the initiation of HPV vaccine in males if they received the seasonal influenza

vaccine (Hechter et al., 2013). Additionally, males whose mothers received Pap testing

were more likely to receive the HPV vaccine than males whose mothers without a history

of genital papillomatosis were more likely to receive HPV vaccine (Hechter et al., 2013).

The authors concluded that maternal use of preventive health services might influence

HPV vaccination uptake in males (Hechter et al., 2013). Rahman, Laz, McGrath, and

Berenson (2014) found a similar association with the uptake of the HPV vaccination in

older adolescent females who received a seasonal influenza vaccination.

Some researchers argued that the disparity of HPV vaccinations may be related to

underreporting due to parental recall. Attanasio & McAlpine (2014) implied that parental

recall might inaccurately depict HPV uptake rates. The researchers evaluated parental

recall of HPV vaccination compared to clinical records while also evaluating social

characteristics of the accuracy by the parents surveyed. Researchers used data from the

2009-2010 NIS-Teen. The NIS-Teen survey consists of household interviews and a

provider-completed immunization history to compare responses to patient records. The

results showed parental underreporting of HPV uptake associated non-White, lower

household income, and lower education attained adolescent mothers (Attanasio &

McAlpine, 2014). Limitations of the study were related to the timing of the change the

ACIP recommendations for the HPV vaccination for males and the survey depended on

households that participated that also had a complete provider report (Attanasio &

McAlpine, 2014). This research showed that parental recall might cause a significant

limitation in vaccine coverage studies because some parents based on multiple

sociodemographic factors underreported the number of HPV vaccinations given to their

adolescent teen (Attanasio & McAlpine, 2014).

Malkowski (2014) studied the gender impact of the rollout of the HPV vaccine by Merck

Pharmaceuticals in 2006. The initial promotion of Gardasil presented a solution for a

woman-only issue despite the evidence that HPV infected both men and women

(Malkowski, 2014). The author implied that the initial advertisements for the vaccine

focus were not on the soon to be released vaccine nor did it inform the public of anything

related to the sexual transmission of HPV (Malkowski, 2014). The second advertisement

campaign was more focused on the teenage target audience, this time; the focus of the ad

campaign was on the vaccine and the disease without mentioning the mode of

transmission of the virus or even the virus itself (Malkowski, 2014). The third

advertisement campaign launched four years after the initial advertisement offering, and

the target audience was women. Merck used personal testimonies of people infected with

HPV a different tactic from previous campaigns where they targeted women not yet

exposed to the virus. Women in this campaign were portrayed to be the guardians of

public health despite the fact that HPV virus infects both males and females (Malkowski,

2014). Through analysis of all three campaigns, the researcher concluded that Merck

targeted a specific audience and persuaded them to assume a disproportionate burden for

a public health problem that affects men and women (Malkowski, 2014). The author

recommended a retooling of efforts to deconstruct parts of the message to repackage

HPV and the disease process as a more inclusive disease that does not solely place the

burden of protection on women (Malkowski, 2014). There are several theories in the field

of HPV vaccine uptake research. Most of these theories have been used in research to

determine the leading barriers to the uptake of the vaccine (Savoy, 2014). The most

common reasons for decreased uptake are knowledge of the vaccine, cost of the vaccine,

safety, efficacy, and risks of increased promiscuity. Savoy (2014) theorized that the

infrequent visits to the doctor as adolescents than as toddlers for vaccinations to be a

possible cause for the lack of vaccinations. Another rationale considered for decreased

uptake was the parental fear the vaccination would lead to promiscuity which various

researchers have evaluated (Savoy, 2014). Smith, Kaufman, Strumpf, and Lévesque

(2015) evaluated a cohort of over 260,000 females and found no evidence of perceived

promiscuity based on pregnancy and other sexually transmitted diseases. This study was

limited because it only evaluated females up to age 17 and the high attrition rate of the

survey as over 131,000 of the returned questionnaires was ineligible for the analysis

(Smith et al., 2015). Two years before this research, Zimet, Rosberger, Fisher, Perez, and

Stupiansky (2013) also investigated the promiscuity hypothesis and evaluated sexual risk

compensation related to HPV vaccination. The researchers reviewed several selected

published behavioral and social science articles on HPV vaccine acceptance and attitudes

and found no evidence of increased sexual risk-taking in adolescents taking the vaccine

(Zimet et al., 2013). Brown, Blas, Heidari, Carcamo, and Halsey (2013) evaluated

changes in sexual behavior and HPV knowledge after an education and vaccination

intervention in Peruvian female sex workers. The researchers noted that the participants

had a significant decrease in new clients over a 30-day period and utilized at least one

preventative strategy against other sexually transmitted infection upon the seven-month

follow-up survey (Brown et al., 2013). This evidence was corroborated by Zimet et al.

(2013), who came to similar conclusions in a study reviewed that was done on 13 to 21-

year-old females. Ruiz et al. (2012) hypothesized that proximity of first sexual experience

to menarche or the start of menstruation was associated with increased risk of cervical

intraepithelial neoplasia grade 2/3. In their research, they evaluated 1009 Colombian and

1012 Finnish females aged 16 to 23 that enrolled in an HPV vaccination trial that had

accurate data concerning the onset of menstruation and their first sexual experience. Of

the women included in the study, the statistics showed the mean age of menarche as 12.4

years, and the mean age of first sexual intercourse was 16 years (Ruiz et al., 2012). The

results of this study showed that women who had their first sexual intercourse less than

three years after menarche had a higher risk of cervical cytological abnormalities

compared to women who waited beyond three years after menarche (Ruiz et al., 2012).

Ruiz et al. (2012) concluded with the emphasis on the importance of primary prevention

through early vaccination and sexual education of adolescent females. The perception of

the HPV vaccine contributing to infertility was another hypothesis investigated by

researchers. Schuler, Hanley, and Coyne-Beasley (2014) researched parent’s concerns

about infertility as a barrier to accepting the HPV vaccine in adolescent males. 39% of

respondents reported that they were concerned about vaccine acquired infertility (VAI)

(Schuler et al., 2014). Additional analysis showed that this group had no less knowledge

than other parents surveyed indicating an increased need for conversations concerning the

side effects of the HPV vaccine to parents rather than having parents read the vaccine

information sheet (VIS) (Schuler et al., 2014).

This study provided new information that could impact future HPV vaccination

interventions. Based on the evidence discussed in the literature above, increasing the

uptake of the HPV vaccine series decreases the overall cervical cancer risk. As stated in

the research by Crowcroft et al. (2012) high HPV vaccination coverage improves

community health, increases equity, and reduces the absolute risk of cervical cancer. This

research added additional empirical evidence to support the development of future HPV

vaccination intervention programs by way of predicting tendencies of parents of

adolescent males and females to vaccinate their children with the HPV vaccine series

based on the theoretical concepts of the health belief model. My research enhanced

knowledge of decision-making based on maternal income, maternal education, and

maternal age.

Definitions

Cervical cancer: A type of cancer that begins in the cells lining the cervix at the

lower portion of the uterus (American Cancer Society, Inc., 2014). Cervical cancer is the

second most common female cancer worldwide, and there are nearly 500,000 cases per

year contributing to >250,000 deaths each year (Union for International Cancer Control

(UICC),

2015).

Human papillomavirus (HPV): Genital HPV is the most common sexually

transmitted infection (CDC, 2014c). There are over 100 HPV types identified, and there

are more than 40 HPV types that can infect the genital area (Hariri, Dunne, Saraiya,

Unger, & Markowitz, 2011). HPV types are classified by their association with cancer.

Non-oncogenic or low-risk strains of HPV can cause genital warts while oncogenic or

high-risk HPV can cause cervical cancer (Hariri et al., 2011).

HPV vaccine: Two vaccines are available to prevent persistent infection with

oncogenic strains of HPV. One vaccine is effective against four HPV strains, two high

risk, and two low risk and both vaccines are effective at protecting against the types that

cause 70% of cervical cancers (CDC, 2013; U.S. Food and Drug Administration, 2013).

Papanicolaou (Pap) test: A screening test for cervical cancer. The test looks for

abnormal cells on your cervix that could potentially turn into cancer. Early identification

of cancerous cells can improve the overall success of treatment. All women should start

getting regular Pap tests starting at age 21 (Techakehakij & Feldman, 2008).

Gardasil: The first HPV vaccine released in the United States in 2006. Gardasil

immunizes against HPV serotypes 6,11, 16 and 18 (U.S. Food and Drug Administration,

2013).

Cervarix: The second HPV vaccine released in the United States in 2009.

Cervarix immunizes against HPV serotypes 16 and 18 (CDC, 2010).

Maternal income: The value of the participant’s income reported on the NIS-Teen

survey by postal code.

Maternal education: The level of maternal educational reported by the 2014 NIS-

Teen survey participants (Centers for Disease Control and Prevention [CDC], National

Center for Immunization and Respiratory Diseases [NCIRD}, & National Center for

Health Statistics [NCHS], 2015).

Race: The 2014 NIS-Teen Survey defines race as White, Black or African

American, American Indian, Alaska Native, Asian, and Native Hawaiian (CDC, NCRID

and NCHS, 2015).

Ethnicity: The 2014 NIS-Teen Survey defines ethnicity as participants of

Hispanic or non-Hispanic origin (CDC, NCRID & NCHS, 2015).

Assumptions

This study was based on several assumptions. The most critical assumption was

that the instrument for data collection is valid and reliable based on the previous use of

the survey for HPV vaccination research. The NIS-Teen launched in 2006 provides the

most current, household, population-based, state and local area estimates of vaccination

coverage among children and teens using a standard survey methodology (CDC, 2016).

Additionally, there was the assumption that the random digit dial [RDD] sampling

method used to collect the NIS-Teen data resulted in a representative sample of telephone

households in Columbus, Ohio metropolitan area. Another assumption was that surveyors

collected data in a nonbiased manner and the participants provided the most honest and

accurate responses on the survey. The target audience for the NIS-Teen were adolescents

13-17 years living in households in the United States at the time of the survey. Lastly, it

was assumed the questions in the NIS-Teen survey are reliable and valid measures of

gathering information. This assumption was based on several years of use by the CDC

and in multiple previously published research studies. These assumptions are necessary to

conduct this research using this secondary data source. Due to the multiyear collection of

data by the National Immunization Survey (NIS), it was assumed that this source of data

was valid and reliable to be used in research. These assumptions are critical to the

research and to the analysis of the data provided so that conclusions can be made on the

population surveyed.

Scope and Delimitations

This study was limited to the analysis of selected SES variables and uptake of the

HPV vaccine series by postal code in the Columbus, Ohio metropolitan area. The

findings cannot be generalized to other vaccines. This survey data was limited by

information recall of parents who participated in the study. The sample population

interviewed for the 2014 NIS-Teen may not be generalizable to other populations. The

study did not analyze the variants of insurance coverage plans, or the state-related

variances in the VFC although some these variances may affect the conclusions and

should be considered for future research. Eligibility of the survey participants was

determined by the self-reporting by parents or guardians of adolescent children in the

household ages 13-17 years old via random digit dialing phone interviews. The NIS-Teen

is a large national representative sample that estimates vaccination coverage for the 50

States (CDC, 2016). Lastly, this research analyzed the responses from participants living

in the Columbus, Ohio metropolitan area and excluded postal codes not associated with

this location.

Significance and Potential for

Social Change

Significance of Study

The significance of this research was that it could add potentially valuable

evidence that could contribute to the improvement of HPV intervention programs based

on maternal income and other selected SES variables examined in this study. Potential

evidence discovered through this research could enhance future researchers’

methodological approach to the implementation of community intervention programs by

tailoring HPV vaccination programs to fit selected communities based on maternal

income, education, maternal age, and ethnicity. Additionally, evidence found in the

analysis could potentially exclude some SES factors that were thought to influence the

uptake of the vaccine.

This research is an original contribution to field as there are many research studies

published exploring the barriers contributing to the decreased uptake of the HPV vaccine

series. None of the published contributions have explored maternal income, maternal

education and maternal age by community (defined by postal code) to explore for an

association within a community’s maternal SES status indicators and uptake of the HPV

vaccine series in adolescent females and males ages 13-17.

Social Change

The potential for significant social change related to this study was based on the

potential evidence of an association between uptake of the HPV vaccine series and

maternal influences as it relates to measurable maternal SES factors (maternal income

and maternal education). Such results of the research could potentially be used to reduce

the burden of cervical cancer in women through the enhancement of vaccination

programs contributing to the decreased incidence of a significant health disparity for

women. As a potential result of this research, more women could live longer and reach

their full potential through the improvement and enhancement of HPV vaccine series

interventions. Additionally, as a secondary result, this research could potentially change

the recommended screening schedule for cervical cancer screening decreasing the

frequency of exposure to invasive screening tests. Both men and women are reservoirs

for the HPV virus, universal HPV vaccination of all adolescents could lower the

incidence of HPV infection in women and the progression to cervical cancer.

Summary

HPV infection is the most common sexually transmitted infection (CDC, 2013).

HPV can progress to cervical cancer, and cervical cancer is responsible for over 4,000

deaths in the United States annually and a much higher burden worldwide, especially for

developing countries (American Cancer Society, Inc., 2014; Union for International

Cancer Control (UICC), 2015). To prevent cervical cancer, the enhancement of HPV

vaccination programs to meet the healthy people 2020 goal of 80% HPV infection is

critical (Savoy, 2014). HPV infection reduction can be accomplished by the use of

Gardasil, which is effective against HPV serotypes 6, 11, 16 and 18, and Cervarix, which

protects women from HPV types 16 and 18 (CDC, 2013; U.S. Food and Drug

Administration, 2013). The identification of additional barriers to HPV acceptance could

contribute to the continued improvement of vaccine intervention programs. The use of

the health belief model as the framework for the study may bring to light additional

measures to improve and enhance HPV vaccination programs. This study focused on

maternal SES factors by exploring for a correlation based on maternal income, maternal

education, as well as the maternal age of the participants through the analysis of the 2014

NIS-Teen survey data.

Conclusion

In conclusion, there was a gap in the literature for research exploring associations

between maternal community-level SES influences and the voluntary uptake of the HPV

vaccine series. Improving HPV vaccination rates among adolescents ages 13-17 was an

issue that must be addressed (Moss et al., 2014). The protection offered by this vaccine

can keep women from acquiring strains of oncogenic HPV that account for 70 % of all

cervical cancers in the U.S.(Harper et al., 2014). As both men and women are reservoirs

for the HPV virus, universal HPV vaccination of all adolescents could lower the

incidence of HPV infection and the progression to cervical cancer. The following chapter

provided the rationale behind the research design and data collection methods to reinforce

the significance and need of this scholarly project.

Section 2: Research Design and Data Collection

Introduction

The purpose of this doctoral study was to assess if there was an association

between maternal SES variables of income, education, age, and ethnicity and uptake of

the HPV vaccine in adolescent females and males ages 13-17 in Columbus, Ohio. In this

section, I explain my research design and the rationale for the choice of design. This

section also provides a comprehensive explanation of the methodology used for the study

in the event future researchers may want to replicate this research. Next, I elucidate my

choice of instrumentation, its purpose, and how I operationalized the constructs. Lastly, I

describe threats to validity and ethical procedures, to include the protection of data, and

close by summarizing the pertinent details in this section.

Research Design and Rationale

This was a cross-sectional quantitative observational research study in which I

explored associations between maternal socioeconomic influences based on income and

education and uptake of the HPV vaccine series. Additional analyses covered the

association between maternal age and ethnicity. The dependent variable in this study was

uptake of the HPV vaccine series, which I defined as a dichotomous (yes/no) response to

whether the adolescent (male or female) received at least one dose of the vaccine. The

four independent variables were maternal income, maternal education, maternal age, and

ethnicity. The first independent variable, maternal income, was defined as the income

reported by the respondents living within a certain postal code within the Columbus,

Ohio metropolitan area. The second independent variable, maternal education, was

defined by the highest level of education reported by the mother divided into the

following categories: no high school diploma; high school graduate or GED; completed a

vocational, trade, or business school program; some college credit but no degree;
associate degree (AA, AS); bachelor’s degree (BA, BS, AB); master’s degree (MA, MS,

MSW, MBA); and doctorate (PhD, EDD) or professional degree (MD, DDS, DVM, JD).

The third independent variable, maternal age, was defined by the mother’s age at the time

of survey completion divided into the following categories: 18-22, 23-27, 28-31, 31-36,

37-41, 42-45, and 46 and above. Ethnicity was measured by the following six categories:

White, Black/African-American, Native American, Asian, Native Hawaiian, and Pacific

Islander.

I chose a cross-sectional research design for this study. Cross-sectional design is

the predominate method for survey research and can be used sufficiently to examine

associations between properties and dispositions (Frankfort-Nachmias & Nachimas,

2008). There were no time or resource constraints related to using a cross-sectional

design approach, as the data analyzed was from a secondary analysis of the 2014 NIS-

Teen survey. Cross-sectional design was an optimal choice for this study as it allowed an

analysis of the dependent variables, uptake of the HPV vaccine series with multiple

independent variables, maternal income, maternal education and maternal age. By using

the cross-sectional design, the findings could be helpful in predicting outcomes based on

the SES variables analyzed in the study. The information gathered could enhance

intervention strategies based on any discoveries noted from the analysis of the SES

variables. A cross-sectional design was the best option because I was trying to elicit a

pattern of a relationship between the SES variables and uptake of the HPV vaccine series.

The statistical plan for my study was multiple logistical regression, which is used

when there is one categorical dependent variable and two or more independent variables

(McDonald, 2014). I chose multiple logistic regression because of its appropriateness for

seeking a functional association between the independent variables and the dependent

variable. This statistical plan can be used to predict probabilities of an effect of multiple

independent variables on a categorical dichotomous dependent variable and in some

circumstances can be used to make inferences about which independent variables have a

larger effect on or stronger association with the dependent variable (McDonald, 2014).

Methodology

In this section, I describe how I performed the research by defining the study

population, sampling techniques, access to secondary data, instrumentation,

operationalization of constructs, threats to validity, and ethical considerations.

Study Population

Columbus is the capital of the State of Ohio. The population of Columbus is

approximately 850,106 (Department of Commerce, 2016b). The target population for the

study was parents or guardians of adolescent teens who participated in the 2014 NIS-

Teen survey and live in the Columbus, Ohio metropolitan area. Since the participants of

the survey are asked specifically about HPV vaccine series uptake, I included all

participants of the survey who live in the geographic region when conducting the analysis

of the data. The sample size needed to be at least 1188 participants based on the

calculations using G*Power 3.1 Statistical Power Analysis for a logistic regression two-

tailed analysis with 0.95 power (1-β err prob; Faul et al., 2007). Optimally, if participants

resided across different postal codes, comparing uptake of the HPV vaccine series among

multiple SES variables would have improved the analysis. The target population for the

2014 NIS-Teen was adolescents aged 13–17 years living in non-institutionalized

households in the United States at the time of the interview (CDC, 2015b). Researchers

conduct the 2014 NIS-Teen concurrently with the 2014 NIS. The 2014 NIS-Teen

identified households containing one or more adolescents who was 13-17 years of age at

the time of the survey. Interviews were conducted with the household adults who were

the most knowledgeable about the teenager’s record of vaccinations (CDC, 2015b). Upon

completion of the survey and after obtaining consent from the parent or guardians of the

teenagers surveyed, the 2014 NIS-Teen surveyors also contacted the teenager’s

vaccination providers to request information on their vaccination records (CDC, 2015b).

The criteria for inclusion in this research study were as follows:

• Being an adolescent male or female between 13-17 years of age by the time of

the interview in 2014,

• Live in one of the Columbus, Ohio metropolitan postal codes

• The first HPV shot was received between 9 -17 years of age.

The ACIP recommends routine vaccination at age 11 or 12 years with HPV4 or

HPV2 for females and with HPV4 for males, although the vaccination series can be

started as early as age 9 (CDC, 2014b). For those unvaccinated at the routine age, the

vaccine is recommended for males aged 13 through 21 years and females aged 13

through 26 years who have not been vaccinated previously or who have not completed

the 3-dose series (CDC, 2014b).

Sampling and Sampling Procedures

The 2014 NIS -Teen Survey was used by surveyors to collect data from

households with adolescents 13-17 years and the teens’ vaccination providers (CDC,

2015). The NIS-Teen survey was conducted in two parts. The first part of the survey was

the random-digit-dialing (RDD) telephone survey of parents and guardians of randomly

selected households in all 50 states and the District of Columbia and the second part of

the survey was the survey of the teen’s vaccination providers (CDC, 2015). The NIS-

Teen surveyors obtained the consent to survey from the parents or guardians of eligible

teenagers so that contact could be made with their vaccination providers (CDC, 2015).

Researchers mailed a survey questionnaire to participants’ vaccine providers to perform a

check of their medical records (CDC, 2016). The goal of the mail survey of vaccination

providers was to confirm the accuracy and recall from the parents as compared to the

actual vaccination records and to assure the accuracy and precision of overall vaccination

coverage estimate (CDC, 2015c). The 2014 NIS-Teen survey included 59 geographic

strata for which vaccination coverage levels could be estimated, including seven mostly

urban cities and county areas (including the District of Columbia). Lastly, the remaining

52 estimation areas were either entire states or territories (including U.S. Virgin Islands

and Guam) or “rest of state” areas (CDC, NCRID & NCHS, 2015). According to CDC,

NCRID and NCHS (2015), this design makes it feasible to produce yearly predictions of

vaccine coverage levels for each state or territory (including U.S. Virgin Islands and

Guam) and for each of the seven sub-state estimation areas with a specified degree of

precision (a coefficient of variation of approximately 7.5%). Additionally, using the same

data collection methods and survey instruments researchers use the NIS-Teen to produce

results that are comparable to predict vaccination coverage levels among estimation areas

and through subsequent years (CDC, NCRID & NCHS, 2015). However, on the 2014

NIS-Teen survey, there was a change of the definition of adequate provider data (Reagan-

Steiner et al., 2015). As of 2014 on the NIS-Teen survey, adequate provider data was

achieved if the adolescent had vaccination history data from one or more of the named

vaccination providers or if the parent reported that the adolescent was completely

unvaccinated (Reagan-Steiner et al., 2015). Prior to 2014, there were more criteria

associated with the definition of adequate provider data, and it was based on a

comparison between provider reports of vaccination history and parental reports of

vaccination history, either by shot record report or recall (Reagan-Steiner et al., 2015).

This change means that future studies using the NIS-Teen survey data cannot be

compared to the previously published vaccine coverage estimates (Reagan-Steiner et al.,

2015).

Access to Secondary Data

The procedure to gain access to the 2014 NIS-Teen data was detailed at the CDC

website. The 2014 NIS-Teen public-use secondary dataset can be downloaded from this

website. In addition to the 2014 NIS-Teen dataset, users can download other pertinent

documents such as the readme file, data user’s guide, household interview questionnaire,

provider-immunization history questionnaire data documentation, codebook and

frequencies, SAS input statements, and R Input Statements (CDC, NCRID & NCHS,

2015).

Permissions to use the 2014 NIS-Teen are clarified in the readme file. The

permissions for the use of data was strictly used only for the purpose of health statistical

reporting and analysis and any attempts to ascertain the identities of the participants were

prohibited by law (CDC, NCRID & NCHS, 2015). To comply with permissions for the

usage of the 2014 NIS-Teen, users of the data must be in compliance with the following:

1. Use the data in these data files for statistical reporting and analysis only.

2. Make no use of the identity of any person or establishment discovered

inadvertently and advise the Director, NCHS, of any such discovery (301-458-

4500)

3. Not link these data files with individually identifiable data from other NCHS

or non-NCHS data files (CDC, NCRID & NCHS, 2015).

Instrumentation and Operationalization of Constructs

Instrumentation

The NIS-Teen was launched in 2006. The target population for the NIS-Teen was

adolescents 13-17 years living in the United States at the time of the interview (CDC,

2015b). The NIS, developed in 1994 are a group of phone surveys used to monitor

vaccination coverage among children 19-35 months, teens 13-17 years, and flu

vaccinations for children 6 months-17 years (CDC, 2015b). The first NIS survey began in

April 1994 to examine vaccination coverage in the United States after measles outbreaks

in the early 1990s (CDC, 2015b). Researchers at the NCIRD of the CDC developed the

NIS by using a culmination of research experience using several different survey

methodologies (Zell et al., 2000). The NIS is an annual survey designed to provide

current and continuous estimates of vaccine coverage, provide reliable and valid

estimates of vaccination coverage in 78 separate areas (all 50 states, the District of

Columbia and 27 urban areas considered to be at risk of under-vaccination), provide

timely estimates, and produce estimates using reasonable resources (Zell et al., 2000).

Operationalization

Uptake of the HPV vaccine series: A person’s receiving one or more vaccinations

of the HPV vaccine series. This was the dependent variable. Based on the ACIP

recommendation, all participants of the HPV vaccination series are eligible to accept the

vaccine at 11-12 years of age but can be given as early as age 9 (CDC, 2014b). The

recommended age of initiation of the vaccine is well within the parameters of the NIS-

teen survey which gathers data of adolescents aged 13-17.

Maternal income: The household income reported based on the postal code of the

survey participants. Maternal income was measured by the income reported income per

postal code on the 2014 NIS-Teen survey. Maternal educational is measured by the

reported level of maternal educational reported during the survey by postal code.

Maternal education: The level of maternal education based on the postal code of

the survey participants, defined by the highest level of education attained, separated into

the following categories: no high school diploma; high school graduate or GED;

completed a vocational, trade, or business school program; some college credit but no

degree; associate degree (AA, AS); bachelor’s degree (BA, BS, AB); master’s degree

(MA, MS, MSW, MBA); and doctorate (PhD, EDD) or professional degree (MD, DDS,

DVM, JD).

Maternal age: The age of the mother at the time of survey participation. Maternal

age was measured by dividing the mothers’ age into the following groups: less than 25

years old, 25-34 years old, 35-44 years and 45+ years old.

Ethnicity: The grouping of the major divisions of humankind, having distinct

physical characteristics. Ethnicity was measured by dividing participants into the

following six categories: White, Black/African-American, Native American, Asian,

Native Hawaiian, and Pacific Islander.

Data Analysis Plan

I analyzed the 2014 NIS-Teen secondary data using SPSS® version 21 (IBM

Corp., 2016). I validated the analyses using the built-in validation functions in SPSS®

v.21. I conducted simple descriptive analyses of the variables. I recoded the identified

variables, categorized and manipulated them to fit the variables in line with the research

questions. I conducted normality testing and performed binary analyses, bivariate

analysis, followed by multiple logistic regression using SPSS® on the independent

variables to search for statistically significant associations with the dependent variable.

Research Question(s) and Hypotheses
The objective of this research study was to explore the association between

maternal SES and uptake of the HPV vaccine series. The research questions are as

follows:

Q1: what is the association between maternal income and uptake of the HPV

vaccine series in adolescent males and females 13-17 in communities with postal codes in

the Columbus, Ohio metropolis?

HO: there is no association exists between maternal income and uptake of the

HPV vaccine series in adolescent males and females 13-17 after controlling for ethnicity

and maternal age based on postal codes in Columbus, Ohio.

HA: There is an association exists between maternal income and uptake of the

HPV vaccine series in adolescent males and females 13-17 after controlling for ethnicity
and maternal age based on postal codes in Columbus, Ohio.
Q2: What is the association between maternal education and uptake of the HPV

vaccine series in adolescent males and females 13-17 in the postal codes within the

Columbus, Ohio metropolis?

HO: There is no association between maternal education and uptake of the HPV

vaccine series in adolescent males and females 13-17 in the postal codes within the

Columbus, Ohio metropolis

HA: There is an association between maternal education and uptake of the HPV

vaccine series in adolescent males and females 13-17 in the postal codes within the
Columbus, Ohio metropolis.

Q3: What is the association between maternal age and uptake of the HPV vaccine

series in adolescent males and females 13-17 in communities with postal codes in the

Columbus, Ohio metropolis?

HO: There is no association between maternal age and uptake of the HPV vaccine

series in adolescent males and females 13-17 in communities with postal codes in the

Columbus, Ohio metropolis

HA: There is an association between maternal age and uptake of the HPV vaccine

series in adolescent males and females 13-17 in communities with postal codes in the
Columbus, Ohio metropolis.
Q4: What is the association between ethnicity and uptake of the HPV vaccine

series in adolescent males and females 13-17 in the postal codes within the Columbus,

Ohio metropolis?

HO: There is no association between ethnicity and uptake of the HPV vaccine

series in adolescent males and females 13-17 in the postal codes within the Columbus,
Ohio metropolis.

HA: There is an association between ethnicity and uptake of the HPV vaccine series in

adolescent males and females 13-17 in the postal codes within the Columbus, Ohio

metropolis.

A cross-sectional design study using multiple logistical regression analysis was

performed to predict the most parsimonious model of HPV vaccine series uptake. The

variables of the study were measured by using responses from the 2014 NIS-Teen. This

research explored the association between SES variables of maternal income, maternal

education, maternal age and ethnicity in communities within communities of Columbus,

Ohio and uptake of the HPV vaccine series.

Threats to Validity

The data produced by the NIS are considered the gold standard for public health

surveillance on immunization rates. The NIS is one of the largest telephone surveys and

produces high-quality estimates of vaccine coverage in the United States (NORC at the

University of Chicago, 2016). Due to the method in which this survey was conducted,

and the theoretical framework the analysis, there are several external and internal threats

to validity involving the participants, location, and their reported income via postal code.

The analysis was performed on participants who completed the 2014 NIS-Teen in

communities within Columbus, Ohio metropolitan and the results are not generalizable to

different populations. There also could be a temporal association between the effects of

SES related to the length of time participants resided in the sampled postal codes. Newer

residents may not have the full effect of access or lack of access to money, knowledge,

prestige, power and supportive social networks where they were surveyed. This threat

was addressed by the random digit dialing sampling of selected households during the

survey. Additionally, there is evidence that parental reporting of vaccination statuses can

be inflated as compared to provider records (Lu, Dorell, Yankey, Santibanez, &

Singleton, 2012). The NIS-teen survey compares the vaccination status reports from

parents to the reports from the adolescent’s providers only when the parents or guardians

have given consent (CDC, 2015c). Lastly, up-to-date vaccine information, individual or

vaccine series was drawn from provider-reported data. There was no recheck of

households or reconciliation of data that might be different from the report of parents or

guardians, and the NIS-Teen surveyors do not re-contact households or providers to

attempt to reconcile potential discrepancies in provider-reported vaccination dates or to

resolve date-of-birth reporting errors (NORC at the University of Chicago, 2016).

Ethical Considerations

Human Subjects

I conducted the research using the 2014 NIS public-use data file for this

secondary data analysis study. The 2014 NIS -Teen staff and contractors are subject to

strict federal laws in regards to protecting the participants and the provider’s privacy

(CDC, 2015e). Employees working on the NIS are required to sign a legal document

saying that they will keep all information private as well as details the consequences of

the illegal disclosure of the information (CDC, 2015e). All information in the 2014 NIS-

Teen was collected under strict confidentiality and can be used only for research as

outlined in [Section 308(d) of the Public Health Service Act, 42 U.S. Code 242m(d), the

Privacy Act of 1974 (5 U.S. Code 552a), and the Confidential Information Protection and

Statistical Efficiency Act (5 U.S. Code)] (CDC, NCRID & NCHS, 2015). Prior to the

release of the public-use data file, the contents of file go through extensive review by the

NCHS Disclosure Review Board to ensure that participant privacy was protected as well

as the protection of data confidentiality (CDC, NCRID & NCHS, 2015). The information

collected in the NIS-Teen are used only for reporting of important statistical health

information in the United States and its territories, and the organization has taken

precautions to protect the privacy of individuals, families, and businesses participating in

the survey (CDC, 2015e).

Ethical Issues

During the data collection phase, many of the telephone numbers are randomly

selected by a computer so listed, and unlisted phone number receive phone calls

requesting permission to conduct the survey. Additionally, potential participants are

mailed a letter from the Director of the National Center for Immunization and

Respiratory Diseases, which describes the survey before a telephone interview was

conducted (CDC, 2015c). These steps were taken to protect participants’ confidentiality

and to make certain they understand that their participation was voluntary (CDC, 2015c).

Summary

In summary, this research was a cross-sectional quantitative study that explored

for a correlation between maternal socioeconomic influences and uptake of the HPV

vaccine series in the communities within the cities of Columbus, Ohio metropolis. By

using multiple logistic regression analysis, this study analyzed the dependent variables of

uptake of the HPV vaccination series with the independent variable of maternal income

measured by postal code. Additionally, the research analyzed the additional independent

variables of maternal education, and maternal age as well as ethnicity. There are several

threats to validity, but due to the RDD nature of gathering participants to be surveyed, the

threats to validity should be minimal. The research was ethical and should satisfy the

requirements of protecting the privacy of human subjects as all information in the 2014

NIS-Teen was collected under strict confidentiality and can be used only for research as

outlined in [Section 308(d) of the Public Health Service Act, 42 U.S. Code 242m(d), the
Privacy Act of 1974 (5 U.S. Code 552a), and the Confidential Information Protection and

Statistical Efficiency Act (5 U.S. Code)]. Additionally, the employees collecting

information for the 2014 NIS-Teen are under strict federal laws in regards to protecting

the participants and the provider’s privacy. The next section of this doctoral study

discussed the results and the findings of the research.

Section 3: Presentation of the Results and Findings

Introduction

The purpose of this study was to assess if there was an association between

maternal SES variables of maternal income and maternal education as well maternal age

and ethnicity and uptake of the HPV vaccine in adolescent males and females ages 13-17

based on postal codes within communities within the city of Columbus, Ohio. I provided

evidence of an association between maternal SES influences and uptake of the HPV

vaccine series. Four research questions were answered as a result of this study: (a) What

is the association between maternal income and uptake of the HPV vaccine series in

adolescent males and females 13-17 in communities with postal codes in the Columbus,

Ohio metropolis, and (b) What is the association between maternal education and uptake

of the HPV vaccine series in adolescent males and females 13-17 in communities with

postal codes in the Columbus, Ohio metropolis, and (c) What is the association between

maternal age and uptake of the HPV vaccine series in adolescent males and females 13-

17 in communities with postal codes in the Columbus, Ohio metropolis, and (d) What is

the association between ethnicity and uptake of the HPV vaccine series in adolescent

males and females 13-17 in communities with postal codes in the Columbus, Ohio

metropolis? The null hypothesis stipulated that there is no association between maternal

socioeconomic factors of income and education and uptake of the HPV vaccination

series, nor between the variables of age and ethnicity and uptake of the HPV vaccination

series in adolescent males and females 13-17 in communities with postal codes in the
Columbus, Ohio metropolis.

In this section, I present the results of a secondary data analysis. I analyzed the

2014 NIS-Teen secondary data using SPSS® version 21 (IBM Corp., 2016). I validated

the analyses using the built-in validation functions in SPSS® v.21. I conducted simple

descriptive analyses of the variables. I recoded the identified variables and categorized

and manipulated them to fit the variables in the research questions. I conducted normality

testing and performed binary analyses and bivariate analysis, followed by multiple

logistic regression using SPSS® on the independent variables to search for statistically

significant associations with the dependent variable. I conclude this section with a

summary of the findings from the data analysis.

Data Collection of Secondary Data Set

The annual NIS-Teen survey is conducted as an adjunct to the NIS. The overall

goal of the NIS is to estimate vaccination coverage rates among 19- to 35-month-old

children in the United States. The NIS uses a random digit dialing (RDD) telephone

survey to identify households with children aged 19 to 35 months and interviews the

adult who was the most familiar with the child’s vaccination history. When such a

household was identified, and the NIS interview was completed, the household was

further screened for the presence of 13- to 17-year-old adolescents. Households without

19- to 35-month-old children are not administered the NIS interview but are further

screened for the presence of a 13- to 17-year-old adolescent. If a household containing

one or more adolescents aged 13 to 17 years was identified, one of those adolescents was

randomly chosen from within the household, and the adult who was most knowledgeable

about the teen’s vaccinations was interviewed. The household interviews for the 2014

NIS-Teen landline and cell-phone samples began on January 9, 2014, and ended on

February 8, 2015 (CDC, National Center for Immunization and Respiratory Diseases, &

National Center for Health Statistics, 2015). The samples were drawn independently from

RDD phone numbers from within the 58 selected geographical regions of the annual NIS.

Data obtained from the teen’s vaccine provider were collected from February 2014

through April 2015 for both landline and cell-phone sample sources. The response rates

for the 2014 NIS-Teen were as follows: resolution rate of 82.6%, screener completion

rate 87.2%, interview completion rate 83.8%, CASRO response rate 60.3%, and teens

with adequate provider data rate 57.1% (CDC, National Center for Immunization and

Respiratory Diseases, & National Center for Health Statistics, 2015).

Discrepancies

There were some discrepancies from the use of this secondary data set. Upon

review of the data, there were no specific methods to identify survey respondents by

postal code for the Columbus, Ohio metropolis. Additionally, I noted that the State of

Ohio (n = 754) did not have the required sample size to conduct multiple logistic

regression analysis. Based on this discrepancy, I modified my location of research into

two cities with differing levels of per capita income using the same methods I planned to

use in the research on Columbus, Ohio. My revised research plan analyzed New York

City, New York (n = 616) and Houston, Texas (n = 679). The combination of these cities

gave me a sample size of 1,295, which was adequate to perform multiple logistic

regression analysis. However, after reviewing the responses concerning HPV uptake for

the cities selected, I found that 170 respondents were unaware of HPV vaccine series

uptake in their adolescent teen and those respondents were excluded from the analysis.

The final sample size analyzed in this study was n = 1,125 which lowered the achieved

power to 0.939 power (1-β err prob) based on calculations using G*Power 3.1 Statistical

Power Analysis for a logistic regression two-tailed analysis from the original 0.95 power

(1-β err prob)(Faul et al., 2007). My study examined two large metropolitan areas with

vast differences in per capita income. In 2015, the per capita personal income (PCPI) in

New York City was $ 63,196 (Department of Commerce, 2017), whereas the PCPI in

Houston for 2015 was $36,913 (Department of Commerce, 2016a). The states that

encompass these cities have comparable HPV-related cervical cancer rates: New York at

7.57 and Texas at 8.27 per 100,000 (CDC, 2017). Both New York and Texas have the

highest cervical cancer rates for the United States. The change in my research plan

prompted a necessary change of my research questions.

My revised research questions are as follows:

Q1: What is the association between maternal income and uptake of the HPV

vaccine series in adolescent males and females 13-17 in communities within the cities of

New York City,

New

York and Houston, Texas?

HO: There is no association exists between maternal income and uptake of the

HPV vaccine series in adolescent males and females 13-17 after controlling for ethnicity

and maternal age within the cities of New York City, New York and Houston, Texas.

HA: There is an association exists between maternal income and uptake of the
HPV vaccine series in adolescent males and females 13-17 after controlling for ethnicity
and maternal age within the cities of New York City,

New

York and

Houston, Texas.

Q2: What is the association between maternal education and uptake of the HPV

vaccine series in adolescent males and females 13-17 within the cities of New York City,

New York and Houston, Texas?
HO: There is no association between maternal education and uptake of the HPV
vaccine series in adolescent males and females 13-17 within the cities of New York City,

New York and Houston, Texas.

HA: There is an association between maternal education and uptake of the HPV
vaccine series in adolescent males and females 13-17 within the cities of New York City,
New York and Houston, Texas.
Q3: What is the association between maternal age and uptake of the HPV vaccine

series in adolescent males and females 13-17 in communities within the cities of New

York City, New York and Houston, Texas?

HO: There is no association between maternal age and uptake of the HPV vaccine
series in adolescent males and females 13-17 in communities within the cities of New

York City, New York and Houston, Texas.

HA: There is an association between maternal age and uptake of the HPV vaccine
series in adolescent males and females 13-17 in communities within the cities of New
York City, New York and Houston, Texas.
Q4: What is the association between ethnicity and uptake of the HPV vaccine

series in adolescent males and females 13-17 within the cities of New York City, New

York and Houston, Texas?
HO: There is no association between ethnicity and uptake of the HPV vaccine

series in adolescent males and females 13-17 within the cities of New York City, New
York and Houston, Texas.

HA: There is an association between ethnicity and uptake of the HPV vaccine

series in adolescent males and females 13-17 within the cities of New York City, New

York and Houston, Texas.

There were also some slight differences in my expectations in how the variables

were listed in the 2014 NIS-Teen data. The differences of the listing of the variables were

incorporated into the multiple logistic regression analysis. Maternal income was reported

on the 2014 NIS-Teen as: $0-$7500, $7501-$10000, $10001-$17500, $17501-$20000,

$20001-$25000, $25001-$30000, $30001-$35000, $35001-$40000, $40001-$50000,

$50001-$60000, $60001-$75000, and over $75000. Ethnicity was reported on the 2014

NIS-Teen as Hispanic, non-Hispanic White only, non-Hispanic Black only, and non-

Hispanic other + multiple race. Maternal education was reported on the 2014 NIS-Teen

as: less than 12 years education, 12 years of education, more than 12 years non-college

graduate, and college graduate. Maternal age was reported on the 2014 NIS-Teen as: less

than 34 years old, 35-44, and over 45 years. Lastly, HPV acceptance was reported on the

2014 NIS-Teen as Yes, No or I don’t know (CDC et al., 2015). The “I don’t know”

respondents were excluded from further analysis.

Additionally, there were other discrepancies from the use of this secondary data

as the NIS-Teen is a telephone survey and the results are weighted to be representative of

all children ages 19-35 months, and even with the statistical adjustments to account for

non-response and households without telephones, there may be some residual bias.

National estimates of vaccination coverage are precise, but state and local estimates

should be interpreted with caution because of limited sample size and widened

confidence intervals than for national estimates of vaccination coverage (CDC et al.,

2015).

The total sample, including the U.S. territory of Puerto Rico, contained

approximately 8.1 million telephone numbers (5.0 million landline and 3.1 million cell-

phone) and created household interviews for 38,703 teens (20,030 landline and 18,673

cell-phone), 21,057 of whom (11,353 landline and 9,704 cell-phone) had vaccine

provider data adequate to conclude whether the teen was current with the recommended

vaccination schedule (CDC et al., 2015). The NIS-Teen RDD telephone survey phase

used independent, quarterly samples of telephone numbers. Sampling frames for the NIS

were provided by Marketing Systems Group (MSG) and the target sample size of

completed interviews in each estimation area was designed to approximately achieve the

equal coefficient of 6.5% of the estimated vaccine coverage from provider reported

histories, given a true coverage parameter of 50% (CDC et al., 2015).

In order to best represent the general population, the 2014 NIS-Teen survey

weights for landline and cell-phone samples were combined in order to weight the full

population of teens aged 13 to 17 years. Teens that resided in landline-only households

(from the landline sample) and cell-phone-only households (from the cell-phone sample)

within the estimation areas were weighted to represent teens in landline-only and cell-

phone-only households. Additionally, because landline and cell-phone sampling frames

sometimes overlap in coverage of teens in landline and cell-phone dual-use households,

dual-users from both samples are combined based on the most effective number of teens

with completed household interview within each phone sample type (landline, cell-

phone), and were weighted to represent teens in dual-use households within each

estimation area. Lastly, teens who lived in houses without phones were excluded from the

dual-frame sample but were accounted for by using controls derived from combining the

2013 census population estimates and the public-use single year 2011 and the 2011-2013

American Community Survey (ACS) data for the United States and Puerto Rico. The

representation within the estimation areas was derived by using small area statistical

modeling techniques used by Blumberg et al. 2012. The modeled telephone estimates

were applied to the control total for the estimation area to approximate the control totals

by detailed telephone status within each estimation area. Additionally, sampling

variability was reduced, and precision of estimation was improved by trimming extreme

weights within an estimation area. RDD sampling weight values that surpassed the

median weight plus three times the interquartile range of the weights within an estimation

area were truncated to that threshold. This weight trimming prevented teens with

unusually large weights from having an unusually large effect on vaccination coverage

estimates (CDC et al., 2015).

Univariate Analysis

Descriptive Characteristics of the Sample Population

A total sample of 1295 respondents in the estimation areas of New York City,

New York (NYC) and Houston, Texas completed the 2014 NIS-Teen survey (NYC 522

and Houston 603). Of the 1,295 respondents, 170 responded that they did not know if the

teen had received any HPV vaccinations and were excluded from the analysis. Table 2-5

depict the sample sizes of the maternal SES variables that were evaluated in the study.

Table 2

Maternal Age

Age Groups Frequency Percent Valid

Percent

Cumulative

Percent

<= 34 YEARS 80 7.1 7.1 7.1

35 TO 44 YEARS 410 36.4 36.4 43.6

>= 45 YEARS 635 56.4 56.4 100.0

Total 1125 100.0 100.0

Table 3

Maternal Income

Frequency Percent Valid

Percent
Cumulative

Percent

$0 – $7500 35 3.1 3.1 3.1

$7501 – $10000 40 3.6 3.6 6.7

$10001 – $17500 74 6.6 6.6 13.2

$17501 – $20000 56 5.0 5.0 18.2

$20001 – $25000 58 5.2 5.2 23.4

$25001 – $30000 51 4.5 4.5 27.9

$30001 – $35000 38 3.4 3.4 31.3

$35001 – $40000 44 3.9 3.9 35.2

$40001 – $50000 55 4.9 4.9 40.1

$50001 – $60000 37 3.3 3.3 43.4

$60001 – $75000 60 5.3 5.3 48.7

$75001+ 436 38.8 38.8 87.5

DON’T KNOW 51 4.5 4.5 92.0

REFUSED 90 8.0 8.0 100.0

Total 1125 100.0 100.0

Table 4

Maternal Education

Frequency Percent Valid
Percent
Cumulative
Percent

LESS THAN 12
YEARS

213 18.9 18.9 18.9

12 YEARS 192 17.1 17.1 36.0

MORE THAN 12
YEARS, NON-
COLLEGE GRAD

217 19.3 19.3 55.3

COLLEGE
GRADUATE

503 44.7 44.7 100.0

Total 1125 100.0 100.0

Table 5

Ethnicity

Frequency Percent Valid
Percent
Cumulative
Percent

HISPANIC 352 31.3 31.3 31.3

NON-HISPANIC
WHITE ONLY

409 36.4 36.4 67.6

NON-HISPANIC
BLACK ONLY

231 20.5 20.5 88.2

NON-HISPANIC
OTHER +
MULTIPLE RACE

133 11.8 11.8 100.0

Total 1125 100.0 100.0

Tables 6-8 depict the sample sizes of HPV vaccine series uptake, the sample size of the

two communities sampled by the 2014 NIS-Teen survey and a summary depicting the

number of the males and females who received or did not receive the HPV vaccine.

Table 6

HPV Vaccine Series Uptake

Frequency Percent Valid
Percent
Cumulative
Percent

YES 610 54.2 54.2 54.2

NO 515 45.8 45.8 100.0

Total 1125 100.0 100.0

Table 7

Estimation Area Of Residence

Frequency Percent Valid
Percent
Cumulative
Percent

NYC 522 46.4 46.4 46.4

Houston 603 53.6 53.6 100.0

Total 1125 100.0 100.0

Table 8

Gender of Child

HPV
YES

HPV
NO

Total

Male

257 289 546

47.1% 52.9% 100%

Female

353 226 5

61% 39% 100%

Total

610 515 1125

Bivariate Analysis

A bivariate analysis of the respondents’ four SES variables and uptake of the

HPV vaccine series revealed a significant finding related to ethnicity. The other three

variables, maternal age, maternal education, and maternal income were found to be non-

significant. The complete results of all bivariate analyses are shown in the following

tables. I conducted a bivariate Pearson χ2 test of the crosstabulation on the variables of

HPV vaccination uptake and maternal age. HPV uptake and maternal age were not

significantly related, Pearson χ2 (2, N = 1125) = .751, p = .69. For HPV uptake and

maternal education, I also conducted a bivariate Pearson χ2 test of cross tabulation of the

variables of HPV vaccination uptake and maternal education. HPV uptake and maternal

education were also showed no significant relationship Pearson χ2 (2, N = 1125) = 4.06,

p = .25. For HPV vaccination uptake and ethnicity, I conducted a bivariate Pearson χ2

test of the crosstabulation of the variables of HPV Vaccination Uptake and ethnicity.

There was a significant pattern of association between HPV uptake and ethnicity, Pearson

χ2 (2, N = 1125) = 8.37, p = .039. The results of the analysis showed that there was a

significant relationship between uptake of the HPV vaccine series and ethnicity of the

respondents living within the estimation areas analyzed in the study. Lastly, an

independent-samples t-test was conducted to evaluate HPV vaccine uptake and maternal

income. The test was non-significant, t(982) = -1.38, p = .17. The patterns of association

between HPV vaccine uptake and the maternal demographic variables are seen below in

Tables 9-12.

Table 9

Crosstabulation Ethnicity of Sample and HPV Vaccine Uptake

Ethnicity* HPV Uptake
Yes NO

Total

Hispanic 213 (60.5%) 139 (39.5%)
White 213

(52.1%)
1

(47.9%)

409
(36.4%)

Black 116
(50.2%)

115
(49.8%)

231
(20.5%)

Other 68
(51.1%)

65
48.9%

133
(11.8%)

Total Responses to
Survey for Maternal

Ethnicity

610
(54%)

515

(46%)

*Indicates p <.05

Table 10

Crosstabulation Maternal Age of Sample and HPV Vaccine Uptake

Maternal Age HPV Uptake
Yes NO

Total

<= 34 YEARS

41
(51.2%)

39
(48.8%)

409
(36.4%)

35 TO
44
YEARS

218
(53.2)

192
(46.8)

231
(20.5%)

>= 45
YEARS

351

(55.3%)
2

(44.7%)

133

(11.8%)

Total Responses to
Survey for Maternal
Age

610
(54%)
515
(46%)

Table 11

Crosstabulation Maternal Education of Sample and HPV Vaccine Uptake

Maternal Education HPV Uptake
Yes NO

Total

LESS THAN 12 YEARS

123
(57.7%)

90
(42.3%)

213
(19%)

12 YEARS

(50.5%)

95
(49.5%)

192
(17%)

12 YEARS, NON-
COLLEGE GRAD

109
(50.2%)

108
(49.%)

217
(19%)

COLLEGE GRADUATE
281

(55.9%)
222

(44.1%)
503

(45%)

Total Responses to
Survey for Maternal

Education

610

(54%)
515

(46%)

Table 12

Crosstabulation Maternal Income and HPV Vaccine Uptake

Maternal
Income

HPV Uptake

Yes No Total

$0 – $7500 20 (57.1%) 15 (42.9%) 35 (3.6%)

$7501 – $10000 19 (47.5%) 21 (52.5%) 40 (4.1%)

$10001 –
$17500

54 (73%) 20 (27%) 74 (7.5%)

$17501 –
$20000

32 (57.1%) 24 (42.9%) 56 (5.7%)

$20001 –
$25000

32 (55.2%) 26 (44.8%) 58 (5.9%)

$25001 –
$30000

28 (54.9%) 23 (45.1%) 51 (5.2%)

$30001 –
$35000

18 (47.4) 20 (52.6%) 38 (3.9%)

$35001 –
$40000

27 (61.4%) 17 (38.6%) 44 (4.5%)

$40001 –
$50000

27 (49.1%) 28 (50.9%) 55 (5.6%)

$50001 –
$60000

18 (48.6%) 19 (51.4%) 37 (3.8%)

$60001 –
$75000

20 (33.3%) 40 (66.7%) 60 (6.1%)

$75001+ 245 (56.2) 191 (43.8%) 436 (44.3%)

Total Responses
to Survey for
Maternal
Income

610
(54%)
515

(46%)
1125

(100%)

Logistic Regression Analysis

Multiple logistic regression was performed on all four variables simultaneously

to analyze the predictors for uptake of the HPV vaccine series in the cities of New York

City, New York and Houston, Texas. The parsimonious analysis of each predictor

variable corresponds with each of the research questions reviewed later in the analysis.

In addition, sex of the child (male/female) and city of residence (New York

City/Houston) were included as control variables. The outcome variable was ‘Ever

received any HPV Vaccinations (yes/no).’ The main goal of the logistic analysis was to

determine the role of several critical predictors in explaining the dichotomous outcome

(yes or no HPV vaccination). The critical predictors were: maternal education, maternal

income, maternal age, ethnicity and uptake of the HPV vaccine series. City of residence

and child sex were also included, as mentioned above. Linearity of the continuous

variables with respect to the logit of the dependent variable was assessed via the Box-

Tidwell (1962) procedure. Based on this assessment, the sole continuous independent

variable, maternal income was found to be linearly related to the logit of the dependent

variable. The model containing all four independent variables (maternal income,

maternal education, maternal age, and ethnicity) as well as the two control variables

(child sex and city of residence) was statistically significantly related to HPV

vaccination uptake (χ2(8) = 34.867, p = .0005.

Pseudo- R2 (Nagelkerke) was .047. Uptake of the HPV vaccine series was

correctly classified for 58.1% of cases by the combination of variables in the logistic

regression equation. Sensitivity was 73.33%, specificity was 39.63%, positive predictive

value was 59.64%, and negative predictive value was 55.00%. Of the six predictor

variables, two were statistically significant: ethnicity and child sex. Non-Hispanic

Whites, Non-Hispanic Blacks, and Non-Hispanic Others all had approximately half the

odds to of being vaccinated than Hispanics (as shown in Table 13). Difference in odds

ratio when comparing the racial groups indicated that an increase in one unit of the

independent variable (0 for Hispanic to 1 for White patients, for example) decreases the

odds of receiving the HPV vaccine series such that Whites were 1.535 times – Exp(B) –

less likely to receive the HPV vaccine series than Hispanics. The reason that the

comparison was seen as a decrease in odds of receiving the HPV vaccine series despite

the fact that the beta value was positive was that the designation of the outcome variable

was (arbitrarily) coded in the reverse direction, with the “yes HPV” category represented

by a lower number (‘1’) than the “no HPV” category (‘2’). When comparing Black teens

to Hispanic teens, the odds of Blacks receiving the vaccine were 1.799 less likely than

Hispanics to receive the vaccine, and teens who were identified as multi-race or “other”

were 1.796 times less likely than Hispanics to receive the vaccine.

Maternal Income, Maternal Education and Maternal age were not significant

predictors of uptake of the HPV vaccine series as their associations with the dependent

variable were non-significant Maternal income β = .013, p > .05, Maternal Age β = –

.096, p > .05, and Maternal Education, β = -.026, p > .05. City of residence was not a

significant predictor of HPV vaccination, β = .004, p > .05.

Child sex was a significant predictor of HPV vaccination, β = -.571, p > .0005.

Females were more likely (61 percent of females) than males (47 percent of males) to

have received the HPV vaccination. The designation of female in the dataset was 2 and

the designation of male was 1, which explains the negative beta coefficient for this result.

Table 13

Logistic Regression Results for Maternal Education, Maternal Age, Maternal

Race/Ethnicity, and Maternal Income as Predictors of Teens’ HPV Vaccine Series

Uptake

Predictor Variable B SE p
Odds
Ratio

95 % Confidence Intervals
for Odds Ratio

Lower Upper

Ethnicity

White .453 .194 .02 1.535 1.046 2.254
Black .534 .187 .004 1.799 1.239 2.612

Multi-
race/
Other .509 .227 .025 1.796 1.141 2.827

Maternal
Age -.096 .112 .387 .908 .730 1.130
Maternal
Education -.026 .07 .734 .974 .839 1.132
Maternal
Income .013 .024 .595 1.013 .966 1.062
City of
Residence .004 .003 .170 1.004 .998 1.010
Child Sex -.571 .131 .0005 .565 .437 .730
Note: For the categorical variable “ethnicity,” Hispanic is the reference category. The
outcome variable was coded as 1-yes, 2-no in the dataset; sex was coded 1-male, 2-
female in the dataset.

Results

One logistic regression analysis was conducted that tested the specific

associations between the variables in the research questions and the outcome variable,

HPV vaccination. The results of the overall logistic regression are seen in Table 13. Each

component of the equation, corresponding to the individual research questions, is

presented below.

Research Question 1

Logistic regression analysis was conducted to investigate the association between

maternal income and uptake of the HPV vaccine series in adolescent males and females

13-17 in communities within the cities of New York City, New York, and Houston,

Texas. The outcome variable of interest was uptake of the HPV vaccine series, and the

possible predictor variable was maternal income. The predictor variable, maternal income

was found to be non-significant β = .013, p > .05. Therefore, no statistically significant

association was found between maternal income and uptake of the HPV vaccine series in

adolescent males and females 13-17 in communities within the cities of New York City,

New York and Houston, Texas.

The null hypothesis that there is no association exists between maternal income

and uptake of the HPV vaccine series in adolescent males and females 13-17 after

controlling for ethnicity and maternal age within the cities of New York City, New York,

and Houston, Texas could not be rejected.

Research Question 2

A logistic regression analysis was conducted to investigate the association

between maternal education and uptake of the HPV vaccine series in adolescent males

and

females 13-17 in communities within the cities of New York City, New York and

Houston, Texas. The predictor variable, maternal education was found to be non-

significant β = -.026, p > .05. Therefore, no statistically significant association was found

between maternal education and uptake of the HPV vaccine series in adolescent males
and females 13-17 in communities within the cities of New York City, New York and
Houston, Texas.

The null hypothesis that there is no association exists between maternal education

and uptake of the HPV vaccine series in adolescent males and females 13-17 after

controlling for ethnicity and maternal age within the cities of New York City, New York

and Houston, Texas could not be rejected.

Research Question 3

A logistic regression analysis was conducted to investigate the association

between maternal age and uptake of the HPV vaccine series in adolescent males and

females 13-17 in communities within the cities of New York City, New York and

Houston, Texas. The predictor variable, maternal age was found to be non-significant β =

-.096, p > .05. Therefore, no statistically significant association was found between

maternal age and uptake of the HPV vaccine series in adolescent males and females 13-

17 in communities within the cities of New York City, New York and Houston, Texas.

The null hypothesis that there is no association exists between maternal age and

uptake of the HPV vaccine series in adolescent males and females 13-17 within the cities

of New York City, New York, and Houston, Texas could not be rejected.

Research Question 4

A logistic regression analysis was conducted to investigate the association

between ethnicity and uptake of the HPV vaccine series in adolescent males and females

13-17 in communities within the cities of New York City, New York, and Houston,

Texas. The predictor variable, ethnicity was found to be significant: Non-Hispanic White

β = .429, p = .029, Non-Hispanic Black β = .587, p = .002, and Non-Hispanic Other β =

.586, p =.011. Therefore, the odds of Hispanic mothers reporting that their child had been

vaccinated were 1.535, 1.799, and 1.796 times that of Whites, African-Americans, and

those identified as multiracial or other race, was found between ethnicity and uptake of

the HPV vaccine series in adolescent males and females 13-17 in communities within the

cities of New York City, New York and Houston, Texas.

The null hypothesis that there is no association exists between ethnicity and

uptake of the HPV vaccine series in adolescent males and females 13-17 within the cities

of New York City, New York, and Houston, Texas was rejected. The evidenced

displayed in the above table suggested that the existence of a relationship between the

ethnicity and uptake of the HPV vaccine series was supported, hence rejecting the null

(Ho) hypothesis.

Summary

In summary, I presented the results of the 2014 NIS-Teen survey as it pertains to

uptake of the HPV vaccine series and maternal SES covariates. A total of 1125

respondents in the estimation areas of New York City, New York and Houston Texas

completed the survey. I used logistic regression analysis to evaluate the relationship

between the four variables of maternal income, maternal education, maternal age, and

ethnicity, and uptake of the HPV vaccine series in adolescent males and females 13-17 in

communities within the cities of New York City, New York and Houston, Texas. Based

on the results of the analysis, I found a statistically significant relationship between the

ethnicity and HPV vaccination uptake such that Hispanic teens were approximately twice

as likely to receive at least one dose of HPV vaccine series than teens from all other

ethnicities. The association between ethnicity and HPV vaccination uptake was

statistically significant; therefore, I rejected the null hypothesis. Bivariate analysis of

these variables also showed a relationship between ethnicity and HPV vaccination

uptake. Logistic regression analysis of the other three predictor variables (maternal

income, maternal education, and maternal age) resulted in a non-significant relationship;

therefore, I failed to reject the null hypotheses of these variables. In the next and final

section of the study, I discussed the findings of my research, their potential application to

professional practice, and the implications for social change.

Section 4: Application to Professional Practice and Implications for Social Change

Introduction

HPV vaccination coverage in the United States does not meet the Healthy People

2020 goals of an 80% vaccination rate. As a means to gather information that could

improve vaccination programs, I investigated the association between maternal SES

variables and uptake of the human papillomavirus (HPV) vaccine in male and female

adolescents ages 13-17 within the estimation areas of New York City, New York, and

Houston, Texas. The study was designed to provide evidence about maternal SES factors

and their association with HPV vaccine series acceptance. I conducted an analysis of

secondary data from the 2014 NIS-Teen public-use survey dataset. The analysis of the

secondary data was done using SPSS version 21 where univariate, bivariate and

multivariate analyses were done.

Concise Summary of Findings

By analyzing the 2014 NIS-Teen data, I found that ethnicity was a significant

predictor of being vaccinated with the HPV vaccine series. HPV uptake and ethnicity

were found to be significantly related, Pearson χ2 (2, N = 1125) = 8.37, p = .039. The

results of the analysis showed a significant relationship between uptake of the HPV

vaccine series and the ethnicity of the respondents living within the analyzed estimation

areas. The odds of Hispanic mothers reporting that their child had been vaccinated were

1.535, 1.799, and 1.796 times that of Whites, African-Americans, and those identified as

multiracial or other race, respectively. Additionally, a child’s sex was a significant

predictor of HPV vaccination, β = -.571, p > .0005. Males had lower odds of being

vaccinated compared to females.

Interpretation of the Findings

Ethnicity

Ethnicity was associated with HPV vaccination uptake in the cities of New York

City, New York, and Houston, Texas. This evidence confirmed the findings of

Bednarczyk et al. (2014), who noted that Hispanic adolescents were consistently higher

in the initiation of the HPV vaccine. Additionally, Kumar & Whynes (2011) found that

ethnicity, childhood immunizations, and usage of preventive and primary care and

cervical screening were predictive of the uptake of the HPV vaccine. Lastly, Lechuga et

al. (2011) indicated that the predictors of HPV vaccine intentions varied by cultural group

and that culture moderated the influence of norms on intentions. The evidence discovered

through my research illustrates a need for additional research to more clearly explain and

find further evidence of the association of ethnicity and the initiation of the HPV

vaccination series in order to improve HPV vaccination initiation across all racial/ethnic

groups.

Maternal Age

Maternal age was not associated with HPV vaccination uptake in the cities of

New York City, New York, and Houston Texas. This non-significant maternal age

association with HPV vaccination uptake disconfirmed the findings of Watson-Jones et

al. (2012), who found that parents who refused vaccination of their daughters tended to

be older household members with less education. As there were not many studies in the

literature examining maternal age as a predictor for HPV vaccine uptake, my research

results indicate a need for further research to elucidate the association or lack thereof

between maternal age and HPV vaccine uptake.

Maternal Income

Maternal income was not associated with HPV vaccination uptake in the cities of

New York City, New York, and Houston, Texas. This non-significant maternal income

association with HPV vaccination uptake in my study disconfirms other research. For

example, Musto et al. (2013) found that the participant’s neighborhood SES was related

to the likelihood of being HPV vaccinated. Additionally, Bednarczyk et al. (2014) found

that since 2008, adolescents living below the poverty level had higher HPV vaccination

initiation than adolescents above the poverty level. These previous studies showed an

association with maternal income not found in my study but also conflicting maternal

income associations in regards to level of SES. My research extends the knowledge of

maternal income as a possible predictor variable of HPV vaccine uptake and justifies

further research on this variable. As found in the above mentioned studies, maternal

income was found to be inconsistently related to levels of HPV vaccine uptake.

Maternal Education

Lastly, maternal education was not associated with HPV vaccination uptake in the

cities of New York City, New York, and Houston, Texas. This non-significant

association between maternal education and HPV vaccine uptake disconfirms findings by

Dorell et al. (2014), who noted that females that delayed HPV vaccination tended to be

White, come from higher income homes, and have mothers with college degrees.

Similarly, Feiring et al. (2015) found an association between higher maternal education

and a lower probability of initiation of the vaccine series, whereas lower education was

associated with a higher likelihood of initiation of the vaccine series. Other studies,

however, suggested that a higher level of education was associated with increased uptake

of the HPV vaccine. Yu et al. (2016) found increased vaccine acceptability to be

associated with older daughters, higher income, and higher level of education. Cullen,

Stokley, and Markowitz (2014) also found that increasing parent education could increase

uptake of the HPV vaccine. My research disconfirms the overall association of maternal

education and HPV vaccine uptake found in these earlier studies. Based on these

contradictory findings in the literature, combined with my finding of a non-significant

association of maternal education and HPV uptake, I recommend further research to

expand knowledge about the role of maternal education as a predictor of HPV

vaccination uptake.

Conceptual Framework

Ethnicity, along with gender, age, personality, socioeconomics, and knowledge

can influence or moderate relationships between health beliefs and health behaviors

(Skinner et al., 2015). Applying the HBM to this study, I found evidence that ethnicity

was more associated with HPV vaccination uptake than the other SES variables

examined. According to the HBM constructs, the various sociodemographic variables of

age, sex, race, education or socioeconomic issues possibly moderate relationships

between health beliefs and health behaviors (Skinner et al., 2015). This concept seemed

consistent in my research as the modifying factor; ethnicity was a significant predictor of

HPV vaccine uptake. As this was a secondary data analysis, indirect analysis showed that

ethnicity could affect the perception of susceptibility to cervical cancer. More research is

necessary to directly test the perception of susceptibility to cervical cancer and the

potential moderation of the perception by ethnicity. According to (Skinner et al., 2015),

perceived susceptibility was a major component for the adoption of preventative health

behaviors. My study revealed that ethnicity was a significant factor predicting HPV

vaccination uptake in New York City, New York, and Houston, Texas. The effective use

of the HBM constructs during an HPV immunization program redesign could improve

HPV vaccination coverage across different ethnic groups and across different regions in

the United States. The effective tailoring of HBM construct-driven vaccination programs

towards communities and regions based on ethnic/racial cultural considerations and

barriers could provide a positive impact and enhance HPV vaccination coverage within

the broad range of diverse communities across the United States.

Limitations of the Study

The data used for this study was secondary data originally obtained as part of the

2014 NIS-Teen survey for immunization coverage estimates of 13-17-year-old adolescent

males and females in the United States. The findings of this study cannot be generalized

to the entire U.S. population as the study sample populations were only in the estimation

areas of New York City, New York, and Houston, Texas and not adequate to be fully

representative of the entire U.S. population. Secondary data can create limitations to a

study as well, as the data were not originally collected for the purpose of this research.

Another limitation of the study was that it was based solely on parental recall and if the

teen received at least one HPV vaccination. This study did not analyze whether the

respondent’s teen completed the HPV vaccination series. Lastly, this study did not

analyze responses from participants’ vaccine providers to confirm the accuracy and recall

from the parents as compared to the actual vaccination records and to assure the accuracy

and precision of overall vaccination coverage estimate.

Recommendations

My current secondary data analysis only looked at two estimation areas in the

states of New York, and Texas. To be more comprehensive in the research process, other

estimation areas should be quantitatively researched to compare or provide additional

information about maternal SES variables in different regions. Additionally, as this was a

secondary data analysis, further research using an HPV vaccine tailored instrument

grounded in the HBM constructs to analyze parents and medical providers’ responses

regarding maternal SES variables and uptake of the HPV vaccine is recommended.

Another consideration for further research would be in-depth interviews and focus group

discussions to qualitatively analyze participants’ responses. Evidenced by the

contradictions in maternal SES associations in previous research, more research is

necessary to improve the knowledge of maternal SES associations and to minimize or

even eliminate some of the contradictions. Lastly, a study using mixed methods,

qualitative and quantitative with responses gathered by the primary researchers

examining the same maternal SES variables could provide more information that could

further advance the goals of this research.

Implications for Professional Practice and Social Change

The study has shown that ethnicity could have a positive or negative effect on

HPV vaccination uptake in New York City, New York, and Houston, Texas. My study

examined researchable modifying factors of the HBM via secondary data analysis to look

at factors that may influence HPV vaccination uptake in large metropolitan areas in

States with high levels of cervical cancer.

Professional Practice

This study provides valuable information gathered through the process of

secondary data research. In regards to professional practice, the findings from this study

could be used to develop and test strategies to improve the uptake of the HPV vaccine

series across the different racial/ethnic groups. The findings of this research could also be

used in the development or the enhancement of culturally-sensitive educational programs

for parents and adolescent teens for use by primary care practitioners. The evidence

found in this study could be used to target the evidenced-based predictors, such as

race/ethnicity as seen in my study in HPV vaccine series educational programs. Lastly,

the evidence found in this study could be used to enhance the knowledge of primary care

health providers about the importance of race/ethnicity sensitive education and literature

to improve the HPV vaccination uptake within their patient populations.

Implications for Research

Findings from this research study showed that future researchers should attempt

to expand the knowledge of the impact of ethnicity on the uptake of the HPV vaccine

series. The study results also indicated that future research should be performed to

improve the knowledge and strength of the relationship between ethnicity and uptake of

the HPV vaccine series and to also examine possible regional influences as my study

looked at two cities with different levels of per capita income in different locations in the

United States. These results provided information that can further advance the field of

HPV vaccination research. The study depicted secondary data analysis as a low cost,

effective means of testing relevant hypotheses concerning HPV vaccination uptake and as

a means to explore for associations that may act as facilitators or barriers to HPV

vaccination. This analysis should be repeated to analyze other regions or States in the

U.S. to test the hypotheses of this study. A strength of this study was it was a low-cost

analysis of a public-use secondary data collected annually to check immunization

coverage across the U.S. within the 58 selected geographical regions of the annual NIS.

Positive Social Change

The research was conducted to narrow the gap of previous research concerning

maternal SES variables and their association with uptake of the HPV vaccine series. In

this study, ethnicity was a significant predictor of uptake of the HPV vaccine series, and

more predictive than maternal education, maternal age, and maternal income. As of a

result of this research, this information could contribute to the improvement of HPV

vaccination programs aimed at increasing the coverage to meet the Healthy People 2020

goals. Using these findings to redesign, supplement or enhance HPV vaccination

programs could ultimately reduce overall morbidity and mortality from cervical cancer in

the U.S.

Conclusion

The findings from this study revealed that: (a) there is an association between

ethnicity and uptake of the HPV vaccine series in adolescent males and females 13-17

within the cities of New York City, New York, and Houston, Texas, (b) there is not an

association between maternal income and uptake of the HPV vaccine series in adolescent

males and females 13-17 within the cities of New York City, New York, and Houston,

Texas, (c) there is not an association between maternal education and uptake of the HPV

vaccine series in adolescent males and females 13-17 within the cities of New York City,

New York, and Houston, Texas, and (d) there is no association between maternal age and

uptake of the HPV vaccine series in adolescent males and females 13-17 within the cities

of New York City, New York, and Houston, Texas. Vaccination to prevent HPV

infection and subsequent cervical cancers should remain a public health priority, and

more research is necessary to further enhance the knowledge gaps in the uptake of the

HPV vaccine series associations within different communities in the U.S. Additionally,

based on conflicting evidence found in previous research, there is a need for more

research to decrease the contradictory evidence found in the literature on HPV

vaccination acceptance. Although there have been significant improvements in the

identification and treatment of cervical cancer, it is still a significant medical and

financial burden for those affected with the disease. Cervical cancer is much more costly

to treat than to prevent through the uptake of the HPV vaccine series. The evidence found

in the study could be used to potentially enhance educational programs designed to

improve vaccination rates, increase prevention, and reduce the overall incidence of

cervical cancer. The use of the results to enhance HPV vaccination programs have the

potential for positive social change by way of improving the lives of individuals,

families, and communities by increasing HPV vaccination and reducing cervical cancer

in the U.S.

References

American Cancer Society, Inc. (2014). What is cervical cancer? Retrieved July 23, 2016,

from http://www.cancer.org/cancer/cervicalcancer/detailedguide/cervical-cancer-

what-is-cervical-cancer

Apte, G., Pierre-Joseph, N., Vercruysse, J. L., & Perkins, R. B. (2015). Could poor

parental recall of HPV vaccination contribute to low vaccination rates? Clinical

Pediatrics, 54(10), 987–991. https://doi.org/10.1177/0009922815590115

Attanasio, L., & McAlpine, D. (2014). Accuracy of parental reports of children’s HPV

vaccine status: implications for estimates of disparities, 2009-2010. Public Health

Reports (Washington, D.C.: 1974), 129(3), 237–244.

Bednarczyk, R. A., Curran, E. A., Orenstein, W. A., & Omer, S. B. (2014). Health

disparities in human papillomavirus vaccine coverage: Trends analysis from the

National Immunization Survey-Teen, 2008-2011. Clinical Infectious Diseases: An

Official Publication of the Infectious Diseases Society of America, 58(2), 238–

241. https://doi.org/10.1093/cid/cit707

Blackman, E., Thurman, N., Halliday, D., Butler, R., Francis, D., Joseph, M., … Ragin,

C. C. (2013). Multicenter study of human papillomavirus and the human

papillomavirus vaccine: Knowledge and attitudes among people of African

descent. Infectious Diseases in Obstetrics and Gynecology, 2013, 428582.

https://doi.org/10.1155/2013/428582

Bresse, X., Goergen, C., Prager, B., & Joura, E. (2014). Universal vaccination with the

quadrivalent HPV vaccine in Austria: Impact on virus circulation, public health

and cost-effectiveness analysis. Expert Review of Pharmacoeconomics &

Outcomes Research, 14(2), 269–281.

https://doi.org/10.1586/14737167.2014.881253

Brown, B., Blas, M. M., Heidari, O., Carcamo, C., & Halsey, N. A. (2013). Reported

changes in sexual behaviour and human papillomavirus knowledge in Peruvian

female sex workers following participation in a human papillomavirus vaccine

trial. International Journal of STD & AIDS, 24(7), 531–535.

https://doi.org/10.1177/0956462412473891

Btoush, R. M., Brown, D. R., Fogarty, S., & Carmody, D. P. (2015). Initiation of human

papillomavirus vaccination among predominantly minority female and male

adolescents at inner-city community health centers. American Journal of Public

Health, 105(11), 2388–2396. https://doi.org/10.2105/AJPH.2015.302584

Centers for Disease Control and Prevention. (2010). FDA licensure of bivalent human

papillomavirus vaccine (HPV2, Cervarix) for use in females and updated HPV

vaccination recommendations from the Advisory Committee on Immunization

Practices (ACIP). Retrieved July 24, 2016, from

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5920a4.htm

Centers for Disease Control and Prevention. (2011). Recommendations on the use of

quadrivalent human papillomavirus vaccine in males—Advisory Committee on

Immunization Practices (ACIP), 2011. Retrieved May 14, 2016, from

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6050a3.htm

Centers for Disease Control and Prevention. (2013). Vaccine information statement |

HPV Cervarix | VIS | CDC. Retrieved March 16, 2016, from

http://www.cdc.gov/vaccines/hcp/vis/vis-statements/hpv-cervarix.html

Centers for Disease Control and Prevention. (2014a). Human papillomavirus

vaccination: Recommendations of the Advisory Committee on Immunization

Practices (ACIP). Retrieved May 14, 2016, from

http://www.cdc.gov/mmwr/preview/mmwrhtml/rr6305a1.htm

Centers for Disease Control and Prevention. (2014b). STD Facts–Human papillomavirus

(HPV). Retrieved April 3, 2016, from http://www.cdc.gov/std/hpv/stdfact-

hpv.htm

Centers for Disease Control and Prevention. (2015a). Human papillomavirus (HPV)

infection–2015 STD treatment guidelines. Retrieved January 3, 2016, from

http://www.cdc.gov/std/tg2015/hpv.htm

Centers for Disease Control and Prevention. (2015b). National, regional, state, and

selected local area vaccination coverage among adolescents aged 13–17 years —

United States, 2014. Retrieved September 11, 2016, from

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6429a3.htm

Centers for Disease Control and Prevention. (2015c). About the National Immunization

Surveys. Retrieved April 5, 2016, from http://www.cdc.gov/vaccines/imz-

managers/nis/about.html

Centers for Disease Control and Prevention. (2016). NIS | About the National

Immunization Surveys | Vaccines | CDC. Retrieved July 19, 2016, from

http://www.cdc.gov/vaccines/imz-managers/nis/about.html

Centers for Disease Control and Prevention, National Center for Immunization and

Respiratory Diseases, & National Center for Health Statistics. (2015). 2014

National Immunization Survey – Teen.

Centers for Disease Control and Prevention, National Center for Immunization and

Respiratory Diseases, & National Center for Health Statistics. (2015b). National

Immunization Survey-Teen A User’s Guide for the 2014 Public-Use Data File.

Retrieved from

ftp://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NIS/NIST

EENPUF14_DUG

Cowburn, S., Carlson, M., Lapidus, J., Heintzman, J., Bailey, S., & DeVoe, J. (2014).

Insurance continuity and human papillomavirus vaccine uptake in Oregon and

California federally qualified health centers. American Journal of Public Health,

104(9), e71–e79. https://doi.org/10.2105/AJPH.2014.302007

Crowcroft, N. S., Hamid, J. S., Deeks, S. L., & Frank, J. (2012). Human papilloma virus

vaccination programs reduce health inequity in most scenarios: a simulation

study. BMC Public Health, 12, 935–935. https://doi.org/10.1186/1471-2458-12-

935

Cullen, K. A., Stokley, S., & Markowitz, L. E. (2014). Uptake of human papillomavirus

vaccine among adolescent males and females: Immunization Information System

sentinel sites, 2009-2012. Academic Pediatrics, 14(5), 497–504.

https://doi.org/10.1016/j.acap.2014.03.005

Daniel-Ulloa, J., Gilbert, P. A., & Parker, E. A. (2016). Human papillomavirus

vaccination in the United States: Uneven uptake by gender, race/ethnicity, and

sexual orientation. American Journal of Public Health, 106(4), 746–747.

https://doi.org/10.2105/AJPH.2015.303039

Department of Commerce. (2016a). Houston, Texas. Retrieved from

https://www.bea.gov/regional/bearfacts/pdf.cfm?fips=48225&areatype=48225&g

eotype=4

Department of Commerce. (2016b). Population estimates, July 1, 2015, (V2015).

Retrieved January 8, 2017, from

//www.census.gov/quickfacts/table/PST045215/3918000

Department of Commerce. (2017). NYCdata: Per Capita Income. Retrieved June 2, 2017,

from http://www.baruch.cuny.edu/nycdata/income-taxes/per_cap.htm

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3:A flexible

statistical power analysis program for the social, behavioral, and biomedical

sciences. Behavior Research Methods, 39, 175–191.

Feiring, B., Laake, I., Molden, T., Cappelen, I., Håberg, S. E., Magnus, P., … Trogstad,

L. (2015). Do parental education and income matter? A nationwide register-based

study on HPV vaccine uptake in the school-based immunisation programme in

Norway. BMJ Open, 5(5), e006422. https://doi.org/10.1136/bmjopen-2014-

006422

Frankfort-Nachmias, C., & Nachimas, D. (2008). Research methods in the social sciences

(7th ed.). New York, NY: Worth Publishers.

Goldberg, D. S. (2014). The implications of fundamental cause theory for priority setting.

100

American Journal of Public Health, 104(10), 1839–1843.

https://doi.org/10.2105/AJPH.2014.302058

Green, S., & Salkind, N. J. (2014). Using SPSS for Windows and Macintosh: Analyzing

and understanding data (7th ed.). Upper Saddle River, NJ: Pearson Education.

Hariri, S., Dunne, E. F., Saraiya, M., Unger, E. R., & Markowitz, L. (2011). Human

Papillomavirus. In S. W. Roush & L. M. Baldy (Eds.), Manual for the

surveillance of vaccine-preventable diseases (5th ed.). Atlanta, GA: Centers for

Disease Control and Prevention.

Harper, D. M., Irons, B. B., Alexander, N. M., Comes, J. C., Smith, M. S., Heutinck, M.

A., … Ahern, D. A. (2014). Quantifying the decisional satisfaction to accept or

reject the Human Papillomavirus (HPV) vaccine: a preference for cervical cancer

prevention. Plos One, 9(2), e88493–e88493.

https://doi.org/10.1371/journal.pone.0088493

Hechter, R. C., Chao, C., Sy, L. S., Ackerson, B. K., Slezak, J. M., Sidell, M. A., &

Jacobsen, S. J. (2013). Quadrivalent human papillomavirus vaccine uptake in

adolescent boys and maternal utilization of preventive care and history of sexually

transmitted infections. American Journal of Public Health, 103(9), e63-68.

https://doi.org/10.2105/AJPH.2013.301495

Hofman, R., van Empelen, P., Vogel, I., Raat, H., van Ballegooijen, M., & Korfage, I. J.

(2013). Parental decisional strategies regarding HPV vaccination before media

debates: a focus group study. Journal of Health Communication, 18(7), 866–880.

https://doi.org/10.1080/10810730.2012.757390

101

IBM Corp. (2016). IBM SPSS Statistics for Macintosh, Version 24.0 (Version 24).

Armonk, NY: IBM Corp. Retrieved from http://www.ibm.com/support,

https://www-304.ibm.com/support/docview.wss?uid=swg21476197

Kumar, V. M., & Whynes, D. K. (2011). Explaining variation in the uptake of HPV

vaccination in England. BMC Public Health, 11, 172.

https://doi.org/10.1186/1471-2458-11-172

Lechuga, J., Swain, G. R., & Weinhardt, L. S. (2011). The cross-cultural variation of

predictors of human papillomavirus vaccination intentions. Journal of Women’s

Health (2002), 20(2), 225–230. https://doi.org/10.1089/jwh.2010.1993

Link, B. G., & Phelan, J. (1995). Social conditions as fundamental causes of disease.

Journal of Health and Social Behavior, Spec No, 80–94.

Lu, P., Dorell, C., Yankey, D., Santibanez, T. A., & Singleton, J. A. (2012). A

comparison of parent and provider reported influenza vaccination status of

adolescents. Vaccine, 30(22), 3278–3285.

https://doi.org/10.1016/j.vaccine.2012.03.015

Luque, J. S., Raychowdhury, S., & Weaver, M. (2012). Health care provider challenges

for reaching Hispanic immigrants with HPV vaccination in rural Georgia. Rural

and Remote Health, 12(2), 1975–1975.

Malkowski, J. (2014). Confessions of a pharmaceutical company: voice, narrative, and

gendered dialectics in the case of Gardasil. Health Communication, 29(1), 81–92.

https://doi.org/10.1080/10410236.2012.719178

McDonald, J. H. (2014). Multiple logistic regression – Handbook of Biological Statistics.

102

Retrieved January 28, 2017, from

http://www.biostathandbook.com/multiplelogistic.html

Millen, J. C., Ginde, A. A., Anderson, A. T., Fang, P., & Camargo, C. A. (2009).

Multicenter Study of Knowledge about Human Papilloma Virus and Attitudes

Among Emergency Department Patients. Journal of Pediatric and Adolescent

Gynecology, 22(6), 356–359. https://doi.org/10.1016/j.jpag.2009.03.001

Moss, J. L., Feld, A. L., O’Malley, B., Entzel, P., Smith, J. S., Gilkey, M. B., & Brewer,

N. T. (2014). Opportunities for increasing human papillomavirus vaccine

provision in school health centers. The Journal of School Health, 84(6), 370–378.

https://doi.org/10.1111/josh.12158

Musto, R., Siever, J. E., Johnston, J. C., Seidel, J., Rose, M. S., & McNeil, D. A. (2013).

Social Equity in Human Papillomavirus Vaccination: a natural experiment in

Calgary Canada. BMC Public Health, 13, 640. https://doi.org/10.1186/1471-2458-

13-640

National Institutes of Health. (2013). NIH Fact Sheets – Cervical Cancer. Retrieved April

11, 2016, from https://report.nih.gov/nihfactsheets/viewfactsheet.aspx?csid=76

Nettleman, M. D., & Garcia-Chen, V. (2013). Vaccinating Through a Lifetime: Adult

Priorities. South Dakota Medicine: The Journal of The South Dakota State

Medical Association, Spec no, 73–79.

NORC at the University of Chicago. (2016). National Immunization Survey (NIS) |

NORC.org. Retrieved October 9, 2016, from

http://www.norc.org/Research/Projects/Pages/national-immunization-survey.aspx

103

Orji, R., Vassileva, J., & Mandryk, R. (2012). Towards an Effective Health Interventions

Design: An Extension of the Health Belief Model. Online Journal of Public

Health Informatics, 4(3). https://doi.org/10.5210/ojphi.v4i3.4321

Phelan, J. C., & Link, B. G. (2013). Fundamental Cause Theory. In C. W. Cockerham

(Ed.), Medical Sociology on the Move: New Directions in Theory (pp. 105–125).

Dordrecht: Springer Netherlands. Retrieved from http://dx.doi.org/10.1007/978-

94-007-6193-3_6

Rahman, M., Laz, T. H., McGrath, C., & Berenson, A. B. (2014). Correlates of Human

Papillomavirus Vaccine Series Completion Among Young Adult Female

Initiators. Human Vaccines & Immunotherapeutics, 10(8), 2163–2167.

https://doi.org/10.4161/hv.29633

Reagan-Steiner, S., Yankey, D., Jeyarajah, J., Elam-Evans, L. D., Singleton, J. A., Curtis,

C. R., … Stokley, S. (2015). National, Regional, State, and Selected Local Area

Vaccination Coverage Among Adolescents Aged 13-17 Years – United States,

2014 (pp. 784–792). Atlanta, United States: U.S. Center for Disease Control.

Retrieved from

http://search.proquest.com.ezp.waldenulibrary.org/healthcomplete/docview/17041

78569/abstract/7D6D3B944448470CPQ/3

Rosenstock, I. M. (1974). Historical Origins of the Health Belief Model. Health

Education Monographs, 2(4), 328–335.

https://doi.org/10.1177/109019817400200403

Ruiz, Á. M., Ruiz, J. E., Gavilanes, A. V., Eriksson, T., Lehtinen, M., Pérez, G., …

104

Haupt, R. M. (2012). Proximity of first sexual intercourse to menarche and risk of

high-grade cervical disease. The Journal of Infectious Diseases, 206(12), 1887–

1896. https://doi.org/10.1093/infdis/jis612

Savoy, M. (2014). An opportunity for improvement: adolescent administration of the

human papillomavirus vaccine. Delaware Medical Journal, 86(2), 53–55.

Schmidt, S., & Parsons, H. M. (2014). Vaccination interest and trends in human

papillomavirus vaccine uptake in young adult women aged 18 to 26 years in the

United States: an analysis using the 2008-2012 National Health Interview Survey.

American Journal of Public Health, 104(5), 946–953.

https://doi.org/10.2105/AJPH.2013.301828

Schuler, C. L., Hanley, C. J., & Coyne-Beasley, T. (2014). Misconception: human

papillomavirus vaccine and infertility. Clinical Pediatrics, 53(2), 158–165.

https://doi.org/10.1177/0009922813504026

Skinner, C. S., Tiro, J., & Champion, V. L. (2015). The Health Belief Model. In K.

Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health behavior: Theory, research,

and practice (5th ed., pp. 75–94). San Francisco, CA: John Wiley & Sons, Inc.

Smith, L. M., Brassard, P., Kwong, J. C., Deeks, S. L., Ellis, A. K., & Lévesque, L. E.

(2011). Factors associated with initiation and completion of the quadrivalent

human papillomavirus vaccine series in an Ontario cohort of grade 8 girls. BMC

Public Health, 11, 645–645. https://doi.org/10.1186/1471-2458-11-645

Smith, L. M., Kaufman, J. S., Strumpf, E. C., & Lévesque, L. E. (2015). Effect of human

papillomavirus (HPV) vaccination on clinical indicators of sexual behaviour

105

among adolescent girls: the Ontario Grade 8 HPV Vaccine Cohort Study. CMAJ:

Canadian Medical Association Journal = Journal De L’association Medicale

Canadienne, 187(2), E74–E81. https://doi.org/10.1503/cmaj.140900

Techakehakij, W., & Feldman, R. D. (2008). Cost-effectiveness of HPV vaccination

compared with Pap smear screening on a national scale: a literature review.

Vaccine, 26. https://doi.org/10.1016/j.vaccine.2008.09.036

Trim, K., Nagji, N., Elit, L., Roy, K., Trim, K., Nagji, N., … Roy, K. (2011). Parental

Knowledge, Attitudes, and Behaviours towards Human Papillomavirus

Vaccination for Their Children: A Systematic Review from 2001 to 2011,

Parental Knowledge, Attitudes, and Behaviours towards Human Papillomavirus

Vaccination for Their Children: A Systematic Review from 2001 to 2011.

Obstetrics and Gynecology International, Obstetrics and Gynecology

International, 2012, 2012, e921236. https://doi.org/10.1155/2012/921236,

10.1155/2012/921236

Union for International Cancer Control (UICC). (2015). Cervical Cancer Initiative (CCI) |

UICC. Retrieved December 3, 2015, from

http://www.uicc.org/programmes/cervical-cancer-initiative-cci/issue

U.S. Food and Drug Administration. (2013). Press Announcements – FDA Approves

New Indication for Gardasil to Prevent Genital Warts in Men and Boys – Gardasil,

October 16, 2009 [WebContent]. Retrieved March 16, 2016, from

http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm187003.ht

106

Vanderpool, R. C., Van Meter Dressler, E., Stradtman, L. R., & Crosby, R. A. (2015).

Fatalistic Beliefs and Completion of the HPV Vaccination Series Among a

Sample of Young Appalachian Kentucky Women. The Journal of Rural Health :

Official Journal of the American Rural Health Association and the National Rural

Health Care Association, 31(2), 199–205. https://doi.org/10.1111/jrh.12102

van Keulen, H. M., Otten, W., Ruiter, R. A. C., Fekkes, M., van Steenbergen, J.,

Dusseldorp, E., & Paulussen, T. W. G. M. (2013). Determinants of HPV

vaccination intentions among Dutch girls and their mothers: a cross-sectional

study. BMC Public Health, 13, 111–111. https://doi.org/10.1186/1471-2458-13-

111

Watson-Jones, D., Tomlin, K., Remes, P., Baisley, K., Ponsiano, R., Soteli, S., … Hayes,

R. J. (2012). Reasons for Receiving or Not Receiving HPV Vaccination in

Primary Schoolgirls in Tanzania: A Case Control Study. PLOS ONE, 7(10),

e45231. https://doi.org/10.1371/journal.pone.0045231

Yu, Y., Xu, M., Sun, J., Li, R., Li, M., Wang, J., … Xu, A. (2016). Human

Papillomavirus Infection and Vaccination: Awareness and Knowledge of HPV

and Acceptability of HPV Vaccine among Mothers of Teenage Daughters in

Weihai, Shandong, China. PLOS ONE, 11(1), e0146741.

https://doi.org/10.1371/journal.pone.0146741

Zimet, G. D., Rosberger, Z., Fisher, W. A., Perez, S., & Stupiansky, N. W. (2013).

Beliefs, behaviors and HPV vaccine: correcting the myths and the

misinformation. Preventive Medicine, 57(5), 414–418.

107

https://doi.org/10.1016/j.ypmed.2013.05.013

Walden

University

ScholarWorks

Walden Dissertations and Doctoral Studies
Walden Dissertations and Doctoral Studies

Collection

2017

Maternal Health Literacy, Antenatal Care, and
Pregnancy Outcomes in Lagos, Nigeria
Olubunmi Adanri

Walden University

Follow this and additional works at: http://scholarworks.waldenu.edu/dissertations

Part of the Public Health Education and Promotion Commons

http://www.waldenu.edu/?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F3651&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F3651&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu/dissertations?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F3651&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu/dissanddoc?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F3651&utm_medium=PDF&utm_campaign=PDFCoverPages

http://scholarworks.waldenu.edu/dissertations?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F3651&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/743?utm_source=scholarworks.waldenu.edu%2Fdissertations%2F3651&utm_medium=PDF&utm_campaign=PDFCoverPages

mailto:ScholarWorks@waldenu.edu

Walden University

College of Health Sciences

This is to certify that the doctoral dissertation

Olubunmi Adanr

has been found to be complete and satisfactory in all respects,

and that any and all revisions required by

the review committee have been made.

Review Committee

Dr. Diana Naser, Committee Chairperson, Public Health Faculty

Dr. Shari Jorissen, Committee Member, Public Health Faculty

Dr. Egondu Onyejekwe, University Reviewer, Public Health Faculty

Chief Academic Officer

Eric Riedel, Ph.D.

Walden University

2017

Abstract

Maternal Health Literacy, Antenatal Care, and Pregnancy Outcomes in Lagos, Nigeria

Olubunmi A. Adanri

MA, Illinois State University, 200

BS, University of Lagos, Nigeria, 198

Dissertation Submitted in Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Public

Health

Walden University

June 2017

Abstract

Maternal mortality, an example of poor maternal health outcomes, is widely accepted as

an indicator of the overall health of a population. One of the Millennium Development

Goals was reduction in maternal mortality by 3 quarters by 2015. These goals were not

met in Nigeria and it is important to look at some of the reasons why. Education has been

shown to have positive impact on pregnancy outcomes; however, the characteristics of

pregnant women, their health literacy level, their usage of antenatal care services and

how these impact pregnancy outcomes are yet to be analyzed in

Lagos,

Nigeria.

Guided

by the social cognitive theory and health belief model, the purpose of this cross-sectional

quantitative study was to determine if there is a relationship between maternal

health

literacy, antenatal care visits, development of medical conditions during pregnancy, and

pregnancy outcomes (measured by healthy or unhealthy baby) in Lagos, Nigeria. The

research question for this study tested if there was a relationship between these variables.

Lisa Chew’s health literacy assessment tool was used in a sample of 130 women in

Shomolu local government in Nigeria who met the inclusion criteria. Using binary

logistic correlations, only problems developed during pregnancy is statistically significant

with pregnancy outcomes (p < .05). The results suggested an increase in

problems

developed during pregnancy most likely will increase the chance of having negative

pregnancy outcomes. Results from this study could promote positive social change by

helping health professionals identify the characteristics of at-risk women during

antenatal

education sessions. The results could also help health professionals in the develop

ment of

targeted antenatal care interventions.

Maternal Health Literacy, Antenatal Care, and Pregnancy Outcomes in Lagos,

Nigeria

by
Olubunmi A. Adanri

MA, Illinois State University, 2000

BS, University of Lagos, 19

Dissertation Submitted in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
Public Health

Walden University
June 201

Dedication

This study is dedicated to my husband, Dr. Adebayo A Adanri, for taking this

journey with me, supporting my efforts, and encouraging me all the way. This is also

dedicated to my children, Adetayo, Adejoke and Tolulope Adanri, for your support and

understanding, especially when I could not give you my undivided attention due to

studying and researching; you guys are simply the best. What I want you to take out of

this journey is that you can achieve anything you set out to do if you are committed and

have the support of your family. Understand your support system as a solid foundation,

nurture the relationship, and put God as your foundation. Also, this study is dedicated to

women all over the world-those who have successfully carried their pregnancy to term,

those who tried and could not conceive, those still trying to conceive, and those who have

lost their lives trying to conceive or give birth. You are more than mothers; you are

providers and great contributors to Mother Earth. May your hearts and homes be filled

with happiness, love, peace, and contentment.

Acknowledgments

I know it came as no surprise that I would embark on this doctoral journey; what

could be surprising is the time it took me to finally go back to school. On this journey, I

appreciate the help and support of my committee chair, Dr. Diana Naser, my committee

member, Dr. Shari Jorissen, and my URR, Dr. Egondu Onyejekwe, for their patience,

guidance, and thoughtful feedback. Thank you for questioning and pushing me to

challenge my ideas. I also want to thank all the faculty members for their diligence and

commitment to excellence. My family deserves my appreciation for their unrelenting

support. I specially want to thank Eng. and Mrs. Falobis for their contribution to this

journey. Thank you to all my sisters and friends for the nickname, “Prof.” I just couldn’t

let you all down, could I? Thank you to my brother, Dr. Blessing Adeoye, and to all who

have invested in me in different ways and have helped me to reach this point in my

academic journey. This has been an interesting and rewarding journey, and I look

forward (expectantly) to what the future offers.

Table of Contents

List of Tables ………………………………………………………………………………………………………

List of Figures ……………………………………………………………………………………………………….v

Chapter 1: Introduction to the Study …………………………………………………………………………

Background of the Study …………………………………………………………………………………..

Problem Statement ……………………………………………………………………………………………

Purpose of the Study …………………………………………………………………………………………5

Research Question(s) ………………………………………………………………………………………..

Theoretical Foundation ……………………………………………………………………………………..7

Nature of the Study …………………………………………………………………………………………..

Definitions……………………………………………………………………………………………………..

Assumptions …………………………………………………………………………………………………..

Scope and Delimitations ………………………………………………………………………………….1

Limitations …………………………………………………………………………………………………….

Significance of the Study …………………………………………………………………………………1

Summary ……………………………………………………………………………………………………….

Chapter 2: Literature Review …………………………………………………………………………………

Introduction ……………………………………………………………………………………………………18

Literature Search Strategy………………………………………………………………………………..

Theoretical Foundation ……………………………………………………………………………………

Literature Review……………………………………………………………………………………………

Summary and Conclusions ………………………………………………………………………………49

Chapter 3: Research Method ………………………………………………………………………………….

Introduction ……………………………………………………………………………………………………50

Research Design and Rationale ………………………………………………………………………..

Methodology ………………………………………………………………………………………………….

Sampling and Sampling Procedures ……………………………………………………………

Sample Size ……………………………………………………………………………………………..

Procedures for Recruitment, Participation, and Data Collection ……………………..

Data Analysis Plan ………………………………………………………………………………………….

Ethical Procedures ………………………………………………………………………………………….

Summary ……………………………………………………………………………………………………….

Chapter 4: Results ………………………………………………………………………………………………..

Introduction ……………………………………………………………………………………………………77

Data Collection ………………………………………………………………………………………………

Results ………………………………………………………………………………………………………….

Chapter 5: Discussion, Conclusions, and Recommendations ……………………………………

100

Introduction ………………………………………………………………………………………………….100

Limitations of the Study…………………………………………………………………………………

103

Recommendations …………………………………………………………………………………………

105

Implications………………………………………………………………………………………………….

106

Conclusions ………………………………………………………………………………………………….

107

References …………………………………………………………………………………………………………

108

Appendix A: Survey Instrument: ………………………………………………………………………….130

iii

Appendix B: Permission to Use Instrument……………………………………………………………1

Appendix C: Certificate of Completion of NIH Training …………………………………………1

Appendix D: Permission Letter from Lagos State Government ……………………………….1

List of Tables

Table 1. Sample Size Calculation: t tests ……………………………………………………………… 5

Table 2. Operationalization of Study Variables …………………………………………………….. 655

Table 3. Descriptive Variables ………………………………………………………………………………

Table 4. Sample Size Calculation: t tests ……………………………………………………………… 800

Table 5. Survey Participants’ Demographics (N = 130) …………………………………………… 82

Table 6. Marital Status and Pregnancy Outcomes ……………………………………………………

Table 7. Employment Status and Pregnancy Outcomes …………………………………………… 85

Table 8. Level of Education and Pregnancy Outcomes ………………………………………….. 866

Table 9. Household Income and Pregnancy Outcomes ………………………………………….. 866

Table 10. Religion and Pregnancy Outcomes ……………………………………………………….. 877

Table 11. Number of Pregnancy and Pregnancy Outcomes ………………………………….. 87

Table 12. Number of Children and Pregnancy Outcomes ………………………………………… 88

Table 13. Correlation Coefficient for Dependent and Independent Variables ……………..

Table 14. Models for Binary Regression……………………………………………..…..

Table 15. Omnibus Test of Model Coefficient…………………………………………….93

Table 16. Classification Table for Pregnancy Outcome ……………………………………………

Table 17. Model Summary ………………………………………………………………………………….. 94

Table 18. Regression Coefficients for Maternal Health Literacy, Frequency of Antenatal

Visits, Number of Antenatal Visit, Timing of the Start of Antenatal Care, and

Problems Developed During Pregnancy………………………………………………96

List of Figure

Figure 2. Map of Lagos state and research study area ………………………………………………

Chapter 1: Introduction to the

Study

Introduction

The World Health Organization (WHO; 2014c) defined maternal mortality as the

ratio of the number of maternal deaths per 100,000 live births. Complications

from

pregnancy and maternal deaths are common in both developing and developed countries,

and in 1990 approximately 523,000 mothers died of childbirth when compared to

289,000 deaths in 2013 (WHO, 2014b). Although this was a marked decrease in

maternal mortality, this number is still unacceptable. The deaths primarily occurred in

low and middle-income countries and may have been avoidable through the availability

of proper antenatal care (WHO, 2004).

In sub-Saharan Africa, 1 in every 40 women die as a result of childbirth compared

to 1 in 3,300 deaths in Europe (WHO, 2014a). In Nigeria, 1 in 13 pregnancies results in

the death of the mother (United Nations International Children’s Emergency Fund

[UNICEF], n.d.). Although medical reasons such as preeclampsia, hemorrhage, and

sepsis have been the cause of these deaths it is important to find ways to reduce avoidable

maternal deaths through education and antenatal communication (Obiechina et al., 2013).

One of the seven Millennium Development Goals (MDG) set by the United Nations (UN)

was to reduce maternal mortality by 75% by the year 2015 (Say et al., 2014), but this goal

was not achieved in Nigeria (Oye-Adeniran et al., 2014). Although there have been

research studies done on maternal mortality in Nigeria (Bukar et al., 2014; Obiechina et

al., 2013), researchers have focused mainly on medical causes of maternal deaths while

few focused on other underlying causes of maternal mortality (Okereke et al., 2013;

Ozumba & Nwogu-Ikojo, 2008).

To successfully reduce maternal mortality there is a need to improve maternal

pregnancy and delivery outcomes by studying education and other socio-determinants of

health necessary for the improvement of the health of these pregnant women (UN, 2011).

In this study, I specifically focused on health literacy level, maternal characteristics, and

antenatal care usage to see how they are associated with maternal pregnancy outcomes in

order to inform healthcare professionals about how to best convey the messages about

antenatal care to mothers so that maternal complications can be avoided. Results from

this study could be used to promote positive social change by helping health

professionals identify the characteristics of at-risk women during antenatal

education

sessions. The results from the study could also be used in the development of targeted

antenatal care interventions. This may potentially help to reduce maternal health

complications and improve pregnancy

outcomes.

In this chapter, I will provide a brief overview of the study by explaining the

background of the study, the problem statement focused on the gap that I intended to fill

with this study, and the purpose of the study. I will also explain the research question

variables of the study and the theories that were used to guide this study. Different terms

used in the study will be included in this chapter as well as an explanation of the nature of

the study, limitations, and significance.

Background of the Study

Zozulya (2010) indicated that a woman dies every 10 minutes as a result of

childbirth and pregnancy in Nigeria, and the WHO (2014b) found that Nigeria has the

second highest maternal rate in the world with over 40,000 maternal deaths every year.

More than half a million women die every year as a result of complications during

pregnancy and childbirth and it is important to find ways to improve the

pregnancy

outcomes of pregnant women and new mothers in order to reduce maternal mortality

(Veneman, 2007). The causes of high maternal mortality and examples of poor pregnancy

outcomes are many and varied. Ujah (2005) attributed a high maternal mortality to lack

of resources or inability to pay for hospital cost, healthcare personnel’s inability to detect

obstetric emergencies early enough, and the inability to perform cesarean sections when

necessary. The National Population Commission (Nigeria) and ICF International (2014)

reported 31% of women between the ages of 15 and 24 and 54% of women between the

ages of 45 and 49 years were illiterate which indicates that the education level of the

population needs to be addressed. If women are illiterate, this may result in them not

being able to read or understand written health materials that are given to them by health

providers. This has the potential to negatively impact their ability to make informed

pregnancy health decisions, especially if they need to be made quickly (Zozulya, 2010).

Lower education levels have been found to be correlated to higher maternal

mortality (Karlsen et al., 2011). Poor literacy is found to be higher in patients with low

educational attainment and this may lead to communication difficulties that may affect

health outcomes (Adhoc Committee on Health Literacy, 1999). However, early antenatal

care and skilled attendance during delivery, access to skilled health workers, and

improvement in basic education are some of the ways to improve maternal health and

reduce maternal mortality (Veneman, 2007). According to Veneman (2007), education

helps to build habits and behaviors that could have a positive impact on a woman’s health

because educated women may have more ability to access health information, to know

what their options are, and be able to better gauge the quality of the care that they are

receiving.

Antenatal visits have been found to reduce the occurrence of maternal deaths

(Taguchi et al., 2004). Normally, when women become pregnant, they visit a medical

facility to talk to a health professional about the state of their health and the child they are

carrying. Ali and Adam (2011) noted that inadequate antenatal care and maternal

education were predictors of maternal mortality in the Sudan. There is a need to

investigate the provision and use of antenatal care services, the literacy level of the

women who use these services, and maternal characteristic of the women who use the

services in relation to maternal health outcomes. Antenatal education has been seen as a

positive approach to preparing pregnant women for the experience of childbirth (Anya,

Hydara, & Jaiteh, 2008). However, Anya et al. (2008) did not examine if there is a

relationship among the number or adequacy of antenatal visits, a pregnant woman’s level

of education, the characteristics of these pregnant women, and pregnancy outcomes in

Nigeria. This study was needed to fill a gap in the literature by examining the

relationship between maternal characteristics, particularly health literacy level; antenatal

care service usage; and pregnancy outcomes for women in Lagos, Nigeria.

Problem Statement

Many women do not receive adequate antenatal care due to factors such as

poverty and lack of knowledge about options (Singh, Bloom, Haney, Olorunsaiye, &

Brodish, 2012). Researchers have looked at how education level is related to maternal

mortality in Nigeria (Mojekwu & Ibekwe, 2012) and how early antenatal visits can help

with detection and treatment of adverse pregnancy related outcomes (Adekanle &

Isawumi, 2008), but I found no study where the researchers examined this education gap,

maternal characteristics, or if how the women utilized antenatal care has any potential

impact on maternal pregnancy outcomes. Therefore, the problem that I addressed in this

study was the high rate of maternal mortality and adverse pregnancy related outcomes by

studying the predictive relationship between maternal health literacy level, the number of

antenatal visits, timing of the first antenatal care visit, development of medical conditions

during pregnancy, and pregnancy outcomes (which is healthy or unhealthy baby) in

Lagos,

Nigeria.

Purpose of the Study

The purpose of this cross-sectional quantitative study was to examine the high

rate of maternal mortality and adverse pregnancy-related outcomes by studying the

predictive relationship between maternal health literacy level; the number of antenatal

visits; timing of the first antenatal care visit; development of medical conditions during

pregnancy; and pregnancy outcomes (birth status of child, which is either healthy or

unhealthy baby) in Lagos, Nigeria. A better understanding of how maternal education

level and maternal characteristics impact pregnant women’s utilization of antenatal care

will help in framing messages about expectations during pregnancy and give a better

understanding of the needs of women that use antenatal services. This information could

be used by the health system to change the way antenatal messages are delivered to

pregnant women. To improve health literacy, health professionals could use this

information to develop powerful antenatal care interventions for women at any level of

education and ensure adequate and necessary care is received, which, ultimately, may

improve maternal pregnancy and health outcomes (Alexander & Kotelchuck, 2001;

Bhutta, Darmstadt, Hasan, & Haws, 2005; Carroli, Rooney, & Villar, 2001; Ickovics et

al., 2007).

Research Question

The following research question and hypotheses guided this study:

Research Question: What is the predictive relationship between maternal health

literacy levels; the number of antenatal visits; timing of the first antenatal

care

visit; development of medical conditions during pregnancy; and pregnancy

outcomes (birth status of child, which is either healthy or unhealthy baby) in

Lagos, Nigeria?

H0: There is no statistically significant predictive relationship between maternal

health literacy level; the number of antenatal visits; timing of the first antenatal

care visit; development of medical conditions during pregnancy; and pregnancy

outcomes (birth status of child, which is either healthy or unhealthy baby) in
Lagos, Nigeria.

HA: There is a statistically significant predictive relationship between maternal

health literacy level; the number of antenatal visits; timing of the first antenatal
care visit; development of medical conditions during pregnancy; and pregnancy
outcomes (birth status of child, which is either healthy or unhealthy baby) in
Lagos, Nigeria.

The independent variables I examined in this study were maternal health literacy,

frequency of antenatal visits, number of antenatal visit, timing of the start of antenatal

care, and problems developed during pregnancy. The dependent variable was pregnancy

outcomes, measured by birth status of child, whether a healthy or unhealthy baby.

Theoretical Foundation

In order to integrate this study within a cohesive body of knowledge, I used two

interrelated theories to explain the relationship between the independent variables

(maternal health literacy, frequency of antenatal visits, number of antenatal visit, timing

of the start of antenatal care, and problems developed during pregnancy) and the

dependent variable (pregnancy outcomes). The two theories were social cognitive theory

(SCT) and the health belief model (HBM). I will provide a brief overview of the theories

in the following subsections but will discuss the theories in more detail in Chapter 2.

Social Cognitive Theory (SCT)

SCT explains how people obtain and maintain certain behavioral patterns and

how these behaviors can be modified (Bandura, 1998). The central tenet of the model

that I focused on in this study was self-efficacy in the participants. Self-efficacy is how a

person assesses his or her ability to exercise control over certain events that affect their

lives or their capability to achieve a level of performance (Bandura, 1989). Specifically,

measuring variables that are related to what they know, what they need to know, how

they can go about adding to their current knowledge, and how they apply the knowledge.

The concept of outcome expectations, which is an expectation that adoption of a behavior

would lead to desired outcomes (Bandura, 1998), is one of the concepts from SCT that I

explored in this study. SCT was relevant to the study because of the relationship between

acquisition of learning through skills and self-efficacy and how this impacts human

behavior. I anticipated that the participants in this study would value the outcome of a

healthy pregnancy (healthy baby), so there would be an incentive to want to learn how

that could be achieved

The Health Belief Model (HBM)

The HBM is a cognitively-based model that originated from the psychological

theories of stimulus response theory and cognitive theory (Hochbaum, Rosenstock, &

Kegels, 1952). The HBM focuses on mental processes as they pertain to why people

accept preventive health services and why people would not adhere to health regimens

(Hochbaum et al., 1952) and the theory has been used to explain many health education

practices and changes in health behaviors (Glanz, 2008). The likelihood of someone

taking preventive action, as proposed by the HBM, depends on the person’s perception of

vulnerability to the condition, the perception that consequences of the condition would be

serious, the perception that taking precautionary behavior would effectively prevent the

condition, and that the benefit of reducing the threat of the condition exceeds the cost of

taking the action (Redding et al., 2003). The constructs of the HBM are perceived

susceptibility, perceived severity, perceived benefits, perceived barriers, cues to action,

and self-efficacy (Hochbaum et al., 1952). The HBM provided insight into factors that

the pregnant women participants perceived as threatening to the outcome of their

pregnancy and helped to explain the characteristics these women possessed that cued

them to taking action.

Nature of the Study

I used a quantitative cross-sectional study design to collect data from woman who

has been pregnant or given birth in the Shomolu local government area, Lagos, Nigeria.

Data were collected through the use of surveys since they can be used to gain reliable and

practical information concerning the wellbeing and functional health of a community or

an individual from a patient or individual point of view (Quality Metric, 2013). I

administered the survey to women over 18 years old who had been pregnant or given

birth in the Shomolu local government area in Lagos, Nigeria. To increase the response

rates, face-to-face administration of the survey was conducted, but only women who

could speak and/or read English were surveyed.

I assessed the health literacy level of participants by asking them the 16 questions

from Dr. Chew’s Literacy Screening Questions (Chew, Bradley, & Boyko, 2004) and

measured antenatal care usage by Kotelchuck’s Adequacy of Prenatal Care Utilization

(APNCU) Index (Koroukian & Rimm, 2002). Antenatal care was measured as adequate

plus, adequate, intermediate, or inadequate maternal care visits based on the

month

antenatal care began and the number of antenatal visits. The dependent variable was

pregnancy outcome, measured by birth status of child which included either a healthy (1)

or unhealthy (0) baby. I used logistic regression and multiple linear regression to analyze

data.

Definitions

Antenatal care: Antenatal care is also known as prenatal care, and it is the care

that a pregnant woman receives from organized health care services (Banta, 2003). These

two terms are used interchangeably but antennal care was used in this

study.

Frequency of antenatal visit: For this study, frequency of visits referred to how

often the women attended antenatal care.

Health literacy: The degree to which individuals have the capacity to obtain,

process, and understand basic health information and services needed to make

appropriate health decisions (Ratzan & Parker, 2000).

Maternal characteristics: In this study, maternal characteristics included the

following: education level, age, occupation, income, religion, and cultural beliefs

Maternal health: The overall physical health of the mother during pregnancy and

during the postnatal period (WHO, 2014). This includes the antenatal care a

pregnant

woman receives and the postnatal care of the woman.

Maternal health literacy: Maternal health literacy is “the cognitive and social

skills (that) determine the motivation and ability of women to gain access to, understand,

and use information to ensure positive health outcomes for them and their children”

(Renkert & Nutbeam, 2001, p. 381). This term, as used in this study, is comparable with

the term health literacy.

Maternal mortality: The WHO 10th International Statistical Clarification of

Health Disease defines maternal mortality as

The death of a woman from pregnancy-related causes while pregnant or within

days of termination of the pregnancy, irrespective of the duration of pregnancy

and the site of pregnancy, from any cause related to or aggravated by the

pregnancy or its management but not from accidental or incidental causes. (WHO,

2014c, para. 2)

Maternal mortality ratio (MMR): The number of maternal deaths in a given year

per 100,000 live births during the same year (WHO, 2014).

Maternal pregnancy complications: Health problems that occur during pregnancy

that can impact the mother’s health (CDC, 2014). Maternal pregnancy complications

include anemia, urinary tract infection, hypertension, gestational diabetes, mental health

issues (CDC, 2014) and complications from malaria.

Number of antenatal visit: For this study, number of antenatal visit referred to the

total number of antenatal visits a pregnant made during the participant’s last pregnancy.

Pregnancy outcomes: In this study, the birth status of child, which was either a

healthy or unhealthy baby.

Assumptions

One of my assumptions in this study was that the Shomolu local government area

was representative of Lagos State demographic and socio-economic characteristics due to

its central location and inclusion of upper, middle, and lower class residents (Lagos State

Government, 2011). The assumption that led to this study was that all pregnant women

would want to have positive pregnancy outcomes and would want to protect themselves

and unborn babies from harm (Anya, Hydara, & Jaiteh, 2008). The extent to which

women do this may depend on their education and characteristics. The survey

instruments were self-administered and the administration format was face-to-face. It was

also assumed that the responses from the participants were true and accurate and that they

were not coerced to participate in the study. Lastly, it was assumed participants would

include both educated and uneducated women.

Scope and Delimitations

The results of this study were specific to Lagos, Nigeria, which could limit the

generalizability of the results. It may be possible to generalize the interventions

regarding culturally appropriate messages developed using health literacy tools to other

parts of Nigeria and even other parts of Africa with similar socio-economic, ethnic, and

religious distributions. In addition, my analysis for this study was specific to females

who had been pregnant and/or have given birth in the past, who resided in Shomolu,

Lagos, Nigeria. Lagos State is considered one of the most affluent states in the country

(Lagos State Government, 2014) and, although there have been studies conducted in the

northern and eastern parts of Nigeria, the results from these studies cannot be generalized

to an urban population like Lagos. I did not conduct this study in a hospital but focused

on women who had been pregnant to gather information about their antenatal practices. I

was interested in the characteristics pregnant women have in common and the differences

in their antenatal usage and education. In analyzing theories to use for the study, I

considered the theory of planned behavior developed in 1980 by Ajzen and Fishbein

because it explains individuals as rational decision makers who consider options and

implications of a behavior before engaging in it (Glanz, Rimer, & Viswanath, 2008).

However, because there are many motivational factors that may determine whether a

pregnant woman performs a specific behavior (like using antenatal care services) or not

and because attitude, perceived behavioral control, and subjective norms were not being

studied, the theory was not used.

Limitations

Potential limitations associated with this study included the use of self-reported

data, which can introduce recall bias (Creswell, 2007). In this study, I used a face-to-

face, self-administered survey because of the ability of ensuring feedback and completion

of the questions and assisting participants who may have questions (Creswell, 2007).

Although this method may have reduced nonresponse bias, it did not provide anonymity

and participants might not have truthfully answered sensitive questions (Creswell, 2007).

To guard against this, Pannucci and Wilkins (2010) suggested the use of a validated

scale. The scale I used was already validated and I also dropped off questionnaires for

participants who were busy or did not want to complete the survey form with me

there.

These were picked up in a sealed envelope, and this could have reduced nonresponse

bias. Also, Frankfort-Nachmias and Nachmias (2008) suggested researchers ensure that

the

questions asked are not threatening by asking participants to rate how uneasy they felt

other people would feel about the questions or rate the degree of difficulty in answering

the questions. Although not all biases in the study could be controlled, the awareness of

the presence of bias allowed thorough scrutiny of the results (Sica, 2006).

Another limitation for this study was the lack of data from women who had died

giving birth. Although this data would have provided rich information about what they

went through, data cannot be collected from a dead person and collecting data from the

family of a dead, pregnant woman may have brought unpleasant memories and the

information provided may have only been hearsay. Furthermore, the study was limited to

women 18 years and over because women under the age of 18 are considered minors and

would have needed adult consent to participate in this type of study.

Significance of the Study

There have been studies conducted in the northern and eastern parts of Nigeria

where researchers looked at the etiology of maternal deaths and the association between

education level and maternal mortality (Idris, Gwarzo, & Shehu, 2007; Ifenne et al.,

1997; Ikeako, Onah, & Iloabachie 2006). In my review of the literature, I did not find

studies that were focused on the Shomolu local government area of Lagos State, which is

located in the southern part of Nigeria. Also, some studies have been conducted in rural

areas of Nigeria (Gazali et al., 2012; Kabir, Iliyasu, Abubakar, & Sani, 2005; Okereke et

al., 2013), but the results from these studies cannot be generalized to an urban population

like Lagos. None of the existing studies mentioned the characteristics of women and

adequacy of antenatal visits or how these could impact pregnancy outcomes. Therefore,

the findings from this study could bring attention to the importance of maternal literacy

and access to antenatal care as some of the ways to reduce maternal death in the south-

western part of Nigeria. When these variables are studied along with the numbers of

antenatal care visits, it could help to understand how a woman’s education level, health

literacy level, and adequacy of antenatal care, and characteristics could impact pregnancy

outcomes. Furthermore, I did not find a study that had been carried out in Nigeria on

maternal literacy level and pregnancy outcomes using a health literacy instrument. This

study could be a pioneering study using a validated tool for maternal health literacy on

the impact maternal characteristics and maternal literacy have on the way women attend

to antenatal messages and how this could impact pregnancy

outcomes.

The study is significant to theory because it builds on the existing theory and

explored potentials for future theory development as it relates to women health care and

pregnancy outcomes in the developing countries. In the study, I explored the reliability

and validity of the instrument used in a different geography and culture.

This study offered a unique opportunity to advance practice and knowledge in

public health education and health delivery services. The results from this study may

help to infer the potential impact of a pregnant woman’s characteristics on birth

outcomes

and influence on the MMR in this area; this could lead to suggestions for ways of

reducing maternal mortality. The positive social implications are that the results of this

study could help health practitioners develop culturally appropriate educational messages

during antenatal and postnatal sessions and also find a way to ensure pregnant women

receive adequate antenatal care services.

Understanding the health literacy level and health care needs of pregnant women

may lead to educational and training opportunities to empower women in the study area

and Nigeria in general.

Results from this study could promote positive social change by helping health

professionals identify the characteristics of at-risk women during antenatal education

sessions and assisting with the development of targeted antenatal care interventions. This

may potentially help to reduce maternal health complications and improve pregnancy

outcomes.

Summary

Maternal mortality remains a great concern around the world despite the fact that

there are preventable measures available. Women, especially in Nigeria and in other

developing countries, are still dying as a result of pregnancy-induced complications. In

this study, I examined the association between maternal characteristics including

education level, adequacy of antenatal visits, the development of medical conditions

during pregnancy, and maternal pregnancy outcomes.

Researchers have shown that there is correlation between women’s education

level and health care decisions ((Babalola & Fatusi, 2009). In this study, I examined the

participants’ education level and how that could impact their health decisions. I also

looked at other maternal characteristics to see how these affect the way a pregnant

woman attends to antenatal care and how this could impact pregnancy outcomes. The

independent variables were maternal health literacy, frequency of antenatal visits, number

of antenatal visit, timing of the start of antenatal care, and problems developed during

pregnancy. The dependent variable was pregnancy outcomes, and it was measured by the

birth status of the child, which is either a healthy or unhealthy baby.

In Chapter 1, I provided an overview of the purpose of the study, the research

question and how the question would be answered, outlined the scope and theoretical

framework for the study, the research design, assumptions, limitations, and significance

of the study. In the next chapter, I will provide a literature review, which included the

review and analysis of the existing research to find the gap that I attempted to fill with

this study. Chapter 2 will also include my thought process behind the selection of the

literature I used in the study and the databases used to search for literature relevant to the

study.

Chapter 2:

Literature Review

Introduction

One of the major concerns of the global community is maternal mortality (Hogan

et al., 2010). Researchers have indicated that maternal age, marital status, education

level, and occupation all play significant roles in identifying women who are at high risks

for pregnancy complications that could lead to maternal deaths (Romero-Gutierrez et al.,

2007). It is unfortunate that in this age of improved technology and medical advances

that some women are still having negative pregnancy outcomes that may result in death

(Oye-Adeniran et al., 2010). Characteristics such as education level, age, occupation,

income, religion, and cultural beliefs could be factors in whether or not a woman uses

antenatal care services during her pregnancy, and it has been reported that having

antenatal care and access to antenatal information throughout pregnancy increases the

chance of having a successful pregnancy and healthy child (Anya et al., 2008). Antenatal

care provides the opportunity for pregnant women to receive information and education

about pregnancy and how to have successful pregnancy outcome. Researchers have

shown that educated women when compared to uneducated women have better

pregnancy outcomes (Harrison, 1985).

I conducted this study to examine the association between maternal health

literacy, antenatal care, development of medical conditions during pregnancy, and

pregnancy outcomes in Lagos State, Nigeria. While researchers in the northern and

eastern parts of Nigeria have shown a positive relationship between maternal education

and better pregnancy outcome (Idris et al., 2007; Ifenne et al., 1997; Ikeako et al., 2006),

this relationship remains to be examined in the southwestern part of the country. Also,

health literacy is critical in health communication, and the health literacy level may limit

a patient’s understanding of their medical conditions, thereby creating a barrier in

discussing their health risks and treatment options (Davis et al., 2008).

In 2000, 149 UN members adopted eight MDGs as international agenda for the

21st century to ensure a just, peaceful, prosperous world (UN, 2015)). The leaders

committed to reduce extreme poverty and to strive to achieve these goals by year 2015

(UN, 2015). The MDGs were adopted to:

Eradicate extreme poverty and hunger; achieve universal primary education;

promote gender equality and women empowerment; reduce child mortality;

improve maternal health; combat HIV/AIDS, malaria and other diseases; ensure

environmental sustainability; and develop a global partnership for development.

(UN, 2015, para. 4)

As a result of the importance attached to maternal mortality, the goal of MDG 5 was to

reduce the MMR by three quarters, between 1990 and 2015 (UN, 2015) and

improvements were made worldwide. In 2013, there were 289,000 maternal deaths

globally and this estimate showed a 45% decline from the figures obtained in 1990

(UNICEF, 2013). The WHO indicated that maternal mortality rate globally was 210

maternal deaths per 100,000 live births in 2013 compared to 380 maternal deaths per

100,000 live births in 1990 (UNICEF, 2013). While this reduction shows progress and a

movement towards the achievement of the MDG 5, these numbers should be concerning

especially since two continents, Africa and Asia, make up 86% of global deaths with 62%

and 24% respectively (UNICEF, 2013).

According to the 10th revision of the International Statistical classification of

Health Diseases, maternal death is defined as:

The death of a woman while pregnant or within 42 days of termination of

pregnancy, irrespective of the duration and site of the pregnancy, from any cause

related to or aggravated by the pregnancy or its management but not from

accidental or incidental causes (UNICEF, 2013, p. 3).

Causes of maternal deaths have been classified as direct and indirect. The direct

causes of maternal deaths are deaths that occur as a result of complications due to

pregnancy or management of the pregnancy (Hogan et al., 2010). In essence, the death

would not have occurred if the woman had not been pregnant. Indirect causes of

maternal mortality are a result of preexisting conditions aggravated by the pregnancy

state (Hogan et al., 2010). Many reasons have been advanced for maternal deaths, so the

knowledge of the final cause or causes of maternal death is important to study to ensure

provision of resources to help improve the incidence of maternal mortality.

In this study, I explored the influence of maternal characteristics, antenatal care

utilization, and maternal health literacy on pregnancy outcomes in Shomolu local

government area of Lagos state, Nigeria. This topic is important because a woman’s

characteristics and health literacy level may mean the difference between understanding

and attending to antenatal messages that could impact pregnancy outcomes and can also

make a difference in the reproductive behavior and health of the woman (Schnell-Anzola,

Rowe, & LeVine, 2005).

The chapter will begin with an overview of the literature search followed by a

discussion of the theoretical constructs for this study, SCT and the HBM. I will also

provide a historical and geographical overview of Nigeria as a nation and the

demography of the people of Lagos state before moving on to discuss maternal mortality

in Africa and in Nigeria. Also, in this chapter I will examine the historic overview of

health services and health provision in Nigeria, the role of women in the society,

women’s education, and antenatal education. In the rest of the chapter, I will focus on the

discussion of the variables addressed.

Literature Search Strategy

Google Scholar, PubMed, ProQuest, Science Direct through Walden University

and Ebscohost were the primary library databases I used to search for materials for this

literature review. Various search terms like maternal health and Africa, maternal

mortality and Africa, antenatal care, pregnancy and disabilities, maternal deaths in

Nigeria, and pregnant women and education level in Nigeria were used, but the ones that

yielded the most results were maternal health combined with Africa and women’s health

combined with maternal mortality and antenatal care. Although my search was limited

to articles published in English, the date of publication was not necessarily limited to the

last 10 years. This was because of the dearth in current literature on the topic and the

need to look at what research was conducted historically on maternal mortality in

Nigeria.

When sorted by topic, I retrieved 2,307 articles, and when filtered to the last 10

years, the search yielded 2,023 articles and sources. Only academic and peer-reviewed

articles were reviewed for this study, and when sorted with those criteria, 574 articles

were retrieved. A search through dissertations and theses was also conducted and three

dissertations were retrieved from the Walden Library database of thesis and dissertations.

A similar search from all dissertation and theses databases produced 78 results. The

search terms I used were maternal health and Africa, maternal mortality and Africa,

antenatal care, and education. Since only 78 dissertations were retrieved with those

terms, the references of the dissertations were searched for additional articles. The final

articles selected for review were included if they addressed maternal mortality; maternal

health; antenatal care; postnatal care; and women’s health in developing countries, in

Africa and especially in Nigeria.

My search on maternal morbidity in Nigeria yielded 3,281 results, but when it

was filtered by journal as the content type 111 articles were found. Also, Medline was

searched for articles published anytime with a combination of the search terms of

maternal health literacy, maternal education level, and maternal mortality level

worldwide. There were 25,005 materials retrieved for global maternal mortality. When I

conducted the same search limited to Africa, 1,291 articles were retrieved. A search for

maternal mortality in Nigeria yielded 668 articles. For maternal health search, only 303

articles were retrieved, most of which were not applicable to the study. A combined

search of maternal mortality and maternal health literacy on Medline produced six

articles for Nigeria with only two articles applicable for the study; a similar search for

Africa produced five results with only one article on literacy. When maternal health

literacy was used as search term for Nigeria only 11 articles were retrieved and only three

articles were applicable for the study.

I needed background information on Nigeria to explain the history, culture,

location, and social economic factors in the country that affect maternal health. In order

to find this background information, I used the specific search terms of Africa, Nigeria,

Lagos, Lagos state, Shomolu, Lagos Island, Ministry of Health in Nigeria, Health system

in Lagos state, the most populous city in Africa and location of Lagos state. The Google

search engine was used for information on Lagos and Nigeria because I found more

information with credible sources this way. However, there was limited information on

the history and demography of the Shomolu local government, so the information I

retrieved from the Lagos State government website was used. Information from the

Lagos State ministry of health was also used. Shomolu local government maintains a

website, but at the time of the study, there was no information relevant to this topic found

there.

To search for a health literacy tool, I reviewed many websites, but the most

productive search was through Health and Psychosocial Instruments database accessed

through the Walden University Library. My search for health literacy assessments

yielded 53 results of articles where some sorts of health literacy assessment tools were

used. However, when a limiter of adult health literacy was used, only 14 results were

retrieved. There were no results for maternal health literacy. Of the14 results retrieved,

11 were in English and of the 11 results that were in English, five of them used a short or

abbreviated version of the health literacy tool.

Theoretical Foundation

In the development of a research study, theory provides a fundamental ground to

start from and which to determine the methods and direction for the research study

(McEachan, Lawton, Jackson, Conner, and Lunt (2008). According to McEachan,

Lawton, Jackson, Conner, and Lunt (2008), theory can guide how behavior may be

changed and allows a researcher to evaluate the reason(s) for the occurrence of the

change and whether the changes in behavior are due to changes in one theory or another.

The choice of a theory used in public health research shapes the way a study is

constructed. I used SCT and the HBM for the theoretical framework of this study.

Social Cognitive Theory (SCT)

SCT has been used to explain behavior as how interaction between behavior,

interpersonal factors, and a person’s environment all interact and determine each other

(U.S. Department of Health and Human Services, 1989). Determinants of a behavior in

SCT are not only the intrinsic factors, which makes a person a product of the

environment, rather SCT posits humans have influence on what they do, what their

individual characters are, and how they respond to their environment, thereby defining

their environment (U.S. Department of Health and Human Services, 1989). Because

these factors work together, change in one is expected to affect the others. SCT explains

how personal, behavioral, and environmental factors influence each other how these

behaviors can be modified (National Cancer Institute, 2005).

Bandura (1998) posited that self-efficacy, outcome expectancies, and goals will

change a health behavior. This is because behavioral change and the maintenance of the

new behavior is as a result of the expectations of the outcome that could result from the

new behavior and the expectations of an individual’s capability to perform the new

behavior (Strecher, DeVellis, Becker, & Rosenstock, 1986). This theory can be used as a

conceptual framework that helps in understanding factors influencing human beings and

how learning occurs (Glanz et al., 2008). Human functioning is a mutual interaction

between environmental factors, behavior, and personal factors like outcome expectancies

and self-efficacy (Ford et al., 2002).

Since the interplay of personal, environmental, and behavioral factors are

explained by SCT in program intervention, I used the theory to analyze how the

environmental factors such as home, work place, and community; behavioral factors such

as following doctors’ instruction, asking for clarification of instruction, and antenatal care

visits; and personal factors such as hygiene, health of the mother, literacy level, and

ability to comprehend written and oral instructions all work to affect a pregnant woman’s

behavior. Availability of health care facilities, access to care, transportation availability,

and education of young girls are environmental factors that I explored using SCT in this

study. When a community places emphasis on the education of their girls and efforts are

made to ensure that pregnant women understand the antenatal messages and early

warnings of pregnancy complications, there is possibility of reduction in maternal deaths

due to awareness and literacy level (Jain & Bisen, 2012). Social support during

pregnancy may reduce anxiety and encourage the use of medical facilities, which could

lead better pregnancy outcomes (Elsenbruch et al., 2007).

Since SCT focuses on acquisition and maintenance of certain behavioral patterns,

I used components of the theory such as self-efficacy, outcome expectations, and enactive

learning, in this study. The focus on acquisition of learning through self-efficacy and

skills was expected to lead to change in pregnant women’s behavior because knowing

that they have the power to protect their pregnancy by following some guidelines would

be empowerment for the pregnant women. Observational learning is one of the concepts

of SCT (Glanz et al., 2008) and by watching the actions and behaviors of other pregnant

women, pregnant women may be more likely to adopt their behavior, which is the

expectation of the intervention. According to Bandura (1998), to change a behavior, a

person will have to believe that such a change will lead to certain outcome.

One of the tenets of the SCT is that human behaviors are determined by incentives

and expectations (Bandura, 1998). These expectancies are self-efficacy expectations.

There are environmental cues and outcome expectancies, which is belief about how an

individual’s behavior will influence outcomes. In order to improve pregnancy outcomes

and support the pregnant women’s own health and well-being antenatal/antenatal care

programs should be available to all pregnant women. Ford et al. (2002) concluded that

women who are unmarried, from a low social economic background, under the age of 20

and who have less than a high school education are often at risk for obtaining inadequate

antenatal care. This study looked at the education level and social economic status of the

women attending antenatal care to evaluate the impact of education on the usage of

antenatal care services.

Health Belief Model (HBM)

HBM was used to understand how individuals relate to health-related matters.

The model was originally developed by Rosenstock and Kegels in the 1950s to explain

poor usage of preventive care services (Edberg, 2007). The HBM seeks to explain that

for an individual to take actions that would prevent an adverse health condition, the

individual must perceive they are susceptible and they must perceive the condition as

serious (Edberg, 2007). Thus, for a pregnant woman to utilize the available antenatal

care services, they must perceive their condition, pregnancy, as serious and that they are

susceptible to pregnancy complications.

The HBM theory is based on the assumptions that if a person feels that taking an

action could help to improve a negative health condition, the person would likely take

action; a person has a positive expectation that taking the recommended action would be

effective at preventing negative health conditions and the person believes she can take the

actions (Glanz et al., 2008). There are six constructs in the model: a) self-efficacy–a

person’s confidence in his or her ability to take action in respect to that behavior, b)

perceived barriers to taking the suggested action, c) perceived severity in one’s opinion

about the seriousness of the health matter and what the consequences are, d) perceived

susceptibility is the chances of getting the condition, e) perceived benefits of taking an

action, and f) cues to action and self-efficacy in respect to that action (Glanz et al., 2008).

Austin, Ahmad, McNally, and Stewart (2002) examined factors influencing breast

and cervical cancer screening behavior in Hispanic women in Toronto and found

embarrassment, fear of cancer, cultural beliefs, and limited ability to speak English as

major perceived barriers to cancer screening. Although many women understand the

usefulness of mammography in successfully detecting breast cancer early they do not see

themselves as vulnerable if they have no family history of the disease or do not have the

symptoms, thus they consider screening for breast and cervical cancer as unnecessary

(Austin et al., 2002). This is an important area to focus on because many women may not

go for antenatal care if they are physically healthy or unaware of the importance of

antenatal care (Gazali, 2012). The HBM was used in this study to assess perceived

susceptibility of pregnancy complications by the pregnant women as a result of not

knowing what to look for during pregnancy. Also, perceived severity of the complication

awareness would be measured by asking questions relating to pregnant women’s

knowledge of what could happen during pregnancy and how to find answers. Perceived

barriers to attending antenatal clinics would also be included to gain understanding of

how these beliefs and barriers affect maternal health.

Literature Review

The review of literature gathered information on the maternal health in Africa,

characteristics of Nigerian women, causes of maternal health problems in Nigeria, health

literacy, and how early antenatal care can be beneficial to both the mother and the unborn

child.

Maternal Mortality in Africa

One of the MDGs adopted by the international community in 2000 is improving

maternal health (WHO, 2014). Although a number of countries have reduced their

maternal death level since 1990 the decrease is still far from the intended numbers needed

to meet the MDG5 (WHO, 2014). The MDG5 was not met as a result of the high number

of maternal mortality, in developing world as 99% of maternal deaths occur in

developing countries but over half these deaths occur in sub-Saharan Africa (WHO,

2014). The trend in maternal mortality in Africa is concerning. The lifetime risk of

maternal deaths ranges in 1 in 4,000 in developed countries, but the risk is 1 in 38 women

in sub-Saharan Africa (UNICEF, 2013).

Sub-Saharan Africa alone accounted for 62% of global deaths in 2013. Africa

has a 100% to 200% increase in rates in maternal death with a ratio of 1:15 women dying

from pregnancy (Abdoulaye, 2005; UNICEF, 2013). Some of the reasons that have been

suggested for the high rate in maternal deaths in Africa include hemorrhage (30%), sepsis

(18%), eclampsia (13%), and complications from abortion (10%; Abdoulaye, 2005). This

is different from the developed countries where hypertensive disorders such as eclampsia

are the primary causes of maternal mortality and hemorrhage accounts for only 13% and

(Alvarez, Gil, Hernandez, & Gil, 2009). Indirect causes, which are preexisting disease

not caused by pregnancy but are made worse by pregnancy, represent 20% of the total

deaths (Alvarez et al., 2009). Alvarez et al. (2009) found that in countries where women

had higher education, the MMR were lower and in these countries, there was overall

increase in the general health of the population. This could be because education shapes

the way a person thinks or acts. Smith-Greenaway (2013) also found a relationship

between a mother’s former education and lower risk of child mortality.

Maternal Mortality in Nigeria

Several reasons have been put forward for the high rate of maternal mortality in

Nigeria. Many of these reasons deal with medical causes of maternal mortality and

morbidity. Obeichina et al. (2013) showed that hemorrhage, sepsis, pre-eclampsia, and

ruptured uterus were the major direct causes of maternal mortality in this area and Bukar

et al. (2014) showed that most maternal in deaths in Yola, northern part of Nigeria, were

caused by preeclampsia/eclampsia, obstetric hemorrhage, and severe anemia. It is not

unusual for pregnant women to be anemic during pregnancy, but severe anemia has been

found to contribute significantly to maternal morbidity and mortality (Broek, 2003). The

World Bank data revealed that 58% of pregnant women in Nigeria suffer from anemia.

Although this medical cause will not be included in this study, it is important to be aware

of this, learn how health care workers address this during antenatal visits, and identify

implications for those who do not utilize antenatal services.

Pregnancy Complications/Disability

According to the WHO (2005), maternal deaths can result from pregnancy related

complication which can occur during pregnancy, child birth or postpartum period. While

11% to 17% of the deaths occur during childbirth, approximately 50% to 70% of the

deaths occur in postpartum period (WHO, 2015, p. 5) and this makes it important to

know what the pregnant women are taught and how the messages are

developed during

antenatal period to handle complications that could occur during pregnancy. Ashford

(2002) noted that the leading causes of death and disability of women in the productive

age, between 15-44 years, in developing countries is maternal complications. This is

because the disabilities that affect these women in the prime of their lives are preventable

and can occur during or after birth and can last as long as life time (Koblinsky, 2012).

These complications could be described as direct and indirect causes. Direct

causes are mainly as a result of complications resulting from error of omission,

intervention or incorrect treatment which could results from the following four major

direct causes: hemorrhage, infection, eclampsia and obstructed labor (WHO, 2005). In

Nigeria, one of the leading maternal complications is hypertensive disorder which may be

preexisting or induced by pregnancy and this could have severe consequences on the

mother and child (Singh et al., 2014). It has been reported that 5% to 10% of pregnancies

in Nigeria are complicated by hypertensive disorder and is a major cause for

hospitalization during the antenatal period (Singh et al., 2014). Early antenatal

intervention will help to reduce incidence of progression of preeclampsia to eclampsia

and better care for women who may be at risk of developing further diseases. Indirect

causes of pregnancy complication that may lead to disability include malaria, hepatitis,

diabetes, cardiac disease, malaria and sexually transmitted diseases and anemia; they

pose health risk to the mother and unborn child (Ashford, 2002).

Nigeria

With a population of 134 million, Nigeria is 2.04% of the world population

(National Population Commission [NPC] and ORC Macro, 2013). Nigeria is the most

populated country in Africa and one of the 10 countries with the highest population

(NPC) [Nigeria] and ORC Macro, 2013). Nigeria lies between latitudes 4º16′ and 13º53′

north and longitudes 2º40′ and 14º41′ on the eastern side. Nigeria is bordered by

Cameroon in the east, Niger in the north, Benin in the west and by Badagry in the South

and approximately 850 kilometers of the Atlantic Ocean (Nigerian Demographic and

Health Survey, 2008). The fourteenth largest country in Africa, Nigeria has a total land

area 923,768 square kilometers (Nigerian Demographic and Health Survey, 2008). In

1960, Nigeria received independence from Britain (Adetunji, 1992). Nigeria has an

estimated 374 ethnic groups with different dialects but these groups can be grouped into

three major groups, Yoruba, Igbo, and Hausa. Currently, there are 36 states in Nigeria

and a Federal Capital Territory (FCT; Nigerian Demographic and Health Survey, 2008).

Population Distribution

The population of Nigeria is not evenly distributed across the country. While the

country is divided into north, south, east and west mainly because of language and

culture, some areas are densely populated while others are sparsely populated. According

to the Nigeria Demographic and Health Survey (NDHS; 2008), the Nigerian average

population density in 2006 was estimated at 150 people per square kilometer and states

like Lagos (South), Abia, Akwa Ibom, Imo, and Anambra (all in the Eastern part of

Nigeria) were considered most densely populated states (NDHS, 2008). The colonial era

and pre-establishment of medical services pointed to the fact that getting antenatal care is

essential during pregnancy, labor, and post-delivery. As far back as 1930 the usage of

health provider either by faith organizations or traditional healers have been associated

with positive maternal outcomes but there are social and economic reasons that made

access to these services not accessible to all (Adetunji, 1992).

Lagos State, Nigeria

Nigeria is divided into three regions: the northern region, the eastern region and

the southern region. The study will be carried out in the southern part of Nigeria,

specifically in Lagos state. Lagos state has been chosen because it is the most industrial

state in the southwestern part of Nigeria and it has more mixed population than any part

of Nigeria (Lagos State Government, 2011). Many residents of Lagos state are from the

north, east, west, and southern part of Nigeria and foreigners also mainly reside in Lagos

State.

Figure 1. Map showing some of the local government area boundaries of Lagos State.

Reprinted with permission from Lagos State government.

Lagos state is the smallest state in Nigeria but the most populous in the nation

(Lagos State Government, 2011). According to the UN (2012), it is one of the one of the

world’s five largest cities by 2005. Metropolitan Lagos is home to over 85% of the State

population (Lagos State Government, 2011). Lagos State is different and unique in

characteristics when compared to the rest of the nation. Unlike other parts of Nigeria,

90% of the Lagos State population has access to electricity and the city consumes 45% of

the energy of the country (UN, 2012). The city is naturally surrounded by water but in

spite of this there is acute and worsening water supply and inadequate sewerage disposal

with much the city’s human waste disposed of by the drainage of rainwater through open

Study

Area

ditches that releases onto the tidal flats (UN, 2012). There are 17,552,942 people in

Lagos with 9,115,041 males and 8,437,901 females all in the 20 local government

councils of the state (Lagos State Government, 2011). As at 2006, the population of

Shomolu Local government area was one million, with a population projection of 1.3

million in 2012 (Lagos State Government, 2011).

This study was located in the Shomolu local government in Lagos state. Located

in the south-western part of Nigeria, Lagos use to be the capital of Nigeria before the

capital was moved to Abuja in 1992 (Lagos State Government, 2014). Lagos State is

considered the economic nerve center of Nigeria due to the large concentration of

industries, major seaports, financial institutions and tourist sites (Lagos State

Government, 2014). Geographically, Lagos is the smallest out of the 36 states in Nigeria,

with an area mass of 3,577.28 km2, which is further reduced by creeks, rivers, lagoons,

and swamps, but it is the most populated with a population of over 21 million (Lagos

State Government, 2011). As would be expected in most urban areas, Lagos State is

highly congested and over crowded with poor roads, limited amenities, and access to

health care (Aluko, 2010). Although the current government is working hard to provide

amenities there are limited resources to go around and it is also interesting to note that

Lagos State also has areas that could be referred to as elite area with low-density

population and occupied by elites, foreign businesses, and high-income group

(Expatarrival, 2015).

There are 20 local government councils and 37 local council development areas.

The study will focus on Shomolu local government council, an area that could be

described as low-to-middle income residential area which has one of the highest levels of

literacy in those who are age 15 to 49 years (Lagos State Government, 2011). In a recent

study by the Lagos state government in 2011, 87% of the participants were found to be

literate in English and 75% are also literate in other languages (Lagos State Government,

2011).

Maternal Health in Nigeria

Maternal health is the overall health of a woman during pregnancy, during

childbirth, and during the postpartum period (WHO, 2014). When a woman is pregnant,

there should be specialized care for the safety of the mother and the child. Even in the

historical society of Africa, before education or civilization, traditional healers were

responsible for ensuring safe delivery of pregnant women and providing obstetric care

and divination (Adetunji, 1992). At the time of birth, the female members of the

household would assist the pregnant woman with the birth and if there were

complications the traditional healers would be call upon to assist (Adetunji, 1992). With

the advent of Christianity there was a shift to mission houses for healthcare because

Christian converts did not want to be seen with herbalists or traditional healers in order

not to compromise their faith (Harrison, 2009). This is an important development

because church organizations became pioneers of modern health care services in Africa

in the colonial era (Adetunji, 1992).

The Church Missionary Society of Nigeria partnered with the colonial

government to raise midwifery standard to the British standard by meeting the healthcare

needs of the people and improving their ways of live (Harrison, 2009). Although there

was no record of mortality or fatality, it could be assumed that there must have been a

high percentage of maternal mortality in the pre-colonial era and this was why the church

introduced the concept of maternal care. The aim of obstetrics care was to reduce

maternal mortality and this was done by providing antenatal care, identifying danger

signs that could lead to adverse pregnancy outcomes, and facilitating early intervention of

such maternal complications (Adetunji, 1992).

The influence of churches in this era provided a way of looking at the future of

obstetric care. The introduction of Christ Apostolic Church (CAC) pregnancy care in

1930 led to the creation of school of midwifery and midwives were trained in England by

the Church Missionary Society of Nigeria (Adetunji, 1992). By 1949 there were

maternity homes to handle 6500 deliveries in a year and the ratio of maternal mortality

was 46 per 100,000 births (Harrison, 2009). Within CAC, it was mandatory for women

to register their pregnancy early and attend weekly meetings. The mandatory weekly

meeting was used for teaching the pregnant women, praying for them, and praying for

their unborn babies (Adetunji, 1992). Topics covered in these meetings included sexual

education such as safe sex during pregnancy and when to stop, pregnant women’s state of

mind and when to be concerned, and work habits including posture, physical exercise,

weight lifting, health, and hygiene (Adetunji, 1992). Large-scale community health

education was combined with the establishment of maternity homes operated by trained

midwives with thorough supervision and established measures to ensure efficiency and

boost staff morale (Harrison, 2009).

Antenatal Care

Antenatal visits have been found to lower the chance of maternal deaths (Taguchi

et al., 2004). Antenatal or antenatal care visits are the visits women make to the health

care practitioners during pregnancy and these visits are important because pregnant

women have the opportunity to learn more about pregnancy, what to expect during labor,

and how to take care of their newborn (Jennings, Yebadokpo, Affo, & Agbogbe, 2010).

Antenatal visits also provide an opportunity for screening, routine examination,

treatment, counseling, and access to information to help pregnant women have a

successful pregnancy and prepare them for what to expect in the next months (Jennings et

al., 2010). Access to this information and services during pregnancy is effective for

improving maternal health. The need for antenatal care is further justified by the number

of deaths that occur among women who are not registered for delivery (Ekott, Ovwigho,

& Ehiegiegba, 2012). Antenatal care visits present an avenue for the women to be

screened for disease treatment and prevention. Also, attending antenatal care program

provides the opportunity for pregnant women to be aware of pregnancy complication and

to seek help early when needed (Ekott et al., 2012).

The timing of when a pregnant woman goes in for the first antenatal visit is

important to diagnose any abnormalities and to ensure appropriate and early intervention

and treatment (WHO, 2002). According to the WHO (2002), the first antenatal

visit

should occur before or during the twelfth week in order to get accurate medical and

obstetric history that would allow for adequate follow up and identification of supports

needed for successful pregnancy. With 920 deaths per 100,000 live births, sub Saharan

African is noted as the region with the highest level of maternal mortality (Alvarez et al.

2009; McTavish et al. 2010). This statistic is not too different from UNICEF’s claim that

Nigeria loses about 145 women of childbearing age to pregnancies every day (UNICEF,

n.d., para 2). Many reasons have been suggested for this high number of maternal deaths

and they include direct causes like hypertension, bleeding/hemorrhage, obstructive labor

and hypertensive disorder (Alvarez et al. 2009). According to Harrison (1997),

fundamental issues of poverty, illiteracy, unbooked emergencies (pregnant women who

did not register for antenatal services but who were rushed to the hospital as a result of

pregnancy complications), and governmental economic policies were the major causes of

maternal mortality 15 years ago. Much has changed in 18 years but the statistics on

maternal mortality in Nigeria are becoming worse and baffling to researchers (Mojekwu

& Ibekwe, 2012).

Role of Religion and Religious Beliefs on Pregnancy Health

Religion and religious beliefs have been shown to impact how health services are

used and perceptions of the efficacy of modern services (Gazali et al., 2012). Gazali et

al. (2012) found that women tend to remain in the house and rely on their Muslim

religious leaders’ ritualistic reading of the Quran to protect the pregnant woman and the

unborn child instead of going to health facility. This author revealed religious and

traditional beliefs still influence maternal mortality in Nigeria and it is important to

address this to evaluate if religious practices of a woman still impact their utilization of

antenatal clinics and their attendance to the message received. The importance of formal

education in these instances cannot be over emphasized because education creates

awareness that will create knowledge. The ability to read and understand antenatal

message may have a protective effect on acceptance or rejection of traditional practices

that could affect maternal mortality.

Researchers found that uneducated women may not be able to obtain the help of

professional health services because of lack of awareness of availability of such services

or as a result of fear or alienation they feel with modern health care facility while women

who are educated have better understanding of health system process and are less

disposed to accept traditional practices (Gazali et al., 2012). Since religious beliefs and

taboos can impact a woman’s choice of health and treatment during pregnancy (Gazali et

al., 2012), they can also affect early detection of warning signs of pregnancy

complications and response to such signs or warnings since antenatal visits help to detect

and treat adverse pregnancy related outcomes (Adekanle & Isawumi, 2008).

Religious and traditional beliefs are factors affecting maternal mortality in Sub-

Saharan Africa and it is important to address these to maintain maternal health. Chiwuzie

and Okolocha (2001) noted that bleeding during pregnancy was considered normal and

many taboos associate with pregnancy include not eating eggs, snails, or sweet foods like

milk to prevent pregnancy complications. Religion and religious beliefs have been

shown to influence the use of health services and perceptions of the efficacy of modern

services (Gazali et al., 2012). These authors illustrated how reliance on traditional taboos

and religion can impact a pregnant woman’s decision to seek medical help during

pregnancy and the type of treatment they receive.

Social Economical Status and Antenatal Services Utilization

Eighteen percent of females in urban areas and 7% in rural areas have completed

secondary school level education compared to 22% of males in urban areas and 11% in

rural areas that completed secondary school education (NPC & ICF Macro, 2009). This

can lead to gaps in socioeconomic status as more education is correlated to more income.

As a result of poverty, many women may not be able to afford antenatal care during

pregnancy or medical care for their baby after the child is born. Babalola and Fatusi

(2009) found there is low use of medical facilities by women with low literacy level in

Nigeria. This association between poverty and literacy was also supported by McTavish,

Moore, Harper, and Lynch (2010) who reported that in countries with lower literacy

levels, women in poor households were less likely to use maternal care.

Ogunlesi (2004) found that women from higher social economic backgrounds are

more likely to value and use antenatal care and delivery services during pregnancies.

These women would most likely be able to take prompt and cost effective health

decisions when compared to women from lower social economic background.

Furthermore, when women are employed they tend to have higher economic status,

which leads to better usage of antenatal services (Kabir et al., 2005).

Pregnant women are expected to visit medical facilities to have medical personnel

check the growth of the growing baby and their overall health. Unfortunately, some

women do not do this and it is evidenced in the findings that most deaths happen within

24 hours of admission as a result of delay in going to health facility (Bukar et al., 2014).

A recent report on the progress of MDG 5 concluded that “proven health-care

intervention can prevent or manage these complications, including antenatal care in

pregnancy, skilled care during childbirth and care and support in the weeks after

childbirth” (UN, 2015, p. 39).

Nigeria, unlike developed countries, does not track statistics relating to maternal

mortality and most of the data available are based on hospital research conducted by

other researchers (Adegoke, Lawoyin, Ogundeji, & Thomson, 2007). Because most

births are not recorded or take place in a medical center there are no accurate statistics on

number of pregnancy and childbirth related deaths (Adegoke et al., 2007).

Maternal Age and Pregnancy Outcomes

According to the World Bank Report (2014), women constitute 49% of the

Nigerian population and 23% of these women were teenage mothers between 15-19 years

of age. It is important to consider the maternal age of women and how this could impact

maternal mortality because of the implication for maternal maturity and biological

readiness. Age and parity are considered major biological determinants of rate of

maternal mortality in some parts of the world.

In Sweden, the optimum child bearing age is considered between 20 and 29 years

and risk of maternal death for women younger than 20 years and older than 29 years are

two to six times higher (Högberg & Wall, 1986). Researchers in Pakistan found women

under 19 or over 39 years of age are at a greater risk of maternal death (Midhet, Becker,

& Berendes, 1998). Having a child early in life, during the teenage years, could limit a

young girl’s opportunity for better education or ability to get a good job, but delaying

motherhood to later years could lead to pregnancy complications as well (World Bank,

2014). Jacobson, Ladfors, and Milson (2004) found that the risks of miscarriage,

gestational diabetes, and preeclampsia increased in women between 40 to 45 years.

Age

would then be an important variable to consider as having an impact on education,

standard of living, pregnancy complications, and pregnancy outcomes.

Status of Women, Literacy, and Pregnancy Outcomes

Status of women in Nigeria. In the traditional Nigerian society, women were

important in the economic sphere because of the roles they play in virtually all areas of

farming activities such as land clearing for farming, produce processing and marketing

(Anugwom, 2009). Women are often the ones who keep the household while holding

down a job or trade, and they sometimes may not have the time to take care of themselves

or may not consider it a priority. However, with education, a pregnant woman would be

sensitized to the importance of caring for herself during pregnancy so she could have the

time and energy to care for her family. Thus, educating women could be viewed as an

investment not just for the family but also for the nation (Anugwom, 2009).

Generally, a healthy pregnancy outcome is a result of the care a mother receives

during and immediately after antenatal visits. Most significant is the care received during

the antenatal period because during the antenatal care visit, a new or seasoned mother

would get information about pregnancy care and how to care for themselves and their

children. Babalola and Fatusi (2009) found that most Nigerian women have a lower usage

of maternal health facilities and a lower usage of skilled assistance during delivery than

those from other African countries. Low usage of maternal facilities during pregnancy

could be connected to the education level of these women. These authors found a

statistically significant positive association between education and use of maternal

services (Babalola & Fatusi, 2009). Education leads to information seeking behavior

which could lead to higher level of health awareness and better knowledge of available

health services and resources (Babalola & Fatusi, 2009). A low educational level leads to

a low literacy level which is associated with less optimal health outcomes and

information seeking behavior (Shieh, Mays, McDaniel, & Yu, 2009).

Literacy. When women are educated and literate they are more likely to use

antenatal care services during pregnancy, visit health facilities more often during

pregnancy, and most likely deliver at a health facility (Iyaniwura & Yusuf, 2009). As a

result of antenatal care and health facility usage there would be access to information that

these women could use and this could help in reducing maternal mortality and also

improve health in general. Women’s educational level attainment has been linked to their

comprehension of health message and understanding of medical instruction (LeVine,

LeVine, Rowe, & Schnell-Anzola, 2004) and also to birth and death rates (LeVine et al.,

2004). Ferguson (2008) reported that individuals without a high school education are

more likely to have lower literacy than those with a high school education. Deficiency in

literacy affects health seeking behavior and information seeking abilities (Shieh et al.,

2009). This is because there are many information sources currently and the ability to

seek and use information is dependent on a woman’s education level. In examining the

relationship between comprehension of health information and literacy skills among

young mothers in Zambia, Stuebing (1997) found that school-acquired literacy skills, and

especially oral “decontextualized” language skill, are indeed a missing link in the

relationship between maternal schooling adult comprehensions of health information.

Although conducted in 1997, the results are still relevant today because of the findings of

the relationship between maternal schooling and ability to comprehend and act upon new

information outside the home.

Okereke et al. (2013) found a significant association between education,

knowledge of maternal danger signs, and knowledge of antenatal visits. They confirmed

the importance of education to understanding maternal danger signs because there was

significance in having some form of education and being aware of safe motherhood

practices. However, this study was carried out in rural settings and findings may not be

generalized to urban setting like Shomolu. Also, antenatal visits in this study did not

include the kind of messages women receive and how these messages help to shape the

women’s antenatal behaviors.

It is generally said in Africa that “if you educate a man you educate an individual,

but if you educate a woman you educate a family (nation)” (Suen, 2013, p. 66). This is

also true of maternal education. When mothers are able to read, they are able to

comprehend written health information and even auditory health messages used in

developing countries (Levine et al., 2004). Understanding of these messages is important

to adopting healthy behavior that has been found to lead to children survival which may

also translate into maternal survival (Levine et al., 2004). Although Levine et al. (2004)

showed an association between a mother’s literacy skills and the survival of the child,

there is a need to confirm in Nigeria if there is such an association between a mother’s

education level, maternal health literacy, and pregnancy outcomes. This study provided

investigation of these associations.

Understanding basic information about one’s health may include written

directions or reading materials and the ability to process this information may be critical

to making life and death decisions (LeVine et al., 2004). Mojekwu and Ibekwe (2012)

found there is a need to improve maternal education since this is one of the factors

responsible for high maternal mortality in the past in Nigeria. Maternal education could

lead to empowerment of women in making decisions about issues that concern them and

the infants they are carrying.

Decision-making skills are needed to navigate pregnancy, childbirth, and child

rearing and individuals with limited education may find it difficult to carry out such

functions without encountering difficulties (Ferguson, 2008; Shieh & Halstead, 2009).

When given materials containing information on how to implement their care, it becomes

difficult for people with low education to understand and carryout the instructions in their

care plan (Ferguson, 2008). Nikiéma, Beninguisse, and Haggerty (2009) researched the

extent to which women in 19 countries in sub-Saharan Africa remembered receiving

information about complications in pregnancy during their antenatal visit and if it

impacted the likelihood of delivering their babies in health institutions or at home. They

found that pregnant women were not routinely provided with information about

pregnancy complications during their antenatal care visits or the information was not

provided in a way that the women remembered the information. It is important to look at

the information being given to pregnant women to see if education level is one of the

reasons for not remembering this information and if not remembering antenatal

information has implications for unmet needs and what the impact is on maternal

outcomes.

Antenatal care visits should be used to inform and educate pregnant women about what to

do and what to expect during pregnancy, care for the newborn, and necessary postnatal

care. Anya, Hydara, and Jaiteh (2008) found that, in spite of repeated visits to the clinics,

pregnant women in their study did not benefit from effective information, education, or

communication since most women reported spending few minutes with the care providers

and could not recall being informed about important subjects as nutrition and diet, family

planning, care of the baby, place of birth, or what to do in case of complication.

Rothman et al. (2004) determined that it is difficult for women with low literacy to read

directions, to follow medical recommendations, and they have poorer knowledge which

can lead to worse clinical outcomes. In a study in Sudan, it was observed that inadequate

antenatal care and maternal education were predictors of maternal mortality (Ali &

Adam, 2011); this leads to the question if this can also be true in Nigeria. There is a need

to investigate the provision and use of antenatal care services, the education level of the

women who use this service and socio demographic characteristics of the women who

use the service in relation to maternal deaths.

Previous researchers who have studied maternal health care have found

associations between maternal health care and literacy level (McTavish, Moore, Harper,

& Lynch, 2010). The concept of maternal education as it affects women’s health and

specifically women’s maternal health has been issue of discussion for researchers in the

developed world. According to the WHO’s definition of health literacy, as reported in

Renkert and Nutbeam (2001),

Health literacy represents the cognitive and social skills that determine the

motivation and ability of individuals to gain access to, understand, and use the

information in ways that promote and maintain good health. Health literacy means

more than being able to read pamphlets and successfully make appointments….

(p. 381)

Thus, health literacy enables individuals to process and understand health

messages in order to make better health choices. This ability to read and process health

messages is said to improve a woman’s reproductive behavior, which would lead to an

improvement in their children’s health and survival (Levine et al., 2012). Literacy skills,

which includes the ability to speak and understand spoken language, read, write, and

understand written language, and ability to understand the use of numbers is very

important in understanding health related behaviors in different community (Ciampia et

al., 2012). The United States Department of Health and Human Services Health noted

that nine out of ten American adults find it difficult to understand information about their

health (U.S. Department of Health and Human Services, 2014) though 99% of Americans

are said to be somewhat literate (Central Intelligence Agency[CIA] Factbook, 2014). It

would be interesting to find out what percentage of Nigerians can understand

information

about their health though 68% of the population age 15 and over can read and write (CIA

Factbook, 2014). Also, the Shomolu local government is one of the areas with high

number of educated women in Lagos State (Lagos state government, 2014) I sought to

find out if the women’s literacy level correlates with their health literacy level, and how

this impacts health mortality.

Summary and

Conclusions

While there have been studies on importance of antenatal visits (Adekanle &

Isawumi, 2008; Anya et al., 2008; Jennings, Yebadokpo, Affo, & Agbogbe, 2010;

Okereke et al., 2013; Taguchi et al., 2004), there has not been a study focusing mainly on

the impact of maternal characteristics, maternal health literacy, and antenatal visits on

pregnancy outcomes in Nigeria. It is important to know what women learn during

antenatal visits and if their characteristics and health literacy levels in anyway affect their

understanding of antenatal services and utilization of information received. The next

chapter focuses on methodology, data collection, and data analysis plans. The purpose of

this cross-sectional quantitative study was to examine the high rate of maternal mortality

and adverse pregnancy related outcomes by studying the predictive relationship between

maternal

health literacy level, the number of antenatal visits, timing of the first antenatal

care visit, development of medical conditions during pregnancy and pregnancy outcomes

(birth status of child, which are healthy or unhealthy baby) in Lagos. Chapter 3 focused

on the research design and methodology, population under study, sample size, sample

selection process, description of the instrument used for data collection, and the process

of data collection and methods used in analyzing the data.

Chapter 3: Research Method

Introduction

In this chapter, I will discuss the research design, methodology, and rationale for

this study; the population under study; sample size; method and procedure for data

collection; the instruments used for data collection; and how the data were analyzed. The

purpose of this cross-sectional quantitative study was to examine the high rate of

maternal mortality and adverse pregnancy related outcomes by studying the predictive

relationship between maternal health literacy level, the number of antenatal visits, timing

of the first antenatal care visit, development of medical conditions during pregnancy and

pregnancy

outcomes (the birth status of the child, which is either a healthy or unhealthy

baby) in Lagos, Nigeria.

Poor maternal health outcomes, especially the high incidence of maternal

mortality rate in Nigeria, create an unacceptable major public health problem. It has been

established that an approximately 287,000 women died giving birth worldwide and 99%

of these deaths are from developing nations (Say et al., 2014). While some factors are

medical factors, there are nonmedical and avoidable factors that could help in reducing

maternal mortality. Factors such as delay in seeking medical care, delay in recognizing

an obstetric problem, following medical advice, and non-utilization of antenatal care

services (Ozumba & Nwogu-Ikojo, 2007) can all be attributed to insufficient education

and the characteristics of the pregnant women. In this study, I examined the predictive

relationship between maternal health literacy level, antenatal care, development of

medical conditions during pregnancy, and pregnancy outcomes among women that are 18

years and older who had been pregnant before, but were not pregnant at the time of the

study, in Lagos State, Nigeria.

To achieve this, I used a survey to gather information about the number of

antenatal visits, maternal health literacy level, timing of the first antenatal care visit,

development of medical conditions during pregnancy, and the outcome of the pregnancy.

Also, information on maternal characteristics of these women that included their age,

religion, number of pregnancies, number of children, education level, household income,

and employment status were collected in order to examine the predictive relationships

between these factors and pregnancy outcomes. The theoretical application of SCT and

the HBM were utilized to explain how pregnant women make decisions relating to their

pregnancy and the impact of the environment on their decision-making powers.

The research question that guided the study was: What is the predictive

relationship between maternal health literacy levels, the number of antenatal visits,

timing of the first antenatal care visit, development of medical conditions during

pregnancy and pregnancy outcomes (the birth status of the child, which is either a healthy

or unhealthy baby) in Lagos,

Nigeria?

The H0 developed to answer this question was:

There is no statistically significant predictive relationship between maternal health

literacy level, the number of antenatal visits, timing of the first antenatal care visit,

development of medical conditions during pregnancy, and pregnancy outcomes (the birth

status of the child, which is either a healthy or unhealthy baby) in Lagos, Nigeria (H0: r =

0).

I used descriptive statistics and logistic regression to analyze the data. As noted

by Mertler and Vannatta (2013), the appropriate measure of the degree of relationship

between two or more variables is correlation and/or regression analysis. Bivariate

statistics were used to examine the correlation between the variables. According to

Bhattacherjee (2012), the most common type of bivariate statistics used is bivariate

correlation, which is expressed as a number between -1 and +1. The strength of the

correlation was measured by how close the number is to -1 or +1. A positive correlation

implies that an increase in the independent variables by one unit would result in an

increase in the dependent variable, while a negative correlation implies an inverse

relationship, an increase in the independent variable would result in a decrease in the

dependent variable (Bhattacherjee ,2012). The null hypothesis would be rejected if the

correlations between the dependent and the independent variables are statistically

significant, the null hypothesis would be rejected if the p values were equal or less than

0.05 (p ≤ 0.05).

Research Design and Rationale

Quantitative, qualitative, and mixed methods are the three research designs

common in social sciences studies. The quantitative method is used to test a theory or

hypothesis and to operationalize variables derived from theory and a research question

that a researcher is attempting to solve (Creswell, 2009). The quantitative research

method applies survey design to observe and measure opinion, attitudes, and the

behavioral patterns of a population sample and then generalize the findings back to the

whole population so that inferences can be made about the characteristics, attitudes, or

behaviors of the population (Creswell, 2009). The type of survey designs common in

social sciences include self-administered surveys or questionnaires, interviews, structured

interviews, and structured observations (Adanri, 2016). Compared to other methods,

survey designs are cheaper and faster to collect, but they do not do a good of a job of

studying or understanding complex social phenomenon (Adanri, 2016).

. The strategy of inquiries that are common among qualitative studies include

narrative, phenomenology, ethnography, case study, and grounded theory. The

qualitative research method requires “learning the meanings that the participants hold

about the problem or issue, not the meaning that the researcher bring to the research or

writer express in literature” (Creswell, 2009, p. 175). While qualitative research method

provides opportunity to study complex social phenomenon, it is prone to researcher’s bias

and subjectivity.

The research method I selected for this study was a quantitative research

approach. I chose a quantitative research method based on its alignment with the

research question and purpose of the study. There are four types of quantitative research

designs: experimental research design, quasi-experimental research design, cross-

sectional research design, and longitudinal research design (Creswell, 2009). I used a

cross-sectional research design in this study because it allowed for the observation of

patterns or relationships between two or more variables.

This study was a cross-sectional quantitative study of women who had been

pregnant before but were not pregnant at the time of the study. Survey instruments were

administered to women 18 years and older who had been pregnant or had given birth in

Shomolu local government area in Lagos, Nigeria. The independent variables examined

in this study were maternal health literacy, frequency of antenatal visits, number of

antenatal visits, timing of the start of antenatal care, and problems developed during

pregnancy. The dependent variable was pregnancy outcomes and they were measured by

the birth status of the child, which was either a healthy or unhealthy baby. The level of

education and socio-economic characteristics of the women were assessed through the

survey. The education level of the women was assessed by asking them what grade they

completed and their socio-economic status was assessed by asking questions about their

annual income. Because of the nature of the design, I collected data face-to-face, and I

spent considerable time and resources on this project. The variables used in the study

were also analyzed using the Statistical Package for Social Sciences (SPSS), Version 21.

Variables

Independent variables. The independent variables were maternal health literacy,

frequency of antenatal visits, number of antenatal visits, timing of the start of antenatal

care, and problems developed during pregnancy. I assessed health literacy level by

asking questions from Dr. Chew’s Literacy Screening Questionnaire (Chew et al., 2004),

and antenatal care usage was measured by Kotelchuck’s APNCU Index (Koroukian &

Rimm, 2002). This was measured as adequate plus, adequate, intermediate, or inadequate

maternal care visits based on the month antenatal care began and the number of antenatal

visits.

Dependent variables. The dependent variable was pregnancy outcomes, which

was measured by the birth status of the child, either a healthy or unhealthy baby.

Covariates. The covariates I used in the study included socio-economic status of

family, number of previous pregnancies, religion, age, number of children, education

level, household income, and employment status.

Methodology

Population to Sample

According to Frankfort-Nachmias and Nachmias (2008), the population of a study

refers to the complete set of relevant units of analysis, while a population sample refers to

a subset of the population that is used to generalize back to the population. The target

population for this study was Nigerian women that had been pregnant before but were not

pregnant at the time of the survey. The sample population was drawn from women 18

years and over who had been pregnant before but were not pregnant at the time of the

study. Although women between the ages of 15 to 49 are considered to be in their

reproductive years (WHO, 2006), women 18 years and older were allowed to participate

due to their ability to consent to be in the study without parental permission. I used the

Shomolu local government area of Lagos State as the study area because it best

represented the various characteristics that constitute the independent variables in this

study. The Shomolu local government is a high-density residential area and is one of the

local government areas with the highest level of literacy in the population (Lagos State

government, 2011). The state government in 2011 showed that 87% of the participants in

a survey they conducted were literate in English and 75% were also literate in other

languages (Lagos State Government, 2011). Therefore, participants for this study needed

to be able to speak and read English.

Sampling and Sampling Procedures

There are five strategies that can be used to select a representative sample from a

population (Creswell, 2010). The strategies include simple random sampling, systematic

sampling, stratified sampling, cluster sampling, and stage sampling (Creswell, 2010). In

a simple random sampling, each member of a population has an equal chance of being

selected (Creswell, 2010). Systematic sampling, although similar to random sampling,

uses a sampling frame from which participants are selected at regular intervals (Creswell,

2010). Stratified sampling is similar to systematic sampling; the only difference is that

one or more characteristics of the sample frame can be altered at the same time ensure

that the same percentage of participants are selected from the population (Adanri, 2016).

I used a combination of systematic sampling and purposive sampling methods in

this study. Purposive sampling is a form of nonprobability sampling that allows

researchers to make decisions on who would be included in a study based on knowledge

of the study issue or capability or willingness to participate in the research (Jupp, 2008).

I purposively chose neighborhoods in the local government area because it was not

possible to have the address of all the women that had been pregnant in the local

government area or the streets they resided on. I sampled women who were available on

the streets and met the inclusion criteria. Systematic sampling was introduced in order to

ensure that the sample was a fair representation of the population studied. The sample

frame consisted of women who were 18 years and over, not pregnant at the time of the

study but who had been pregnant at least once, able to read and comprehend the English

language, and who resided in the Shomolu local government area of Lagos State, Nigeria.

The participants were systematically selected by alternating the side of the road from

which participants were selected and limiting the number of survey per each street block

to five.

I conducted the study in the Shomolu local government area where there are two

government health centers and 35 private hospitals (Lagos State Government, 2011).

Sampling units can be challenging in Nigeria because of the problems of incomplete

frames, clusters of elements, and foreign elements (Frankfort-Nachmias & Nachmias,

2008). It is likely that some women do not use the medical services when they are

pregnant or when they have the babies because of a lack of education or access to

medical services. If they do not have complications, these women may end up not going

to a hospital at all. For these reasons, women who had been pregnant before were

surveyed by me going door-to-door to various residential areas in the Shomolu local

government area of Lagos State using a purposive convenience sampling method to select

the streets and houses.

Sample Size

Sample size estimation and statistical power analyses are important because it

would not be ethically acceptable to conduct a study by recruiting thousands of

participants when sufficient data could be obtained with hundreds of participants instead

and research ethics committees often ask for justification of the study based on sample

size estimation and statistical power (Prajapati, Dunne, & Armstrong, 2010). According

to Bartlett, Kotrlik, and Higgins (2001), sample size is one of the four inter-related

features that can influence the detection of significant differences, relationships or

interactions in a research study. It is considered as one of the important features of a

survey research design (Bartlett et al., 2001). There are three steps involved in

determining the sample size required for a research study.

The first step is to determine the alpha which is the for Type 1 error. The higher

the alpha level the more likely for researcher to reject a true null hypothesis (type I error)

and the lower the alpha level the more likely the researcher will accept a false null

hypothesis, which will be a Type II error (Tomczak, Tomczak, Kleka, & Lew, 2014). In

most social science and behavioral studies, alpha is often set at 0.05. The second step is

to determine the effect size which is the limit for Type II errors and based on literature

review in the social science and behavioral studies is to set effect size as either small,

medium or large but the convention is to set it at 0.20 (Fritz, Morris, & Richler, 2011).

The third step in determining sample size is to determine the power. A commonly

recommended power is 0.80 according to Fritz et al. (2011). The sample size can then be

manually calculated or a computer software like G*Power can then be used to calculate

the sample size.

I used the G* Power calculator v. 3.1.9.2 to calculate the sample size taking into

consideration the purpose of the study, the research question, and the hypothesis to be

tested. The family of test selected for the study was t-test. Statistical test selected was

correlation: point biserial model. Type of power analysis was a priori: compute required

sample size given alpha, power and effect size. Type of tail(s) was two tails. The effect

size was set at .20 which is a medium effect size; alpha was set at 0.05 and power was set

at .80 which is typical in social science studies (Frits et al., 2011). The estimated sample

size computed by G*Power was 191. The sample size calculation is shown in Table 1.

Table 1

Sample Size Calculation: t tests

t tests – Correlation: Point biserial model

Analysis: A priori: Compute required sample size

Input: Tail(s) = Two

Effect size |ρ| = 0.20

α err prob = 0.05

Power (1-β err prob) = 0.

Output: Noncentrality parameter δ = 2.8210518

Critical t = 1.9725951

Df = 189

Total sample size = 1

Actual power = 0.80142

Procedures for Recruitment, Participation, and

Data Collection

The data collection relied on the survey instrument developed by Chew et al.

(2004) with permission to use granted by Dr. Chew. Permission to use the instrument

was sought and granted by Dr. Chew. The survey instrument can be found in Appendix A

and the permission letter in Appendix B. I did not modify the instrument and it was

approved by the research committee. Walden University’s approval number for this

study was IRB# 06-20-16-0290994. The approved instrument was used to collect social

and demographic data, level of education and antenatal visits during pregnancy. The data

provided reliable and practical information concerning the well-being and functional

health of a community or an individual from a patient or individual’s point of view

(Quality Metric, 2013, para. 1).

Surveys are usually cost effective in collecting information from a large number

of individuals who are most likely representative of the target audience (Cengage

Research Methods Workshops, 2005). Surveys can be useful when a researcher wants to

collect data from the general public that cannot be directly observed and the data would

be collected through the use of questionnaires. There was no controlled setting and this

made the setting less artificial. Survey data can be collected through the mail where the

survey is sent to participants by mail, filled out by participants, and mailed back to

researcher; face to face survey where participants are met in person by an interviewer, the

interviewer administers the survey and records the responses to survey questions; and

telephone survey where participants are contacted over the telephone by an interviewer,

the interviewer administers the survey and records the responses to survey questions

(Kansas State University, n.d.). Internet research methods usually allows researcher to

have access to rich data, especially for those working in sensitive areas because of

anonymity (Hine, 2011).

However, because the population this research was focused on may not have

access to the Internet, the use of Internet survey would likely not yield many responses.

In addition, in Nigeria the house addresses and mailing system is not as effective as we

have in developed countries (personal experience) so the use of mail survey has been

determined to not be an effective manner of surveying potential participants. Therefore,

a face-to-face self-administered or interviewer-administered survey was used in this study

because it ensured access to participants and assist in the completion of the surveys so the

needed sample size could be more easily attained. Although this method could reduce

nonresponse bias, it did not provide anonymity and responders may not have truthfully

answer sensitive questions (Cengage Research Methods Workshops, 2005). In order to

counteract this, I did not collect the names of participants and informed the participants

that their responses would remain confidential.

Procedures for Recruitment

Once approval was granted by Walden University Institutional Review Board

(IRB), I proceeded to collect the data from participants residing in the study area. Since

there is no bank of information about women and their pregnancies, I purposively chose

neighborhoods and streets in the study area in order to access women who may be

eligible to participate in the study. I sampled women I approached in public and/or live

on the street who met the inclusion criteria of 18 years and older, have been pregnant in

the past, could speak /read English, and not currently pregnant. This sampling method

was used due to the need to be able to approach women who may be eligible for

participation.

Because of the nature of this study, face-to-face interaction was used and the self-

administered surveys were distributed to the participants in their homes and on the

streets. If requested, I dropped off the survey and picked up the completed survey at a

later date and time that the participant and I found to be convenient. If there were no

women that fit the criteria (18 years and older, can speak/read English, and have been

pregnant before, not currently pregnant) at a particular house, I moved to the next house

on the street until the number of participants on the street have been satisfied. Each

participant was told the reason for the study and were asked to read the participant

informed consent. The participants were given the options to participate or not to

participate in the study. It is implied that by completing the survey or responding to

questions in the survey, the participants gave their consent. Completed surveys were

returned in sealed envelope that was provided and included in the survey packet to ensure

confidentiality and anonymity of the participants. As a token of appreciation, I included

a N100 phone card (an equivalent of $1.00) in the survey packet and this was for the

participants to keep whether they completed the survey or not. The survey packet

consisted of the survey instrument, a return envelope, and a N100 phone card.

Confidentiality of the completed instrument and participants was ensured by not

disclosing the result or participants and by ensuring I am the only one who have access to

the raw data. The data were entered straight into SPSS file and processed by me for

analysis.

Ethical Concerns

I followed the standard practices and ethical compliance checklist of the APA

Manual, Sixth Edition, the Walden University IRB and protecting the confidentiality of

the research participants. I also emphasized to the participation in the study was

voluntary and that there will be no retribution for refusal to participate. The participants

were assured of the confidentiality of the study; the survey did not identify the

participants by name of by code. Although I was present when many of the surveys were

completed by the participants, I offered no assistance nor did I influence the participants

in their response to the survey. The survey was free of personal bias. I collected the

completed surveys in person and personally processed the data; no other person or

persons had access to the data and only aggregate results were used in the reporting.

Instrumentation and Operationalization of Constructs

Instruments

Health Literacy Screening Questions (Chew, Bradley, & Boyko, 2004).

To measure the health literacy level of the participants in this study, the Health

Literacy Screening Questions developed by Chew et al. (2004) was used. Chew et al.

(2004) developed the health literacy screening questions to assess people’s ability to

comprehend information or to perform tasks usually encountered in health care setting.

This instrument consists of 16 questions that use a 5-point Likert scale. Out of the sixteen

questions asked, three questions were found to be effective in detecting inadequate health

literacy. In the following questions “How often do you have someone help you read

hospital materials?” “How confident are you filling out medical forms by yourself?” and

“How often do you have problems learning about your medical condition because of

difficulty understanding written information?” the area under the receiver operating

characteristic curve are 0.87, 0.80, and 0.76, respectively. Other researchers have used

these three questions in comparison to Short Test of Functional Health Literacy in Adult

(S-TOFHLA) to assess literacy skills. Schwurtz et al. (2013) used these items and the

area under receiver operating characteristic curve (AUROC) for all the questions was

0.90 (95% CI, 0.85-0.95). Wallace et al. (2006), in order to evaluate the three screening

questions by Chew et al. (2004), compared the questions to the Rapid Estimate of Adult

Literacy in Medicine (REALM) and also found that the questions were effective at

identifying individuals with marginal or limited health literacy skills. Ohl et al. (2010)

also used the screening questions to identify people with low health literacy and accuracy

of provider perception in two HIV primary care clinics. For this study, all 16 questions

from the Health Literacy screening tool were used in assessing health literacy in the study

population. Permission to use the instrument was sought and granted by Dr. Chew. The

survey instrument can be found in Appendix A and the permission letter in Appendix B.

Operationalization of Variables

Based on American Congress of Obstetricians and Gynecologists (ACOG)

recommendations, antenatal care was analyzed in terms of frequency, which would be

coded as 1, 2–3, or 4 visits or more, timing of the visit, whether antenatal care was

initiated in the first, second, or third trimester of pregnancy. To understand if women

who were pregnant received antenatal care, it is important to know how many of these

women really registered for antenatal care. This was assessed by a dichotomous variable:

0-did not receive antenatal care; 1- received antenatal care. According to the United

Nations Population Fund (UNFPA; n.d. a), there should be at least four antenatal

visits

early in the pregnancy to educate women, identify and manage either current or potential

pregnancy risks, and develop a birth plan on how to reach a medical care in case of

emergency. UNFPA (n.d. b) also recognized this visit could be the pregnant woman’s

first visit, therefore some health educational activities on sexual and reproductive health,

family planning, birth spacing, and care of the newborn should be incorporated into these

visits (UNFPA, n.d. a). In assessing antenatal messages received, the women in this

study were asked whether or not they received information on these topics during their

visits. Maternal education level was measured by the number of years of formal

education received by the mother. For this study, education level was divided into six

levels as follows: 1- No education; 2-Elementary education (1-6th grade): 3. Junior

Secondary School (6-9th grade); 4- Secondary School (Complete 12th grade); 5- Post-

secondary (some College); and 6- College degree and above. Table 2 consists of the

operationalization of the study variables.

Table 2

Operationalization of Study Variables

Variable name Variable

Source

Variable Type Value Labels Level of

Measurement

Pregnancy Outcome RQ, SQ26 Dependent 0- Unhealthy

baby

1- Healthy baby

Nominal

Health

Literacy

Number of Antenatal

Care Visits

Timing of Antennal

Care

Survey Q11

Survey Q12

Survey Q13

RQ,

Survey

Q22

RQ, Survey

Q20

Survey Q25

Independent

1-Always

2-Often

3-Sometimes

4-Occasionally

5-Never

1-Always
2-Often
3-Sometimes
4-Occasionally
5-Never

0–3 visits

4–6 visits

7–9 visits

0=0–3

months

Ordinal

Categorical

Medical Problem

developed during
pregnancy

Independent

1=4–6 months

2=later than 6

months

0-Gestational

diabetes

1-Depression

2-Heart

problem

3-High blood

pressure

4-Bleeding

5-Other health

issues

Categorical

Covariates that were included in this study were: maternal age,

employment

status, income, and marital status; descriptive statistics would be used for these maternal

characteristics.

Table 3

Descriptive Variables

Variable name Variable
Source

Variable Type Value Labels Level of
Measurement

Last time pregnant?

Survey Q

Descriptive 0= 1–11 months

ago

1=12–23 months

ago

2= 24–35

months ago

3=36 + Months

ago

Interval

Received antenatal

care?

Survey Q
18

Descriptive

0- No
1-

Yes

Dichotomous

Likely reasons for not

getting antenatal care

Survey Q
19

Descriptive 0- I didn’t know

I was pregnant

1- I didn’t know

where to go

Categorical

2-I didn’t have

enough money

to pay for my

visits

3- I didn’t want

anybody to

know I was

pregnant

4- I’ve given

birth before and

I didn’t feel I

needed antenatal

care

5- I had no way

of getting to a

medical center

6- It is against

my religion

7- Other

reasons?

Information received

Frequency of antenatal

visit

Survey

Q23

Survey Q
21

Descriptive

1- Health

educational

materials on

sexual and

reproductive

health

2- Family

planning and

birth spacing

information

3- Care of the

newborn

4- What to

expect during

pregnancy

5- No

information

0-Weekly

1- Every two

weeks

2- Once per

month

3- Every few

months

4- Only when

there was a

Categorical

Ratio

How many Visits

Marital status

Number of children

Number of pregnancy?

Employment status

Religion

Age

Education Level

Survey
Q22

Survey

Q28

Survey

Q34

Survey Q
33

Survey

Q30

Survey

Q33

Survey

Q27

Survey Q
29

Descriptive

Demographic

Descriptive

problem

1- 1

Visit

2- 2 Visits

3- 3 Visits

4- 4 Visits

5- 5 Visits

6- 6+Visits

0- Dating

1- Married

2-Separated

3-Divorce

4- Widowed

1–15

0-No

employment

outside the home

1-Part

employment

2- Full time

employment
outside the home

1- Christianity

2- Islamic

3- Traditional

4- No religion

1-15–20 years

2-21–25 years

3-26–30 years

4-31–35 years

5-36–40 years

6-40–50 years

1- No formal

education

2- Primary

school

Ratio

Nominal

Ratio

Ordinal

Nominal

Interval

Nominal

Household Income

Survey
Q34

Descriptive

3- Junior

Secondary

School

4- Secondary

School

certificate

5- Ordinary

diploma or

college of

education

certificate i.e.

OND, NCE

6- College

degree or above

1- Less than

N25,000

2- N26,000–

N50,000

3- N51,000–

N75,000

4- More than

N75,000

Interval

(table continues)

Data Quality and Potential Bias by Mode of Survey Administration

According to Bowling (2005), data quality can be measured by response rate and

the accuracy of the responses, absence of bias, and the completeness of information

obtained from respondents. Bowling (2005) noted that a low response rate could affect

the reliability of the survey and result in study bias which could weaken the external

validity. Some of the factors that could affect data quality include impersonality,

cognitive burden, and the order of the questionnaire items and the way the participants

respond to the questions (Bowling, 2005). Aside from answering generic questions

which the participants might have had, I did not provide any assistance or attempt to

influence the decision of the participants. The participants did not appear to experience

or suffer any embarrassment which could affect their response to any of the questions.

Although some methods of survey create cognitive burden on the participants, face-to-

face or drop off and pick are less burdensome. The participants voluntarily completed

the survey; they were motivated to be part of the solution to a national problem as a result

the legitimacy of the study was established. The participants were in control of the

questionnaire; I did not offer any help other than answer questions that are generic to the

survey. Therefore, I am confident that my personal bias or data collection method do not

affect the quality of the data.

Data Analysis Plan

The data analysis included screening the data for errors and missing data. As

errors could result from missing values or incorrect coding, one of the ways this study

guarded against this was by replacing missing values. I manually replaced missing data

based on the mean or median value from frequency. The plan also included testing the

data for parametric test assumptions. Descriptive statistics was conducted to describe the

sample and population demographics used in the study, this include frequency, mean,

median, standard deviation (measures of central tendency), and measure of variability.

SPSS version 21was used to analyze data. A variety of statistical methods were used to

test the hypotheses based on the characteristics of the dependent and independent

variables. Descriptive statistics were used to explore frequency distributions and

percentages for nominal and ordinal level. Output tables were created to reflect the

number of participants, percentages, cumulative frequencies, percentages for maternal

characteristics. Descriptive analyses were used for Questions 18-20 in order to determine

the percentage of the participants who received antenatal care and if they encountered

any problems receiving antenatal care. Question 22 helped to identify the kind of health

conditions women developed during their last pregnancy and question seven was

assessed to determine the kind of information the women received during their antenatal

care.

Binary logistic regression was used to assess the predictive relationship between

maternal literacy level and maternal pregnancy outcomes. A multivariate analysis was

performed to test if there is correlation between the dependent and the independent

variables. Regression was used to estimate the influence of maternal characteristics on

the odds of pregnancy outcomes. The relationship between maternal health literacy

levels was computed, using SPSS 21, to determine if there was a correlation to pregnancy

outcomes. The dependent variable was dichotomized into 0 and 1 where 0 meant

unhealthy baby and 1 denoted healthy baby. The null hypothesis was rejected if the p

value was less than or equal to .05 (P ≤ .05). Pregnancy outcomes were measured by

birth status of child, which were healthy or unhealthy baby.

Research Question: What is the predictive relationship between maternal health

literacy levels, the number of antenatal visits, timing of the first antenatal care

visit, development of medical conditions during pregnancy and pregnancy

outcomes (birth status of child, which are healthy or unhealthy baby) in Lagos,

Nigeria?

H0: There was no statistically significant predictive relationship between

maternal health literacy level, the number of antenatal visits, timing of the first antenatal
care visit, development of medical conditions during pregnancy and pregnancy outcomes

(birth status of child, which are healthy or unhealthy baby) in Lagos, Nigeria.

HA: There was a statistically significant predictive relationship between maternal

health literacy level, the number of antenatal visits, timing of the first antenatal

care visit, development of medical conditions during pregnancy and pregnancy

outcomes (birth status of child, which are healthy or unhealthy baby) in Lagos,
Nigeria.

Threats to Validity

In any research study, there are certain factors that can affect the reliability and

validity of a study. Threats to internal validity include selection, history, maturation,

mortality, testing, regression, and compensatory demoralization (Creswell, 2009).

According to Creswell (2010), selecting participants with certain characteristics which

enables them to have certain outcomes will affect the outcome of the research. Also,

history, which is events that occur during the research that can affect the outcome of the

research, can also threaten the validity if some unanticipated events occur that affect the

outcome of the research (Creswell, 2010). As a result of normal development,

participants may mature during the research and this maturation may affect how they

respond overtime (Creswell, 2010). Experimental mortality occurs when participants

drop out of the research and this has the potential of affecting the study since the result of

those who dropped out would not be known (Creswell, 2010). Creswell (2010) also

noted that it is possible for participants to become sensitized to the measurement

instruments during pretest and their post-test response result may be due to their exposure

to the outcome measurement earlier. Statistical regression occurs when participants are

selected as a result of their extremes cores (Creswell, 2010). Compensatory

demoralization occurs when one group receives the treatment and the other group does

not; this could be demoralizing and could lead to resentment as well (Creswell, 2010).

The threats to validity anticipated in this study are incorrect recollection, response

bias, and measurement bias. Since I was asking participants to think back to their

experience, it is possible that some participants may not recollect the information clearly

and give false information. The study included only women who have been pregnant

before and who can speak/read English and or understand spoken English in order to

assess their literacy level; women who do not speak or understand English are excluded

from the survey.

Response bias was another type of bias anticipated in this study. Participants may

be providing answers they think I was looking for and may not answer truthfully or they

could be guessing on the answers. Measurement bias was also anticipated. This is

because participants were asked to read and respond to certain questions in the literacy

instrument. In order not to appear illiterate, the respondent might have made up the

answers. This study randomly recruited participants from Shomolu local government of

Lagos state. Since this study was specifically aimed at women in Lagos state the study

may not be generalized to other parts of Nigeria (Frankfort-Nachmias & Nachmias,

2008), though data may be compared to look for similarities and areas needing

improvement. To guide against response bias, I explained to the participants that the

outcome would not affect them directly but the information could help other women in

their pregnancy.

Ethical Procedures

The study approval was sought from Walden University IRB to ensure

compliance with the Institution’s ethical guidelines. The study ensured that individuals

within the protected class were excluded from the study this include anyone under the age

of 18 and any woman that was pregnant at the time of the survey. The study also ensured

confidentiality of the participants by not identifying them by name or coding system.

The participants were informed that by voluntarily complete the survey or respond to the

survey questions, they have given their informed consent therefore the participants were

not required to sign any document. Participants were offered a one hundred Naira

(N100.00) phone card which is an equivalent of one dollars ($1.00) as a token of my

appreciation for the time taken to complete the survey. Since the survey was

administered face to face (paper and pencil) it was important to ensure that no individual

name was used on any documents in order to maintain confidentiality. Informed consent

was shared with the participant and completion of the survey constituted granting

informed consent.

I was aware that participants cannot be forced to participate, so there was a need

to let the individuals know that their participation was entirely voluntary and that their

refusal would not in any way have a negative impact on them. They could refuse to

participate without explanation and participants were free to withdraw from the study at

any point in time (Crosby, DiClemente, & Salazar, 2006). Also, since the participants

were likely to be from different educational backgrounds, it was my responsibility to

ensure that oral and written directions were communicated in such a way that the

participants could understand (Crosby et al., 2006). Participants were adults (18 years

and older) and not pregnant so they were not classified as a protected group.

The data collected for this study were used solely for this research and the identity

of the participants was not released to any organization. To that extent, I entered survey

data collected into a password protected laptop and the original questionnaire collected

was kept in a locked safe in my home. The completed questionnaires will be shredded

five years after the study has been completed. I was the only one to have access to the

data collected. The study was conducted for academic purposes therefore no conflict of

interest was declared. The time and resources that were used for this research study were

at my expense in order to fulfill academic requirements.

Summary

In this cross-sectional research study, I examined the relationship between

pregnancy outcomes (dependent variable) and maternal health literacy, frequency of

antenatal visits, number of antenatal visit, timing of the start of antenatal care and

problems developed during pregnancy (independent variables). The dependent variables

were measured by birth status of child (healthy or unhealthy baby). An existing

instrument that was slightly modified and approved by the researcher’s dissertation

committee was used for the data collection. The survey participants were systematically

purposively selected. Any woman that was under the age of 18 or pregnant at the time of

the survey was excluded from the survey. Women that do not speak or comprehend

English language were also excluded from the survey. I used drop-and-pick as well as

face-to-face interview methods to administer the survey. In chapter 3, I described the

choice of research design and method that guide the study. The choice of research design

was based on research question and research purpose. Quantitative research design lends

itself to questions that explore the relationship between two or more variables and allows

for the generalization of the research findings to the population that was being studied. In

chapter 3 I also provided information on the how the questionnaire was administered and

the plan for the data analysis. In chapter 4 I described how I processed and analyzed the

and I also included the method of analysis and findings from the study.

Chapter 4: Results

Introduction

The purpose of this cross-sectional research study was to determine if there is a

relationship between maternal health literacy, frequency of antenatal visits, number of

antenatal visit, timing of the start of antenatal care, and problems developed during

pregnancy (independent variables) and pregnancy outcomes (dependent variable) in

Lagos, Nigeria. I measured pregnancy outcomes by the birth status of the child (which

was either a healthy or unhealthy baby). A review of the literature did not reveal any

previous researchers who had examined the relationship between maternal characteristics

and use of antenatal care on maternal pregnancy outcomes or any study carried out in

Nigeria that examined the relationship between maternal literacy level and pregnancy

outcomes using a health literacy instrument. Therefore, the results of this study could be

used to fill the identified research gap of the relationship between health literacy

(measured with a validated tool), maternal characteristics, utilization of antenatal care,

and potential pregnancy outcomes in Nigeria.

The following research question and hypotheses were formulated to guide this

study:

Research Question: What is the predictive relationship between maternal health
literacy levels, the number of antenatal visits, timing of the first antenatal care

visit, development of medical conditions during pregnancy, and pregnancy

outcomes (the birth status of the child, which is either a healthy or unhealthy

baby) in Lagos, Nigeria?

H0: There is no statistically significant predictive relationship between maternal
health literacy level, the number of antenatal visits, timing of the first antenatal

care visit, development of medical conditions during pregnancy, and pregnancy

outcomes (the birth status of the child, which was either a healthy or unhealthy

baby) in Lagos, Nigeria.

HA: There is a statistically significant predictive relationship between maternal
health literacy level, the number of antenatal visits, timing of the first antenatal
care visit, development of medical conditions during pregnancy, and pregnancy
outcomes (the birth status of the child, which was either a healthy or unhealthy

baby) in Lagos, Nigeria.

In this chapter, I will review the methods for data collection, present the

descriptive statistics of the dependent and independent variables, as well as the study

covariates. I will also discuss the issues I identified, the inferential statistics analysis and

hypothesis testing for the research question, the results of the statistical analyses, and

provide a summary of the answers to the research questions.

Data Collection

Following the approval of my research proposal by the Walden University IRB, I

initiated data collection on June 24, 2016 in Bariga, Shomolu local government area in

Lagos State, Nigeria and ended data collection on July 15, 2016. I used a combination of

traditional face-to-face data collection as well as drop-off and pickup methods to

administer the questionnaire. The questionnaires were administered to 150 participants.

Since a purposive convenience sampling method was used, I was able to go into homes

and shops as well as survey who I encountered on the street who met the recruitment

criteria. Women under the age of 18 years old and women that were pregnant at the time

of the survey were excluded from the study. When the women were busy, they asked me

to provide them the questionnaire and pick it up later in the evening or the following

morning. I made sure to explain the reason for the study and read the consent form to the

women before I left. I was able to go to different neighborhoods in the local government

area and speak with different groups of women. There were no requirements that the

participants sign an informed consent form as they were advised that by completing the

survey or responding to the survey questions they implied informed consent. The

participants were assured of the confidentiality and anonymity of the study, and I was

careful in the way the questions were asked in the face-to-face survey to ensure that the

participants were not influenced in their responses.

As I noted in Chapter 3, the sample size calculated necessary to meet the power,

effect, and alpha sizes for this study was 191, but because of time constraints and other

logistic reasons, I only administered 150 surveys and collected 130 completed surveys.

Therefore, a post hoc power analysis involving estimating the effect size based on the

obtained study data was conducted. This analysis took into account the sample size, alpha

level, two-tailed nature of the statistical test, and .20 effect size to derive the power

(Volker, 2006). The effect size was 0.2 and power was 0.64. The result of the post hoc

analysis is displayed in Table 4. This calculation shows 64% power to detect a

statistically significant association between an outcome variable and predictor variable

using regression techniques. The return rate was 86.7%, which is greater than the

between 32% and 50% return rate in most behavioral science research (Adanri, 2016;

Baruch & Holtom, 2008; Cycyota & Harrison, 2006).

Table 4

Sample Size Calculation: t tests
t tests – Correlation: Point biserial model

Analysis: Post hoc: Compute achieved power

Input: Tail(s) = Two
Effect size |ρ| = 0.20
α err prob = 0.05

Power (1-β err prob) = 0.80

Output: Noncentrality parameter δ = 2.3273733

Critical t = 1.9786708

Df =

128

Total sample size =

130

Actual power = 0.64

Data Preparation and Screening

Missing data. According to Bhattacherjee (2012), it is inevitable to have missing

data in any empirical data set because participants may not answer some questions for

any reason and a researcher needs to detect and correct this before data analysis.

Therefore, before I started my data analysis, I screened the data for accuracy to ensure

my data were imputed correctly, to determine what to do with missing data, to look for

outliers, and to test if assumptions of normality were violated. The data showed that

missing values existed, but there was no pattern to the organization of missing values.

Almost all (99.10%) of the values were completed data, while 0.9% of the data had

incomplete values. I chose to replace the missing values using the mean value of the

variables instead of removing the cases from my analysis. Mertler and Vannatta (2013)

suggested calculating the mean of the missing value and replacing the missing values

before starting the analysis. I ran descriptive statistics in SPSS to get the mean value in

each of the cases and manually inputted the missing values. According to Mertler and

Vannata, the mean value is the best estimate of missing data when there is no other

information available. Using the mean value does not change the overall mean of the

data or affect the variance because the number of missing values in the data set were few

(Mertler & Vannatta,2013).

However, not all missing data required values to be assigned. I excluded two of

the questions with follow-up questions from the missing value analysis because the

questions were expected to have missing values as they were not applicable to everyone.

One of the questions was “Did you develop any health problems during your

pregnancy?” and another asked “Did you get any antenatal care during your pregnancy?”

If the participants did not have any issues during their pregnancy, the follow-up questions

about the problems were not applicable.

Grouping of variables and values. The first 16 items on the instrument I used in

this study consisted of Likert-type items developed by Chew et al. (2004) to measure

health literacy. The remaining 18 items or questions in the instrument were developed to

capture data necessary to answer the research question. Likert-type items are single

questions that use some of the aspects of the Likert response alternatives, while a Likert-

scale consists of a summation of multiple Likert-type items into a single composite score

to create a new variable during the data analysis process (Boone & Boone, 2012). I used

an ordinal level scale as the scale of measurements in this study. I tested the predictor

variables for normality, linearity, and homoscedasticity assumptions, and the result

showed that the data were not normally distributed. Frequency analyses were performed

to check for skewness and kurtosis. According to Mertler and Vannatta (2013) and Field

(2009), skewness and kurtosis values should be between -1 and +1. The closer to zero

the skewness and kurtosis, the more the data are said to be normally distributed (Mertler

& Vannatta, 2013). In this study, the skewness values ranged from .04 and 1.529, while

the values of the kurtosis ranged between -.038 and 3.873. The skewness and kurtosis in

the data are far from zero which indicates the data were not normally distributed.

Therefore, I used nonparametric tests to analyze the data.

Results

In this section, I will focus on discussing the results of the statistical analyses, and

it be divided by the type of analyses I conducted. Descriptive statistics were used to

analyze the demographics of study participants in this study. I will describe the

demographic variables and correlational analysis. I will also discuss the binomial

regression analysis I conducted.

Demographic Variables

This study included 130 participants. I used SPSS, Version 21.0 to generate

demographic percentages, frequencies, means, standard deviation, and kurtosis.

Table 5

summarizes the descriptive analysis and characteristics of the study participants.

Table 5

Survey Participants’ Demographics (N = 130)

Variables

Frequency

Percentage

Marital Status

Single 11 8.5

Ongoing dating relationship with father 16 12.3

Married 93 71.5

Separated 5 3.8

Divorced 4 3.1

Widowed 1 0.8

Education Level

No formal education 12 9.2

Primary/elementary school 6 4.6

Junior secondary school 6 4.6

Secondary school certificate 27 20.8

Ordinary diploma or college of education certificate 30 23.1

College degree/advanced degree 49 37.7

Religion

No religion 1 0.8

Christianity 94 72.3

Islam 31 23.8

Traditional 4 3.1

Employment Status

No employment outside home 37 28.5

Part time employment outside of home 38 29.2

Full time employment outside of home 55 42.3

Household Income

Less than N25,000 42 32.3

N26,000–N50,000 42 32.3

N51,000–N75,000 18 13.8

More than N75,000 28 21.5

Developed Health Problems

No 102 78.5

Yes 28 21.5

Type of Health Problem Developed

Gestational diabetes 5 3.8

Depression 5 3.8

Heart problem 3 2.3

High blood pressure 12 9.2

Bleeding 6 4.6

Information Received

Health education material on sexual/reproductive health 14 10.8

Family planning and birth spacing 11 8.5

Care of newborn 21 16.2

What to expect during pregnancy 28 21.5

More than one choice 44 21.5

No information 12 9.2

Outcome of Pregnancy

Healthy baby 109 83.8

Unhealthy baby 21 16.2

(table continues)

Overall, respondents were between age 18 and 42 years and the mean age was 30

years. 71.5% of respondents were married, 8.5% were single. Almost half of the

respondents (42%) were employed outside the home and 29.2% of respondents had part

time employment. A majority of respondents (72.3%) were Christians, 23.8% were

Muslims and in terms of education, 37.7% of respondents had a college degree or above,

23.1% had a diploma or college of education certificate, 20.8% had a secondary school

certificate. The household income for the participants ranged from less than N25,000 to

more than N75,000. Close to a quarter of the participants said they developed some sort

of medical problems during their last pregnancy (n = 28) and majority of the participants

received antenatal information during their last pregnancy, and only 9.2% of the

participants stated they did not revive any information. Only 10% of the participants did

not receive antenatal care and 40% of these participants who did not receive antenatal

care said they did not know they were pregnant.

Descriptive Statistics for Covariates and Pregnancy Outcomes

I analyzed the associations between pregnancy outcomes with the covariates of

marital status, employment status, education level, income level, religion, and number of

pregnancies using cross tabulation. As displayed in Table 6, out of the 109 healthy baby

outcomes, participants who were married reported the most positive outcomes (n = 85).

Table 6

Marital Status and Pregnancy Outcomes

Marital Status Healthy

Baby

Unhealthy

baby

Total

Single (Never married)

5 (3.8%)

6 (4.6%)

11 (8.5%)

On-going Dating

Relationship with Father

of the Child

15 (11.5%) 1 (.8%) 16 (12.3%)

Married 85 (65%) 8 (6.15) 93 (71.5%)

Separated 1 (.8%) 4 (3.1) 5 (3.8)

Divorced 2 (1.53) 2 (1.53) 4 (3.1)

Total 109 (83.85) 21 (16.15) 130 (100%)

Income also seemed to have positive effect on pregnancy outcomes. About three fourths

of the participants who are somewhat employed reported positive birth outcomes as can

be seen in Table 7.

Table 7

Employment Status and Pregnancy Outcomes

Employment Status Healthy

Baby
Unhealthy
baby

Total

No Employment

29 (22.3%)

8 (6.15)

37 (28.46%)

Part-Time Employment 34 (26.15%) 4 (3.1) 38 (29.23)

Full-Time Employment 46 (35.38%) 9 (6.92%) 55 (42.31%)

Total 109 (83.85) 21 (16.15) 130 (100%)

Level of education and birth outcomes shows an interesting dynamic. While it

would be expected that higher education would serve as a protective factor against

negative birth outcomes, the result did not indicate that. Participants with little or no

education reported fewer negative birth outcomes compared to participants with higher

education. Table 8 shows that almost half of the participants with senior secondary

school education reported unhealthy birth outcomes, while participants with elementary

education did not report any unhealthy birth outcomes.

Table 8

Level of Education and Pregnancy Outcomes

Level of Education Healthy

Baby
Unhealthy
baby
Total

No Formal Education

11(8.5%)

1 (.8%)

12 (9.2%)

Primary School 6 (4.61%) 0 6 (4.61%)

Junior Secondary School 4 (3.1) 2 (1.53 6 (4.61%)

Secondary School

Certificate

19 (14.61%) 8 (6.15) 27 (20.76)

Ordinary Diploma

(OND) or National

Certificate of Education

(NCE)

26 (20%) 4 (3.1) 30 (23.07%)

College Degree or

Above

43 (33.07%) 6 (4.61%) 49 (37.69)

Total 109 (83.85) 21 (16.15) 130 (100%)

As would be expected, participants with more income had better

pregnancy outcomes.

Participants earning between N25,000–N50,000 reported more negative birth outcomes

when compared to participants earning between N51,000 and over N75,000. This is

reported in table 9.

Table 9

Household Income and Pregnancy Outcomes

Household Income Healthy

Baby
Unhealthy
baby
Total

Less than N25,000

35 (26.92%)

7 (5.38%)

42 (32.30%)

N26,000–N50,000 33 (25.38%) 9 (6.92%) 42 (32.30%)

N51,000–N75,000 17 (13.07%) 1 (.8%) 18 (13.84%)

More than N75,000 24 (18.46%) 4 (3.1%) 28 (21.53%)

Total 109 (83.85) 21 (16.15) 130 (100%)

Table 10 shows that participants for this study were mainly Christians, and out of 94

Christians, 81 reported positive birth outcomes. Out of the 31 Muslim participants, 25

reported positive birth outcomes. Religion does not seem to be related to pregnancy

outcomes.

Table 10

Religion and Pregnancy Outcomes

Religion Healthy

Baby
Unhealthy
baby
Total

No Religion

1 (.8%)

Christianity 81 (62.31%) 13 (10%) 94 (72.31%)

Islamic 25 (19.23%) 6 (4.61%) 31 (23.84%)

Traditional 3 (2.31%) 1 (.8%) 4 (3.1%)

Total 109 (83.85) 21 (16.15) 130 (100%)

In table 11, participants who have had more than one pregnancy reported fewer unhealthy

babies when compared to participants who reported one pregnancy.

Table 11

Number of Pregnancy and Pregnancy Outcomes

No. of

Pregnancy

Healthy Baby Unhealthy

baby
Total

25 (19.23%)

7 (5.38%)

32 (24.61%)

2 18 (13.84%) 3 (2.31%) 21 (16.15%)

3 39 (30%) 3 (2.31%) 45 (34.61%)

4 24 (18.46%) 4 (3.1%) 28 (21.53%)

5 14 (10.77%) 4 (3.1%) 18 (13.84%)

6 4 (3.1%) 1 (.8%) 5 (3.85%)

7 4 (3.1%) 0 4 (3.1%)

Total 109 (83.85) 21 (16.15) 130 (100%)

As noted in table 12, only 11 respondents (9.2%) reported having more than four children

and out of this number, only one (.8%) reported an unhealthy baby as their pregnancy

outcome.

Table 12

Number of Children and Pregnancy Outcomes

No. of Children Healthy

Baby
Unhealthy
baby
Total

26 (20%)

7 (5.38%)

33 (25.38%)

2 23 (17.69%) 4 (3.1%) 27 (20.72%)

3 35 (26.92%) 6 (4.61%) 41 (31.53%)

4 14 (10.77%) 3 (2.31%) 17 (13.07%)

5 8 (6.15%) 1 (0.8%) 9 (6.92%)

6 1 (.8%) 0 1 (.8%)

7 2 (1.53) 0 2 (1.53)

Total 109 (83.85) 21 (16.15) 130 (100%)

Correlational Analyses

Using a bivariate correlational analysis, the relationship between maternal health

literacy, number of antenatal visits, timing of first antenatal care visit, problems during

pregnancy, and pregnancy outcomes were examined to determine if any of the variables

were too highly correlated to leave in the regression analyses due to multicollinearity.

The analysis showed no multicollinearity in the variables so all the variables were

included in the analysis. To assess the intervariable relationship, correlation analysis was

used to examine the relationship between the dependent and independent variables. Table

13 shows the results of the analysis. The convention in behavioral studies such as this is

to interpret correlation value of .10 to .29 as small correlation; .30 to .49 as medium

correlation and .50 to 1.0 as large correlation (Green & Salkind, 2011).

Table 13

Correlation Coefficient for Dependent and Independent Variables

Develop

ment of
health
problems
Health
Literacy

Frequency

of Visit

Number

of
antenatal
Visit

Timing

of first

antenatal
visit

Pregnancy
outcomes

Development of

health problems

1.000

Health Literacy .005 1.000

Frequency of

Visit

-.016 -.032 1.000

Number of

antenatal Visit

.107 .224* -.380** 1.000

Timing of first

antenatal visit

.050 .035 .081 -.106 1.000

Pregnancy
outcomes

-.177* .097 -.204* .145 -.111 1.000

Note. *Correlations are statistically significant at the .05 level (2-tailed)

**Correlations are statistically significant at the .01 level (2-tailed)

Also, the result from this study shows small to medium correlations among the

variables. The result shows that there was a medium correlation between the level of

education and number of antenatal visits, r = 0.27, p = .002 (2-tailed); education also

correlated with development of medical condition, r = 0.18; timing of first antenatal care

visit, r = -0.23; employment status, r = 0.48 and household income, r = 0.64 (all with p <

0.05). There was statistically significant relationship between pregnancy outcomes and

frequency of antenatal care visit (r = -0.219, p < 0.005. There is a medium correlation

and statistically significant relationship in number of antenatal visits and the frequency of

the visit, r =.0.38, p = .000; there is also a small correlation but statistically significant

relationship between number of antenatal visits made and health literacy (r = .20, p <.05).

This indicates that women with better health literacy skills make more antenatal visits.

This could be because of awareness of the need to have medical checkup or knowledge of

the importance of medical examination during pregnancy. There is also a small

correlation between pregnancy outcomes and development of medical problems during

pregnancy but a statistically relationship, r = -.18, p = .04. This could be because when

pregnant women go for antenatal visit their medical conditions are noted and monitored.

This will enable the women utilizing health serviced have better pregnancy outcomes if

they any develop medical problems. Using bivariate correlation coefficient, maternal

health literacy, number of antenatal visit, and timing of the start of antenatal care are not

statistically significant. This implies that there may be other actors affecting pregnancy

outcome beyond health literacy, antenatal care, and timing of the start of care

Binomial Logistic Regression

I used logistic regression to assess how well the independent variables predicted

the dependent variable. The variables used in the analysis were pregnancy outcome

(dependent variable), as measured by whether a mother delivered either a healthy or

unhealthy baby. The independent variables were maternal health literacy, frequency of

antenatal visits, number of antenatal visit, timing of the start of antenatal care and

problems developed during pregnancy. To start the analysis, I collapsed cells and recoded

the variables but there was no multicollinearity that would negatively impact the analysis

so all the variables were included.

Using a stepwise logistic regression approach, all independent and dependent

variables were entered into the equation one at a time. Frequency of visits was entered

first because it has the highest correlation. This was followed by problem develop during

pregnancy, number of visits, timing of the first visits and health literacy. During this

process, the best regression equation decreased the variables accordingly to improve the

logistic regression model. There were five steps created and there were no big differences

in the results from the steps. At the first step, when frequency of visit was entered, the

model explained 84% of the results; when development of medical problems was entered

in the second block, 85% of the model was correctly presented; when antenatal visit was

entered in block 3 85% of the results was predicted; at block 4 when timing of visit was

removed, 85% was correctly classified and by the fifth block when health literacy was

entered, 85% of the cases were correctly classified. The five models completed for the

research question did not show any significant difference. Table 14 shows how the

models were created.

Table 14

Models for Binary Regression

Blocks

Variables included

1 Frequency of visit

1, 2 Frequency of visit, Develop problem

1,2,3

Frequency of visit, Develop problem,

Number of antenatal visit

1,2,3,4 Frequency of visit, Develop problem,

Number of antenatal visit, Timing of

visit

1,2,3,4,5 Frequency of visit, Develop problem,

Number of antenatal visit, Timing of

visit, Health Literacy

The results presented in Table 15 shows that the omnibus goodness-of-fit test for

the regression model was statistically significant and reliable in distinguishing or

differentiating pregnancy outcome which is whether a healthy or unhealthy baby is

delivered (χ2 = 13.23; p = .040).

Table 15

Omnibus Test of Model Coefficient

Chi-square

Sig.

Step 2.699 2 .259

Model 2.699 2 .259

Block 13.229 6 0.040

The classification table as can be seen in Table 16 shows that 18 cases are classified as

having unhealthy baby and 108 cases are classified as having healthy baby, implying that

the model correctly classified about 85.4% of the cases.

Table 16

Classification Table for Pregnancy Outcome

Observed Pregnancy Outcome

Predicted

Pregnancy

Outcome

(Healthy Baby)

Percentage

Correct

No Yes

Yes

3
1
18
108

14.3

99.1

Overall percentage 85.4

The Cox and Snell R Square and Nagelkerke R square values, in Table 17, were

used to explain variation in the dependent variable. Based on the outcomes, the amount

of variance in the dependent variable accounted for by the model is about 17%.

Table 17

Model Summary

Step -2 Log

Likelihood

Cox & Snell

R Square

Nagelkerke

R Square

101.747a

.165

A correlation analysis between the covariates and independent variables shows

that education level is correlated to timing of the start of antenatal care, health literacy,

household income, number of antenatal care visits, employment status and marital status

(p < .05). While it is not strange that education correlates with employment status,

number of antenatal visits and number of children, medical problems developed during

pregnancy is only correlated with pregnancy outcomes. The number of antenatal care

visits is highly correlated with frequency of visits (p < =.01), employment (p =.05),

education (p =.01).

According to Szumilas (2010), an odds ratio is used to measure the relationship

that exists between an exposure and an outcome. This means that given a particular

exposure an outcome will occur compared to the odds that the outcome occurs without

the exposure. In other words, odds are “the ratio of probability that an event will occur

divided by the possibility that the event will not occur” (Mertler & Vannatta, 2013, p.

298). The regression coefficients are presented in Table 18. The Wald statistics shows

only problems developed during pregnancy is related to pregnancy outcomes (p =.047) at

statistically significant levels and there is no statistical significance among the rest of the

variables. The odds ratio for problem developed showed that for every increase of 1 in

the pregnancy problem developed that there is a 0.316 less chance for a healthy baby.

Since the value Exp(β) is less than 1 this indicates a negative odds or change, Thus, an

increase in pregnancy problems developed during pregnancy will lead to decrease

possibility of healthy pregnancy.

From the correlations model in Table 18, problems developed during pregnancy

add significantly to the prediction of the pregnancy outcome or whether the mother has a

healthy or unhealthy baby (p = <.05). Frequency of visits is also related at a statistically

significant level to pregnancy outcome and this makes sense because the more visit a

pregnant woman makes to a health facility the more likely health personnel would be able

to detect any medical issues. The number of antenatal visits, health literacy, and timing of

the first antenatal care visit did not add significantly to the prediction of pregnancy

outcome. The odds ratio (Exp(B) indicates how the independent variables change the

likelihood of the pregnancy outcomes, that is whether a healthy or unhealthy baby will be

delivered. The findings presented in Table 18 reveal high odds for the effects of the

different categories of problems developed during pregnancy. Although timing of

antenatal visit is not statistically significant, it is important to note that the odds ratios for

the categories for this variable shows that the women who attended antenatal visit at least

once in a month are three times likely to affect their pregnancy outcome.

Table 18

Regression Coefficients for Maternal Health Literacy, Frequency of Antenatal Visits, Number of Antenatal Visit, Timing of the

Start of Antenatal Care and Problems Developed During Pregnancy.

β S.E. Wald df Sig. Exp(β)
95% C.I. for EXP(β)

Lower Upper

Frequency of antenatal Visit

6.947 3 .074

Frequency of antenatal Visit (1) .613 1.212 .256 1 .613 1.846 .171. 19.871

Frequency antenatal Visit (2) -.882 1.240 .506 1 .477 .414 .036 4.704

Frequency antenatal Visit (3) -1.517 1.548 .960 1 .327 .219 .011 4.563

Pregnancy problem developed -1.151 .580 3.944 .047 .316 .102 .985

Number of Antenatal Visit

2.448 2 .294

Number of Antenatal Visit (1) -.903 .779 1.343 1 .247 .406 .088 1.867

Number of Antenatal Visit (2) .825 1.015 .662 1 .416 2.283 .313 16.674

Timing of first antenatal visit

2.830 2 .243

Timing of first antenatal visit (1) 1.140 .776 2.158 1 .142 3.126 .683 14.305

Timing of first antenatal visit (2) 1.293 .795 2.646 1 .104 3.645 .767 17.320

Health Literacy .408 .397 1.060 .303 1.504 .691 3.272

Constant -.101 1.858 .003 1 .956 .904

Note. Variable(s) entered on Step 1: Health Literacy, Pregnancy Problems, Frequency of antenatal visit, number of Antenatal

Visits, Timing of first visit

Results in Relationship to Null Hypothesis

The research question and hypotheses that guided this study were: What is the

predictive relationship between maternal health literacy level, the number of antenatal

visits, timing of the first antenatal care visit, development of medical conditions during

pregnancy and pregnancy outcomes (birth status of child, which are healthy and

unhealthy baby) in Lagos, Nigeria? The H0 was: There was no statistically significant

predictive relationship between maternal health literacy level, the number of antenatal
visits, timing of the first antenatal care visit, development of medical conditions during
pregnancy and pregnancy outcomes (birth status of child, which are healthy and

unhealthy baby) in Lagos, Nigeria. The H0 was rejected if p-values were p ≤ 0.05.

Based on my research question I used binary logistic regression to determine the

odds of having healthy or unhealthy baby as a pregnancy outcome. To begin I checked to

be sure the assumptions of the logistic regression were met, so the result would analysis

would be valid. According to Mertler and Vannatta (2013) the dependent variable needs

to be binary or binomial. Pregnancy outcome was categorized as binomial variable and I

coded healthy pregnancy as 1 and unhealthy pregnancy as 0. Also, there should be no

multicollinearity. According to Mertler and Vannatta, this can lead to understanding

which independent variable contributes to the explained variance. A preliminary multiple

linear regression was conducted to evaluate multicollinearity in the variables. As can be

seen in table 18, multicollinearity was not violated because tolerance statistics for all the

five independent variables were greater than 0.1. Also, the Chi Square test of association

shows there is no multicollinearity. There are no outliers and the test of goodness of fit to

assess the fit of the model to the data (Mertler & Vannatta, 2013) shows the model is a

good fit. This was confirmed by the Hosmer-Lemeshow not being statistically significant

(p = 0.641). A logistic regression model was used to assess the importance of each

independent variable and to predict which variables had the strongest relationship with

each other. Of the predictor variables, only problems developed during pregnancy was

found to be statistically significant (p =.05). Overall only one variable was a statistically

significant predictor of relationship between the independent and dependent variables

(Chi square = 13.23, p < 0.05 with df = 6) 13.23 therefore, the result is not conclusive.

Summary

The purpose of this study was to examine the relationship between maternal

health literacy, antenatal care visits, development of medical conditions during pregnancy

(independent variables), and pregnancy outcomes (dependent variable) for women in the

city of Lagos, Nigeria. The population studied is diverse and evenly distributed. In this

study, I examined if there is a statistically significant relationship between the dependent

and the independent variables or the extent to which the independent variables (i.e.,

maternal health literacy, antenatal care visits, and development of medical conditions

during pregnancy) predict the dependent variable (i.e., pregnancy outcomes) in Lagos,

Nigeria based on social cognitive theory and the health belief model. In chapter 4 I

reported the data collection procedure, return rate, data screening and treatment of

missing data and data analysis methods. Findings from the study show that there is

statistically significant relationship between pregnancy outcomes and antenatal care

visits.

Next, the findings from further analysis that included the regression of

independent variable maternal health literacy, antenatal care visits, frequency of antenatal

care, and timing of antenatal care, and development of medical problems during

pregnancy on the dependent variable pregnancy outcome showed that only one variable

was a statistically significant predictor of relationship between the independent and

dependent variables. The results show that only development of medical problems

during pregnancy play significant role in women pregnancy outcomes. However, the

findings should be interpreted with caution because the participants might have

responded in a manner beyond my control. The data did not meet the parametric tests

which limit the extent to which it can be generalized to the larger population but the

findings from the study can be used to develop educational materials and to develop

programs and services that will better improve women’s pregnancy outcomes. In chapter

5 I discussed the research findings, limitation of the study, recommendations,

implications, and conclusion.

100

Chapter 5: Discussion, Conclusions, and

Recommendations

Introduction

The purpose of the study was to determine if there is a relationship between

maternal health literacy, antenatal care visits, and development of medical conditions

during pregnancy (independent variables) and pregnancy outcomes (dependent variable)

in Lagos, Nigeria. In the study, I identified antenatal care utilization measured by the

timing of first antenatal care visit. I also assessed the relationship of the independent

variables on pregnancy outcomes measured by the birth status of the child, which was

either a healthy and unhealthy baby. In this chapter, I will discuss the results of the study

and the public health implications of the findings and provide my conclusions and

recommendations for further research. A binary correlation was conducted to assess the

relationship between the dependent and independent variables. Results showed that the

variable problems developed during pregnancy were significantly correlated with

pregnancy outcomes.

Interpretation of Findings

Overall, the findings of this study indicated that problems developed during

pregnancy are significantly correlated to pregnancy outcome. I performed a binary

logistic regression to find the predictable relationship between maternal health literacy

level, the number of antenatal visits, the timing of the first antenatal care visit,

development of medical conditions during pregnancy, and pregnancy outcomes (which

was measured by having either a healthy baby or unhealthy baby). The binary logistic

regression results showed statistical significance for medical problems developed during

101

pregnancy. The model explained 20% (Nagelkerke R2) of the variance in pregnancy

outcome and correctly classified about 87% of the cases. The sensitivity test for those

who had a healthy baby was 96% and those who did not have healthy baby as 38%.

About 21% of the participants reported they had or developed medical problems during

their pregnancy. Medical complications may exist before pregnancy or worsen with

pregnancy if they are not treated (WHO, 2016). Some of the complications mentioned by

the participants were bleeding, high blood pressure, malaria, anemia, and other

complications. This is generally consistent with WHO’s (2016) report that major health

complications account for 75% of maternal deaths. Unlike the WHO (2016) reports, in

this study I did not look at women who died from these complications but tried to see if

medical problems predict birth outcomes. Medical conditions during pregnancy highlight

the need for antenatal care to ensure healthy pregnancy. This is important for coordinated

care among health care providers to effectively manage the health of the pregnant women

and ensure high likelihood of successful outcomes (Iezzoni, Yu, Wint, Smeltzer, &

Ecker, 2014).

In this study, I did not find a statistically significant predictive relationship

between maternal health literacy level and pregnancy outcomes in the study population.

Past researchers have identified an association between a mother’s literacy skills and the

survival of the child (Levine et al., 2004). Shieh, Mays, McDaniel, and Yu (2009)

reported that low health literacy affects the women’s pregnancy knowledge which will

affect the health of the baby. Rothman et al. (2004) also showed it was difficult for

women with low literacy to read directions and to follow medical recommendations,

102

which leads to them having poorer knowledge that can contribute to worse clinical

outcomes. However, these researchers’ results could not be vindicated in this study

because no association was found.

As a result, I looked at the education level of the women participants to see how

their education level could impact their birth outcomes. Past studies have found level of

education an important factor in birth outcomes (Idris et al., 2007; Ifenne et al., 1997;

Ikeako et al., 2006). A study by Auger, Luo, Platt, and Daniel (2008) in Canada found an

association between not having a high school diploma and low birth weight. However,

the result from this study shows an interesting dynamic. While it would be expected that

higher education would serve as a protective factor against negative birth outcomes,

participants in this study with little or no education reported fewer negative birth

outcomes compared to participants with higher education. This finding is different from

other studies that suggest an association between education level and pregnancy outcome

(Kohan, Ghasemi, & Dodangeh, 2007).

Skilled obstetric assistance during delivery and adequate antenatal care are

important to reducing maternal mortality and morbidity (Iyaniwura & Yusuf, 2009), since

pregnant women will receive information that will help them in making the right

decisions concerning their pregnancy, receive treatment for existing medical conditions,

and also receive screening for risk factors. The results from this study showed that a

majority of the pregnant women in this study received adequate antenatal care but

receiving adequate antenatal care may not be enough because a majority of the fatal

complications may occur during or shortly after delivery (Iyaniwura & Yusuf, 2009).

103

Most of the women started antenatal care after the first trimester and this could be due to

a lot of reasons that were not explored in this study. Therefore, there may be a need to

look at other socio-cultural factors which may act as barriers to starting antenatal care

early which could lead to successful pregnancy outcomes.

The results from this study are in accord with the HBM. The correlation in

education, income, health literacy level, and timing of first antenatal visit are products of

the environments. The socio-cultural factors which are associated with low antenatal

care utilization can be viewed as modifying factors and perceived barriers to health

seeking behavior for the women in the study. Furthermore, results from this study

indicate the need for prevention focused programs that specifically target and encourage

women to seek medical care as soon as they discover they are pregnant. This may help to

promote healthy practices in these women. Self-efficacy can be used to help these women

to set short term goals from one visit to another.

Limitations of the Study

Potential limitations associated with this study included the use of self-reported
data, which can introduce recall bias (Creswell, 2007). In this study, I used a face-to-
face, self-administered survey because of the ability of ensuring feedback and completion
of the questions and assisting participants who may have questions (Creswell, 2007).

Although this method may reduce nonresponse bias it did not provide anonymity and

participants might not have truthfully answered sensitive questions (Creswell, 2007). To

guard against this, Pannucci and Wilkins (2010) suggested the use of only validated

scale. The scale I used was already validated and I also dropped off questionnaire for

104

participants who were busy or did not want to complete the survey with me there. These

were picked up in a sealed envelope and this could have reduced non-response bias. Also,

Frankfort-Nachmias and Nachmias (2008) suggested researchers ensure that the

questions asked are not threatening by asking participants to rate how uneasy they felt

other people would feel about the questions or rate the degree of difficulty in answering

the questions. Although not all bias in the study could be controlled but the awareness of

the presence of bias allowed thorough scrutiny of the results (Sica, 2006).

Another limitation for this study was lack of data from women who have

died giving birth. Although this would have provided rich information about

what they went through, data cannot be collected from a dead person and

collecting data from the family of a dead pregnant woman will bring unpleasant

memories and the information provided may be only hearsay. Furthermore, the

study was limited to women over 18 years because women 17 years and under are

considered minors who will need adult consent to participate. Pregnant women

under age 18 may have different reactions to the questions in this study and their

experience or lack of experience with antenatal care could have provided robust

data.

Block (2002) noted there are intrinsic limitations with the self-reporting research

method and analysis that relies on empirical measures alone; this is because cognitive and

situational factors could affect the validity of the self-administered questionnaires.

According to Gliem and Gliem (2003), the use of single item question in a construct is

not a reliable way to generate conclusion. The dependent variable in this study was a

105

single item construct and this could constitute a limitation to the study and present an

opportunity for future studies and the development of stronger instruments to support

future study.

Recommendations

As a result of dearth of literature showing evidence of use of a standardized

instrument to measure health literacy and also to find predictable relationships between

health literacy, maternal health characteristics, and pregnancy outcomes, I conducted this

study to add to the body of literature. The SCT of interaction between the environment,

behavior, and what the pregnant women participants acquired through their education and

information received during antenatal visit could empower women to make informed

decisions about their health during pregnancy and beyond. Exposure to antenatal care and

information provided during these visits could also help pregnant women in their decision

making.

One of the seven MDGs, the reduction of maternal mortality by 75% by the year

2015 (Say et al., 2014), was not achieved (Oye-Adeniran et al., 2014), but women in this

study were aware of the importance of antenatal visits. Seventy percent of women

attended six or more antenatal visits; more than the four visits or more as proposed by

WHO (2002). Although there were more than six visits for more than half of the women

surveyed, only 47% of these women started antenatal care before the end of the first

trimester (3 months). It is important to encourage early start of antenatal visits. This will

allow early detection of likely medical problems and will reduce pregnancy

complications that can lead to death or debilitation. Future studies should consider what

106

makes women wait longer to start antenatal care and how media campaigns could work to

remedy this. The women who did not receive antenatal care at all were either not aware

they were pregnant or did not want others to know they were pregnant. The women

participants in this study were not pregnant at the time of the study but a comparison with

women who are pregnant at the time of the study may give better insight. Also, future

studies using the health literacy tool should try to consolidate the questions. Some of the

participants reported the questions as repetitive or not very clear.

Health practitioners should be proactive in educating pregnant women, during

each visit, on the importance of self-care and the early start of antenatal care. The

participants in the study volunteered unsolicited information about the kind of care they

received and their expectations for their care. A mixed methods approach of qualitative

and quantitative research methods would be able to gather more information than this

quantitative study and would help to bring out other reasons of lack of attendance or late

attendance of antenatal care and what can be done to get women to use health facilities

before, during, or after pregnancy.

Implications

Reducing maternal mortality is an important goal for public health professionals

globally. The results from this study showed an interplay of different variables and

underscore the challenges health professionals in a developing country, like Nigeria,

might have in reducing maternal pregnancy and promoting better pregnancy outcomes.

Data from the study showed Nigerian women have a high level of education; to improve

maternal literacy, antenatal education should be introduced in high school and students

107

should be encouraged to take health-related courses emphasizing this subject area even if

they are not majoring in health sciences. Community health promotion and awareness

should emphasize early antenatal care to safeguard the health of the mother and child.

Although this was not an area of focus in this study, information the women receive

during their routine antenatal visits should address the importance of healthy living for a

successful pregnancy outcome. This combination of practices is essential for a successful

pregnancy outcomes and preventive health education could be useful for behavioral

change and promoting antenatal care among future pregnant women.

Conclusions

Of the five predictor variables, problems developed during pregnancy was

statistically significant. An increase in problems developed during pregnancy most likely

will increase the chance of having negative pregnancy outcomes. It has been established

in the literature that having antenatal care and care throughout the pregnancy increases

the chance of having a successful pregnancy and healthy child. There is still much to be

done to encourage early antenatal usage. The practice of registering late for antenatal

care and having to go for biweekly visits, as a result of a late start, needs to be

discouraged. Uniformity of care for pregnant women in both private and public health

facilities needs to be encouraged to allow women who use government hospitals to have

confidence in the care they are receiving. The long term positive social change

implication of this study is that educational efforts and healthcare outreach could be

focused on those women that normally would not pursue antenatal health care on their

own.

108

References

Abdoulaye, D. (2006). Maternal mortality in Africa. Internet Journal of Health, 5.

Retrieved

from https://ispub.com/IJH/5/1/12701

Adanri, A. A. (2016). The relationship between Nigerian local government administrative

leadership styles and organization outcomes (Order No. 10134346). Available.

Retrieved from Dissertations & Theses @ Walden University. (UMI

no.1803939357).

Retrieved from

http://ezp.waldenulibrary.org/login?url=http://search.proquest.com/docview/1803

939357?accountid=14872

Adegoke, A. A., Lawoyin, T. O., Ogundeji, M. O., & Thomson, A. M. (2007). A

community-based investigation of the avoidable factors of maternal mortality in

Nigeria: The pilot experiences. African Health Sciences, 7(3), 176-–181.

Retrieved from http://www.ajol.info/index.php/ahs/article/viewFile/7011/58272

Adekanle D., & Isawumi A. (2008). Late antenatal care booking and its predictors

among pregnant women in south western Nigeria. Online Journal of Health Allied

Sciences, 7(1), 4. Retrieved from http://cogprints.org/6084/1/2008-1-

4 ?origin=publication_detail

Adetunji, J. A. (1992). Church-based obstetric care in a Yoruba community, Nigeria.

Social Science & Medicine, 35(9), 1171-–1178. Retrieved from

http://www.sciencedirect.com/science/article/pii/027795369290229J?via%3Dihub

Adewunmi, A., Rabiu, K., & Tayo, A. (2008). Gestational age at antenatal

booking in Lagos, southwest Nigeria. Internet Journal of Gynecology and

https://ispub.com/IJH/5/1/12701

109

Obstetrics, 12(1). Retrieved from http://ispub.com/IJGO/12/1/5220

Adhoc Committee on Health Literacy. (1999). Health literacy: Report of the Council on

Scientific Affairs, America Medical Association. Journal of the American

Medical Association, 281, 582-–557. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/10022

112

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human

Decision Processes, 50(2), 179-211. Retrieved from

http://www.sciencedirect.com/science/article/pii/074959789190020T

Ali, A. A., & Adam, I. (2011). Lack of antenatal care, education, and high maternal

mortality in Kassala hospital, eastern Sudan during 2005–2009. Journal of

Maternal-Fetal & Neonatal Medicine, 24(8), 1077-1078.

doi:10.3109/14767058.2010.545908

Aluko, O.E. (2010). The impact of urbanization on housing development: The Lagos

experience, Nigeria. Ethiopian Journal of Environmental Studies and

Management, 3(3), 64-74. Retrieved from

http://www.ajol.info/index.php/ejesm/article/view/63967

Alvarez, J. L., Gil, R., Hernández, V., & Gil, A. (2009). Factors associated with maternal

mortality in Sub-Saharan Africa: An ecological study. BMC Public Health, 9(1),

462. Retrieved from http://www.biomedcentral.com/1471-2458/9/462

Aniebue, U. U., & Aniebue, P. N. (2010). Women’s perception as a barrier to focused

antenatal care in Nigeria: The issue of fewer antenatal visits. Health Policy and

110

Planning, 11(18). Retrieved from

http://heapol.oxfordjournals.org/content/early/2010/11/18/heapol.czq073.full

Anugwom, E. E. (2009). Women, education and work in Nigeria. Educational Research

and Review, 4(4), 127-134. Retrieved from http://www.resourcedat.com.ng/wp-

content/uploads/2013/02/women-education3

pdf

Anya, S. E., Hydara, A., & Jaiteh, L. E. (2008). Antenatal care in the Gambia: Missed

opportunity for information, education and communication. BMC Pregnancy and

Childbirth, 8(1), 9. Retrieved from

http://www.biomedcentral.com/content/pdf/1471-2393-8-9

Auger, N., Luo, Z. C., Platt, R. W., & Daniel, M. (2008). Do mother’s education and

foreign born status interact to influence birth outcomes? Clarifying the

epidemiological paradox and the healthy migrant effect. Journal of Epidemiology

and Community Health, 62(5), 402-409. Retrieved from

http://www.jstor.org/stable/pdf/40665740

Austin, L. T., Ahmad, F., McNally, M. J., & Stewart, D. E. (2002). Breast and cervical

cancer screening in Hispanic women: A literature review using the health belief

model. Womens Health Issues, 12(3), 122-128. Retrieved from

http://www.whijournal.com/article/S1049-3867%2802%2900132-9/fulltext

Babalola, S., & Fatusi, A. (2009). Determinants of use of maternal health services in

Nigeria-looking beyond individual and household factors. BMC Pregnancy and

Childbirth, 9(1), 43. Retrieved from http://www.biomedcentral.com/1471-

2393/9/43

http://www.biomedcentral.com/content/pdf/1471-2393-8-9

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Austin%20LT%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Ahmad%20F%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22McNally%20MJ%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed?term=%22Stewart%20DE%22%5BAuthor%5D

http://www.ncbi.nlm.nih.gov/pubmed/12015184

111

Baker, D. W., Williams, M. V., Parker, R. M., Gazmararian, J. A., & Nurss, J. (1999).

Development of a brief test to measure functional health literacy. Patient

Education and Counseling, 38(1), 33-42.

http://www.sciencedirect.com/science/article/pii/S0738399198001165

Bandura, A. (1997). Self-efficacy: Toward a unifying theory of behavioral change.

Psychology Review, 84, 191-215. Retrieved from

http://www.uky.edu/~eushe2/Bandura/Bandura1977PR

Bandura, A. (1998). Health promotion from the perspective of social cognitive theory.

Psychology and Health, 13, 623-649. Retrieved from

https://www.uky.edu/~eushe2/Bandura/Bandura1998PH

Bandura, A. (2001). Social cognitive theory: An agentic perspective. Retrieved from

http://www.uky.edu/~eushe2/Bandura/Bandura2001ARPr

Banta, D. (2003). What is the efficacy/effectiveness of antenatal care and the financial

and organizational implications? Copenhagen: WHO regional office for Europe;

2003. (Health Evidence Network Report). Retrieved from

http://www.euro.who.int/__data/assets/pdf_file/0007/74662/E82996

Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and trends in

organizational research. Human Relations, 61(8), 1139-1160.

Bhattacherjee, A. (2012). Social science research: Principles, methods, and practices.

Textbooks collection. Book 3: University of South Florida Scholar Commons.

Retrieved from http://scholarcommons .usf.edu/oa_textbooks/3

https://www.uky.edu/~eushe2/Bandura/Bandura1998PH

http://www.euro.who.int/__data/assets/pdf_file/0007/74662/E82996

http://scholarcommons/

112

Bukar, M., Kunmanda, V., Moruppa, J. Y., Ehalaiye, B., Takai, U. I., & Ndonya, D. N.

(2014). Maternal mortality at federal medical centre Yola, Adamawa State: A

five-year review. Annals of Medical and Health Sciences Research, 13(3), 568-

Carrara, V. I., Hogan, C., De Pree, C., Nosten, F., & McGready, R. (2011). Improved

pregnancy outcome in refugees and migrants despite low literacy on the Thai-

Burmese border: Results of three cross-sectional surveys. BMC Pregnancy and

Childbirth, 11(1), 45. Retrieved from http://www.biomedcentral.com/1471-

2393/11/45

Carroll, S. (n.d.). Types of research designs you can use for your dissertation. Retrieved

from http://www.dissertation-statistics.com/research-designs.html

Ceesay, S. M., Prentice, A. M., Cole, T. J., Foord, F., Poskitt, E. M., Weaver, L. T., &

Whitehead, R. G. (1997). Effects on birth weight and perinatal mortality of

maternal dietary supplements in rural Gambia: 5 year randomised controlled trial.

BMJ, 315(7111), 786-790. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2127544/pdf/9345173

Cengage Research Methods Workshops. (2005). Retrieved from

http://www.wadsworth.com/psychology_d/templates/student_resources/workshop

s/resch_wrk.html

Centers for Disease Control and Prevention. (2014). Pregnancy complications. Retrieved

from http://www.cdc.gov/reproductivehealth/maternalinfanthealth/preg

complications.htm

http://www.cdc.gov/reproductivehealth/maternalinfanthealth/

113

Chew, L. D., Bradley, K. A., & Boyko, E. J. (2004). Brief questions to identify patients

with inadequate health literacy. Health, 11, 12. Retrieved from

https://www.stfm.org/fmhub/fm2004/september/lisa588

Creswell, J. W. (2007). Qualitative inquiry & research design: Choosing among five

approaches. (2nd ed.). Thousand Oaks, CA:

SAGE.

Cycyota, C. S., & Harrison, D. A. (2006). What (not) to expect when surveying

executives: Ameta-analysis of top manager response rates and techniques over

time. Organizational Research Methods, 9(2), 133-160.

Edberg, M. (2007). Essentials of health behavior: Social and behavioral theory in public

health. Sadbury, MA: Jones and Bartlett.

Ekott, M. M., Ovwigho, U. U., Ehigiegba, A. A., Fajola, A. A., & Fakunle, B. B. (2013).

Perception of pregnant women about antenatal care in a cottage hospital in Port

Harcourt, Nigeria. Journal of Community Health, 38(2), 360-365. Retrieved from

https://link.springer.com/article/10.1007/s10900-012-9625-1

Elsenbruch, S., Benson, S., Rücke, M., Rose, M., Dudenhausen, J., Pincus-Knackstedt,

M. K., … Arck, P. C. (2007). Social support during pregnancy: Effects on

maternal depressive symptoms, smoking and pregnancy outcome. Human

Reproduction, 22(3), 869-877. Retrieved from

http://humrep.oxfordjournals.org/content/22/3/869.full

Ewbank, D.C. (1994). Maternal education and theories of health behaviour: A cautionary

note. Health Transition Review, 4(2), 215-223. Retrieved from

http://www.jstor.org/stable/40652137

https://link.springer.com/article/10.1007/s10900-012-9625-1

114

Expatarrival. (2015). Moving to Lagos. Retrieved from

http://www.expatarrivals.com/nigeria/lagos/moving-to-lagos

Frankfort-Nachmias, C., & Nachmias, D. (2008). Research methods in the social sciences

(7th ed.) New York, NY: Worth Publisher.

Fritz, C. O., Morris, P. E., & Richler, J. J. (2012). Effect size estimates: Current use,

calculations, and interpretation. Journal of Experimental Psychology: General,

141(1), 2. Retrieved from

http://comp.uark.edu/~whlevine/psyc5133/fritz.morris.richler.2012.xge

Gazali, W. A., Muktar, F., & Gana, M. M. (2012). Barriers to utilization of maternal

health care facilities among pregnant and non-pregnant women of child bearing

age in Maiduguri Metropolitan Council (MMC) and Jere Lgas of Borno State.

Continental Journal of Tropical Medicine, 6(1), 12-21. Retrieved from

http://docslide.us/documents/maternalthesis.html

Geller, P. A. (2004). Pregnancy as a stressful life event. CNS Spectrums, 9(3), 188-197.

Ghana Maternal Health Survey. (2008). Retrieved from

http://pdf.usaid.gov/pdf_docs/PNADO492

Gil-González, D., Carrasco-Portiño, M., & Ruiz, M. T. (2006). Knowledge gaps in

scientific literature on maternal mortality: A systematic review. Bulletin of the

World Health Organization, 84(11), 903-909.

Glanz, K., Rimer, B. K., & Viswanath, K. (Eds.). (2008). Health behavior and health

education: Theory, research, and practice (4th ed.). San Francisco, CA: Jossey-

Bass.

http://pdf.usaid.gov/pdf_docs/PNADO492

115

Harrison, K. A. (1997). Maternal mortality in Nigeria: The real issues. African Journal of

Reproductive Health/La Revue Africaine de la Santé Reproductive, 1(1), 7-13.

Harrison, K. A. (2009). The struggle to reduce high maternal mortality in Nigeria.

African Journal of Reproductive Health, 13(3). Retrieved from

http://www.ajol.info/index.php/ajrh/article/viewFile/55723/44189

Harrison, K. A., & Rossiter, C. E. (1985). Child-bearing health and social priorities: A

survey of 22774 consecutive hospital births in Zaria northern Nigeria. 7. Multiple

pregnancy. British Journal of Obstetrics and Gynaecology, 92, 49-60. Retrieved

from http://www.popline.org/node/425631

Hine, C. (2011). Internet research and unobtrusive methods. Retrieved from

http://sru.soc.surrey.ac.uk/SRU61

Hochbaum, G., Rosenstock, I., & Kegels, S. (1952). Health belief model. United States

Public Health Service. Retrieved from

http://www.infosihat.gov.my/infosihat/artikelHP/bahanrujukan/HE_DAN_TEORI

/DOC/Health%20Belief%20Model

Högberg, U., & Wall, S. (1986). Age and parity as determinants of maternal mortality—

impact of their shifting distribution among parturient in Sweden from 1781 to

1980. Bulletin of the World Health Organization, 64(1), 85. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2490922/

Iezzoni, L. I., Yu, J., Wint, A. J., Smeltzer, S. C., & Ecker, J. L. (2014). General health,

health conditions, and current pregnancy among US women with and without

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2490922/

116

chronic physical disabilities. Disability and Health Journal, 7(2), 181-188.

Retrieved from http://pubmedcentralcanada.ca/pmcc/articles/PMC3971389

Ikeako, L. C., Onah, H. E., & Iloabachie, G. C. (2006). Influence of formal maternal

education on the use of maternity services in Enugu, Nigeria. Journal of

Obstetrics & Gynecology, 26(1), 30-34. Retrieved from

http://www.sciencedirect.com/science/article/pii/S0277953606002280

Iyaniwura, C. A., & Yussuf, Q. (2009). Utilization of antenatal care and delivery services

in Sagamu, southwestern Nigeria. African Journal of Reproductive Health, 13(3),

111-122. Retrieved from

http://www.ajol.info/index.php/ajrh/article/viewFile/55719/44186

Jain, G., & Bisen, V. (2012). Female literacy & its relevance with maternal and infant

mortality rates. International Journal of Management, 3(2), 65-79. Retrieved from

https://www.researchgate.net/publication/256042423_Female_Literacy_its_Releva

nce_with_MaternalAnd_Infant_Mortality_Rates

Jennings, L., Yebadokpo, A. S., Affo, J., & Agbogbe, M. (2010). Antenatal counseling in

maternal and newborn care: Use of job aids to improve health worker

performance and maternal understanding in Benin. BMC Pregnancy and

Childbirth, 10(1), 75.

Jupp, V. (2006). The Sage dictionary of social research methods. Thousand Oaks, CA:

SAGE.

117

Kabir, M., Iliyasu, Z., Abubakar, I. S., & Sani, A. A. (2005). Determinants of utilization

of antenatal care services in Kumbotso village, Northern Nigeria. Tropical

Doctor, 35(2), 110-110.

Kansas State University. (n.d.). Research Design. Retrieved from http://www-

personal.ksu.edu/~goe/lec08sl/sld002.htm

Karlsen, S., Say, L., Souza, J. P., Hogue, C. J., Calles, D. L., Gülmezoglu, A. M., &

Raine, R. (2011). The relationship between maternal education and mortality

among women giving birth in health care institutions: Analysis of the cross-

sectional WHO global survey on maternal and perinatal health. BMC Public

Health, 11(1), 606. Retrieved from http://www.biomedcentral.com/1471-

2458/11/606

Kerber, K. J., de Graft-Johnson, J. E., Bhutta, Z. A., Okong, P., Starrs, A., & Lawn, J. E.

(2007). Continuum of care for maternal, newborn, and child health: From slogan

to service delivery. Lancet, 370(9595), 1358-1369.

Kirchengast, S., Mayer, M., & Voigt, M. (2007). Pregnancy outcome is associated with

maternal marital status in Austria—even at the beginning of the 21st century.

Anthropologischer Anzeiger, 415-426. Retrieved from

http://www.ncbi.nlm.nih.gov/pubmed/18196764

Kirsch, I. S. (1993). Adult literacy in America: A first look at the results of the National

Adult Literacy Survey. Washington, DC: U.S. Government Printing Office,

Superintendent of Documents, Washington, DC 20402 (Stock No. 065-000-

00588-3). Retrieved from http://nces.ed.gov/pubs93/93275

118

Koblinsky, M., Chowdhury, M. E., Moran, A., & Ronsmans, C. (2012). Maternal

morbidity and disability and their consequences: Neglected agenda in maternal

health. Journal of Health, Population, and Nutrition, 30(2), 124–130. Retrieved

from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3397324/

Kohan, S., Ghasemi, S., & Dodangeh, M. (2007). Associations between maternal health

literacy and prenatal care and pregnancy outcome. Iran Journal of Nursing and

Midwifery Research, 12(4), 146-52. Retrieved from

http://irandanesh.febpco.com/FileEssay/ftpvm-Autumn2007-87-6-1-mm7

Koroukian, S., & Rimm, A. (2002). The “Adequacy of Prenatal Care Utilization”

(APNCU) index to study low birth weight: Is the index biased? Journal of

Clinical Epidemiology, 55, 296-305.

Kotrlik, J. W. K. J. W., & Higgins, C. C. H. C. C. (2001). Organizational research:

Determining appropriate sample size in survey research appropriate sample size in

survey research. Information Technology, Learning, and Performance Journal,

19(1), 43. Retrieved from

Lagos State Government. (2011). Digest of statistics. Retrieved from

http://www.lagosstate.gov.ng/images/pageimages/downloadfiles/docs/DIGEST_O

F_STATISTICS2011

Lagos State Government. (2014). Population. Retrieved from

http://www.lagosstate.gov.ng/pagelinks.php?p=6

Lanthier, E. (2002). Correlation. Retrieved

from http://www.nvcc.edu/home/elanthier/methods/correlation.htm

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3397324/

119

LeVine, R. A., LeVine, S. E., Rowe, M. L., & Schnell-Anzola, B. (2004). Maternal

literacy and health behavior: A Nepalese case study. Social Science & Medicine, 58(4),

863-877.

McBride, C. M., Emmons, K. M., & Lipkus, I. M. (2003). Understanding the potential of

teachable moments: The case of smoking cessation. Health Education Research,

18(2), 156-170.

McEachan, R., Lawton, R., Jackson, C., Conner, M., & Lunt, J. (2008). Evidence, theory

and context: Using intervention mapping to develop a worksite physical activity

intervention. BMC Public Health, 8, 326. Retrieved from

http://www.biomedcentral.com/1471-2458/8/326

Midhet, F., Becker, S., & Berendes, H. W. (1998). Contextual determinants of maternal

mortality in rural Pakistan. Social Science & Medicine, 46(12), 1587-1598.

Mojekwu, J. N., & Ibekwe, M. U. (2012). Maternal mortality in Nigeria: Examination of

intervention methods. Retrieved from

http://www.ijhssnet.com/journals/Vol_2_No_20_Special_Issue_October_2012/13

National Cancer Institute, Rimer, B. K., & Glanz, K. A. (2005). Theory at a glance: A

guide for health promotion practice (2nd ed.). Washington, DC: U.S. Department

of Health and Human Services, National Institutes of Health. Retrieved from

http://www.cancer.gov/cancertopics/cancerlibrary/theory

120

National Population Commission [Nigeria], & ICF International. (2014). Nigeria

demographic and health survey 2013. Abuja & Rockville: NPC & ICF

International. Retrieved from https://dhsprogram.com/pubs/pdf/FR293/FR293

National Population Commission (NPC) [Nigeria] and ICF Macro. (2009). Nigeria

demographic and health survey 2008. Abuja, Nigeria: National Population

Commission and ICF Macro. Retrieved from

http://www.measuredhs.com/pubs/pdf/FR222/FR222

National Population Commission (NPC) [Nigeria] and ORC Macro. (2013). Nigeria

demographic and health survey 2013. Calverton, MD: National Population

Commission and ORC Macro. National Population Commission and ICF Macro.

Retrieved from

http://www.population.gov.ng/images/Vol%2003%20Table%20DSx%20LGAPop

%20by%20SDistrict-PDF

Nikiéma, B., Beninguisse, G., & Haggerty, J. L. (2009). Providing information on

pregnancy complications during antenatal visits: Unmet educational needs in sub-

Saharan Africa. Health Policy and Planning, 24(5), 367-376. Retrieved from

http://heapol.oxfordjournals.org/content/24/5/367.full +html

Obiechina, N. J., Okolie, V. E., Okechukwu, Z. C., Oguejiofor, C. F., Udegbunam, O. I.,

Nwajiaku, L. S. A., … Egeonu, R. (2013). Maternal mortality at Nnamdi Azikiwe

University Teaching Hospital, Southeast Nigeria: A 10-year review (2003–2012).

International Journal of Women’s Health, 5, 431. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3726439/

121

Ogbera, A. O., & Ekpebegh, C. (2014). Diabetes mellitus in Nigeria: The past, present,

and future. World Journal of Diabetes, 5(6), 905. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4265879/

Ogunlesi, T. A. (2005). The pattern of utilization of prenatal and delivery services in

Ilesa, Nigeria. Internet Journal of Epidemiology, 2(2). Retrieved from

http://ispub.com/IJE/2/2/6953

Ohl, M., Harris, A., Nurudtinova, D., Cai, X., Drohobyczer, D., & Overton, E. T. (2010).

Do brief screening questions or provider perception accurately identify persons

with low health literacy in the HIV primary care setting? AIDS patient care and

STDs, 24(10), 623-629. Retrieved from

http://online.liebertpub.com/doi/full/10.1089/apc.2009.0319?src=recsys&

Okereke, E., Aradeon, S., Akerele, A., Tanko, M., Yisa, I., & Obonyo, B. (2013).

Knowledge of safe motherhood among women in rural communities in northern

Nigeria: Implications for maternal mortality reduction. Reproductive Health,

10(1), 57. Retrieved from http://www.reproductive-health-

journal.com/content/10/1/57

Olusegun, O. L., Ibe, R. T., & Micheal, I. M. (2012). Curbing maternal and child

mortality: The Nigerian experience. International Journal of Nursing and

Midwifery, 4(3), 33-39. Retrieved from

http://www.academicjournals.org/article/article1379408909_Ogunjimi%20et%20

al.pd

http://ispub.com/IJE/2/2/6953

122

Ozumba, B. C., & Nwogu-Ikojo, E. E. (2008). Avoidable maternal mortality in Enugu,

Nigeria. Public Health, 122(4), 354-360. Retrieved from

Pannucci, C.J., & Wilkins, E.G. (2002). Identifying and avoiding bias in research.

Plastic Reconstruction Surgery, 126(2), 619–625. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2917255/

Prajapati, B., Dunne, M., & Armstrong, R. (2010, July 16). Sample size estimation and

statistical power analyses. Optometry Today. Retrieved from

http://www.psycho.uni-

duesseldorf.de/abteilungen/aap/gpower3/literature/Dokumente

Quality Metric. (2013). SF health surveys. Retrieved from

http://www.qualitymetric.com/WhatWeDo/GenericHealthSurveys/tabid/184/Defa

ult.aspx?gclid=CPDsmaXWgrUCFY9DMgod2UsAiw

Rashad, W., & Essa, R. (2010). Women’s awareness of danger signs of obstetrics

complications. Journal of American Science, 6(10), 1299-306. Retrieved from

http://www.jofamericanscience.org/journals/am-

sci/am0610/149_3926am0610_1299_1306

Rothman, R. L., DeWalt, D. A., Malone, R., Bryant, B., Shintani, A., Crigler, B., &

Pignone, M. (2004). Influence of patient literacy on the effectiveness of a primary

care–based diabetes disease management program. JAMA: The Journal of the

American Medical Association, 292(14), 1711-1716. Retrieved from

http://jama.jamanetwork.com/article.aspx?articleid=199582

123

Say, L., Chou, D., Gemmill, A., Tunçalp, Ö., Moller, A. B., Daniels, J., … Alkema, L.

(2014). Global causes of maternal death: A WHO systematic analysis. Lancet

Global Health, 2(6), e323-e333. Retrieved from

http://www.sciencedirect.com/science/article/pii/S2214109X1470227X

Schnell-Anzola, B., Rowe, M. L., & LeVine, R. A. (2005). Literacy as a pathway

between schooling and health-related communication skills: A study of

Venezuelan mothers. International Journal of Educational Development, 25(1),

19-37.

Schwartz, K. L., Bartoces, M., Campbell-Voytal, K., West, P., Monsur, J., Sartor, A., &

Neale, A. V. (2013). Estimating health literacy in family medicine clinics in

metropolitan Detroit: A MetroNet study. Journal of the American Board of Family

Medicine, 26(5), 566-570. Retrieved from

http://www.jabfm.org/content/26/5/566.short

Scott-Emuakpor, A. (2014). The evolution of health care systems in Nigeria: Which way

forward in the twenty-first century. Niger Medical Journal, 51, 53-65. Retrieved

from: http://www.nigeriamedj.com/text.asp?2010/51/2/53/70997

Shieh, C., Mays, R., McDaniel, A., & Yu, J. (2009). Health literacy and its association

with the use of information sources and with barriers to information seeking in

clinic-based pregnant women. Health Care for Women International, 30(11), 971-

988. Retrieved from

http://www.tandfonline.com/doi/full/10.1080/07399330903052152.

http://www.nigeriamedj.com/text.asp?2010/51/2/53/70997

124

Sica, G. T. (2006). Bias in research studies. Radiology, 238(3), 780-789. Retrieved from

http://pubs.rsna.org/doi/full/10.1148/radiol.2383041109

Singh, K., Bloom, S., Haney, E., Olorunsaiye, C., & Brodish, P. (2012). Gender equality

and childbirth in a health facility: Nigeria and MDG5. African Journal of

Reproductive Health, 16(3). Retrieved from

http://www.jstor.org/stable/23485700?seq=1#page_scan_tab_contents

Singh, S., Ahmed, E. B., Egondu, S. C., & Ikechukwu, N. E. (2014). Hypertensive

disorders in pregnancy among pregnant women in a Nigerian teaching hospital.

Nigerian Medical Journal: Journal of the Nigeria Medical Association, 55(5),

384. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4178334/

Smith, P. J., Humiston, S. G., Marcuse, E. K., Zhao, Z., Dorell, C. G., Howes, C., &

Hibbs, B. (2011). Parental delay or refusal of vaccine doses, childhood

vaccination coverage at 24 months of age, and the Health Belief Model. Public

Health Reports, 126(Suppl 2), 135. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3113438/

Strecher, V. J., DeVellis, B. M., Becker, M. H., & Rosenstock, I. M. (1986). The role of

self-efficacy in achieving health behavior change. Health Education & Behavior,

13(1), 73-92. Retrieved from

http://deepblue.lib.umich.edu/bitstream/handle/2027.42/68171/10.1177_1090198

18601300108 ?sequence=2

Stuebing, K. W. (1997). Maternal schooling and comprehension of child health

information in urban Zambia: Is literacy a missing link in the maternal schooling-

http://pubs.rsna.org/doi/full/10.1148/radiol.2383041109

125

child health relationship? Health Transition Review, 151-171. Retrieved from

http://htc.anu.edu.au/pdfs/stuebng1

Suen, S. (2013). The education of women as a tool in development: Challenging the

African maxim. Hydra, 1(2), 60-76. Retrieved from

http://journals.ed.ac.uk/hydra/article/view/720 w

Sword, W. (1999). A socio‐ecological approach to understanding barriers to prenatal

care for women of low income. Journal of Advanced Nursing, 29(5), 1170-1177.

Syrjälä, A.M.H., Knuuttila, M.L.E., & Syrjälä, L.K. (2001). Self-efficacy perceptions in

oral health behavior. Acta Odontologica Scandinavica, 59, 1-6. Szumilas, M.

“Explaining Odds Ratios.” Journal of the Canadian Academy of Child and

Adolescent Psychiatry, 19.3 (2010): 227–229. Retrieved from

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2938757/

Tayebi, T., Zahrani, S. T., & Mohammadpour, R. (2013). Relationship between adequacy

of prenatal care utilization index and pregnancy outcomes. Iranian Journal of

Nursing and Midwifery Research, 18(5), 360. Retrieved from

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3877457/

The World Bank. (2014). World report: Teenage mothers. Retrieved from

http://data.worldbank.org/indicator/SP.MTR.1519.ZS

U.S. Department of Health and Human Services, National Institutes of Health, National

Library of Medicine (NLM). (2000). In Seiden, C.R., Zorn, M., Ratzan, S., et al;

eds. Health Literacy, January 1990 through 1999. February 2000; NLM Pub. No.

http://journals.ed.ac.uk/hydra/article/view/720

126

CBM 2000–1. Bethesda, MD: NLM. February 2000; NLM Pub. No. CBM 2000–

U.S. Dept. of Health and Human Services, Public Health Service, National Institutes of

Health, Office of Cancer Communications, National Cancer Institute. (1989).

Making health communication programs work: A planner’s guide. Retrieved from

http://www.cancer.gov/cancertopics/cancerlibrary/pinkbook/page1

Ujah, I. A. O., Aisien, O. A., Mutihir, J. T., Vanderjagt, D. J., Glew, R. H., & Uguru, V.

E. (2005). Factors contributing to maternal mortality in north-central Nigeria: A

seventeen-year review. African Journal of Reproductive Health, 9(3),27-40.

Retrieved from

https://tspace.library.utoronto.ca/bitstream/1807/4990/1/rh05037

United Nations Educational, Scientific and Cultural Organization. (2011). International

Standard Classification of Education-ISCED 2011: Unesco. Retrieved from

http://www.uis.unesco.org/Education/Documents/isced-2011-en

United Nations International Children’s Emergency Fund. (n.d.). Maternal child health.

Retrieved from http://www.unicef.org/nigeria/children_1926.html

United Nations International Children’s Emergency Fund. (2013). United Nations

Population FundUNFPA, World Bank (2012). Trends in maternal mortality: 1990

to 2013. Geneva: World Health Organization. Retrieved from

http://data.unicef.org/files/MMR2013

United Nations Population Fund (n.d. a). Maternal health. Retrieved from

http://www.unfpa.org/maternal-health

https://tspace.library.utoronto.ca/bitstream/1807/4990/1/rh05037

127

United Nations Population Fund. (n.d. b). Safe motherhood. Retrieved from

https://www.unfpa.org/emergencies/manual/3.htm#Annexes

United Nations. (2011). Discussion Paper: A social determinant approach to maternal

health. New York, NY: United Nations. Retrieved from

http://www.undp.org/content/dam/undp/library/Democratic%20Governance/Disc

ussion%20Paper%20MaternalHealth

United Nations. (2015a). Millennium summit (6-8 September 2000). Retrieved from

http://www.un.org/en/events/pastevents/millennium_summit.shtml

United Nations. (2015b). The millennium development goals report. Retrieved from

www.un.org/millenniumgoals

Veneman, A. M. (2007). Education is key to reducing child mortality: The link between

maternal health and education. UN Chronicle, 44(4), 58-59. Retrieved from

http://www.who.int/pmnch/topics/mdgs/2008unchronicle_aveneman

Wallace, L. S., Rogers, E. S., Roskos, S. E., Holiday, D. B., & Weiss, B. D. (2006). Brief

report: Screening items to identify patients with limited health literacy skills.

Journal of General Internal Medicine, 21(8), 874-877. Retrieved from

http://onlinelibrary.wiley.com/doi/10.1111/j.1525-1497.2006.00532.x/full

Widmer, M., Betran, A. P., Merialdi, M., Requejo, J., & Karpf, T. (2011). The role of

faith‐based organizations in maternal and newborn health care in

Africa. International Journal of Gynecology & Obstetrics, 114(3), 218-222.

Retrieved

https://www.unfpa.org/emergencies/manual/3.htm#Annexes

http://www.un.org/millenniumgoals

128

https://pdfs.semanticscholar.org/e5ab/8fd0a6cfce1e8465a3661a00ed9403250dff.p

Wiwa, K., & Böhm, M. (2014). Incidence and prevalence of pregnancy-related heart

disease. Cardiovascular Research, 101(4), 554-560. Retrieved from

http://cardiovascres.oxfordjournals.org/content/101/4/554

World Health Organization. (2002). WHO antenatal care randomized trial: Manual for

the implementation of the new model. WHO Programme to mark best

reproductive health practices. Geneva, Switzerland: World Health Organization.

Retrieved from http://whqlibdoc.who.int/hq/2001/WHO_RHR_01.30

World Health Organization. (2004). Beyond the numbers: Reviewing maternal deaths and

complications to make pregnancy safer. Retrieved from

http://apps.who.int/iris/bitstream/10665/42984/1/9241591838 ?ua=1

World Health Organization. (2006). Reproductive health indicator: Guidelines for their

generation, interpretation and analysis for global monitoring. Retrieved

https://extranet.who.int/iris/restricted/bitstream/10665/43185/1/924156315X_eng.

pdf

World Health Organization. (2014a). Health topics: Maternal health. Retrieved from

www.who.int/topics/maternal_health/en/

World Health Organization (2014b). Saving mothers’ lives. WHO infographic. Retrieved

from

http://www.who.int/reproductivehealth/publications/monitoring/infographic/en/

http://whqlibdoc.who.int/hq/2001/WHO_RHR_01.30

http://www.who.int/reproductivehealth/publications/monitoring/infographic/en/

129

World Health Organization. (2014c). Health statistics and information systems. Retrieved

from http://www.who.int/healthinfo/statistics/indmaternalmortality/en/

World Health Organization. (2016). Media Center: Maternal mortality. Retrieved from

http://www.who.int/mediacentre/factsheets/fs348/en/

World Health Organization & United Nations International Children’s Emergency Fund..

(2014). Trends in maternal mortality: 1990 to 2013: estimates by WHO, UNICEF,

UNFPA, The World Bank and the United Nations Population Division: executive

summary. Retrieved from

http://apps.who.int/iris/bitstream/10665/112697/1/WHO_RHR_14.13_eng

Zozulya, M. (2010). Maternal mortality in Nigeria: An indicator of women’s status.

Retrieved from

http://www.consultancyafrica.com/index.php?option=com_content&view=article

&id=358:maternal-mortality-in-nigeria-an-indicator-of-womens-

status&catid=59:gender-issues-discussion-papers&Itemid=267

http://www.who.int/healthinfo/statistics/indmaternalmortality/en/

http://www.who.int/mediacentre/factsheets/fs348/en/

130

Appendix A: Survey Instrument

Qualifying Question:

1. Are you currently pregnant? 0=no; 1=yes—If no, continue with survey.

Survey:

Please answer each of the following items based on your experience:

1. How often are appointment slips written in a way that is easy to read and

understand?

(1) Always (2)

Often (3) Sometimes (4)

Occasionally (5) Never

2. How often are medical forms written in a way that is easy to read and understand?

(1) Always (2) Often (3)

Sometimes (4) Occasionally (5) Never

3. How often are medication labels written in a way that is easy to read and

understand?

(1) Always (2)

Often (3)

Sometimes (4) Occasionally (5) Never

4. How often are patient educational materials written in a way that is easy to read

and understand? (1) Always (2) Often (3) Sometimes (4) Occasionally (5) Never

5. How often are hospital or clinic signs difficult to understand?

(1) Always (2) Often (3) Sometimes (4) Occasionally (5) Never

6. How often are appointment slips difficult to understand?

(1) Always (2) Often (3) Sometimes (4) Occasionally (5) Never

7. How often are medical forms difficult to understand and fill out?

(1) Always (2) Often (3) Sometimes (4) Occasionally (5) Never

8. How often are directions on medication bottles difficult to understand?

(1) Always (2) Often (3) Sometimes (4) Occasionally (5) Never

9. How often do you have difficulty understand written information your health care

provider (like a doctor, nurse, nurse practitioner) gives you?

(1) Always (2) Often (3) Sometimes (4) Occasionally (5) Never

10. How often do you have problems getting to your clinic appointments at the right

time because of difficulty understanding written instructions? (1) Always (2)

Often (3) Sometimes (4) Occasionally (5) Never

11. How often do you have problems completing medical forms because of difficulty

understanding the instructions? (1) Always (2) Often (3) Sometimes (4)

Occasionally (5) Never

12. How often do you have problems learning about your medical condition because

of difficulty understanding written information? (1) Always (2) Often (3)

Sometimes (4) Occasionally (5) Never

131

How often are you unsure of how to take your medication(s) correctly because of

problems understanding written instructions on the bottle label? (1) Always (2)

Often (3) Sometimes (4) Occasionally (5) Never

13. How confident are you filling out medical forms by yourself?

(1) Extremely (2) Quite a bit (3) Somewhat (4) A little bit (5)

Not at all

14. How confident are you filling out medical forms by yourself?

(1) Extremely (2) Quite a bit (3) Somewhat (4) A little bit (5) Not at all

15. How confident do you feel you are able to follow the instructions on the label of a

medication bottle? (1) Extremely (2) Quite a bit (3) Somewhat (4) A little bit (5)

Not at all

16. How often do you have someone (like a family member, friend, hospital/clinic

worker, or caregiver) help you read hospital materials? (1) Always (2) Often (3)

Sometimes (4) Occasionally (5) Never

For the following survey items please answer the question(s) thinking only about

your last pregnancy.

17. When was the last time you were pregnant?

0=1–11 months ago

1=12–23 months ago

2=24–35 months ago

3=36 months or more

18. Did you get any antenatal care during your pregnancy? (Antenatal care is the care

that a pregnant woman receives from organized health care services).

0=no

1=yes

2=not sure

If the answer to #18 (Did you get any antenatal care during your pregnancy?) is no,

display the following items (What was the main reason you get any antenatal care

during your pregnancy):

19. What was the main reason you get any antenatal care during your pregnancy?

0=I didn’t know I was pregnant

1=I didn’t know where to go for this type of care

2=I didn’t have enough money to pay for the care

3=I didn’t want anybody to know I was pregnant

4=I didn’t believe I needed this type of care as I had been pregnant before

5=I had no way of getting to a medical center for this type of care

6=This type of care is against my religion

132

7=Other. (allow open ended response)

If the answer to #18 (Did you get any antenatal care during your pregnancy?) is yes,

display the following items (#20, #21, #22, #23):

20. How far along in your pregnancy were you when you started antenatal care visit?

0=3 months or less

1=4 to 6 months

2=later than 6 months

21. How often did you go for antenatal visit?

0=Weekly

1=Every 2 weeks

2=Once per months

3=Every few months

4=Only when there were problems or I had questions

22. Approximately, how many antenatal visits did you make in total during your last

pregnancy? (1, 2, 3, 4, 5, 6, 7, 8, 9+=more than 9)

23. What information did you receive during your antenatal visits? (check all that
apply)
0=Health educational materials on sexual and reproductive health
1=Family planning and birth spacing information
2=Care of the newborn
3=What to expect during pregnancy
4=No information

Display the following question (Did you develop any health problems during your

pregnancy?) to all participants who have been pregnant (those who did and did not

get antenatal care)

24. Did you develop any health problems during your pregnancy? 0=no; 1=yes

If the answer to the previous item (Did you develop any health problems during your

pregnancy?) is yes, display the following item:

25. Which of the following were health problems you developed during your
pregnancy? (check all that apply)

0=Gestational diabetes

1=Depression

2=Heart problem

3=High blood pressure

4=Bleeding

133

5=Other health issues

26. Display the following questions (26-34) to all participants who have been

pregnant (those who did and did not get antenatal care) When your child was

born the child…

0=was a healthy baby

1=was premature

2=had a lower than normal birthweight

3=was stillborn

4=had another health issue

27. How old were you when you became pregnant with this child? (Give actual age)

28. What was your marital status when you became pregnant with this child?
0=Single

1=Ongoing dating relationship with father of child

1=Married

2=Separated

3=Divorced

4=Widowed

29. What was your highest level of education attained when you became pregnant
with this child?

0=No formal education

1= Primary school

2=Junior Secondary School

3=Secondary School certificate

4=Ordinary diploma or college of education certificate i.e. OND, NCE

5=College degree or above

30. What was your employment status when you became pregnant with this child?
0=No employment outside of home

1=Part time employment outside of home

2=Full time employment outside of home

31. What was your religion when you became pregnant with this child?
0=No religion

1=Christianity

2=Islam

3=Traditional

32. What was your household income when you became pregnant with this child?
0=Less than N25,000

134

1=N26,000– N50,000

2=N51,000–N75,000

3=More than N75,000

33. How many times have you been pregnant? (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,
14, 15)

How many children do you have (including the child of your last pregnancy, if alive)? (1,

2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)0

2 A

135

Appendix B: Permission to Use Instrument

E-mail Permission from Dr. Lisa Chew to use the questions on her survey

On Friday, September 25, 2015 10:25 AM, Lisa Chew < XXXXXXXX> wrote:

Yes, feel free to use the survey.

The above email may contain patient identifiable or confidential

information. Because email is not secure, please be aware of associated

risks of email transmission. If you are a patient, communicating to a UW

Medicine Provider via email implies your agreement to email communication;

see http://www.uwmedicine.org/Global/Compliance/EmailRisk.htm

The information is intended for the individual named above. If you are not

the intended recipient, any disclosure, copying, distribution or use of the

contents of this information is prohibited. Please notify the sender by

reply email, and then destroy all copies of the message and any attachments.

See our Notice of Privacy Practices at www.uwmedicine.org.

From: olubunmi adanri [XXXXXXXX]

Sent: Friday, September 25, 2015 8:24 AM

To: XXXXXXXX

Subject: Permission to use instrument

My name is Bunmi Adanri and I’m a doctoral student at Walden University. My

dissertation is on “The Association Between Poor Maternal Health Outcomes, Maternal

Characteristics, Antenatal Care Services Usage and Health Literacy in Lagos, Nigeria”.

I’m writing to request for permission to use 16 Health Literacy Screening Questions you

used in your study, “Brief Questions to Identify Patients with Inadequate Health

Literacy”, as a health literacy assessment tool for my research. To the best of my

knowledge, this would be the first to Use Existing Instrument

http://www.uwmedicine.org/Global/Compliance/EmailRisk.htm

http://www.uwmedicine.org/

136

Appendix C: Certificate of Completion of NIH Training

137

Appendix D: Permission Letter from Lagos State Government to Use the Map of Lagos

State

PhD Template

Walden University
Academic Residencies:

Basic Quantitative Research
PhD Residency 3 & 4

Session Learning Objectives
Identify the components of a methodology chapter for quantitative research
Understand criteria for selecting and justifying the research design, research question/hypotheses, data collection, and data analysis methods for the student’s dissertation study
Evaluate how the components of quantitative design and methods are aligned for dissertation research

Daniel W. Salter (DWS) – “how quantitative research aligns with the parts of the dissertation methodology chapter”?

Session Agenda
Overview
From Concepts to Variables
From Research Question to Hypotheses
From Recruitment to Data Collection
Analysis Plan Example
Report Out
Wrap Up

Quantitative Checklist-Chapter 3
Introduction
Research Design and Rationale
Methodology
Population
Sampling and Sampling Procedures
Recruitment, Participation, Data Collection
Instrumentation
Data Analysis Plan
Threats to Validity
Ethical Procedures

Quantitative Checklist
Quantitative Checklist can be found on Center for Research Quality website

Research Design and Rationale
Research design is the plan developed that best answers the research questions.
Two broad categories of research design:
Observational
Interventional
These categories identify the relationship between selected variables or the differences between groups.

Research Design and Rationale
Observational designs are non-experimental.
Cross sectional
Cohort
Case control
Interventional designs look at cause and effect relationships between variables.
Experimental studies
Quasi-experimental studies
Understanding which research design best answers the research questions provides the rationale for your design choice.

Research Design and Research Methodology
Research methodology describes the overall process of conducting research. Includes:
Research design
How the variables will be measured or operationalized
How the data will be collected
Sampling strategy
How the data will be analyzed
Each component of the research methodology needs to be fully discussed in Chapter 3.
Convenience is not a scholarly reason for the design.
It should not be the primary reason for your choice.

Daniel W. Salter (DWS [15]) – This slide seems out of order. Should be #6 in the deck.
Daniel W. Salter (DWS [16]) – I’ve never seen “convenience” used in this way. Its usually tied to sampling.

Instrumentation
Refers to the methods for collecting and measuring the variables in your study.
Preferable to use already published measures, as the reliability and validity have already been established.
Information needed about the measure includes the author, year published, what populations the measure has been used with, and the reliability and validity with each population.
Additional information includes whether permission is needed to utilize the measure.

Instrumentation: Reliability and Validity
Reliability refers to whether the instrument consistently measures what it was designed to measure.
Validity refers to the whether the instrument measures what it was designed to measure. There are different types of validity:
Face
Content
Construct

Finding the Right Instrument(s) for Your Study
How do you find the right instrument for your study?
Databases of instruments exist in some disciplines.
Don’t be afraid to look at databases outside your discipline, as many instruments can be cross-disciplinary.
Reviewing the literature to see what others have used is also a good source of information.
What information should you evaluate?
Year published, author, populations it’s been used with, length, reliability, and validity information.
If the measure contains subscales, knowing the number and names of these will be important.

Example Measure
Three-Dimensional Work Fatigue Inventory
18 items, 6 each that measure:
Physical Fatigue
Mental Fatigue
Emotional Fatigue

PsycTESTS Citation:
Frone, M. R., & Tidwell, M. O. (2015). Three-dimensional work fatigue inventory [Database record]. Retrieved from http://dx.doi.org/10.1037/t43185-000

Source:
Frone, M. R., & Tidwell, M. O. (2015). The meaning and measurement of work fatigue: Development and evaluation of the Three-Dimensional Work Fatigue Inventory (3D-WFI). Journal of Occupational Health Psychology, 20(3), 273–288. doi:10.1037/a0038700

Activity #1: Finding Instrumentation
Demonstration: Log in to the Walden Library at http://academicguides.waldenu.edu/library.
Demonstration of PsycTESTS
Demonstration of Mental Measurements Yearbook
Work individually or in small groups.
Use a test-specific database (i.e., PsycTESTS) or the Mental Measurement Yearbook.

Activity: Finding Instrumentation
Locate an instrument related to your area of research interest. Review the following information:
Year published, including any revised forms
Author
Populations it’s been used with
Reliability and validity information
Scales and subscales used
Next in this session, you will write a research question and hypothesis that might utilize this instrument.

Research Question to Hypotheses—1
Suppose you were interested in studying work fatigue, and you had a theoretical or empirical reason to expect that managers, techies, and regular staff within an organization would differ on each of the three types of work fatigue.
What would the research question and accompanying hypotheses look like?
Research Question: To what extent do managers, techies, and regular staff differ on physical work fatigue, mental work fatigue, and emotional work fatigue?

Research Question to Hypotheses—2
Suppose you also had a measure of “intent to leave the organization” and had a theoretical or empirical reason to expect that the three types of fatigue predict intent to leave.
Research Question: To what extent do managers, techies, and regular staff differ on physical work fatigue, mental work fatigue, and emotional work fatigue?

Research Question to Hypotheses—2
Null Hypothesis 1: The combined effect (Multiple R) of physical work fatigue, mental work fatigue, and emotional work fatigue will not predict intent to leave the organization.
Alternative Hypothesis 1: The combined effect (Multiple R) of physical work fatigue, mental work fatigue, and emotional work fatigue will predict intent to leave the organization.

Research Question to Hypotheses—2
Null Hypothesis 2: The unique effects (i.e., partial correlations) of physical work fatigue, mental work fatigue, and emotional work fatigue will be equal in predicting intent to leave the organization (all pairwise z-tests of Fisher transformed part correlations will be nonsignificant, p > .05)
Alternative Hypothesis 2: The unique effects (i.e., part correlations) of physical work fatigue, mental work fatigue, and emotional work fatigue will not be equal in predicting intent to leave the organization (at least one pairwise z-test of Fisher transformed part correlations will be significant, p < .05) * * Activity #2: Research Question and Hypothesis Work individually or in small groups. Draft a research question and hypothesis based on the instrument you found. * * Population and Sampling Sampling: The method used to select participants from the population. Your choice of sampling method determines how valid it will be to infer that your findings are true for the entire population. * * Population and Sampling The most externally valid sampling technique is: Free from bias (every individual has an equal chance of being selected); and Reliable (you will get the same results each time you study the population) Know the difference between: Probability sampling (random); and Non-probability sampling (convenience) * * Sampling and Sample Size Describe sampling time frame. Describe inclusion and exclusion criteria. What criteria would allow one to participate? Determine the number of participants needed to find a significant result or effect (i.e., sample size or power analysis). Consider the anticipated response rate. Be sure to justify the number of participants chosen. * * Recruitment Recruitment options Participant pool Online recruitment Flyers and other forms of community advertising Recruitment strategies Tailor recruitment to population and type of study Use community partners Provide incentives/compensation Address barriers including language, childcare, transportation * * From Research Question to Population and Sampling Research Question: To what extent do managers, techies, and regular staff differ on physical work fatigue, mental work fatigue, and emotional work fatigue? What is the population? How might you sample from the population? * * Activity #3: Population, Sampling, Recruitment In small groups, using your own research question: Identify your population. Identify your sampling strategy. Identify how you will recruit your sample group. * * Data Collection In-person data collection Advantages Disadvantages Online data collection Advantages Disadvantages Archival data (data that have already been collected) Advantages Disadvantages * * Internal Validity/External Validity Internal Validity: Addresses how well the study was done, how confident you are that the results you found are the “truth.” Control of confounders Information bias (e.g., misclassification) Presence of control group Effects of time (e.g., history, maturation), etc. External Validity: How generalizable are these results beyond the study population? Selection biases * * Ethical Considerations In this section of the proposal, you will outline your plans for ensuring the autonomy, privacy, and confidentiality of participants and corresponding protections for their data (e.g., informed consent procedures, data security). All student research at Walden must obtain Institutional Review Board (IRB) approval prior to data collection. IRB review and approval ensures the protection of human participants and their data. * * Data Analysis Plan Include the following components: Identify the software you are using for analysis. Describe data screening procedures. Describe plans for descriptive statistics, any planned recoding of variables. Restate the research hypotheses and, after each hypothesis, describe how you plan to analyze the data. * * Data Analysis Plan Example Data will be exported from SurveyMonkey to SPSS 23.0 for data analysis. Data will be screened for missing and outlier responses. Descriptive statistics will be calculated for all variables. * * Data Analysis Plan Example, continued A mean composite score for physical work fatigue will be computed and analyzed for reliability, as indexed by Cronbach’s α. Null Hypothesis 1: Managers, techies, and regular staff do not differ on physical work fatigue. Alternative Hypothesis 1: Managers, techies, and regular staff do differ on physical work fatigue. * * Data Analysis Plan Example, continued ANOVA post hoc tests using a Tukey HSD alpha adjustment will be examined for differences on physical work fatigue scores between managers and techies; managers and regular staff; and techies and regular staff. * * Report Out Any volunteers to share: Instrument found Research questions Research hypotheses Population and sampling strategy Analysis plan * * Wrap Up and Questions The purpose of this session was to describe the components of Chapter 3. Use the quantitative checklist to organize your chapter and complete the other sections. Quantitative Checklist can be found on the Center for Research Quality website. * Feedback Survey Please take 2 minutes right now to complete feedback for this session in the Residency App. *

METHODS (NATURE OF STUDY)

Profile of Accra Metropolitan Area

Study Area (Accra Metropolitan Area).

Study design: Unmatched case control study using primary and secondary data from patient Hospital records, next of kin (of cases), and controls selected from Ridge Regional and 37 Military hospitals.

Study Population/Inclusion Criteria

The population is all pregnant women who delivered in health facilities in the Accra Metropolitan Area in 2016.

Cases would be Maternal death that occurred in hospitals in (the Osu Klottey Sub Metro of) Accra Metropolitan Area in 2016.

Controls would be mothers who delivered at the Ridge Regional and 37 Military hospitals in 2016 that were alive at the end of the puerperal periods.

Exclusion Criteria

Maternal deaths outside Osu Klottey Sub Metro and mothers who delivered at health facilities in Accra Metro in 2016 but not at Ridge Regional or 37 Military hospitals and were alive at the end of the puerperal period.

Sampling Methods

Cases would be purposively selected from maternal death in the Osu Klottey Sub Metro of the Accra Metropolitan Area in 2016. Cases are defined as obstetric cases that died or declared dead upon arrival, or after admission (including those who died before the fetus was delivered). Controls would be randomly selected from mothers who delivered at Ridge and 37 Military hospital in 2016 and are alive at the end of puerperal period using simple random sampling.

Sampling Frame: The second post-natal attendants register in both hospitals would be used for control selections.

Sample Size Calculation

The sample size was calculated using EPI INFO STAT CALC. Version 7. Sample size of 189 (63 cases and 126 controls) was determined using 95% confidence interval , power of 80 percent, ratio of 2 control to a case, minimum odds ratio of 2.6.

Data Collection Technique and Tools

Medical records (folders) of all cases from 1st January 2016 to 31st December 2016 would be retrieved from the health facilities and reviewed. Consented next of kins of all cases would be interviewed using structured questionnaire. Medical records of consented and randomly selected controls from the participating hospitals would be retrieved and reviewed. The controls would also be interviewed using structured questionnaire. Health personnel who played significant roles in the management of a case or control would also be interviewed using structured questionnaire.

Data Quality Control

Pre-Testing of questionnaire

Questionnaires (9) would be pre-tested in the Korle Bu Teaching hospital. This would offer an opportunity for coding open ended questions based on the responses received.

Training of Research Assistants

Three research assistants would be recruited and trained to ensure the right information are ascertained from medical records and respondents.

Data Analysis

Data would be entered into EPI INFO 7. Stata/MP 3.0 would be used to describe, cross tabulate and determine association of socio-demographic and service delivery factors with maternal mortality.

Ethical Clearance

Ethical approval would be obtained from the Ghana Health Service review committee, Greater Accra Division. Written informed consent would be obtained from the participating Health Institution and participants. The objectives of the study would be explained and informed consent obtained from them before each interview and medical record review. Data obtained would be confidentially kept and used on for the purpose of the research.

Dissemination of Research Findings

The results will provide information which could inform policy to reduce Maternal Mortality in health institutions in Accra. The findings would be disseminated through institutional and regional workshop, policy brief and publication in scientific journal.

Turn in your highest-quality paper
Get a qualified writer to help you with

“ FOR RAY WRITTER ”

Get high-quality paper

Guarantee! All work is written by expert writers!

Still stressed from student homework?

Get quality assistance from academic writers!

Order now