Sexual Education, Sexual Initiation, and Contraceptive Use Among African American Young Women

Sexual Education, Sexual Initiation, and Contraceptive

Use Among African American Young Women

Chapter 1: Introduction to the Study

Introduction

A nation's health is often judged by its women and children's health, with

unintended pregnancy serving as a key indicator of a population’s reproductive health

(Finer & Zolna, 2014). Unintended pregnancy, whether unwanted or mistimed, is

associated with emotional, social, and economic outcomes which in turn may lead to

further disparities in morbidity and mortality (Carter, Kraft, Hock-Long, &

HatfieldTimajchy, 2013; James & Rashid, 2013; U.S. Department of Health & Human

Services. 2019). Health outcomes among mothers who experience unintended pregnancy

include delayed prenatal care, reduced breastfeeding, depression, delivery of infants with

low birth weight and congenital disabilities, and reduced average income compared to

other women (James & Rashid, 2013; USDHHS, 2019). Infants born due to unintended

pregnancies are at increased risk of poor mental, physical, and behavioral health in

adolescence and lower educational attainment (James & Rashid, 2013; USDHHS, 2019).

These health outcomes associated with unintended pregnancy suggest that reducing its

occurrence may substantially impact public health practices.

Within the United States, two groups among those at the highest risk of

unintended pregnancy are women aged 18–24 and those of African American descent

(USDHHS, 2019). African American women aged 18–24 are at higher risk for

unintended pregnancy and sexually transmitted infections (STIs) due to non-use,

inconsistent, or improper use of contraception, as well as failure to use dual contraceptive

methods (Craig, Dehlendorf, Borrero, Harper, & Rocca, 2014). These are topics covered

during formal sexual health education (FSHE). The purpose of this quantitative study was

to investigate the extent to which knowledge gained through FSHE impacts sexual

initiation and contraceptive use and selection among African American women aged 18–

24 in the United States who have sex with men.

An understanding of the impact of FSHE may help sexual health educators and

practitioners provide African American young women with the required knowledge and

skills to delay sexual initiation and consistently and properly use preference and intention

matched contraceptive methods. African American young women may apply knowledge

and skills to prevent unintended pregnancy, and by extension, may decrease adverse

maternal and child-associated public health outcomes. Doing so may lead to social

change by positively impacting African American young women's socioeconomic status,

and by extension, their children, which may improve associated social, educational, and

public health outcomes for both populations.

Covered in this is chapter is background information about gaps in public health

knowledge regarding African American young women’s sexual initiation and

contraceptive knowledge, selection, use, and unintended pregnancy. The problem

statement, research design, and methodology, which drove this research, will also be

discussed. Key terms are defined, and assumptions, limitations, and significance of the

study are identified and explored. This chapter is followed by the literature review in

Chapter 2 and the discussion of methodology in Chapter 3.

Background

FSHE promotes safer sexual behaviors for young adults, including delayed sexual

initiation and increased and correct contraceptive usage beginning with the first sexual

encounter (Lindberg & Maddow-Zimet, 2012). While teen and adolescent pregnancy and

FSHE in the United States has been the subject of national campaigns, unintended

pregnancy and FSHE in young adult women have received minimal attention. This lack

of attention may be due to the classification of this age group as adults and the false

assumption that they are equipped with knowledge regarding sexual and reproductive

health.

According to the Centers for Disease Control and Prevention (2019), out of the

88% of required U.S. high school students who received FSHE as a course, 76.3%

received education that abstinence as the best way to prevent human immunodeficiency

virus (HIV) and other STIs. In this same cohort, 35% learned how to use a condom

correctly, 52.7% received guidance on how to obtain condoms, and 54.9% learned about

the importance of dual contraceptive methods in preventing pregnancy and STIs (CDC,

2019). Among high schools that offered pregnancy prevention courses, the mean number

of required instruction hours was 4.2(CDC, 2019). Among high school course instructors,

12.6% received professional development on pregnancy prevention during the two years

before the study, while 14.2% wanted to receive professional development on topics

taught in their courses (CDC, 2019d).

This lack of comprehensive sexual health education (CSHE) translates into risky

sexual health behaviors. Nationally, 39% of U.S. high school students had sexual

intercourse, with the number being higher among African American students at 45.8%

(CDC, 2019). Among this cohort, 3.4% experienced sexual initiation before the age of 13

compared to 7.5% of African American students (CDC, 2019). Among this cohort, 46.2%

of sexually active high school students did not use a condom during their last sexual

encounter compared to 47.9% of African American students; 79.3% did not use birth

control pills compared to 86.8% of African American students; 95.9% did not use an

intrauterine device (IUD) or hormonal implant compared to 96.7% of African American

students; 95.3% did not use a hormonal shot, patch, or birth control ring to prevent

pregnancy or STIs compared to 94% of African American students; and 91.2% did not

use dual contraceptive methods compared to 77.5% of African American students (CDC,

2019).

More than 400 risk and protective factors are associated with pregnancy, sexual

behaviors, and contraception use in adolescents; however, their associations among

young adults are undetermined ((Kornides, Kistantas, Lindley, & Wu, 2015). It is

assumed that these factors and lack of contraceptive knowledge seen among high school

students carry over to young adulthood, where individuals explore their sexual freedom

and place themselves at risk for unintended pregnancy. Among college students who

received FSHE, 44.7% received pregnancy prevention information from their university

or college, and 49.5% did not use a contraceptive method the last time they had

intercourse (American College Health Association, 2017). Within this same cohort,

47.2% did not use birth control pills, 71.2% did not use an IUD, hormonal implant, shot,

patch, or birth control ring to prevent pregnancy, and 47.2% did not use dual

contraceptive methods (ACHA, 2017). College students predominantly used birth control

pills (58.2%), male condoms (63.1%), and withdrawal (35.1%) for pregnancy prevention

(ACHA, 2017). Based on data from these two surveys, contraceptive use is similar among

high school and college students surveyed, demonstrating opportunities and a need for

improved initial and supplemental sexual health education among college students.

Young adult pregnancies account for 55% of unintended pregnancies each year,

with one in six non-Hispanic Black women aged 20–24 experiencing an unintended

pregnancy (USDHHS, 2019). Trussell and Wynn (2008) suggested that the three best

strategies for reducing unintended pregnancy are to increase consistent and proper

contraceptive use, methods that do not require strict adherence, and dual contraceptive

methods. African American women aged 18–24 are at higher risk for unintended

pregnancy due to non-use of contraception, inconsistent or improper use of contraception,

and failure to use dual contraceptive methods (Craig et al., 2014). Efforts to reduce

unintended pregnancy should focus on increasing and improving FSHE regarding correct

usage and access to modern contraceptive services and methods, especially long-acting

reversible contraceptives (LARC), which do not require strict adherence (Tibaijuka et al.,

2017). Despite LARC methods being more cost effective, efficient, and not requiring

stricter adherence compared to more common short-acting contraceptives, fewer than 5%

of young adult women (Rubin, Felsher, Korich, & Jacobs, 2016; Tibaijuka et al., 2017).

LARC usage is lower among African American young women, which may be due to

medical mistrust resulting from historical reproductive and medical abuse (Jackson,

Karasek, Dehlendorf, & Foster, 2016).

Problem Statement

Despite recent contributions in contraceptive research, information involving

selection among African American young women has not been fully characterized.

Research examining contraceptive use among African American women has been limited

to either small qualitative studies or large surveys focusing on nonuse. Contraceptive

selection may be inhibited or facilitated by pregnancy intentions, physical and financial

access to methods, personal and vicarious experiences, and partner and social network

preferences. Underuse may be due to misinformation and misperceptions regarding

contraceptive efficacy, side effects, impact on future fertility, lack of social support

regarding contraceptive use, pregnancy ambivalence, and underestimations of pregnancy

risks (Craig et al., 2014). It also may be further exacerbated by historical reproductive

abuse and medical mistrust in communities of color. African American young women

report earlier sexual initiation and are more likely to experience an unwanted or mistimed

pregnancy than their peers (Moilanen, Leary, Watson, & Ottley, 2018). Women who

experience early sexual initiation are less likely to use contraception and are prone to an

increased risk for depression and lower self-esteem (Golden, Furman, & Collibee, 2016;

Lanier, Stewart, Schensul, & Guthrie, 2018). CSHE may have a protective influence on

early sexual initiation (Bourke, Boduszek, Kelleher, McBride, & Morgan, 2014).

Increasing access to CSHE, which highlights modern contraceptive methods, may

mitigate these factors.

The purpose of this study was to investigate the extent to which knowledge gained

through FSHE impacts sexual initiation and contraceptive use and selection among

African American women between 18 and 24 years old in the United States who have sex

with men. FSHE varies by state in terms of medical accuracy and comprehensiveness

regarding abstinence and contraceptive methods. Vicarious and historical experience may

also impact contraceptive use, contraceptive selection, and sexual initiation. Increasing

public health providers’ knowledge regarding the impact of FSHE on sexual initiation

and contraceptive use and selection among African American women between 18 and 24

years old in the United States may lead to the development of interventions designed to

prevent unintended pregnancies through increased consistent preference and intention

matched contraceptive use and improve maternal health and reduce adverse birth

outcomes, continued public health target areas among African

American young women.

Purpose of the Study

The nature of this quantitative cross-sectional analysis of the 2015–2017 National

Survey of Family Growth (NSFG) was to explore receipt of FSHE and the impact it had

on sexual initiation and contraceptive use and selection among African American women

aged 18–24 in the United States who have sex with men. This information may highlight

the importance and increase the provision of FSHE, which may raise the age of sexual

initiation, increase consistent use of preference and intention matched contraceptives, and

reduce contraceptive failures due to improper use.

Research Questions and Hypotheses

The research questions (RQ) for this study are as follows:

RQ1: Is there an association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men?

H01: There is no association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men.

H11: There is an association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men.

RQ2: Is there an association between the type of FSHE received and contraceptive

use at sexual initiation for sexually active African American women aged

18–24 who have sex with men?

H02: There is no association between the type of FSHE received and contraceptive

use at sexual initiation among sexually active African American women aged 18–24 who

have sex with men.

H12: There is an association between the type of FSHE received and contraceptive

use at sexual initiation among African American women aged 18–24 who have sex with

men.

RQ3: Is there an association between the type of FSHE received and the

effectiveness of contraceptive methods selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men?

H03: There is no association between the type of FSHE received and the

effectiveness of contraceptive methods selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men.

H13: There is an association between the type of FSHE received and the effectiveness

of the contraceptive method selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men.

RQ4: Is there an association between the type of FSHE received and the

effectiveness of contraceptive methods primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months?

H04: There is no association between the type of FSHE received and the

effectiveness of the contraceptive method primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months.

H14: There is an association between the type of FSHE received and the

effectiveness of contraceptive methods primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months.

Theoretical Foundation

The reasoned-action approach (RAA) drove this investigation regarding the extent

to which knowledge gained through FSHE impacts contraceptive selection, contraceptive

use, and sexual initiation among African American women aged 18–24 in the United

States who have sex with men. The RAA identifies personal, social, and control-related

factors that impact behavior (Hagger, Polet, & Lintunen, 2018). In this study, the RAA

lent insight into the attitudes, norms, and self-efficacy associated with contraception

intention and decision-making.

Attitudes toward contraceptive behavior include beliefs about benefits or

consequences of performing the behavior, often influenced by knowledge (Fishbein,

2008). Racial/ethnic disparities involving contraceptive knowledge exist, and there is an

association shown between lack of contraceptive knowledge and low contraceptive use.

Not having CSHE is associated with use of less effective contraceptive methods (LECM)

and increased unintended pregnancy (Kim, Dagher, & Chen, 2016).

Perceived social norms toward sexuality, contraceptive use, and pregnancy are

formed based on vicarious experiences. African American women are exposed to higher

teen and single parenthood rates, earlier sexual initiation, and use contraception less

effectively (Barber, Yarger, & Gatny, 2015). African American women who perceive

negative peer norms involving contraceptive use are more likely to be inconsistent in

contraceptive usage and use LECMs (Crosby et al., 2013). Carter et al. (2012) suggested

that anecdotal information and personal experiences may supersede formal education in

influencing African American young women’s contraception decision making.

Self-efficacy involving contraceptive behaviors includes availability of resources,

affordability, access, and skills needed to perform health behavior (Fishbein, 2008;

Nisson & Earl, 2016). Uninsured and underinsured women are more likely not to use

contraception or use LECM and be at increased risk for unintended pregnancy compared

to those with private insurance (Kim et al., 2016). CSHE increases self-efficacy and the

skill needed to select and use more effective contraception.

Conceptual Framework

The purpose of this study was to use the RAA to investigate the extent to which

knowledge gained through FSHE impacts sexual initiation and contraceptive use and

selection among African American women aged 18–24 in the United States who have sex

with men. The first step when using the RAA is to describe and define the behaviors of

interest (Fishbein, 2008). Sexual initiation and contraceptive use and selection, as

selfreported by African American young women who have sex with men, is the behavior

that is the focus of this study. For this study, barrier methods including condoms,

diaphragms, sponges, and spermicides are classified as less effective, with hormonal

methods including oral contraceptives, patch, ring, and injectables classified as

moderately effective (Dehlendorf, Kimport, Levy, & Steinauer, 2014). IUDs, implants,

and sterilization, male or female, are classified as highly effective (Dehlendorf et al.,

2014).

A behavior involves an action directed at a target or performed based on context

within a timeframe (Fishbein, 2008). For this study, I described and defined sexual

initiation and contraceptive use and selection (actions) based on sexual initiation and the

selection of LECMs, MECMs, and HECMs (targets), during vaginal intercourse

(context), within the past 12 months (timeframe; see Figure 1).

Figure 1. Conceptual framework of 2015–2017 NSFG variables. Adapted from “A

reasoned action approach to health promotion” by M. Fishbein, 2008, Medical Decision

Making, 28(6), 834–844.

Nature of Study

The NSFG provides the most comprehensive and nationally representative

assessment of reproductive health in the United States. The NSFG gathers information on

family life, cohabitation, marriage and divorce, pregnancy, and infertility, use of

contraception, medical care, and general and reproductive health. NSFG participants were

civilian noninstitutionalized individuals who were 15 to 49 years of age in the United

States. The NSFG is a continuous, computer-assisted survey used to measure

reproductive health status to determine the need for and effectiveness of health education

programs and provide data on American families.

The 2015–2017 data set includes responses collected from 10,094 participants, of

which 5,554 were women (CDC, 2019, 2019c). I selected the NSFG as the secondary

data set for this study due to its oversampling of African American women and

individuals aged 15–24 (CDC, 2019b, 2019c). For this study, I analyzed data from the

2015–2017 NSFG to evaluate the extent to which FSHE (knowledge; main independent

variable) influences sexual initiation (dependent variable) and the selection of LECMs,

MECMs, and HECMs (dependent variables) among African American women aged 18–

24 who have sex with men. I analyzed FSHE, sexual initiation, and contraceptive use and

section data among African American women aged 18–24 who have sex with men from

the NSFG via descriptive statistics, logistic regression, and multinomial logistic

regression.

Definition of Terms

The literature uses various terms to describe African American young women and

their knowledge, attitudes, and beliefs involving FSHE, sexual initiation, and

contraceptive selection and use.

Abstinence-centered sexual health education (ACSHE): A form of education

which involves teaching abstinence until marriage as the only morally acceptable option

and the only way to prevent premarital pregnancy and STIs (Boyer, 2018).

Abstinence-plus sexual health education (APSHE): A form of education that

stresses abstinence as the best way to prevent pregnancy and STIs but includes

information about contraceptives, including condoms (Boyer, 2018).

Behavioral background/distal factors: Demographic, economic, personality,

attitudinal, and other individual factors that play a role in shaping beliefs and behaviors

(Fishbein, 2008).

Behavioral beliefs: Attitudes, including beliefs and feelings about behaviors and

values attached to outcomes of behaviors (Fishbein, 2008).

Behavioral intention: Readiness to engage in a particular behavior (Fishbein,

2008).

Comprehensive sexual health education (CSHE): A form of education that

involves teaching abstinence as the best method for preventing STIs and unintended

pregnancy; it also covers condoms and other contraceptive methods with demonstrations

(Boyer, 2018).

Emerging adulthood: A distinct period marked by exploration, development of

independence, and personal decision-making. During this period, individuals make key

decisions regarding romantic and sexual relations, which are heavily influenced by social

norms (Olmstead, Koon, Puhlman, Pasley, & Fincham, 2013).

Highly effective contraceptive methods (HECM): IUDs and hormonal implants

(Dehlendorf et al., 2014).

Less effective contraceptive methods (LECM): Condoms, diaphragms, sponges,

and spermicides, including foam and suppositories (Dehlendorf et al., 2014).

Moderately effective contraceptive methods (MECM): Hormonal methods,

including oral contraceptives, patches, rings, and injectables (Dehlendorf et al., 2014).

Mistimed pregnancy: What a woman experiences if conception occurs sooner than

intended (Levi & Quang Dau, 2011).

Normative beliefs: Subjective and social norms resulting from behavior approval

or disapproval and motivation to comply with beliefs (Fishbein, 2008).

Perceived behavior control/Self-efficacy: Perceptions regarding the ability to

perform a behavior outside of individual control. This includes external factors such as

social support and availability of resources, including affordability and access, as well as

internal factors such as skill/availability and self-efficacy (Fishbein, 2008; Nisson & Earl,

2016).

Permanent contraceptive methods: These methods include tubal ligation and

vasectomy (Tibaijuka et al., 2017).

Reasoned action approach (RAA): Encompasses the theory of reasoned action

(TRA) and planned behavior (TPB) to identify personal, social, and control-related

factors that impact behavior (Hagger et al., 2018).

Sexual initiation: The first sexual experience typically defined as the age of first

vaginal intercourse. (Goldberg & Halpern, 2017).

Short-acting contraceptives: These include oral contraceptive pills, condoms,

spermicides, and injectable hormones.

Unintended pregnancy: A pregnancy that is not planned (Levi & Quang Dau,

2011).

Unwanted pregnancy: A pregnancy experienced by a woman who has no desire to

become pregnant. (Levi & Quang Dau, 2011).

Vicarious experiences: Stories of close family and friends that are valued as if

they are one’s own.

Volitional control: How an individual exercises control over behavior (Fishbein,

2008).

Scope and Delimitations

This research involved investigating possible associations between African

American young women’s exposure to FSHE, sexual initiation, and their selection of

LECMs, MECMs, and HECMs. I selected the NSFG to address this area of research as it

provides information on a national probability sample of civilian noninstitutionalized

women between 15 and 49 years of old. It also provides information on family life,

cohabitation, marriage and divorce, pregnancy, and infertility, use of contraception,

medical care, and general and reproductive health. This study only included self-reported,

sexually active African American women aged 18 to 24 who have sex with men and

received FSHE.

Assumptions and Limitations

The purpose of this quantitative cross-sectional analysis of the 2015–2017 NSFG

was to explore the receipt of FSHE and the impact it may have on sexual initiation and

contraceptive use and selection among African American women aged 18–24 who have

sex with men. The NSFG provides general and reproductive health information on a

national probability sample of civilian, noninstitutionalized women age 15 to 49. As such,

when weighted to reflect the United States female population at the interviewing

midpoint, generalizations for the U.S. population may be made (CDC, 2019b, 2019c).

A limitation of this research may be possible non-sampling errors due to the cross-

sectional nature of the NSFG and primary data collection. As such, I could not correlate

the extent to which knowledge impacts contraceptive selection, nor can causal inferences

be drawn from these cross-sectional data. Another limitation of this study involves using

self-reported data, which is subject to participant recall and interviewer biases and over-

and under-reporting. The administration of the majority of the NSFG responses occurred

via audio computer-assisted self-interviewing (ACASI) to reduce limitations and biases

(CDC, 2019b, 2019c).

The sample size was also a limitation of this study. Although oversampling of the

study population occurred to include more African American women, the sample size

was still small; only 1,355 female interviews out of 5,554 were with non-Hispanic Black

women. Another limitation involved the sample size for this study, which included a

subsample from the NSFG containing 499 sexually active, African American women

aged 18–24 who have sex with men. Data cleaning yielded a final sample size of 242

sexually active, African American women aged 18–24 who have sex with men. This

small sample size highlights the need for more robust and targeted studies among African

American women instead of limited qualitative and small studies focusing on nonuse or

large surveys providing percentages of African American women respondents (Mosher et

al., 2015).

Access to data needed to complete this study was another limitation. The NSFG

includes a restricted variable, REGION, that identifies a respondent’s region of residence.

The NSFG website listed the REGION variable as available free of charge upon written

request via encrypted email and confidentiality forms. I followed this procedure but was

informed that access to the REGION variable changed from free to a fee-based model

with limited access at secure onsite facilities, which made the inclusion of this variable

cost-prohibitive. The REGION variable was a crucial part of my initially proposed study

as FSHE requirements vary by state and region. Currently, 24 states and the District of

Columbia mandate sex education, and only 13 require that instruction be medically

accurate (Guttmacher Institute, 2019). Eight states require that instruction be culturally

competent and not biased against any race, sex, or ethnicity, and two states prohibit

promoting religion (Guttmacher Institute, 2019). Twenty-seven states require that

abstinence be stressed, 10 states required that it be covered, and only 18 states and the

District of Columbia require information on contraception (Guttmacher Institute, 2019).

FSHE in southern states overwhelmingly requires that abstinence be stressed or covered

compared to other regions. The south comprises states with historically restrictive

reproductive freedom and justice issues, including eugenics, that do not require FSHE to

be culturally competent or non-biased.

Significance

African American women may receive lower quality family planning care,

contributing to disparities in contraceptive use (Craig et al., 2014). African American

young women report earlier sexual initiation, use LECMs, and are more likely to

experience an unwanted or mistimed pregnancy compared to their peers (Golden et al.,

2016; Lanier et al., 2018; Moilanen et al., 2018). Unintended pregnancy can lead to

significant emotional, social, health, and financial complications, which may perpetuate

cycles of disadvantages experienced by vulnerable populations, especially involving

racial disparities in morbidity and mortality later in life. Bourke et al. (2014) found that

CSHE may have a protective influence on early sexual initiation and unintended

pregnancy. Increasing access to CSHE, which highlights modern contraceptive methods,

may raise the age of sexual initiation, and reduce factors that lead to unintended

pregnancy.

The results found from this study may lead to positive social change by providing

sexual health educators and public health practitioners with an understanding of the

impact of FSHE on sexual initiation and contraceptive use and selection among African

American young women who have sex with men. Their increased understanding of the

impact of FSHE may help sexual health educators, and public health practitioners provide

African American young women with knowledge and skills needed to delay sexual

initiation and consistently and properly use preference and intention matched

contraceptive methods. African American young women may apply knowledge and skills

to prevent unintended pregnancy and decrease adverse maternal and child public health

outcomes. Doing so may positively impact African American young women and

children's socioeconomic status, which may lead to improved associated social,

educational, and public health outcomes for both populations.

Summary

Despite a national decline in pregnancy rates, unintended pregnancies in African

American young women remain high and are a public health concern. Even with recent

contraceptive research contributions, information involving selection among African

American young women has not been fully determined. Despite LARC methods being

more cost effective, efficient, and not requiring strict adherence compared to more

common short-acting contraceptives, few young adult women use them (Rubin et al.,

2016). Usage is low in African American young women, which may be due to medical

mistrust resulting from historical reproductive and medical abuse (Jackson et al., 2016).

Underuse may be due to misinformation and misperceptions regarding contraceptive

efficacy, side effects, impact on future fertility, social support of contraceptive use,

pregnancy ambivalence, and underestimating pregnancy risk (Craig et al., 2014).

Understanding contraceptive selection among African American young women

may assist sexual health educators and public health practitioners with a better

understanding of the impact of FSHE on contraceptive use and selection and sexual

initiation among African American young women. Results from this study may lead to

positive social change by providing African American young women with knowledge and

skills needed to delay sexual initiation and consistently and correctly use preference and

intention matched methods and reduce the occurrence of unplanned pregnancies in this

population. This information may be used to improve the provision of culturally

competent services, such as sexuality and reproductive education, family planning, and

contraceptive counseling, which may increase selection and consistent use of

contraceptive methods, reduce usage failures, and improve birth-related outcomes.

Provided in Chapter 2 is a review of the literature regarding FSHE, knowledge,

attitudes, and beliefs involving contraceptive use and selection and sexual initiation

among African American young women. Provided in Chapter 3 are details regarding how

data from the NSFG was used to investigate the extent to which knowledge gained

through FSHE impacts sexual initiation and contraceptive use and selection among

African American young women who have sex with men. Presented and interpreted in

Chapter 4 are the results of the statistical analysis techniques discussed in Chapter 3. In

Chapter 5, the study findings are compared to existing empirical literature, and

implications and proposed solutions are discussed.

Chapter 2: Literature Review

Introduction

Unintended pregnancy is associated with adverse emotional, social, economic, and

health outcomes, making it a substantial public health concern. Unintended pregnancy

may lead to disparities in morbidity and mortality due to delayed prenatal care, reduced

breastfeeding, depression, delivery of infants with low birth weight or birth defects, as

well as reduced average income compared to other women (Carter et al., 2013; James &

Rashid, 2013; USDHHS 2019). Infants born due to unintended pregnancies experience

increased risks of poor mental, physical, and behavioral health in adolescence and lower

educational attainment compared to Caucasian infants (James & Rashid, 2013; USDHHS,

2019).

African American women aged 18–24 are at higher risk for unintended pregnancy

due to non-use, inconsistent, or improper use of contraception, as well as failure to use

dual contraceptive methods (Craig et al., 2014). These are topics covered during FSHE.

The purpose of this study was to investigate the extent to which knowledge gained

through FSHE impacts sexual initiation and contraceptive use and selection among

African American women aged 18–24 in the United States who have sex with men. This

information may be used to increase the provision of culturally competent sexuality and

reproductive education, family planning, and contraceptive counseling, which may

increase the selection and consistent use of contraceptive methods. This information may

also reduce contraceptive usage failures, raise the age of sexual initiation, and improve

birth related outcomes.

This chapter reviews literature involving FSHE, contraceptive knowledge, beliefs,

decision-making and selection, unintended pregnancy, social and perceived norms, and

sexual initiation associated with African American young. Also, I discuss the RAA,

which successfully predicts and explains increased consistent contraceptive use.

Literature Search Strategy

I conducted a digital literature search through PUBMED, ProQuest Central,

SAGE Journals, and ScienceDirect databases using the following search terms: formal

sexual health education, abstinence-centered, abstinence-plus, and comprehensive sexual

health education, eugenics and historical reproductive justice, contraceptive knowledge,

beliefs, and attitudes, sexual initiation barriers to contraceptive use and access, condoms

and other contraceptives used among teens, young adults, and African American,

perceived and actual contraceptive side effects, and contraceptive social norms. I also

searched for unintended pregnancy, mistimed pregnancy, unwanted pregnancy, older teen

pregnancies, young adult or college pregnancies, young mother, and birth outcomes.

Theoretical Foundation: Reasoned Action Approach The RAA

drove this investigation regarding the extent to which knowledge gained through FSHE

impacts sexual initiation and contraceptive use and selection among African American

women aged 18–24 in the United States who have sex with men. The RAA identifies

personal, social, and control related factors that impact behavior. The RAA integrates

factors from the theory of reasoned action (TRA) and planned behavior (TPB),

focusing on individual motivational factors as determinants of behavior. The TRA and

TPB assume that the best predictor of behavior is intention determined by attitudes and

social norms (Hagger et al., 2018). The TPB includes an additional factor, perceived

control over one’s ability to perform the behavior. Both frameworks have been

successful used by previous researchers to investigate behavioral beliefs such as

attitudes, intentions, and normative beliefs, including vicarious experiences and

subjective norms and their influence on intention (Tyson, Covey, & Rosenthal, 2014).

The TRA asserts that the most important determinant of behavior is behavioral

intention. Through learning, individuals form behavioral beliefs and attitudes. Through

experiences and perceived normative beliefs, intentions are influenced, and individual

behaviors are formed (Fishbein, 2008). Intentions are viewed as readiness to engage in a

behavior (Fishbein, 2008). Attitudes are beliefs and feelings about behaviors and the

values attached to the outcomes of those behaviors. Normative beliefs are subjective and

social norms, including behavior approval or disapproval and motivation to comply with

these beliefs. The TRA involves considering the role that demographic, economic,

personality, attitudinal, and individual factors play in shaping beliefs and behaviors

(Fishbein, 2008). These factors and knowledge are considered background/distal factors

that may influence behavioral or normative beliefs through underlying attitudes and

norms.

The ability to which the TRA can explain behavior depends upon volitional

control. The TPB expounds on TRA factors by suggesting that along with attitude toward

the behavior and subjective norms, perceived behavior control/self-efficacy influences

intention. The TPB is widely used when investigating sexual behaviors as it predicts

behavioral intentions influenced by attitudes, norms, and perceived behavioral control.

Behavior control is a person’s perception regarding his or her ability to perform a

behavior that may be outside of his or her control. Control beliefs include external factors

such as social support and availability of resources, including affordability and access, as

well as internal factors such as skill/availability and self-efficacy.

For this research, I used the RAA, which combines TRA and TBP factors most

successful in impacting behavior (see Figure 2).

Figure 2. Model of a RAA. Reproduced from “A reasoned action approach to health

promotion” by M. Fishbein, 2008, Medical Decision Making, 28(6), 834–844.

Attitudes

Background factors Norms Intention Behavior

Self-Efficacy

Reasoned Action Approach and Contraceptive Selection and Use

According to Hoopes, Gilmore, Cady, Akers, and Ahrens (2016), five key themes

influence contraceptive choices: (a) preferences regarding contraceptive characteristics,

(b) contraceptive knowledge gained from peers, family, or health professionals; (c) gaps

in contraceptive knowledge, (d) preferences for effective or long-acting contraceptives,

and (e) barriers and support including partner approval, social norms, and health care

environment. These themes coincide with the RAA factors that focus on attitudes, social

norms, perceived behavioral control, or self-efficacy. Interventions utilizing the RAA

demonstrate that changing attitudes and subjective norms impact intentions to use

contraceptives, leading to behavior change (Dippel, Hanson, McMahon, Griese, &

Kenyon, 2017). In this study, the RAA lent insight into the attitudes, norms, and

selfefficacy associated with contraception intention and decision-making.

Attitudes toward contraceptive behavior include beliefs about benefits or

consequences of performing the behavior, often influenced by knowledge. Attitudes

regarding why young women do not use or discontinue use of contraceptive methods

involve concerns about weight gain, acne, impact on menstruation, invasiveness, and side

effects (Marshall, Kandahari, & Raine-Bennett, 2018). Attitudes that lead to

contraceptive use include ease of use, fewer side effects, low cost, and easy access.

Knowledge shapes contraceptive attitudes (Guzzo & Hayford, 2018). Racial and ethnic

disparities involving contraceptive knowledge exist with associations shown between less

knowledge and less contraceptive use. Not having CSHE is associated with use of

LECMs and increased unintended pregnancy (Kim et al., 2016). Teal and Romer (2013)

found that 70% had not heard of LARCs in a multiethnic cohort of young women.

Bachorik et al. (2015) found that 40% of young women were familiar with contraceptive

implants. Misconceptions involving LARCs remain a factor in use disparities among

African American young women. Common misconceptions about LARCs include the

impact on future fertility, overestimating associated risk, underestimating oral

contraceptive risk, and overestimating oral contraceptive effectiveness (Sundstrom,

Baker-Whitcomb, & DeMaria, 2015).

Perceived social norms toward sexuality, contraceptive use, and pregnancy are

formed based on familial experiences, which vary by race. African American women are

exposed to higher teen and single parenthood rates, earlier sexual initiation, and use

contraceptives less effectively (Barber et al., 2015). Barber et al. (2015) suggested that

exposure to these factors may cause cognitive dissonance and perpetuate these behaviors.

Anderson, Steinauer, Valenete, Koblentz, and Dehlendorf (2014) found that among

diverse women, negative information regarding LARCs learned from conversations with

female friends and family members and television commercials impacted decision

making. Oral contraceptives are reported as the most acceptable form of contraception

and LARCs, including intrauterine devices, are rarely used (Pritt, Norris, & Berlan,

2017). African American women who perceive negative peer norms surrounding

contraceptive use are more likely to have inconsistent contraceptive usage and use

LECMs (Crosby et al., 2013). Cohen, Sheeder, Kane, and Teal (2017) suggested that

knowledge regarding contraceptive methods alone is insufficient to increase use since

African American women valued anecdotal information just as favorably as factual

information. As such, the stronger predictor for contraceptive method choice in their

study was knowing someone who used, liked, or disliked a particular method. Marshall et

al. (2018) also noted the importance of experiential information from friends and family

in contraceptive decision-making. They described healthcare providers as both trusted

and potentially biased sources of factual only information and noted that there is often

skepticism regarding providers’ ability to provide unbiased reproductive

recommendations. In some instances, participants reported that they made contraceptive

decisions without seeking social support, preferring to learn from personal negative

contraceptive experiences instead of relying on the experiences of their social

communities (Paterno, Hayat, Wenzel, & Campbell, 2017). Anecdotal information and

personal experiences might supersede formal education for African American young

women (Carter et al., 2012).

Self-efficacy involving contraceptive behaviors includes the availability of

resources, including affordability and access, and the skill needed to perform the behavior

(Fishbein, 2008; Nisson & Earl, 2016). Uninsured and underinsured women are more

likely not to use contraception or to use LECMs and be at increased risk for unintended

pregnancy than those with private insurance (Kim et al., 2016). Young women not

covered by a parent’s insurance plan might opt out of electing insurance coverage and

remain uninsured (Pritt et al., 2017). Insured women might not be aware of the benefits

centered around contraceptive costs (Pritt et al., 2017). The costs of LECMs might

influence clinics’ ability to stock them, which could affect patients’ timely access to

contraception (Pritt et al., 2017). Reproductive trends in contraceptive use among African

American women may be impacted by a reproductive history of medical and reproductive

abuse (Jackson et al., 2016). Young women, especially African Americans, may be at risk

of not receiving high-quality family planning care (Dehlendorf et al., 2014). African

American women report experiencing implicit and explicit bias in the form of feeling

coerced to start contraceptive use earlier than desired, written prescriptions provided

without request or consent, refusal to remove implanted methods when requested (Gomez

& Wapman, 2017). They also report feeling as if they could not turn down provider

recommendations and being steered toward more long-acting methods without

considering their reproductive preferences and priorities (Gomez & Wapman, 2017).

Contraceptive dissatisfaction and lack of knowledge regarding alternatives may also

contribute to nonuse or incorrect use of current contraceptive methods (Kornides et al.,

2015). The RAA has been used successfully to investigate self-efficacy associated with

contraceptive use and partner reaction, especially regarding contraception negotiation

(Villarruel et al., 2004). CSHE increases self-efficacy and the skill needed to select and

use more effective contraception.

Conceptual Framework

The purpose of this study was to use a RAA to investigate the extent to which

knowledge gained through FSHE impacts sexual initiation and contraceptive use and

selection among African American women aged 18–24 who have sex with men. The first

step in a RAA is to describe and define the behaviors of interest. The behavior that was

the focus of this study is contraceptive methods selection by African American young

women. For this study, barrier methods including condoms, diaphragms, sponges, and

spermicides have been classified as less effective, with hormonal methods including oral

contraceptives, the patch, the vaginal ring, and injectables classified as moderately

effective (Dehlendorf et al., 2014). IUDs, implants, and sterilization have been classified

as highly effective (Dehlendorf et al., 2014).

In the RAA, behavior involves an action directed at a target, performed based on

context, within a timeframe (Fishbein, 2008). For this study, these constructs are

described and defined as contraceptive selection (action) based on the selection LECMs,

MECMs, and HECMs (target), during vaginal intercourse (context), within the past 12

months (time; see Figure 1).

Literature Review

Formal Sexual Health Education

FSHE is comprised of an evidence based curriculum delivered in a school or

healthcare setting. FSHE may be based on different methods, including an

abstinencecentered, abstinence-plus, or comprehensive approach. ACSHE, often referred

to as sexual risk avoidance programs (SRAP), teaches abstinence until marriage as the

only morally acceptable option before marriage (Boyer, 2018). Participants are taught

that abstinence is the only way certain to prevent premarital pregnancy and STIs and that

sex within a monogamous relationship in the context of marriage is the expected standard

(Boyer, 2018). ACSHE programs also teach that sexual activity outside of marriage is

likely to have harmful psychological and physical effects and impact independence

(Boyer, 2018). ACSHE programs do not include demonstrations, simulations, or the

distribution of contraceptive devices (Boyer, 2018). ACSHE programs typically fail to

provide education and skill-building on critical topics such as healthy relationships,

communication, and consent (Boyer, 2018).

APSHE stresses abstinence as the best way to prevent pregnancy and sexually

transmitted infections but emphasizes safe sex practices, healthy relationships, and

lifestyles by including information about contraception and condoms. CSHE teaches

abstinence as the best method for preventing STIs and unintended pregnancy but also

teaches participants about condoms and other contraceptive methods with demonstrations

(Boyer, 2018). It helps young people explore their values, goals, and options by teaching

interpersonal and communication skills (Boyer, 2018).

Despite concerns, there is no evidence to support that receipt of CSHE is

associated with an earlier sexual initiation, increased risk-taking, or poorer sexual

reproductive health outcomes (Lindberg & Maddow-Zimet, 2012). CSHE, which occurs

before sexual initiation, leads to delayed onset of sexual initiation, greater use of more

effective contraception, especially at the first sexual encounter, and healthier sexual

relationships (Jaramillo, Buhi, Elder, & Corliss, 2017; Lindberg & Maddow-Zimet,

2012). ACSHE does not have a significant impact on sexual initiation or sexual behaviors

and may place young people at increased likelihood of pregnancy and STIs once they

become sexually active by overlooking or downplaying the benefits of contraception

(Boyer, 2018). ACSHE also fails to meet the needs of young people who are already

sexually active, withholding potentially lifesaving information (Boyer, 2018). ACSHE

also perpetuates discrimination and stigmatizes sex, promoting fear, guilt, and shame

around sex (Boyer, 2018). It also ignores systemic biases such as racism, inequality,

discrimination, and trauma, which may impact the sexual health of African American

women (Boyer, 2018). Public health professionals widely reject ACSHE and SRAP

(Santelli et al., 2017).

The goal of FSHE is to prepare young adults to be sexually healthy (Santelli et al.,

2017). Before 2008, federally funded educational programs focused solely on ACSHE

programs, prohibiting discussing contraceptive methods, except to emphasize their failure

rates (Jaramillo et al., 2017; Lindberg & Maddow-Zimet, 2012). In 2010, a reduction in

funding for ACSHE occurred, and the Obama Administration increased CSHE (Jaramillo

et al., 2017). The current administration reverted the content of FSHE to ACSHE without

scientific evidence supporting their effectiveness. As such, the requirements for FSHE

vary by state and region. Currently, 24 states and the District of Columbia mandate sex

education, of which only 13 require that instruction be medically accurate (Guttmacher

Institute, 2019). Eight states require that instruction be culturally competent and not

biased against any race, sex, or ethnicity, and two states prohibit promoting religion

(Guttmacher Institute, 2019). Twenty-seven states require that abstinence be stressed, and

10 required that it be covered, while only 18 states and the District of Columbia require

information on contraception (Guttmacher Institute, 2019, see Table 1). As noted, Region

3, comprised of southern States, overwhelmingly requires that abstinence be stressed or

covered compared to other regions. There are no states in Region 3, historically known

for restrictive reproductive freedom and justice issues, including eugenics, that require

FSHE that is culturally competent and non-biased.

Racial and ethnic disparities involving contraceptive knowledge exist with an

association shown between less knowledge and less contraceptive use. FSHE, for most, is

received through school based programs, which may not be comprehensive, and often in

the South has moral undertones (Guzzo & Hayford, 2018). Receipt of FSHE was lower

among individuals with lower socioeconomic status, who were African American, and

whose mothers had lower educational attainment (Lindberg & Maddow-Zimet, 2012).

Receipt of FSHE after high school is rare. Therefore, knowledge and attitudes formed

about contraception during adolescence may persist throughout life (Guzzo & Hayford,

2018). This highlights the need for CSHE among African American young women who

have poorer sexual reproductive health outcomes, including higher rates of unintended

pregnancy and STIs (Lindberg & Maddow-Zimet, 2012). Shepherd, Sly, and Girard

(2017) found that African American adolescents who received CSHE had more favorable

attitudes about contraception, demonstrated an increase in consistent contraception usage,

and reported less teen pregnancy and vaginal intercourse, highlighting the importance of

early CSHE.

Table 1

U.S. Census Bureau Regions and Divisions with State FIPS Codes and FSHE

Requirements

United States Census Bureau Regions and Divisions with State FIPS Codes

Regional I: Northeast

States Requirement Codes

Connecticut (09)

Maine (23) MSE, MA, AS, CE

Massachusetts (25)

New Hampshire (33)

New Jersey (34) MSE, MA, CC, AS, CE

New York (36)

Pennsylvania (42)

Rhode Island (44) MSE, MA, CC, AS, CE Vermont

(50) MSE, AC, CE

Region 2: Midwest

States Requirement Codes

Illinois (17) MA, AS, CE Indiana

(18) AS

Iowa (19) MSE, MA, CC

Kansas (20)

Michigan (26) MA, AS

Minnesota (27) MSE, AC Missouri

(29)

Nebraska (31)

North Dakota (38) MSE

Ohio (39) MSE, AS

South Dakota (46)

Wisconsin (55) AS

(table continues)

United States Census Bureau Regions and Divisions with State FIPS Codes

Region 3: South

States Requirement Codes

Alabama (01) AS, CE

Arkansas (05) AS

Delaware (10) MSE, AS, CE

District of Columbia (11) MSE, CE Florida

(12) AS

Georgia (13) MSE, AS

Kentucky (21) MSE, AS

Louisiana (22) AS

Maryland (24) MSE, AC, CE

Mississippi (28) MSE, AS

North Carolina (37) MSE, MA, AS, CE

Oklahoma (40) AS

South Carolina (45) MSE, AS, CE

Tennessee (47) MSE, AS

Texas (48) AS

Virginia (51) AC

West Virginia (54) MSE, AC, CE

Region 4: West

States Requirement Codes

Alaska (02)

Arizona (04) AS

California (06) MSE, MA, CC, AC, CE

Colorado (08) MA, CC, AC, CE Hawaii

(15) MSE, MA, AC, CE

Idaho (16)

Montana (30) MSE, AC

Nevada (32) MSE

New Mexico (35) MSE, AC, CE

Oregon (41) MSE, MA, CC, AS, CE

Utah (49) MSE, MA, CC, AS

Washington (53) MA, CC, AS, CE

Wyoming (56) Legend:

Mandated Sex Education (MSE)

Required Medically Accuracy (MA)

Required Culturally Competent and Non-biased instruction (CC)

Abstinence Stressed (AS)

Abstinence Covered (AC)

Required Contraception Education (CE)

Contraceptive Use and Selection

Demographics. Timing and sequencing of experiences across the lifespan are

important as young women make choices based on their experiences. Young women’s

decisions about contraception depend on their life experiences, which are influenced not

only by race but by age (Kusunoki, Barber, Ela, & Bucek, 2016). Women in their late

teens and twenties are in a transition period where they are deciding who they are and

shaping their personalities and beliefs based on familial beliefs, education and

experiences from adolescence, and exploration in a world that is defined by less parental

oversight. For this research, I focused on young adult, African American women aged

18–24.

Women aged 18–19 are late adolescents beginning the transition to adulthood.

Developmental researchers emphasize the importance of this transition as experiences

begin to diverge, which shape adulthood decisions. While some remain at home while

attending college, working, or combining the two, most late adolescents leave home to

attend college or work full-time while living independently (Arnett, 2000). Dating

provides companionship, first loves, and sexual experimentation. Parental supervision

lessens, and there is little normative pressure to marry. Women aged 20–24 are emerging

adults with relationships based on shared values and belief systems, who modify their risk

taking behaviors and make decisions based on future consequences (Arnett, 2000).

Emerging adults have the greatest residential change in any group (Arnett, 2000). While

most continue to live in collegiate residential facilities or independently, some cohabitate

with a romantic partner (Arnett, 2000). Emerging adults reexamine familial beliefs and

form personal beliefs that will shape adulthood (Arnett, 2000). Dating becomes serious,

and the focus shifts to exploring the potential for marriage. While college attendance is at

an all-time high, only 36% of young adults have earned their degrees (U.S. Bureau of

Census, 2016). Many pursue graduate degrees while others begin their occupational path

and possibly marriage and family (Arnett, 2000).

These age groups contracept differently. Contraceptive knowledge is typically

lower, and unintended pregnancies are higher in 18–19 year old women (Wu, Kusunoki,

Ela, & Barber, 2016). Dehlendorf et al. (2014) found that African American women

under 19 years of age were less likely than Caucasians women to use IUDs or implants,

while African American women between the ages of 20–24 were slightly more likely to

use IUDs and implants than Caucasians. Additional contraception differences based on

age are discussed in the next section.

Contraception among African American women. Contraception is used to

prevent pregnancy and the interruption of life goals it could cause (Morse & Moos,

2018). For this research, I categorized contraceptive methods based on the World Health

Organization’s effectiveness recommendations. Barrier methods, including condoms,

diaphragms, sponges, and spermicides, are LECMs, and hormonal methods, including

oral contraceptives, the patch, the vaginal ring, and Depo-Provera injections are MECMs.

IUDs, implants, and sterilization are HECMs. LECMs and MECMs offer varying levels

of effectiveness based on method selection and usage patterns. These methods are the

most used due to lower cost, ease of access, and reversibility, with effect wearing off

quickly once use has ceased or shortly after (USDHHS, 2019). LARC methods provide

hormonal or non-hormonal protection for three to 10 years with one application,

depending on the method, and are HECM. These methods have a higher initial cost but

are more cost effective due to the length of protection provided (USDHHS, 2019).

LARCs are reversible, with effects wearing off once use has ceased or shortly after

(USDHHS, 2019). LARCs include hormonal implants that may be intrauterine hormonal

devices or subdermal implants and non-hormonal IUDs. Permanent contraceptive

methods, including tubal ligation and tubal sterilization, are also HECM used to provide

hormone-free, permanent pregnancy prevention.

Despite most young adults believing that pregnancy should be planned (86–94%),

more than 70% of pregnancies among unmarried women aged 18–24 are unplanned and

likely due to nonuse or inconsistent contraceptive use (Kornides et al., 2015). Half of

unintended pregnancies occur among contraceptive users, with 90% of these pregnancies

resulting from inconsistent or incorrect method use rather than method failure (Wong,

2012). Reasons for nonuse or inconsistent contraceptive use include the belief that one

could not get pregnant (32%), lack of preparedness due to unexpected intercourse (20%),

pregnancy ambivalence (18%), lack of partner desire for contraceptive use (16%), and

concern over contraceptive side effects (15%; Kornides et al., 2015). Paterno et al. (2017)

found that among women aged 20–24, 11.4 to 13% used no contraceptive method. This

number was lower among African American participants despite 60% of African

American women aged 18–24 who believed pregnancy planning is important and 80%

who desired to avoid pregnancy (Paterno et al., 2017).

Barber et al. (2015) found that African American young women tolerated male

partner refusal to use condoms because of the scarcity of potential partners. Sexual health

education focused on condom usage as the best option; however, they are only 79–82%

effective due to incorrect use (Logan et al., 2018; USDHHS, 2019). Condoms are most

effective when used in combination with a dual, hormonal contraceptive method,

providing protection from pregnancy and STIs (USDHHS, 2019). Oral pills have also

been the focus of sexual health education as an ideal contraceptive method, which

provides young women with a false sense of protection due to the overestimation of

effectiveness (Logan et al., 2018). Many are unaware that contraceptive effectiveness

may be impacted by medication, such as antibiotics, and failure to take the pill at the

same time daily. According to the NSFG, African American women have higher oral pill

discontinuation rates than Caucasian women, which may be due to side effect concerns

(Jackson et al., 2016).

Although LARC use has increased over the past five years, use is still low in

young women despite being one of the most effective forms of pregnancy prevention

(Logan et al., 2018). Fewer than 5% of 15–19 year olds select LARCs as their preferred

contraception method (Coates, Gordan, & Simpson, 2018). LARC use varies by age;

however, with women aged 25–34 using LARCs at nearly double the rate of women aged

15–24 (Logan et al., 2018). Despite this increase, young African American women's

usage remains lower than in other racial/ethnic groups (Logan et al., 2018). Coates et al.

(2018) found that in Alabama and Mississippi, two Region 3 states, only 3.7% to 7% of

15–19 year olds used LARCs. Usage may be due to insufficient knowledge and negative

attitudes and beliefs about LARCs stemming from social norms, the knowledge gained

through vicarious experiences, and knowledge of institutional racism and medical

mistrust (Logan et al., 2018). As LARCs are more expensive than LECMs and MECMs

methods and require insertion by a healthcare provider, the lack of insurance may be a

barrier to use (Logan et al., 2018). However, LARC methods are more effective than

LECMs and MECMs, provide better child spacing, are more cost effective, and their

effectiveness is less dependent on user characteristics (Tibaijuka et al., 2017).

Women prefer methods that do not require procedures, which may stem from fear

based on historical reproductive abuse that perpetuates perceptions of institutional racism

and medical mistrust (Jackson et al., 2016; Logan et al., 2018; Tibaijuka et al., 2017).

Even when women want to avoid pregnancy, their contraceptive behaviors are not

necessarily consistent with their intention (Borreo et al., 2015). Potentially modifiable

factors that may contribute to contraceptive use or inconsistent use among African

Americans are perceived low susceptibility of pregnancy, lack of knowledge regarding

contraceptive options, and misperceptions surrounding contraceptive side effects (Borreo

et al. 2015; Kornides et al., 2015). Previous unprotected intercourse without pregnancy

may lead to subfertility or infertility assumptions and give rise to subsequent

contraceptive nonuse or inconsistent use (Borreo et al., 2015). Compared to Caucasian

and Hispanic women, African American women are more likely to report concerns

regarding side effects of contraceptive methods as their rationale for not using

contraception (Kornides et al. 2015). Few women aged 20–24 use LARCs, such as an

IUD implant and injectable hormonal contraception, which may be due to a lack of

knowledge and misconceptions that these methods may be abortifacients (Kornides et al.,

2015). Contraception use, specifically LARC use, might be increased by providing young

women more information on effectiveness, side effects, affordability, and ease of use

(Kornides et al., 2015).

African Americans rely heavily on social networks for sexual and reproductive

norms and information (Hayford & Guzzo, 2013). The social context of pregnancy

includes views of family members or partners, cultural values about pregnancy,

contraception, single parenthood, and life choices resulting from unintended pregnancy

(Kelly, 2014). Low social support during pregnancy is associated with substance use and

abuse, poor nutrition, and lack of exercise (Gray, 2015).

Sexual Initiation

Sexual initiation, considered a critical life transition event, is a significant

milestone towards adulthood as young adults accept increasing responsibility for their

decisions (Goldberg & Halpern, 2017; Moilanen et al., 2018). The mean age of sexual

initiation among American women is approximately 17 years of age, with 15% having

had their first heterosexual vaginal intercourse experience before their 15th birthday

(Guttmacher Institute, 2018; Magnusson, Nield, & Lapane, 2015). African American

young women report earlier sexual initiation and are more likely to experience an

unwanted or mistimed pregnancy than their peers (Moilanen et al., 2018). Current

national norms for the United States consider sexual initiation before age 15 as early,

between 15 and 19 years of age as normative, and after 19 years of age as late (Golden et

al., 2016). Women who experience early sexual initiation are less likely to use

contraception and are prone to an increased risk for depression and lower self-esteem

(Golden et al., 2016; Lanier et al., 2018). Magnusson et al. (2015) found that women with

early sexual initiation also have a higher risk of experiencing multiple unintended

pregnancies compared to women with late sexual initiation.

CSHE may have a protective influence on early sexual initiation (Bourke et al.,

2014). CSHE, grounded in a sex-positive framework, teaches young adults that

consensual sexual behavior can be healthy, normative, and developmentally appropriate

(Bourke et al., 2014; Golden et al., 2016). CSHE equips young adults with the tools

needed to make well-informed and healthy decisions around sexual experiences to reduce

risk and promote healthy development (Golden et al., 2016).

Access to Care

Reproductive trends in contraceptive use among African American women may

be impacted by a reproductive history of medical and reproductive abuse (Jackson et al.,

2016). The United States has a history of social and legislative policies to control the

reproductive health of poor and minority women through forced sterilization (Jackson et

al., 2016). With the introduction of contraceptive implants, many states incentivized or

required usage among women who relied on governmental aid (Jackson et al., 2016).

Combined with historical occurrences such as the Tuskegee experiment, mistrust of

reproductive health practitioners and their recommendations is high. As such, African

American women are more likely than Caucasian women to use LECMs due, in part, to a

desire to preserve reproductive control (Jackson et al., 2016).

Young women, especially those of color, may be at risk of not receiving

highquality family planning care (Dehlendorf et al., 2014). This may contribute to

healthcare disparities, disengagement from healthcare interactions, and perpetuated

institutional mistrust (Gomez & Wapman, 2017). Young women described feelings of

being rushed during appointments, not having questions answered and concerns

addressed, and receiving no information or incomplete information regarding the

method’s side effects provided (Gomez & Wapman, 2017). Contraceptive dissatisfaction

and lack of knowledge regarding alternatives may also contribute to nonuse or incorrect

use of current contraception methods (Kornides et al., 2015). Dissatisfaction and lack of

knowledge may stem from family planning healthcare disparities. Implicit and explicit

bias/pressure, experienced during family planning, may influence contraceptive

decisionmaking, and perpetuate reproductive oppression experienced by minority women

(Gomez

& Wapman, 2017).

Effective patient-centered family planning care should account for medical history

and be engaging, nondirective, interactive, and consider patient preferences and priorities

(Gomez & Wapman, 2017). Provider initiated discussions and recommendations

influenced by race, class, age, or feelings of perceived judgment have a long lasting

impact on young women’s confidence in the healthcare system. This doubt or lack of

confidence may influence decisions to seek and receive contraceptive care (Gomez &

Wapman, 2017). Clinicians are more likely to recommend LARCs to low income women

of color compared to their Caucasian counterparts (Gomez & Wapman, 2017). This type

of provider coercion is seen particularly with patients deemed high-risk, despite their

contraceptive preferences (Morse & Moos, 2018). Provider coercion occurs despite

income, with well-educated patients describing feelings of pressure to be compliant,

discomfort with expressing their opinions and views, and the need to accept provider

recommendations (Joseph-Williams, Edwards, & Elwyn, 2014; Sacks, 2017).

Operationalizing Reproductive Health

The NSFG provides the most comprehensive, nationally representative assessment

of reproductive health in the United States. It provides information on family life,

marriage and divorce, pregnancy, infertility, use of contraception, and general and

reproductive health on a nationally representative sample of the civilian,

noninstitutionalized United States population aged 15– to 49 based on voluntary, inperson

computer assisted personal interviewing (CAPI). The NSGF continuously interviews the

U.S. population, annually over 48 weeks (CDC, 2019b, 2019c). The 2015– 2017 data set

includes responses collected from 10,094 participants, of which 5,554 were women

(CDC, 2019b, 2019c). I chose the NSFG as the data set for this study due to its

oversampling of African American women and individuals aged 15–24 (CDC, 2019b,

2019c).

For this study, I used NSFG variables related to receipt of FSHE, type of

contraceptive method selected for vaginal intercourse at sexual initiation and within the

last 12 months, and age of sexual initiation among African American women aged 18–24

who have sex with men. Participants’ receipt of FSHE before 18 years of age is

categorized as received (yes) or not received (not). If participants self-reported FSHE was

received, the responses collected helped determine if education was ACSHE, APSHE, or

CSHE based on coverage of birth control methods, where to get birth control, how to use

a condom, and waiting until marriage to have sexual intercourse. The stratification of

these variables' results occurred by age into the standard population age groups into the

age brackets of 18–19 and 20–24 years old.

Operationalizing Unintended Pregnancy

Since 1941, unintended pregnancy has been a marker of reproductive health and

dichotomized as one that was either mistimed or unwanted (Guttmacher Institute, 2018).

Mistimed pregnancies, which account for 27% of pregnancies, are characterized as

pregnancies that were not wanted when they occurred, but pregnancy may be desired in

the future (Guttmacher Institute, 2018). Unwanted pregnancies, which account for 18% of

pregnancies, are characterized as pregnancies that were not wanted when they occurred or

in the future (Guttmacher Institute, 2018). An intended pregnancy is one that was wanted

at the time it occurred or sooner (Guttmacher Institute, 2018). Women who were

indifferent about becoming pregnant are counted as intended pregnancies, so that the

unintended pregnancy rate only includes unambiguously unintended pregnancies

(Guttmacher Institute, 2018). Unintended pregnancy mainly results from not using

contraception or inconsistent or incorrect use of effective contraceptive methods

(Guttmacher Institute, 2018). Women more likely to experience unintended births include

unmarried women, African American women, and women with less education or income

(Guttmacher Institute, 2018).

Epidemiology. In 2011, nearly half (45% or 2.8 million) of the United States'

pregnancies were not intended, with 45 unintended pregnancies per every 1,000 women

aged 15–44 (Finer & Zolna, 2016; Guttmacher Institute, 2018). This equates to nearly 5%

of reproductive age women experiencing an unintended pregnancy each year (Finer &

Zolna, 2016). Race disparities in unintended pregnancy exist, with 69% of African

American women identifying their pregnancies as unintended compared to 42% of

Caucasian women (Barber et al., 2015). At 79 per 1,000, the unintended pregnancy rate

for African American women in 2011 was more than double that of non-Hispanic

Caucasian women (33 per 1,000) (Finer & Zolna, 2016; Kusunoki et al., 2016). This

disparity is similar to the percentage of unintended births (Finer & Zolna, 2016; Kusunoki

et al., 2016). African American women have their first child earlier (age 20.9, verse 24.1)

and have more children (2.4 verse 2.2) than Caucasian women (Kusunoki et al., 2016).

The highest unintended pregnancy rates in 2011 were among women in their late teens

and twenties. The age group rate per thousand was 105 (18–19 years old), 101 (20–24

years old), and 69 (25–29 years old) with an overall increased rate of unplanned

pregnancies in women aged 18–25 (Barber et al., 2015; USDHHS 2019).

Rates of unintended pregnancy are generally highest in the south, southwest, and

densely populated states (Kost, 2015). Approximately 55% of unintended pregnancies in

the United States occur in poor and cohabitating young adult women aged 18–24, who

are much less likely to use contraception than older women, more likely not to seek

prenatal care, and have unmet support needs and emotional and financial difficulties

(Finer & Sonfield, 2013; Gray, 2015; Kornides et al., 2015). Kim et al. (2016) found that

African American women were less likely to use and have correct information about

hormonal methods and long-term contraceptive methods than their Caucasian

counterparts.

Implications. In 2011, 42% of unintended pregnancies, excluding miscarriages,

ended in abortion, and 58% ended in birth. This represents a small shift from 2008 when

40% ended in abortion, and 60% ended in birth (Finer & Zolna, 2016). The proportion of

unintended pregnancies ending in birth decreased across all racial and ethnic subgroups

between 2008–2011. The proportion of women experiencing an unintended pregnancy

and choosing to end it in abortion was higher among African American women than

among women in other racial and ethnic groups (Finer & Zolna, 2016). More than half of

all women in the United States will experience an unintended pregnancy by the time they

reach age 45 (Finer & Zolna, 2016).

The estimated cost of unintended pregnancy in the United States as of 2013 was

$4.6 billion annually, with 53%, or $2.5 billion each year, due to inconsistent

contraceptive use (Finer & Sonfield, 2013). By helping women avoid unintended

pregnancies, publicly funded family planning services saved taxpayers $13.6 billion, or

$7.09 for every $1 spent (Frost, Sonfield, Zolna, & Finer, 2014).

Summary

Unintended pregnancy is associated with multiple emotional, social, economic,

and health outcomes, which makes it a substantial public health concern. Unintended

pregnancy may lead to disparities in morbidity and mortality due to delayed prenatal care;

reduced breastfeeding; depression; delivery of infants with low birth weight and birth

defects; as well as reduced average income compared to other women (Carter et al., 2013;

James & Rashid, 2013, USDHHS, 2019). African American women aged 18–24 are at

higher risk for unintended pregnancy due to non-use of contraception, inconsistent or

improper use of contraception, and failure to use dual contraceptive methods (Craig et al.,

2014). These are topics covered during CSHE.

The goal of FSHE is to prepare young adults to be sexually healthy (Santelli et al.,

2017). Misconceptions surrounding long-term contraceptive methods remain a large

factor in use disparities among African American young women. Racial and ethnic

disparities involving contraceptive knowledge exist with an association between less

knowledge and use of LECMs, less knowledge, less self-efficacy, and less knowledge

and increased unintended pregnancy. Reproductive trends in contraceptive use among

African American women may be impacted by a reproductive history of medical and

reproductive abuse. The United States has a history of social and legislative policies to

control poor and minority women's reproductive health through forced sterilization.

FSHE, for most, is received through school based programs, which may not be

comprehensive and often have moral undertones.

African American women are exposed to higher teen and single parenthood rates,

earlier sexual initiation, and use contraception less effectively (Barber et al., 2015).

Perceived social norms involving sexuality, contraceptive selection, and pregnancy

shaped by familial experiences may cause African American women to be more likely to

have inconsistent contraceptive usage and use LECMs. Anecdotal information and

personal experiences learned or experienced in the south may supersede formal education

for African American young women. Current research involving contraception among

African American women is limited to either small qualitative studies or large surveys

focusing on nonuse. The focus of this study was to evaluate the impact of FSHE on

sexual initiation, contraceptive use, and contraceptive selection among African American

young women aged 18–24 who have sex with men. An understanding of the impact of

FSHE may lead to the development of interventions designed to prevent unintended

pregnancies through increased consistent, preference, and intention matched

contraceptive use, as well as improved maternal health and reduced adverse birth

outcomes, all of which public health concerns among young African American women.

Provided in Chapter 3 are details of how the NSFG was used to investigate the

extent to which knowledge gained through FSHE impacts sexual initiation and

contraceptive use and selection among African American women aged 18 24 in the

United States who have sex with men. Presented and interpreted in Chapter 4 are the

results of the statistical analysis techniques discussed in Chapter 3. In Chapter 5 the study

findings are compared to existing empirical literature, and implications and proposed

solutions are discussed.

Chapter 3: Research Method

Purpose of the Study

The purpose of this study was to complete a quantitative analysis of the 2015–

2017 NSFG to explore the impact receipt of FSHE has on sexual initiation and

contraception use and selection among African American young women aged 18–24 in

the United States who have sex with men. Information from this study may lead to the

provision of culturally competent services that increase CSHE, raise the age of sexual

initiation, and increase consistent use of preference and intention matched contraceptive

use among sexually active African American women aged 18–24 in the United States

who have sex men. Information from this study may also decrease unintended

pregnancies, impact maternal health, and reduce adverse birth outcomes public health

concerns in young African American women (Kornides et al., 2015).

In this chapter, I provide information on the study design used for this research

and its supporting rationale. The NSFG population, sampling procedures, procedures for

recruitment, participation, data collection, and operationalization of variables are

discussed. The data analysis plan is outlined, and threats to validity and ethical

procedures are described.

Research Design and Rationale

I conducted this study using a quantitative cross-sectional research approach to

investigate NSFG variables related to FSHE and contraception selection, use, and sexual

initiation among African American young women aged 18–24 in the United States who

have sex with men. Participants' receipt of FSHE before 18 years of age was the

independent variable. If participants self-reported receipt of FSHE, I analyzed responses

to determine if education was ACSHE, APSHE, or CSHE based on coverage of birth

control methods, where to get birth control, how to use a condom, and waiting until

marriage to have sexual intercourse. Age of sexual initiation and selection of no method,

LECMs, MECMs, and HECMs served as dependent variables.

The NSFG provides the most comprehensive and nationally representative

assessment of reproductive health in the United States. I selected the NSFG as the data set

for this study due to its oversampling of African American women and individuals aged

15–24. As part of the federal statistical system, National Center for Health Statistics

(NCHS) supports disseminating the NSFG public use data and documentation files at no

charge to the public. I obtained secondary data from the NSFG public domain website. I

analyzed data from the NSFG through descriptive statistics, logistic regression, and

multinomial logistic regression. Lindberg and Maddow-Zimet (2012) used a similar

strategy to analyze the impact of sexual education on adult sexual behaviors by using

multinomial logistic regression to examine associations between sexual health education

and sociodemographic covariates, and sexual behaviors. Jaramillo et al. (2017) used

multinomial logistic regression to ascertain associations between exposure to sexual

health education topics and contraception used at last sex. Bourke et al. (2014) explored

the receipt of FSHE and sexual initiation using multinomial logistic regression.

Methodology

Population

The NSFG provides information on family life, marriage and divorce, pregnancy,

infertility, use of contraception, and general and reproductive health on a nationally

representative sample of the civilian, noninstitutionalized United States population aged

15-49 based on voluntary, in-person CAPI. The NSGF continuously interviews the U.S.

population, in 12 week intervals, annually over 48 weeks. The 2015–2017 NSFG includes

5,554 interviews conducted with women aged 15–49 from September 2015 to September

2017 (CDC, 2019b). Of this population, 1,355 were non-Hispanic Black women (CDC,

2019b).

Sampling and Sampling Procedures

The NSFG is a multi-staged probability based, nationally representative sample of

individuals aged 15–49. For this study, I used a cross-sectional approach involving

secondary public use data from the 2015–2017 NSFG for African American women aged

18–24 who have sex with men. The NSFG utilizes a sample design that obtains a sample

size of at least 5,000 interviews per year, and oversamples non-Hispanic Blacks,

Hispanics, and teens aged 15–19 (CDC, 2019b). The goal is that interviewees be 20%

non-Hispanic Black, 20% Hispanic, 55% female, and 20% teens aged 15–19, resulting in

a nationally representative sample (CDC, 2019b).

The NSFG uses a stratified five stage area probability sampling method based on

probability proportionate to size (PPS) selection in four domains to ensure the

oversampling of non-Hispanic Blacks, Hispanics, and teens (CDC, 2019b; see Table 2).

Households in domains 2 through 4 are given higher weights to increase the likelihood of

being selected. The first two participant selection stages used these weighted measures

(CDC, 2019b).

Table 2

Probability Proportionate to Size Domain Definition and Characteristics Based on

Household

Domain Definition Total Households Estimated Proportion

Black

Estimated Proportion

Hispanic

1 <10% HH Black

<10% HH Hispanic 65,009,685 .018 .022

2 >=10% HH Black

<10% HH Hispanic 19,871,976 .426 .029

3 <10% HH Black

>= HH Hispanic 20,270,438 .026 .380

4 >=10% HH Black

>=10% HH Hispanic 11,564,194 .301 .299

Note. Reproduced from the 2015-2017 National Survey of Family Growth (NSFG): Sample design documentation by the Centers for

Disease Control and Prevention, 2019b.

Selection of primary sampling units (PSUs).

The NSFG divides the United States into 2,149 PSUs, of which 366 are

considered metropolitan statistical areas (MSAs), and 1,783 are non-MSAs (CDC,

2019b). Each PSU is stratified according to census division, size, and PPS domain. PSUs

with larger populations are assigned a higher selection probability (CDC, 2019b). To

ensure that sampling rates are approximately equal for all households within a sampling

domain, first stage selection probabilities are inversely related to the probabilities of

selection at the second and third stages of selection (CDC, 2019b).

Two-hundred and thirteen PSUs plus two for Alaska and Hawaii are automatically

included in the national probability samples (CDC, 2019b). Of these, 21 PSUs are

considered self-representing (SR) due to population size, and 192 PSUs are considered

non-self-representing (NSR), representing themselves as well as other NSR PSUs (CDC,

2019b). A subset of these 215 PSUs is selected for each 2 year sampling period (CDC,

2019b).

Selection of secondary sampling units (SSUs).

PSUs are segmented into SSUs comprised of one or more census blocks with at

least 50 housing units (HUs) in urban locations and 75 in rural locations (CDC, 2019b).

Most PSUs had 12 SSUs (CDC, 2019b). However, domains 2 through 4 had a higher

selection rate to ensure that African American and Hispanic participants constitute 20%

of all interviews (CDC, 2019b). Segments with fewer than 12 SSUs increased HU size to

60 to 120 based on location (CDC, 2019b).

Listing and selection of HUs within SSUs.

Lists of HUs to be interviewed were generated via the U.S. Postal Service’s

delivery sequence file (DSF) or from a scratch list (CDC, 2019b). The scratch list

consisted of rural and post office box addresses. NSFG personnel physically checked

addresses on the scratch list. Based on these lists, segment maps are developed using

topologically integrated geographical encoding files and geographic information system

(GIS) software. Maps included a view of each segment according to a large scale location

view based on relation to major highways and streets, an intermediate view relative to

major streets and block boundaries, and a detailed small scale view outlining individual

blocks. Maps and DSF addresses are loaded into an electronic listing application and

provided to interviewers one to two months before data collection so that addresses can

be verified and updated if needed. Interviewers also are provided with supplemental

information regarding issues that could impact data collection, such as restricted access

areas and potentially dangerous locations.

Selecting one of the eligible persons within each sampled household.

During the screening activity, the NSFG researchers asked an adult member of the

household to list each resident's gender, age, and race or ethnicity. Information was

requested regarding college students, and those residing in dormitories were listed as part

of their parents’ household, while those with apartments were not. If there was not a

resident between the age of 15 to 49 years of age, the interview was terminated. If more

than one eligible resident was found, the CASI system made the section of which

individual to survey based on meeting target sample sizes for teens and females. Within

household selection rates were designed so that 20% of interviews were conducted with

teens aged 15–19, and 55% of all interviews were females (CDC, 2019b, 2019c).

Two-phase sampling for nonresponses.

During the first 10 weeks of phase-one sampling, unscreened addresses received a

sorry I missed you card and were marked for continuous follow up. Addresses that were

still unscreened during phase two of sampling receive a mailed incentive during the last

two weeks of sampling to encourage participation and more intense follow up. The final

public use data is based on a final weighting design developed from poststratification

adjustment factors and weight trimming. Poststratification was limited to age, gender, and

race and ethnicity based on population estimates from the U.S. Census Bureau and CPS

analysis. Weight trimming was used to reduce the estimated variance without increasing

nonresponse bias. Based on the probability of selection, nonresponse rate, and

poststratification factors, weights at the first and 99th percentiles were conservatively

trimmed. Cases not trimmed had their weights increased so that the sum of weights were

equal to the population control.

Procedures for Recruitment, Participation, and Data Collection

The survey population included college students living in dormitories and

fraternity and sorority housing and military personnel living off base. Excluded from the

survey were individuals residing in prisons, homes for juvenile delinquents, homes for the

intellectually disabled, long-term psychiatric hospitals, and military bases. All HUs

selected used the stratified five stage area probability sampling method were mailed an

advance household letter and NSFG question and answer brochures. These documents

provided information on the survey, explained that completion of the survey was

voluntary, informed potential participants that an interviewer would visit their home

within a few days, and offered a $40 token of appreciation for completion. On the day of

data collection, NSFG interviewers visited HUs selected using the stratified five stage

area probability sampling method, which received the mailed documentation. If no one

was home, a sorry I missed you card was left, and a return visit was made during a

different day or time of day. If the potential participant was home, but unable to complete

the survey, the interviewer answered any questions regarding the survey and offered to

return at a more convenient time.

Individuals willing to participate in the survey completed a 5 minute screener to

ensure they met age and matched the stratified five stage area probability sampling

method requirements. Individuals who passed the screening were provided a respondent

letter, which explained the selection for the survey and received a copy of the informed

consent form; all documentation was provided in English and Spanish. Participants

provided an electronic signature acknowledging informed consent and received a $40

token of appreciation in advance of completing the interview. Per an NCHS ERB granted

waiver of informed consent documentation, participants were not required to sign the

electronic consent form. For those that elected not to sign, the interviewer signed the

consent form acknowledging that the participant received informed consent information

and that the respondent agreed to participate in the survey. NSFG researchers presented

individuals not selected for participation based on the screening process, a $5 token of

appreciation for completing the screener.

Interviews were conducted over 60–80 minutes and administered either by CAPI

or ACASI. In ACASI, the respondent listens to the questions through headphones, reads

them on the screen, or both, and enters the response directly into the computer. The

NSFG researcher administered the majority of surveys via ACASI, as this method avoids

asking the respondent to provide answers to the interviewer, increasing the complete

reporting of sensitive behaviors.

As part of the federal statistical system, NCHS supports disseminating the NSFG

public use data and documentation files at no charge to the public. The NSFG public use

data for the 2015–2017 survey is housed on its website and includes program statements,

syntax guidelines to read the ASCII data into statistical software, and informed consent

documentation. I obtained secondary data from the NSFG public domain website.

Permission for NSFG data use is not required; however, the citation of data is requested.

Operationalization of Constructs

NSFG 2015–2017 variables related to FSHE, age of sexual initiation, and

contraceptive use and method selected for vaginal intercourse at sexual initiation and

within the 12 months before the survey were analyzed to explore the impact receipt of

FSHE has on sexual initiation and contraception use and selection among African

American young women aged 18–24 who have sex with men. Variables analyzed include

(see Table 3).

Demographics. Participants included in the secondary data analysis were African

American (RSCRRACE) and between the ages of 18–24 (RSCRAGE/AGER).

Ever had sex. Participants included in the secondary data analysis had vaginal

intercourse (RHADSEX/HADSEX).

FSHE received. Participants' receipt of FSHE before 18 years of age (SEDNO)

was categorized as received (yes) or not received (no). If received, responses determined

if education was ACSHE, APSHE, or CSHE based on coverage of methods of birth

control (SEDBC), coverage where to get birth control (SEDWHBC), coverage of how to

use a condom (SEDCOND), and waiting until marriage to have sexual intercourse

(SEDABST). FSHE was ACSHE if education on waiting until marriage (SEDABST) was

the only variable answered as yes; APSH, if yes, were the answers to variables SEDBC

and SEDABST; and CSHE, if yes, were the answers to variables SEDBC, SEDWHBC,

SEDCOND, and SEDABST.

Sexual initiation. Participants’ responses to, “The very first time that you had

sexual intercourse with a man, how old were you?” (AGEFSTSX) was categorized as

under 15, 15–17, 18–19, and 20 or over.

Table 3

2015–2017 National Survey of Family Growth Variables Analyzed

NSFG variables Question Responses

RSCRRACE Respondent’s race as reported in the screener

1 Other,

2 Black or

African American

3 White

4 Hispanic

RSCRAGE/AGER Respondent’s age as reported in the screener 18–24

RHADSEX/HADSEX Whether respondent has ever had sex (Vaginal Intercourse) with a

man

1 Yes

2 No

AGEFSTSX The very first time that you had sexual intercourse with a man, how

old were you?”

0–24

SEDNO

Now I’m interested in knowing about formal sex education you

may have had. (Before you were 18, did/have) you ever

(have/had) any formal instruction at school, church, a community

center, or some other place about how to say no to sex?

1 Yes

5 No

SEDABST

Before you were 18, did/have) you ever (have/had) any formal

instruction at school, church, a community center, or some other

place about waiting until marriage to have sex?

1 Yes

5 No

SEDBC

(Before you were 18, did/have) you ever (have/had) any formal

instruction at school, church, a community center, or some other

place about methods of birth control?

1 Yes

5 No

(Before you were 18, did /have) you ever (have/had) any formal 1 Yes

SEDWHBC instruction at school, church, a community center, or some other 5 No

place about where to get birth control?

SEDCOND

(Before you were 18, did/have) you ever (have/had) any formal

instruction at school, church, a community center, or some other

place about how to use a condom?

1 Yes

5 No

Usefstsx Whether R used a method at first sex 1 Yes

5 No

FIRSMETH1 What was the first birth control method you ever used for any

reason?

1 No method

3 Birth control pills

4 Condoms

7 Withdrawal 8

Depo-Provera,

injectables

9 Hormonal implant

(Norplant, Implanon,

or Nexplanon) 10

Calendar rhythm,

standard days, or

cycle beads method

LSTMTHP1 If R used a method at last intercourse with the last partner 12 Diaphragm

in the past 12 months, what method did you or he use? 13 Female condom

14 Foam

15 Jelly or cream

17 Suppository

18 Sponge

19 IUD, coil, or loop

25 Contraceptive patch

26 Vaginal contraceptive ring

Contraceptive selection and use. The NSFG researchers measured contraceptive

use (usefstsx) at sexual initiation. The contraceptive method used at sexual initiation

(FIRSMETH) and last intercourse (FSTMTHP) is outlined in Table 4.

Table 4

2015–2017 National Survey of Family Growth Contraceptive Methods

Value Label

3 Birth Control Pills

4 Condoms

7 Withdrawal

8 Depo-Provera, injectables

9 Hormonal implant (Norplant, Implanon, or Nexplanon)

10 Calendar rhythm, standard days, or cycle beads method

12 Diaphragm

13 Female condom

14 Foam

17 Suppository

18 Sponge

19 IUD, coil, or loop

25 Contraceptive patch

26 Vaginal contraceptive ring

For this study, barrier methods including condoms, diaphragms, sponges, and

spermicides, including foam and suppositories, are classified as LECMs, with hormonal

methods including oral contraceptives, and patch ring, and injectables classified as

MECMs. IUDs and hormonal implants are classified as HECMs. Withdrawal and rhythm

methods are classified as not having used contraceptives. Contraceptive methods

categories are outlined in Table 5.

Table 5

Variables Created for Analysis

Created Variables Based on NSFG Variables Required Responses

ACSHE SEDABST 1 Yes

APSHE SEDABST and SEDBC yes response 1 Yes

CSHE SEDABST, SEDBC, SEDCOND, and

SEDWHBC yes responses 1 Yes

AgeFirstSex agefstsx

0 = 0–13

1 = 14–17

2 = 18–19

3 = 20–24h

Contraceptive

Methods used at

first and last sex

0. No contraceptive used

1 No method

7 Withdrawal

10 Calendar rhythm, standard days, or cycle beads method

1. Less effective method

4 Condoms

12 Diaphragms

14 Foam

15 Jelly or cream

17 Suppository

18 Sponge

2 Moderately effective method

3 Birth control pills

8 Depo-Provera, injectables

3 Birth control pills

8 Depo-Provera, injectables

25 Contraceptive patch

26 Vaginal contraceptive ring

9 Hormonal implant (Norplant, Implanon, or Nexplanon)

3. Highly effective method

19 IUDs, coil or loop

Data Analysis Software, Data Cleaning and Preparation for Public-Use

I used the IBM Statistical Package for the Social Sciences (SPSS 25)

statistical software to complete the data analysis described in this section. NSFG data are

prepared for public use via hard edits, which prevented values that were inconsistent with

answers to related variables, and soft edits, which noted answers that were rare or

improbable (CDC, 2019a, 2019c). Participants were able to correct these responses. If a

participant elected to skip a question, the interview coded the response as not ascertained

with values of 7, 97, 997, 9997, or 99997 (CDC, 2019a, 2019c). If a participant refused to

answer a question, the interviewer coded the response as refusal with values of 8, 98, 998,

9998, or 99998 (CDC, 2019a, 2019c). The NSFG researchers checked the consistency of

responses with recoded variables. Missing or inconsistent variables received new codes

using sequential regression imputation to predict a value for the missing or inconsistent

data or via logical imputation by an NCHS subject matter expert who examined related

variables and assigned a consistent value (CDC, 2019a, 2019c).

The percentage of recoded imputed values did not exceed 2% for any variable

(CDC/NCHS, 2019a, 2019c).

Research Questions and Data Analysis Plan

The RQs for this study are as follows:

RQ1: Is there an association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men?

H01: There is no association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men.

H11: There is an association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men.

To answer RQ1, I used multinomial logistic regression to determine the

association between the self-reported type of FSHE received (abstinence-centered,

abstinence-plus, or comprehensive) and the age category of sexual initiation (0–13, 14–

17, 18–19, and 20 and over) for sexually active African American women aged 18–24

who have sex with men. ACSHE and the 0–13 age category were the referent groups. I

used odds ratios (OR) to measure the strength of the association. OR confidence intervals

(CI) were set at 95%, and the critical p-value significance was .05.

RQ2: Is there an association between the type of FSHE received and contraceptive

use at sexual initiation for sexually active African American women aged

18–24? who have sex with men

H02: There is no association between the type of FSHE received and contraceptive

use at sexual initiation among sexually active African American women aged 18–24 who

have sex with men.

H12: There is an association between the type of FSHE received and contraceptive

use at sexual initiation among African American women aged 18–24 who have sex with

men.

To answer RQ2, I used logistic regression to determine the association between

the self-reported type of FSHE received (abstinence-centered, abstinence-plus, or

comprehensive) and contraceptive use (yes/no) at sexual initiation for African American

women aged 18–24 who have sex with men. ACSHE and no contraceptive use were the

referent groups. ORs were used to measure the strength of the association. OR CIs were

set at 95%, and the critical p-value significance was .05.

RQ3: Is there an association between the type of FSHE received and the

effectiveness of contraceptive methods selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men?

H03: There is no association between the type of FSHE received and the

effectiveness of contraceptive methods selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men.

H13: There is an association between the type of FSHE received and the

effectiveness of the contraceptive method selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men.

To answer RQ3, I used multinomial logistic regression to determine the

association between the type of self-reported FSHE received (abstinence-centered,

abstinence-plus, or comprehensive) and effectiveness of contraceptive method selected

(none, less effective, moderately effective, and highly effective) at sexual initiation by

sexually active African American women aged 18–24 who have sex with men. ACSHE

and no contraceptive method selected were the referent groups. ORs were used to

measure the strength of the association. OR CIs were set at 95%, and the critical p-value

significance was .05.

RQ4: Is there an association between the type of FSHE received and the

effectiveness of contraceptive methods primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months?

H04: There is no association between the type of FSHE received and the

effectiveness of the contraceptive method primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months.

H14: There is an association between the type of FSHE received and the

effectiveness of contraceptive methods primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months.

To answer RQ4, I used multinomial logistic regression to determine the

association between the self-reported type of FSHE received (abstinence-centered,

abstinence-plus, or comprehensive) and effectiveness categories of the contraceptive

method primarily selected (none, less effective, moderately effective, and highly

effective) by sexually active African American women aged 18–24 who have sex with

men at last intercourse within the past 12 months. ORs were used to measure the strength

of the association. OR CIs were set at 95%, and the critical p-value significance was .05.

Threats to Validity

The NSFG sampling procedures addressed external threats to validity. The

researchers selected non-Hispanic African American women and teens aged 15–19 at

higher rates to yield an oversample of the population and weighted the samples to

increase consistency in predictions. NSFG researchers used two-phase sampling methods

to reduce nonresponse bias. Recall bias may have occurred due to the inability to recall

first sexual experiences or the type of sex education they received. Under-reporting and a

social desirability bias may have occurred due to respondents giving what they viewed as

socially desirable instead of truthful answers. The administration of the majority of

surveys was via ACASI to increase the complete reporting of sensitive behaviors. There

were no perceived internal threats to conducting this study, as it was a secondary data

analysis.

Ethical Procedures

Internal Procedures

The process for Walden University's Institutional Review Board (IRB) approval

consisted of the submission of the proposal draft to the assigned University Research

Reviewer (URR). I submitted the description of data sources and partner sites to request

guidance from an IRB staff member on completing the IRB process. I completed and

revised applicable student forms, and approval documentation based on guidance

received according to the preliminary ethics feedback service. After proposal approval, I

completed applicable student forms and approval documentation and revised them as

needed to submit for final IRB approval. The IRB number assigned to this research is 01-

27-20-0107579.

External Procedures

An individual who passed the screening process received a respondent letter,

which explained selection for the survey, along with a copy of the informed consent

form; both were available in English and Spanish. These documents emphasized

confidentiality and provided participants with the NSHS research ethics review board’s

toll-free number and the study number. Also provided was a dedicated website to answer

any questions about the study. Participants provided an electronic signature

acknowledging informed consent. Per an NCHS ERB granted waiver of informed consent

documentation, participants were not required to sign the electronic consent form. For

those that elected not to sign, the interviewer signed the consent form acknowledging the

provision of informed consent information and that the respondent agreed to participate in

the survey.

Proposed NSFG public use files were reviewed by NCHS Disclosure Review

Board (DRB) before release, where several variables were suppressed or modified to

prevent the disclosure of participants' identity. Several suppressed and modified variables

are available for a fee from the NCHS Research Data Center (RDC). Directly identifying

information, including names and addresses, are suppressed from public use files. The

REGION of residence variable is suppressed but is available from the RDC for a fee. Life

events such as marriages, divorces, pregnancies, cohabitations, educational degrees,

military service, and selected health services are suppressed to prevent potential linkages

or use with external data sources to identify respondents (CDC, 2019b, 2019c). The

variable for the Hispanic subgroup (HISPGRP) has been modified and collapsed for

public use, with full detail available from the RDC for a fee. The values of some variables

were randomly altered through statistical perturbation to prevent the identification of

respondents.

The NSFG houses public use archival data on its website from 1973–2017. As

part of the federal statistical system, NCHS supports disseminating the NSFG public use

data and documentation files. Since the inception of the study, variables are suppressed or

modified on the public use data files to protect the survey respondents' identity and

confidentiality. The REGION variable was not suppressed in surveys conducted before

2015–2017.

Summary

In this chapter, I discussed the methodology and research plan for a quantitative

cross-sectional analysis of the 2015–2017 NSFG designed to investigate the extent to

which knowledge gained through FSHE impacts sexual initiation and contraceptive use

and selection sexually active African American women aged 18–24 in the United States

who have sex with men. Descriptive statistics were used to define demographics,

providing the number of self-reported sexually active African American women aged 18–

24. I used multinomial logistic regression to determine the association between the

selfreported type of FSHE received and self-reported age of sexual initiation among

sexually active African American women aged 18–24 who have sex with men. I used

binary logistic regression to determine the association between the self-reported type of

FSHE received and self-reported contraceptive use at sexual initiation for African

American women aged 18–24 who have sex with men. I used multinomial logistic

regression to determine the association between the type of self-reported FSHE received,

self-reported age of sexual initiation, and effectiveness of self-reported contraceptive

method selected at sexual initiation by sexually active African American women aged

18–24. I used multinomial logistic regression to determine the association between the

self-reported type of FSHE received and effectiveness of self-reported contraceptive

methods selected by sexually active African American Women selected at last intercourse

within the past 12 months. OR CI was 95%, and the critical p-value significance was .05.

Racial and ethnic disparities involving contraceptive knowledge exist with an

association between less knowledge and use of LECMs, less self-efficacy, and increased

unintended pregnancy (Kim et al., 2016). Disparities also exist involving African

American young women experiencing earlier sexual initiation. Key results from this

study may lead to an increase in the provision of culturally competent services, such as

sexuality and reproductive education, family planning, and contraceptive counseling. This

may increase African American young women’s selection and consistent use of

contraceptive methods best fit for reproductive intentions, reduce usage failures, decrease

the age of sexual initiation, and improve birth related outcomes. Key results from this

study may also decrease unintended pregnancies and reduce adverse birth outcomes.

Presented and interpreted in Chapter 4 are the results of the statistical analysis

techniques discussed in this chapter. The study findings are compared to existing

empirical literature, and implications and proposed solutions are discussed in Chapter 5.

Chapter 4: Results

Introduction

The nature of this quantitative cross-sectional analysis of the 2015–2017 NSFG

was to explore the receipt of FSHE and the impact it may have on sexual initiation,

contraceptive use and selection among African American women aged 18–24 who have

sex with men. The results of this study may highlight the importance and increase

provisions of FSHE, which may raise the age of sexual initiation, lead to consistent use of

preference and intention matched contraceptives, and reduce contraceptive failures due to

improper use.

In this chapter, I explore the following RQs:

RQ1: Is there an association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men?

H01: There is no association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men.

H11: There is an association between the type of FSHE received and the age of

sexual initiation among sexually active African American women aged 18–24 who have

sex with men.

RQ2: Is there an association between the type of FSHE received and contraceptive

use at sexual initiation for sexually active African American women aged

18–24 who have sex with men?

H02: There is no association between the type of FSHE received and contraceptive

use at sexual initiation among sexually active African American women aged 18–24 who

have sex with men.

H12: There is an association between the type of FSHE received and contraceptive

use at sexual initiation among African American women aged 18–24 who have sex with

men.

RQ3: Is there an association between the type of FSHE received and the

effectiveness of contraceptive methods selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men?

H03: There is no association between the type of FSHE received and the

effectiveness of contraceptive methods selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men.

H13: There is an association between the type of FSHE received and the effectiveness

of the contraceptive method selected at sexual initiation by sexually active

African American women aged 18–24 who have sex with men.

RQ4: Is there an association between the type of FSHE received and the

effectiveness of contraceptive methods primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months?

H04: There is no association between the type of FSHE received and the

effectiveness of the contraceptive method primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months.

H14: There is an association between the type of FSHE received and the

effectiveness of contraceptive methods primarily selected by sexually active African

American women aged 18–24 who have sex with men at last intercourse within the past

12 months.

To answer RQ1, I used multinomial logistic regression to determine the

association between the self-reported type of FSHE received (abstinence-centered,

abstinence-plus, or comprehensive) and age of sexual initiation (0–13, 14–17, 18–19, and

20 and over) for sexually active African American women aged 18–24. ACSHE and the

0–13 age category were the referent groups. ORs were used to measure the strength of the

association. OR CIs were set at 95%, and the critical p-value significance was .05.

To answer RQ2, I used logistic regression to determine the association between

the self-reported type of FSHE received (abstinence-centered, abstinence-plus, or

comprehensive) and contraceptive use (yes/no) at sexual initiation for African American

women aged 18–24. ACSHE and no contraceptive use were the referent groups. ORs

were used to measure the strength of the association. OR CIs were set at 95%, and the

critical p-value significance was .05.

To answer RQ3, I used multinomial logistic regression to determine the

association between the type of self-reported FSHE received (abstinence-centered,

abstinence-plus, or comprehensive) and effectiveness of contraceptive method selected

(none, less effective, moderately effective, and highly effective) at sexual initiation

according to sexually active African American women aged 18–24. ACSHE and no

contraceptive method selected were the referent groups. ORs were used to measure the

strength of the association. OR CIs were set at 95%, and the critical p-value significance

was .05.

To answer RQ4, I used multinomial logistic regression to determine the

association between the self-reported type of FSHE received (abstinence-centered,

abstinence-plus, or comprehensive) and effectiveness of contraceptive methods primarily

selected (none, less effective, moderately effective, and highly effective) according to

sexually active African American women aged 18–24 at last intercourse within the past

12 months. ORs were used to measure the strength of the association. OR CIs were set at

95%, and the critical p-value significance was .05.

In this chapter, I provide information on NSFG data collection and descriptive and

demographic characteristics and statistics of the sample population. I also present,

interpret, and summarize results for the statistical analysis techniques above with

applicable tables and statistics.

Data Collection

The NSGF continuously interviews the U.S. population, in 12 week intervals,

annually over 48 weeks. The 2015–2017 NSFG includes 5,554 interviews conducted with

women aged 15–49 from September 2015 to September 2017. Of this population, 1,355

were non-Hispanic Black women. The NSFG uses a multi staged probability based

nationally representative sample design that minimizes the design effects of women and

men, obtains a sample size of at least 5,000 interviews per year, and oversamples

nonHispanic Blacks, Hispanics, and teens aged 15–19. The goal is that interviewees are

20% non-Hispanic Black, 20% Hispanic, 55% female, and 20% teens aged 15–19,

resulting in a nationally representative sample (CDC, 2019b).

The final public use data are based on a final weighted design developed from

poststratification adjustment factors and weight trimming. Poststratification is limited to

age, gender, and race and ethnicity-based on population estimates from the U.S. Census

Bureau and CPS analysis. As such, when weighted to reflect the U.S. female population

at the interviewing midpoint, generalizations may be made. For this research, I used the

final public use data retrieved online.

Data Cleaning

NSFG data are prepared for public use via hard edits, which prevents inconsistent

values with answers to related variables and soft edits, which noted rare or improbable

answers. Participants were able to correct these responses. If a participant elected to skip

a question, the interview coded responses as not ascertained with values of 7, 97, 997,

9997, or 99997 (CDC, 2019a, 2019c). If a participant refused to answer a question, the

interview coded the response as refusal with values of 8, 98, 998, 9998, or 99998 (CDC,

2019a, 2019c). The team checked the consistency of responses with recoded variables.

An NCHS subject-matter expert recoded missing or inconsistent variables were using

sequential regression imputation to predict a value for the missing or inconsistent data or

via logical imputation who examined related variables and assigned a consistent value.

The percentage of recoded imputed values did not exceed 2% for any variable (CDC,

2019a, 2019c).

For this study, a subsample contained 499 sexually active African American

women aged 18–29 who have sex with men. I cleaned this subsample for missing data

and responses such as don’t know and refused, for variables used in this study.

Demographic analysis showed that the response rate for women aged 25–29 was 2%. As

such, this population did not meet 5% and was not a part of the data analysis. Current

research surrounding sexual health education and contraceptive use among young women

focused on those aged 18–24, which aligned this study with current research.

Data cleaning yielded a final sample size of 242 sexually active African American

women aged 18–24 who have sex with men. Responses from this subpopulation

demonstrated that women who experienced sexual initiation over the age of 20 (3.1%)

were less than 5%, so I removed this population from the data analysis.

Data Analysis Procedure and Statistical Assumptions

I used IBM SPSS 25statistical software to complete the data analysis described in

this section. I used multinomial logistic regression for RQ1. Principle assumptions for

multinomial logistic regression are: (a) The dependent variables should be measured at

the nominal level. Age of sexual initiation, the dependent variable, was derived from

continuous data and collapsed into categorical, nominal data, (b) The independent

variables are nominal, with continuous variables coded as categorical. All independent

variables are nominal or categorical, (c) All variables have independence of observations,

and the dependent variables have mutually exclusive and exhaustive categories. All

variables used in this study met these criteria, (d) There should be no multicollinearity.

Bivariate analysis and the chi-square test tested multicollinearity among the nominal

independent variables, (e) There needs to be a linear relationship between any continuous

independent variables and the logit transformation of the dependent variable. I did not use

any continuous independent variables in the analysis of RQ1, and (f) There should be no

outliers, high leverage values, or highly influential points. I did not use any continuous

independent variables in the analysis of RQ1, so these criteria do not apply.

I used binary logistic regression for research RQ2 through 4. Principle

assumptions for binary logistic regression are: (a) The dependent variable is measured on

a dichotomous scale. For RQ 2 through 4, contraception use at sexual initiation and

effectiveness categories of contraceptive methods are dichotomous, (b) There are one or

more independent variables, which can be either continuous or categorical. For RQ 2

through 4, the independent variables are categorical; (c) There should be independence of

observations, and the dependent variable should have mutually exclusive and exhaustive

categories. For RQ 2 through 4, contraception use at sexual initiation and effectiveness

categories of contraceptive methods are based on mutually exclusive and exhaustive

categories; and (d) there needs to be a linear relationship between any continuous

independent variables and the logit transformation of the dependent variable. For RQ2

through 4, I did not use any continuous independent variables used.

Results

Demographic Characteristics

Univariate analysis. The final population analyzed included 242 sexually active

African American women aged 18–24 who have sex with men. According to Table 6, the

majority were between 20–24 years of age (77.7%), received CSHE (63.2%), and had at

least a high school diploma or general education diploma (GED) (36.8%). When

surveyed (86.8%), the majority had insurance coverage and viewed religion as important

(57.9%). Most were born to single mothers (61.6%) who had at least a high school

diploma or GED (36.8%) and were 20–24 years old at their birth (29.7%). It is important

to note that most participants were born to mothers less than 24 years old (78.5%). Sexual

initiation began between 14–17 years old (70.2%) for most young women who used

contraception at initiation (67.5%). Most used LECMs during sexual initiation (50.6%).

However, most did not use any method at the last sexual encounter (41.3%) despite

stating they would be very upset if they experienced an unintended pregnancy (30.2%).

Table 6

Univariate Analysis of Characteristics of Included Study Participants

Characteristics of participants N = 242 Percentage (%)

Current age (years) 18–

22.3

20–24 188 77.7

Formal sexual health education received Abstinence-centered

12.8

Abstinence-plus 58 24.0

Comprehensive 153 63.2

Education

Less than high school diploma

9.5

High school diploma/GED 89 36.8

2 years of college/associate degree 80 33.1

Bachelor’s degree 38 15.7

Graduate degree 12 5.0

Insurance coverage

13.2

Yes 210 86.8

Importance of religion

Not important

18.6

Very important 140 57.9

Somewhat important 57 23.6

Married parents at participant’s birth

149

61.6

Yes 93 38.4

Maternal education

Less than high school diploma

16.1

High school diploma/GED 89 36.8

2 years of college/associate degree 77 31.8

Bachelor’s degree 37 15.3

(table continues)

Maternal age at first birth

Age of sexual initiation (years) 0–13

9.1

14–17 170 70.2

18–19 50 20.7

Used contraception at sexual initiation

32.5

Yes 132 67.5

Type of contraception used at sexual initiation

None

2.5

Less effective methods 121 50.6

Moderately effective methods 104 43.5

Highly effective methods 8 3.4

< 18 63 27.2

18–19 50 21.6

20–24 69 29.7

25–29 38 16.4

≥ 30 12 5.2

Type of contraception used at last sexual encounter None

100

41.3

Less effective methods 62 25.6

Moderately effective methods 58 24

Reaction to an unintended pregnancy

Would not care

11.8

Very upset 73 3.2

A little upset 72 29.8

A little pleased 41 16.9

Very pleased 28 11.8

Bivariate analysis. A bivariate analysis was used to determine whether a

significant association existed between the independent and dependent variables and to

provide additional characteristics of included study participants. The bivariate analysis in

Table 7 demonstrates that FSHE (X2 = 1.14, p =.83), mother’s age at first birth (X2 = 3.84,

p =.65), mother’s education (X2 = 8.96, p =.25), participants’ parental marital status (X2 =

1.82, p =.4), and importance of religion (X2 = 5.69, p = .2) all had a positive association

with age of initiation, but the associations were not significant. The only significant,

positive association was between age of sexual initiation and reaction to unintended

pregnancy (X2 = 21.08, p =.01). The bivariate analysis in Table 8 demonstrates that FSHE

(X2 = 2.71, p =.26), mother’s age at first birth (X2 = 2.33, p = .51), mother’s education (X2

= 7, p = .07), participants’ parental marital status (X2 = 1.14, p = .29), importance of

religion (X2 = 1.36, p = .51), and reaction to unintended pregnancy (X2 = .30, p = .99), all

had a positive association with contraceptive use at initiation, but the associations were

not significant.

The bivariate analysis in Table 9 demonstrates that FSHE (X2 = 1.56, p = .46),

mother’s age at first birth (X2 = 3.51, p = .32), mother’s education (X2 = 5.74, p = .13),

participants’ parental marital status (X2 = 0.41, p = .52), importance of religion (X2 = 1.78,

p = .41), and reaction to unintended pregnancy (X2 = 3.80, p = .43), all had a positive

association with contraceptive effectiveness category selected at initiation, but the

associations were not significant. The bivariate analysis in Table 10 demonstrates that

FSHE (X2 = 4.45, p = .64), mother’s age at first birth (X2 = 10.07, p =.33), mother’s

education (X2 = 6.04, p =.74), participants’ parental marital status (X2 = 4.08, p = .29), and

importance of religion (X2 = 3.47, p = .72) all had a positive association with

contraceptive effectiveness category selected at last sex, but the associations were not

significant. The only significant, positive association was between contraceptive

effectiveness category selected at last sex and reaction to unintended pregnancy (X2 =

36.40, p < .01).

Table 7

Bivariate Analysis of Age of Sexual Initiation and Independent Variables

Age of sexual

initiation

0–13 14–17 18–19

Study independent variables n % n % n % n % X2 p

Age of sexual initiation

0–13

9.1

14–17 17

70.3 - - - - - - - -

18–19 50 20.7 - - - - - - - -

Formal sexual health education received

Abstinence-centered

12.8

18.2

11.8

14.

Abstinence-plus 58 24.0 6 27.3 40 23.5

12 24.0 1.14 .8

Comprehensive 15

63.2 12 54.6 11

64.7

31 62.0

<18 63 27.2 7 33.3 45 28.0 11 22.0

18–19

20–24

21.6

29.7

28.6

14.3

20.5

29.8

22.0

36.0 3.84

≥25 50 21.6 5 23.8 35

Mother’s education

Less than high school diploma 39 16.1 8 36.4 23 13.5 8 16.0

High school diploma/GED

2 years of college/associate degree

36.8

31.8

27.3

32.8

38.2

32.4

36.0

30.0 8.96

Bachelor’s degree or higher 37 15.3 1 4.6 27 25.9 9 18.0

Mother’s age at first birth

Married parents at participants’ birth

Yes

Importance of religion

Not important

61.6

38.4

18.6

72.7

27.3

22.7

105

61.8

38.2

20.6

56.0

44.0 1.82

10.

Very important 14

57.9 11 50.0 93 54.7 36 72.0 5.69 .2

Somewhat important 57 23.6 6 27.3 42 24.7 9 18.0

Reaction to an unintended pregnancy

Would not care

11.6

27.3

10.0

10.

21.7 10

Very upset

A little upset

30.2 29.8

13.6 36.4

35.9 24.7

18.0

44.0

21.0

A little pleased 41 16.9 3 13.6 33 19.4 5 10.0

Very pleased 28 11.6 2 9.1 17 10.0 9 18.0

Table 8

Bivariate Analysis of Contraceptive Use at Sexual Initiation and Independent Variables

Contraceptive use at sexual initiation

No Yes

Study independent variables n % n % n % X2 p

Contraceptive use at sexual initiation

32.5

Yes 162 67.5 - - - - - -

Formal sexual health education

Abstinence-centered

12.8 12

15.4

11.1

Abstinence-plus 58 24.0 22 28.2 35 21.6 2.71 .26

Comprehensive 153 63.2 44 56.4 109 67.3

Mother’s education

Less than high school diploma

16.1

23.1

High school diploma/GED

2 years of college/associate degree

36.8

31.8

26.9

33.3

41.4

31.5 7.00 .07

Bachelor’s degree or higher 37 15.3 13 16.7 24 14.8

Married parents at participants’ birth

Yes

149

61.6

38.4

56.4

43.6

103

63.6

36.4

Importance of religion

Not important

18.6

14.1

Very important 140 57.9 49 62.8 91 56.2 1.36 .51

Somewhat important 57 23.6 18 23.1 39 24.1

Reaction to an unintended pregnancy

Would not care

11.6

12.8

Very upset

A little upset

30.2

29.8

29.5

28.2

30.9

30.2 0.30 .99

A little pleased 41 16.9 14 18.0 27 16.7

Very pleased 28 11.6 9 11.5 18 11.1

Table 9

Bivariate Analysis of Contraceptive Effectiveness at Sexual Initiation and Independent

Variables

Contraceptive effectiveness category at sexual

initiation

None–less effective Moderately–highly

effective

Study independent variables n % n % n % X2 p

Contraceptive effectiveness category at

sexual initiation

None–Less Effective 127

53.1

- -

Moderately–Highly Effective 112 46.9 - - - - - -

Formal sexual health education

Abstinence-centered 31

12.8 14

11.0

14.

Abstinence-plus 58 24.0 34 26.8 23 20.5 1.56 .46

Comprehensive 153 63.2 79 62.2 73 65.2

Less than high school diploma

16.1

11.0

21.

High school diploma/GED 89 36.8 5 40.2 36 32.1 5.74 .13

2 years of college/associate degree 77 31.8 4

31.5 37 33.1

Bachelor’s degree or higher 37 15.3 2

17.3 15 13.4

Married parents at participants’ birth

Yes 14

61.6

38.4

63.0

37.0

58.9

41.1

0.4

Importance of religion

Not important

18.6

18.9

17.

Very important 14

57.8 6

54.3 70 62.5 1.78 .41

Somewhat important 57 23.5 3

26.8 23 20.5

Reaction to an unintended pregnancy

Would not care 28

11.6

11.0

12.

Very upset

A little upset

30.2 29.8

33.1

30.7

26.8

28.6

3.80 .43

A little pleased 41 17.0 2

17.3 19 17.0

Very pleased 28 11.6 1

7.9 17 15.2

Table 10

Bivariate Analysis of Contraceptive Effectiveness at Last Sex and Independent Variables

Contraceptive effectiveness category at last sex

None

Less

effective

Moderately

effective

Highly

effective

Study independent variables n % n % n % n % n % X2 p

Contraceptive effectiveness

category at last sex

None 10041.3

- - -

Less effective 62 25.6 - - - - - - - - - -

Moderately effective 58 24.0 - - - - - - - - - -

Highly effective 22 9.1 - - - - - - - - - -

Formal sexual health education

Abstinence-centered

12.

9.7

19.

Abstinence-plus 58 24.0 22 22.

18 29.0 14 24.1 4 18.2 4.45 .64

Comprehensive 15363.2 66 66.

38 61.3 33 56.9 1

72.7

Mother’s age at first birth

20-24 69 29.8 23 33.3 27.5 19

≥25 50 21.5 20 40.0 20.0 17

Mother’s education

Less than high school diploma

16.1

14.0

16.1

17.3

22.7

High school diploma/GED 89 36.8 43 43.0 17 27.4 21 36.2 8 36.4 6.04 .74

2 years of college/associate degree 77 31.8 30 30.0 22 35.5 18 31.0 7 31.8

Bachelor’s degree or higher 37 15.3 13 13.0 13 21.0 9 15.5 2 9.1

Married parents at participants’ birth

Yes 149

61.6

38.4

69.0

31.0

56.4

43.6

55.2

44.8

59.1

40.9

4.08

Importance of religion

Not important

18.6

18.0

19.3

22.4

9.1

Very important 140 57.9 61 61.0 35 56.5 29 40.0 15 68.2 3.47 .72

Somewhat important 57 23.5 21 21.0 15 24.2 16 27.6 5 22.7

Reaction to an unintended pregnancy

Would not care

11.5

20.0

4.8

1.7

18.2

Very upset

A little upset

30.2

29.8

22.0

21.0

30.7

35.5

36.2

41.4

50.0

22.7 36.40 <.001

A little pleased 41 17.0 29 19.0 11 17.7 10 17.2 1 4.6

Very pleased 28 11.5 18 18.0 7 11.3 2 3.5 1 4.5

Note.p* < .05

Research Question 1

For RQ1, I explored the association between the type of FSHE received and the

age of sexual initiation among sexually active African American women aged 18–24. To

answer RQ1, I used multinomial logistic regression to determine the association between

the self-reported type of FSHE received (abstinence-centered, abstinence-plus, or

comprehensive) and the age categories of sexual initiation classified as 0–13, 14–17, 18–

19, and 20 years and older for sexually active African American Women aged 18–24. I

did not include the initiation group over 20 years of age and young women 25–29 years

old in the analysis as the percentages of both categories were less than 5%. ACSHE and

the 0–13 age of sexual initiation categories were the referent groups. I used the mother’s

age at first birth, mother’s education, if the participant’s parents were married at their

birth, participant’s religious views, and their reaction to an unintended pregnancy as

covariates. ORs were used to measure the strength of the association with the CIs set at

95%, and the critical p-value significance was .05.

Most young women experienced sexual initiation between 14–17 years old

(70.3%, see Table 6) compared to the ≤13 years old age referent group (9.1%). According

to Table 11, compared to the age of initiation referent group, the odds of experiencing

sexual initiation in the 14–17 years old initiation category was 34% higher (OR = 1.34, p

= .73, 95% CI [0.26, 7.04]) if young women received APSHE and 22% (OR = 1.22 p =

.61, 95% CI [0.57, 2.58] higher if they received CSHE, compared to those who received

ACSHE. The odds of young women in this age of initiation category were 45% (OR =

1.45, p = .64, 95% CI [0.64, 3.29]) higher to be born to mothers between 20–24 years of

age who had an 8.78 (p = .01, 95% CI [1.96, 39.33]) higher likelihood to have a high

school diploma or GED. The odds of their parents being married was 4% (OR =1.04, p

= .95, 95% CI [0.29, 3.73] higher. The likelihood of young women in this age group to

view religion as very important was 81% (OR = 1.81, p = .42, 95% CI [0.43, 7.71])

higher than the referent groups. They felt very upset if they experienced an unintended

pregnancy was 11.05 (p = .01, 95% CI [2.05, 59.62]) times higher than the referent

groups.

Of the young women served, 20.7% experienced sexual initiation between the

ages of 18–19 (see Table 6). According to Table 11, compared to the age of initiation

referent group, the odds of experiencing sexual initiation in the 18–19 years old initiation

category was 5% (OR = 1.05, p = .98, 95% CI [0.16, 7.01]) higher if young women

received APSHE and 5% (OR = .95, p = .98, 95% CI [0.41, 2.23]) lower if they received

CSHE, compared to those who received ACSHE. Compared to the referent group (0–13

years of age), the odds of the young women in this age of initiation group were 2.25 (p =

.8, 95% CI [0.91, 5.59]) higher to be born to mothers between 20–24 years old who had a

7.34 (p = .02, 95% CI [1.30, 41.36]) higher likelihood to have a high school diploma or

GED. The odds of their parent being married were 40% (OR = 1.40, p = .63, 95% CI

[0.35, 3.63]) higher. The likelihood of young women in this age group to view religion as

very important was 5.07 (p = .06, 95% CI [0.92, 28.07]) higher and to feel very upset if

they experienced an unintended pregnancy was 4.10 (p = .16, 95% CI [0.56, 31.03]) times

higher than the referent groups.

Although the odds of being in an older category for the age of sexual initiation

among sexually active African American women aged 18–24 increased with APSHE and

CSHE compared to the referent group (0–13 years of age), the odds were not significant.

The age of initiation was still within a normative age range. As such, I failed to reject the

null hypothesis.

Research Question 2

For RQ1, I explored the association between the type of FSHE received and

contraceptive use at sexual initiation for sexually active African American women aged

18–24. To answer RQ2, I used logistic regression to determine the association between

the self-reported type of FSHE received (abstinence-centered, abstinence-plus, or

comprehensive) and contraceptive use (yes/no) at sexual initiation for African American

women aged 18–24. Young women 25–29 years of age were not included in the analysis

due to being omitted in the previous research question due to the group's percentage being

less than 5%. I used ACSHE and no contraceptive use at sexual initiation as the referent

groups. ORs were used to measure the strength of the association with the CIs set at 95%,

and the critical p-value significance was .05.

At sexual initiation, 67.5% of young women self-reported contraceptive use (see

Table 6). According to Table 12, the odds of contraceptive use at sexual initiation

increased by 14% (OR = 1.14, p = .80, 95% CI [0.42, 3.09]) with receipt of APSHE and

38% (OR = 1.38, p = .17, 95% CI [0.87, 2.19]), compared to those who received ACSHE.

Young women who used contraception at sexual initiation were 45% (OR = 0.55, p

= .19, 95% CI [0.23, 1.35]) less likely to be born to mothers 18–19 years old, the

likelihood that they were born to mothers with at least a high school diploma or GED was

2.91 (p =

.02, 95% CI [1.21, 7.01]) higher. Their parents were 32% (OR = 0.68, p = .23, 95% CI

[0.37, 1.27]) less likely to be married. They were 22% (OR= 0.78, p = .33, 95% CI

[0.48,1.28]) less likely to view religion as somewhat important and their reaction to an

unintended pregnancy was 23% (p = .68, 95% CI [0.46, 3.29]) higher to be very upset

than the referent group.Although the odds of contraceptive use at sexual initiation among

sexually active African American women aged 18–24 increased with receipt of CSHE,

the odds were not significant. As such, I rejected the null hypothesis.

Table 11

Impact of Formal Sexual Health Education on Contraceptive Use at Sexual Initiation

Contraceptive use at sexual initiation

(No – Ref.) uOR

Una

djuste p

95% CI

Adjusted

aOR p 95% CI

Formal sexual health

education (Abstinence-

centered - Ref.)

Abstinence-plus 1.06

.90

[0.43, 2.63] 1.14 .80

[0.42, 3.09]

Comprehensive 1.29 .23 [0.86, 1.93] 1.38 .17 [0.87, 2.19]

Mother’s age at first birth (<18 -

Ref.)

18–19 0.66 .28 [0.31, 1.41] 0.55 .19 [0.23, 1.35]

20–24 0.97 .87 [0.68, 1.39] 0.86 .46 [0.58, 1.28]

≥25 1.03 .82 [0.79,

.35] 0.92

.60

[0.68, 1.26]

Yes Mother’s education (≤11th grade

- Ref.)

High school diploma/GED

2 years of college/associate

degree

Bachelor’s degree or higher

2.87

1.33

1.18

.01

.16

.75

[1.12,

[0.89,

[0.87,

6.42]

1.98]

1.61]

2.91

1.31

1.11

.02

.25

.58

[1.21,

[0.83,

[0.76,

7.01]

2.07]

1.62]

Married parents at participants’

birth (No - Ref.)

Yes 0.74

[0.43,

1.29] 0.68

[0.37, 1.27]

Importance of religion (Not

important - Ref.)

Very important 0.64

[0.30,

1.38] 0.49

[0.21, 1.12]

Somewhat important 0.86 .51 [0.55, 1.34] 0.78 .33 [0.48, 1.28]

Reaction to an unintended

pregnancy (Would not care -

Ref.)

Very upset 1.21 .69

[0.48,

3.03] 1.23

[0.46, 3.29]

A little upset 1.11 .65 [0.70, 1.77] 1.07 .78 [0.66, 1.74]

A little pleased 1.02 .89 [0.73, 1.43] 1.11 .95 [0.71, 1.43]

Very pleased 1.03 .85 [0.78, 1.36] 1.13 .43 [0.84, 1.52]

Research Question 3

With RQ3, I explored the association between the type of FSHE received and the

effectiveness of the contraceptive method selected at sexual initiation for sexually active

African American women aged 18–24. To answer RQ3, I used multinomial logistic

regression to determine the association between the type of self-reported FSHE received

(abstinence-centered, abstinence-plus, or comprehensive) and effectiveness of

contraceptive method selected (none, less effective, moderately effective, and highly

effective) at sexual initiation by sexually active African American women aged 18–24.

At sexual initiation, 50.6% (n = 121) of young women used a LECMs, with a high

percentage, 43.5% (n = 104) using MECMs (see Table 6). None (2.5%, n = 6) and

LECMs categories were combined, as were MECMs and HECMs (3.4%, n = 8)

contraceptive method categories as the percentages of none and HECMs were less than

5%. I did not include young women 25–29 years of age in the analysis due to the

percentage of the group being less than 5%. ACSHE and none–LECM categories selected

were the referent groups. ORs were used to measure the strength of the association with

the CIs set at 95%, and the critical p-value significance was .05.

According to Table 13, the odds of using a MECM to HECM was 48% (OR =

0.52, p = .18, 95% CI [0.20, 1.35]) less likely with receipt of APSHE and 16% (OR =

0.84, p = .4, 95% CI [0.55, 1.27]) less likely with receipt of CSHE. Young women who

used MECMs to HECMs were at 65% (OR = 1.65, p = .22, 95% CL [0.75, 3.64]) higher

odds of being born to a mother 18–19-years old was 58% (OR = 0.42, p = .05, 95% CI

[0.18, 1.00]) less likely to have a high school diploma or GED, which was significant. It

was 39% (OR = 1.39, p = .28, 95% CI [0.76, 2.52]) more likely that the participant’s

parents were married at their birth and was 34% (OR = 1.34, p =.45, 95% CI [0.63, 2.86])

more likely to view religion as important. They were 5% (OR = 1.05, p = .7, 95% CI

[0.80, 1.38]) to be very pleased if they experienced an unintended pregnancy.

The odds of selecting a MECM to HECM at sexual initiation among sexually active

African American women aged 18–24 did not increase with receipt of CSHE. As such, I

failed to reject the null hypothesis.

Research Question 4

For RQ4, I explored the association between the type of FSHE received and the

effectiveness of the contraceptive method primarily selected by sexually active African

American women aged 18–24 at intercourse within the past 12 months. To answer RQ4, I

used multinomial logistic regression to determine the association between the selfreported

type of FSHE received (abstinence-centered, abstinence-plus, or comprehensive) and

effectiveness categories of the contraceptive method primarily selected (none, less

effective, moderately effective, and highly effective) by sexually active African American

women aged 18–24 at last intercourse within the past 12 months. ACSHE and no use of

contraception were referent groups. ORs were used to measure the strength of the

association with the CI set at 95%, and the critical p-value significance was .05.

None (n = 100, 41.3%; see Table 6) was the contraceptive effectiveness category

African American women aged 18–24 primarily selected during intercourse for the last

12 months. Young women used LECMs (n = 62, 25.6%), MECMs (n = 58, 24%) and

HECMs (n = 22, 9.1%) contraceptive effectiveness categories, respectively. According to

Table 14, young women who used LECMs were 2.05 (OR = 2.05, p = .27, 95% CI [0.57,

7.41]) times more likely to have received receipt of APSHE and 18% OR = 1.18, p = .58,

95% CI [0.66, 2.11]) more likely to have received CSHE. Young women who selected

LECMs over the last 12 months were 10% (OR = 1.10, p = .68, 95% CI [0.69, 1.76])

more likely to be born to mothers 20–24 years old who were 1% (OR = 1.01, p = .95,

95% CI [0.66, 1.57]) more likely to have earned a baccalaureate degree and 2.11 (p = .05,

95% CI [0.99, 4.49]) times more likely to be married at the participants’ birth, which was

significant. The odds were 12 % (OR = 1.12, p = .7, 95% CI [0.64, 1.96]) more likely that

they viewed religion as somewhat important and 4.96 (p = .03, 95% CI [1.18, 20.76])

times more likely that they would be very upset if an unintended pregnancy occurred,

which was significant.

Within the previous 12 months, young women were less likely to primarily select

MECMs if they received APSHE (OR = 0.66, p = .5, 95% CI [0.20, 2.29]) or CSHE (OR

= 0.75, p = .2, 95% CI [0.44, 1.27]) (see Table 14). They were 64% (OR = 1.64, p = .06,

95% CI [0.98, 2.74]) more likely to be born to mothers aged 20–24-years of age who

were 68% (OR = 0.32, p = . 05, 95% CI [0.10, 1.02]) less likely to have a high school

diploma or GED, which was significant. The odds were 17% (OR = 1.17, p = .58, 95%

CI [0.67, 2.04]) more likely that they viewed religion as somewhat important and 15.39

(p = .01, 95% CI [1.18, 130.86]) times more likely to be very upset if an unintended

pregnancy occurred, which was significant.

The odds of primarily selecting a HECM for intercourse over the last 12 months

increased by 18% (OR = 1.07, p = .7, 95% CI [0.50, 2.85]) if young women received

CSHE. The odds increased, 7% (OR = 1.07, p = .98, 95% CI [0.15, 7.83]), with receipt of

APSHE. They are 42% (OR = 1.42, p = .3, 95% CI [0.73, 2.78]) more likely to be born to

mothers 20–24years old who were 63% (OR = 0.37, p = .21, 95% CI [0.08, 1.75]) less

likely to have high school diploma or GED, and 76% (OR = 1.76, p = .3, 95% CI [0.60,

5.18]) more likely to have been married at the participants’ birth. The odds were 4.03 (p =

.1, 95% CI [0.75, 21.55]) times more likely that they viewed religion as important and

2.56 (p = .2, 95% CI [0.61, 10.69]) times more likely that they would be very upset if an

unintended pregnancy occurred. Receipt of APSHE nor CSHE increased the odds of

selecting a moderately to highly effective contraceptive as the primary contraceptive

method for the past 12 months among sexually active African American women aged 18–

24. As such, I failed to reject the null hypothesis.

Summary

In Chapter 4, I reported results of a quantitative cross-sectional analysis of the

2015–2017 NSFG that explored the impact that receipt of the various forms of FSHE

(abstinence-centered, abstinence-plus, and comprehensive) may have on the age of sexual

initiation, the contraceptive use at sexual initiation, use of more effective contraceptives

at sexual initiation, and use of more effective contraceptives as the predominate method

during sexual encounters over the previous 12 months among African American women

aged 18–24 who have sex with men. According to the data analysis, the odds of being in

an older category for the age of sexual initiation among sexually active African American

women aged 18–24 increased with APSHE. It increased with CSHE when compared to

ACSHE, the referent group. However, the odds were not significant, and the age of

initiation was still within a normative age range. The odds of contraceptive use at sexual

initiation among sexually active African American women aged 18–24 increased with

receipt of CSHE; however, the odds were not significant. The odds of selecting a

moderately to highly effective contraceptive at sexual initiation among sexually active

African American women aged 18–24 did not increase with receipt of CSHE. The receipt

of APSHE nor CSHE increased the odds of selecting a moderately to highly effective

contraceptive as the primarily selected contraceptive method for the past 12 months

among sexually active African American women aged 18–24.

Chapter 5 includes an interpretation of the analysis results from Chapter 4 and the

limitations of this study and its implications and recommendations for applying the

results to improve the provision of FSHE among African American young women aged

18–24. Chapter 5 includes a comprehensive overview of this research study, a description

of my plans for the future use of study data and results. It also includes possibilities for

continued research on this topic.

Chapter 5: Discussion, Conclusion, and Recommendations

Introduction

The purpose of this quantitative cross-sectional analysis of the 2015–2017 NSFG

was to ascertain the impact receipt of FSHE (knowledge; main independent variable) on

sexual initiation and contraceptive selection and use (dependent variables) among African

American women aged 18–24 who have sex with men. I used descriptive statistics,

logistic regression, and multinomial logistic regression to evaluate the research questions

and accept or reject the hypotheses. Understanding associations between these variables

may lead to better provision of public health services that may lead to delayed sexual

initiation, as well as increased, consistent and proper use of preference and intention

matched methods, reducing occurrences of unplanned pregnancies in this population.

Unintended pregnancy can lead to significant emotional, social, health, and financial

complications, which in turn, may perpetuate cycles of disadvantage experienced by

vulnerable populations, especially in terms of racial disparities in morbidity and mortality

across the. This analysis may lead to positive social change by highlighting the

importance of the provision of FSHE to African American young women aged 18–24

who have sex with men. African American young women may apply knowledge and

skills learned to prevent unintended pregnancies, and by extension, may have fewer

adverse maternal and child public health outcomes.

The goal of FSHE is to prepare young adults to be sexually healthy (Santelli et al.,

2017). A literature review provided evidence that racial and ethnic disparities involving

contraceptive knowledge exist with an association shown between less knowledge and

less contraceptive use. Knowledge and attitudes formed about contraception during

adolescence may persist throughout life (Guzzo & Hayford, 2018). Receipt of FSHE was

lower among individuals with lower socioeconomic status, who were African American,

and whose mothers had lower educational attainment (Lindberg & Maddow-Zimet,

2012). African American young women report earlier sexual initiation and are more

likely to experience unwanted or mistimed pregnancy than their peers (Moilanen et al.,

2018). Contraceptive knowledge is typically lower, and unintended pregnancies are

higher in 18–19-year-old women (Wu et al., 2016). Dehlendorf et al. (2014) found that

African American women under the age of 19 were less likely than Caucasian women to

use IUDs or implants, while African American young women between the ages of 20–24

were slightly more likely to use IUDs and implants than Caucasian women.

For the current study, I analyzed data from the 2015–2017 NSFG to determine the

impact of FSHE on age of sexual initiation, use of contraceptive at sexual initiation, use

of more ECM at sexual initiation, and use of more ECM as the predominate method

during sexual encounters during the 12 months before the survey among African

American women aged 18–24. I compared the impact based on the receipt of ACSHE

(referent group) with that of APSHE and CSHE. Most young women surveyed

experienced sexual initiation between 14–17 years of age (70.3%) compared to the ≤13

years of age referent group (9.1%; see Table 6). The results of my study indicate that the

odds of experiencing sexual initiation in the 14–17-year-old initiation category were 34%

higher (OR = 1.34, p = .73; see Table 11) if young women received APSHE and 22%

higher (OR = 1.22 p = .61) if they received CSHE, compared to those who received

ACSHE. The odds of being in an older category (18–19) for the age of sexual initiation

were 5% higher (OR = 1.05, p = .98) if young women received APSHE and 5% lower

(OR = 0.95, p = .98) if they received CSHE, compared to those who received ACSHE.

The odds of contraceptive use at sexual initiation increased by 14% (OR = 1.14, p = .8;

see Table 12) with receipt of APSHE and 38% (OR = 1.38, p = .17), compared to those

who received ACSHE. The odds of selecting a MECM to HECM over a none to less

effective category at sexual initiation did not increase with receipt of APSHE (OR = 0.52,

p = .18; see Table 13) nor with CSHE (OR = 0.84, p = .4).

The receipt of APSHE nor CSHE increased the odds of selecting a MECM to

HECM as the primarily selected contraceptive method during the 12 months before the

survey among sexually active African American women aged 18–24. Young women who

used LECMs were 2.05 times more likely (OR = 2.05, p = .27; see Table 14) to have

received receipt of APSHE and 18% more likely (OR = 1.18, p = .58) to have received

CSHE. Young women were less likely to primarily select MECMs if they received

APSHE (OR = 0.66, p = .5) or CSHE (OR = 0.75, p = .2). The odds of primarily selecting

a HECM for intercourse over the last 12 months increased by 18% (OR = 1.07, p = .7) if

young women received CSHE. The odds increased, 7% (OR = 1.07, p = .98), with receipt

of APSHE.

Interpretation of the Findings

Formal Sexual Health Education

FSHE promotes safer sexual behaviors for young adults, including delayed sexual

initiation and increased and correct contraceptive usage beginning at first sexual

encounter (Lindberg & Maddow-Zimet, 2012). African American women aged 18–24 are

at a higher risk for unintended pregnancy due to nonuse, inconsistent, or improper use of

contraception, as well as failure to use dual contraceptive methods (Craig et al., 2014).

Shepherd et al. (2017) found that African American adolescents who received CSHE had

more favorable attitudes about contraception, demonstrated an increase in consistent

contraception usage, and reported less teen pregnancy and vaginal intercourse,

highlighting the importance of early CSHE. Therefore, knowledge and attitudes formed

about contraception during adolescence may persist throughout life. This suggests the

need for increased CSHE among African American young women who have poorer

sexual reproductive health outcomes, including higher rates of unintended pregnancy and

STIs.

Based on my analysis of the 2015–2017 NSFG, I found that of the 242 African

American women aged 18–24 included in this study, 63.2%; received CSHE. Twentyfour

percent still received some information about contraception via APSHE, 12.8% receiving

FSHE that was ACSHE. Of the young women surveyed, 61.6% were born to unwed

mothers, of which 36.8% only earned a high school diploma or GED. At participant’s

birth, 29.7% of their mothers were 20–24 years old at their births, while

27.2% were under 18.

Formal Sexual Health Education and Sexual Initiation

Sexual initiation, considered a critical life transition event, is a significant

milestone towards adulthood as young adults accept increasing responsibility for their

decisions. The mean age of sexual initiation among American women is approximately

17, with 15% having had their first heterosexual vaginal intercourse experience before

their 15th birthday (Guttmacher Institute, 2018; Magnusson et al., 2015). African

American young women report earlier sexual initiation (Moilanen et al., 2018). Current

national norms for the U.S. consider sexual initiation that occurs before 15 as early,

between 15 and 19 as normative, and after 19 as late (Golden et al., 2016). The normative

range for this study decreased to age 14, as African American young women report earlier

sexual initiation (Moilanen et al., 2018).

Based on the findings for this study, I could not reject the null hypothesis for

RQ1. This finding aligns with and supports results from the literature that found that

receipt of FSHE had little impact in increasing the age of sexual initiation for African

American young women aged 18–24 past a normative age. Young women were 34%

more likely to experience normative sexual initiation with receipt of APSHE and 22%

more likely with receipt of CHSE (see Table 11). They were, however, only 5% more

likely to wait until age 18–19 to have sexual intercourse with receipt of APSHE and were

5% less likely to wait with receipt of CSHE. The findings for RQ1, although

nonsignificant, are still important because they suggest that FSHE received may provide

knowledge that consensual sexual behavior can be healthy, normative, and provides

young adults with the tools needed to make well-informed, healthy decisions around

sexual experiences (Bourke et al., 2014; Golden et al., 2016).

The findings, when viewed through the lens of the RAA, suggest that

demographics and norms may have also played a role in the age of sexual initiation with

young women within the normative initiation range. African American women are

exposed to higher rates of teen and single parenthood, earlier sexual initiation, and use

contraception less effectively (Barber et al., 2015). Key findings from my analysis

included that young women within the normative initiation range were more likely born

to unwed mothers (aOR = 1.04, p = .95; see Table 11), with only a high school diploma

or GED (aOR = 8.78, p = .01), and under the age of 24 (aOR = 1.45, p = .64). Young

women who delayed initiation until 18–19 years old were more likely to be born to wed

mothers (aOR = 1.40, p = .63), who had a high school diploma or GED (aOR = 7.34, p

= .02), and under the age of 24 (aOR = 2.25, p = .8). Young women in the 18–19 year old

initiation age group were more likely to view religion as very important (aOR = 5.07, p =

.06) than their 14–17 year old counterparts (aOR = 1.81, p = .42).

Formal Sexual Health Education and Contraceptive Selection and Use

Despite most young adults believing that pregnancy should be planned (86-94%),

more than 70% of pregnancies among unmarried women aged 18–24 are unplanned and

likely due to nonuse or inconsistent contraceptive use (Kornides et al., 2015). Paterno et

al. found in their 2017 study that among women aged 20–24, 11.4% to 13% used no

contraceptive method. This number was higher among African American participants,

despite 60% of African American women aged 18-24 who believed pregnancy planning

is important, and 80% desired to avoid pregnancy (Paterno et al., 2017). The findings of

this current study align with their study and emphasize the need for more research around

contraceptive usage among African American women aged 18–19, as, in my analysis,

32.5% did not use contraception at sexual initiation (see Table 6). Among the 67.5% of

young women who self-reported contraceptive use at sexual initiation, 50.6% used a

LECMs, and 43.5% (n = 104) using MECMs. None (2.5%) and LECM categories were

combined, as were MECMs and HECMs (3.4%) categories as the percentages of none

and HECMs were less than 5%. Within the last 12 months, the effectiveness category

selected shows that more young women (41%) opted not to use contraception. Among

those that used contraceptives, LECMs (25.6%) and MECMs (24%) were predominately

used. The use of HECMs (9.1%) methods was scarce. The findings from my study align

with current literature as LARC use remains low in young women despite being one of

the most effective forms of pregnancy prevention (Logan et al., 2018). Fewer than 5% of

15–19 year olds select LARCs as their preferred contraception method, and the usage

among young African American women remains lower than in other racial/ethnic groups

(Coates et al., 2018; Logan et al., 2018).

The results of RQ2, although nonsignificant, demonstrate that contraceptive use at

sexual initiation increased by 14% with receipt of APSHE (OR = 1.14, p = .8) and by

38% with CSHE (OR = 1.38, p = .17) compared to those who received ACSHE (see

Table 12). While FSHE increased use at sexual initiation, odds of selecting a MECM to

HECM at sexual initiation were 48% less likely with receipt of APSHE as well as 16%

less likely with receipt of CSHE (see Table 13). The results from RQ3 suggest there is a

need for more culturally competent contraceptive education to reduce misconceptions

about long-term contraceptive methods. African American women who perceive negative

peer norms involving contraceptive use are more likely to have inconsistent contraceptive

usage and utilize LECMs (Crosby et al., 2013). Compared to Caucasian and Hispanic

women, African American women are more likely to report concerns regarding side

effects of contraceptive methods as their rationale for not using contraception (Kornides

et al. 2015). Contraception use, specifically LARC use, might be increased by providing

young women more information on effectiveness, side effects, affordability, and ease of

use (Kornides et al., 2015).

Similar usage persists into the predominate contraceptive method used within the

last 12 months. My analysis of the data found that young women who received APSHE

were more likely to select LECMs (aOR = 2.05, p = .27) compared to MECMs (aOR =

0.66, p = .5), and HECMs (aOR = 1.07, p = .95) methods (see Table 14). Young women

who received CSHE were more likely to equally select LECMs (aOR = 1.18, p = .58) and

HECMs (aOR = 1.18, p =.7) methods compared to MECMs (aOR = 0.75, p = .2)

methods. It is important to note that within the last 12 months, HECMs were used as

frequently as LECMs among recipients of CSHE. These findings align with those found

by Dehlendorf et al. (2014), who noted that African American women between the ages

of 20–24 were slightly more likely to use IUDs and implants than Caucasian women.

Logan et al. (2018) found that LARC use in young women nearly doubled at the age of

24. Although the results of RQ4 show that use of HECMs is still low, it suggests that

CSHE may be an important factor to increase use in this population by addressing

negative attitudes and beliefs about LARCs stemming from social norms, the knowledge

gained through vicarious experiences, as well as knowledge of institutional racism and

medical mistrust.

The findings, when viewed from the lens of the RAA framework, suggest that

FSHE received shaped attitudes and norms by providing knowledge regarding

contraceptive intention, starting at initiation, and influences the selection of effectiveness

category. In my analysis, young women predominately used LECMs, which may have

been impacted by demographics and norms. Young women who used contraception at

sexual initiation were less likely to be born to mothers 18–19 years old (aOR= 0.55, p

= .19; see Table 12). Most mothers had at least a high school diploma or GED (aOR =

2.91, p = .02), but were less likely to be married (aOR = 0.68, p = .23). Young women

who used contraception at sexual initiation were also less likely to view religion as very

important (aOR= 0.49, p = .09). Young women with mothers 18–19 years old (aOR=

1.65, p = .22) and married (aOR= 1.39, p = .28) at their birth were more likely to use

MECMs to HECMs at sexual initiation, which suggests a protective factor (see Table 13).

They were less likely born to mothers who only had a high school diploma or GED (aOR

= 0.42, p = .05). Young women who selected MECMs to HECMs at sexual initiation

were more likely to view religion as very important (aOR = 1.34, p = .28).

According to my analysis of the data, when selecting a primary contraceptive method

over the last 12 months, young women born to an older mother selected more effective

methods with most selecting MECMs (aOR = 1.64, p = .06) or HECMs (aOR = 1.42, p

= .3; see Table 14). They also were less likely born to women who earned only a high

school diploma or GED (moderately effective: (aOR = 0.32, p = .05) and highly effective

(aOR = 0.37, p = .21). Young women born to married parents selected LECMs (aOR =

2.11, p = .05), which suggests a family-minded norm as they also reported they would be

a little pleased with an unintended pregnancy (aOR = 1.68, p = .04).

Limitations

The purpose of this quantitative cross-sectional analysis of the 2015–2017 NSFG

was to explore the receipt of FSHE and the impact it may have on sexual initiation,

contraceptive use, and contraceptive selection among African American women aged 18–

24 who have sex with men. As a secondary analysis of data, a limitation of this research

may be possible nonsampling errors due to the cross-sectional nature of the NSFG and

primary data collection. As such, I am unable to correlate the extent to which knowledge

impacts contraceptive selection, nor can I draw causal inferences from these

crosssectional data. Another limitation of this study is the use of self-reported data, which

is subject to participant recall and interviewer biases as well as over- and under-reporting.

The majority of the NSFG responses occurred via ACASI to reduce limitations and

biases.

The sample size was also a limitation of this study. Although the oversampling of

the study population occurred to include more African American women, the sample size

was still small; only 1,355 of the 5,554 women interviews were conducted with

nonHispanic Black women. The sample size for this study was further limited because a

subsample used contained 499 sexually active, African American women aged 18–24

who have sex with men. Data cleaning yielded a final sample size of 242 sexually active,

African American women aged 18–24 who have sex with men. This small sample size

highlights the need for more robust, targeted studies among African American women

instead of limited qualitative and small studies focusing on nonuse or large surveys

providing percentages of African American women respondents.

Access to the data needed to complete this study was another limitation.

Requirements for FSHE vary by state; therefore, the region of residence was a key topic

in my initially planned study. The NSFG includes a restricted variable, REGION, that

identifies a respondent’s region of residence. When I started this research study, the

REGION variable for the 2015–2017 NSFG was listed on the NSFG website as available

free of charge, with access grant via an encrypted email with a written request and

submission of confidentiality forms. After submitting a written request for access to the

REGION variable, the variable was changed to a fee-based model with limited access,

which made the inclusion of this variable cost-prohibitive.

Recommendations

Practice Recommendations

The goal of FSHE is to prepare young adults to be sexually healthy (Santelli et al.,

2017). Despite concerns, there is no evidence to support that receipt of CSHE is

associated with an earlier sexual initiation, increased risk-taking, or poorer sexual

reproductive health outcomes. CSHE, which occurs before sexual initiation, leads to

delayed onset of sexual initiation; greater use of more effective contraception, especially

at first sexual encounter; and healthier sexual relationships (Jaramillo et al., 2017;

Lindbery & Maddow-Zimet, 2012). The key results of my data analysis demonstrate that

receipt of FSHE has an impact on sexual initiation and contraceptive use and selection

among African American women aged 18–24. APSHE and especially CSHE lead to

delayed sexual initiation, increased contraceptive use at sexual initiation, and more

effective contraceptive use long term. ACSHE, however, does not have a significant

impact on sexual initiation or sexual behaviors and may place young people at increased

likelihood of pregnancy and STIs once they become sexually active by overlooking or

downplaying the benefits of contraception (Boyer, 2018).

Young women, especially those of color, may be at risk of not receiving

highquality family planning care (Dehlendorf et al., 2014). This disparity may contribute

to healthcare disparities, disengagement from healthcare interactions, and perpetuated

institutional mistrust (Gomez & Wapman, 2017). Based on the results of this study, my

recommendation is an increased provision of CSHE that is culturally aware and

responsive for African American young women in educational, public health, and medical

settings. Reproductive trends in contraceptive use among African American women may

be impacted by a reproductive history of medical and reproductive abuse. As such, CSHE

needs to be tailored to address social and cultural norms to offset vicarious experiences

and misperceptions. CSHE must be tailored to address side-effects and perceived impacts

on fertility specifically. CSHE may help reduce medical mistrust, misperceptions about

contraceptive use, especially LARCs, and help African American young women find a

contraceptive method best matched to their pregnancy intentions. By being aware of

reproductive justice issues, public health sexual health educators and practitioners may be

better able to address concerns raised by African American young women.

Clinicians are more likely to recommend LARCs to low-income women of color

compared to their Caucasian counterparts (Gomez & Wapman, 2017). This type of

provider coercion was seen particularly with patients deemed high-risk, despite their

contraceptive preferences (Morse & Moos, 2018). Provider coercion occurred despite

income, with well-educated patients describing feelings of pressure to be compliant,

discomfort with expressing their opinions and views, and the need to accept provider

recommendations (Joseph-Williams et al., 2014; Sacks, 2017). Public health educators

can fill this communication gap through the provision of CSHE that considers African

American young women’s long-term goals as well as pregnancy intentions. Sexual health

interventions should be tailored to promote favorable attitudes and social norms towards

contraception as young women with positive normative beliefs are more likely to use

contraception. Such efforts might include integrating reproductive planning into all

primary care encounters where questions and information regarding pregnancy intention

as well as effectiveness, ease of use, affordability, and satisfaction with current

contraceptive methods are explored (Kelly, 2014; Kornides et al., 2015). This would

provide the opportunity for public health and clinical practitioners to work together to

improve the reproductive health of African American young women. This joint

partnership would allow public health educators the ability to focus on reducing the

number of sexual partners, increasing the use of hormonal contraception, increase the use

of dual contraceptive methods, and use of LARCs (Buhi et al., 2010; Kornides et al.,

2015). Doing so would allow clinicians to focus on clinical services.

Research Recommendations

A gap in research involving reproductive and contraceptive disparities in African

American young women persists. Targeted research that aggressively and primarily

focuses on African American young women is needed. More extensive quantitative

studies are required to determine the impact of FSHE on reproductive and contraceptive

outcomes, especially based on the region of residence. More nuanced qualitative studies

are needed to explore the effect that background factors, including social norms and

networks, play on reproductive and contraceptive outcomes among African American

young women. My research plans include an exploration of FSHE and reproductive and

contraceptive outcomes among African American young women attending historically

Black colleges and universities. Addressing this gap in literature may lead to the

development of interventions designed to prevent unintended pregnancies through

increased consistent, preference, and intention matched contraceptive use. Well-designed

interventions that focus on CSHE may improve maternal health and reduce adverse birth

outcomes, all of which are public health concerns in young African American women.

An additional plan may include an exploration into the implicit and explicit biases

African American young women experience during family planning.

Implications

African American young women report earlier sexual initiation, use of LECMs,

and are more likely to experience an unwanted or mistimed pregnancy than their peers

(Golden et al., 2016; Lanier et al., 2018; Moilanen et al., 2018). CSHE may have a

protective influence on early sexual initiation (Bourke et al., 2014). Unintended

pregnancy may lead to disparities in delayed prenatal care, reduced breastfeeding,

depression, delivery of infants with low birth weight and birth defects, as well as reduced

average income compared to other women (Carter et al., 2013; James & Rashid, 2013;

USDHHS, 2019). Infants born from unintended pregnancies experience increased risk of

poor mental, physical, and behavioral health in adolescence, as well as lower educational

attainment themselves (James & Rashid, 2013; USDHHS, 2019). Findings from my

study demonstrate that increasing African American young women’s access to FSHE,

which highlights modern contraceptive methods, may mitigate these factors.

Positive Social Change

The key findings of my data analysis may lead to positive social change by

providing public health educators and practitioners with a better understanding of the

impact of FSHE on contraceptive use, contraceptive selection, and sexual initiation

among African American young women. APSHE and especially CSHE lead to delayed

sexual initiation, increased contraceptive use at sexual initiation, and more effective

contraceptive use long term. ACSHE, however, does not have a significant impact on

these behaviors as it withholds essential information to make healthy sexual decisions.

An increased understanding of this impact of FSHE will help public health educators and

practitioners provide African American young women with the required knowledge and

skills needed to delay sexual initiation as well as consistently and properly use preference

and intention matched methods. African American young women may apply the

knowledge and skills learned to prevent unintended pregnancy and may decrease adverse

maternal and child associated public health outcomes. Doing so may positively impact the

socioeconomic status of African American young women and resulting children, which

may improve associated social, educational, and public health outcomes for both

populations.

Conclusion

Through this quantitative, cross-sectional analysis of the 2015–2017 NSFG, I

explored receipt of FSHE and the impact it had on sexual initiation, contraceptive use and

selection among African American women aged 18–24 who have sex with men. Key

findings from this study include that among sexually active African American Women

aged 18–24, neither APSHE nor CSHE extended the age of sexual initiation past a

normative age. Contraceptive use at sexual initiation increased with receipt of CSHE, and

the odds of selecting a MECM to HECM at sexual initiation did not increase with receipt

of CSHE. The receipt of APSHE, as well as CSHE, did not increase the odds of selecting

a MECM to HECM as the primarily selected contraceptive method for the past 12 months

among sexually active African American women. CSHE, however, did lead to young

women selecting HECMs at similar odds to LECMs.

The results of this study add to the literature on the impact of FSHE on the age of

sexual, contraceptive use, and contraceptive selection among African American young

women aged 18–24, which is a public health concern as they impact on unplanned and

mistimed pregnancy. Understanding contraceptive selection among African American

young women may assist public health and clinical practitioners with a better

understanding of the impact of FSHE on contraceptive use and selection and sexual

initiation among African American young women. The key findings of this analysis may

lead to positive social change by providing African American young women with the

required knowledge and skills needed to delay sexual initiation as well as consistently and

properly use preference and intention matched methods, reducing the occurrence of

unplanned pregnancy in this population. The key findings of this analysis may be used to

increase the provision of culturally competent public health services, such as sexuality

and reproductive education, family planning, and contraceptive counseling, which may

increase the selection and consistent use of contraceptive methods best fit for

reproductive intentions, reduce usage failures and improve birth-related outcomes.

Students also viewed