Abstract
Objective:
Nonadherence in sexual risk reduction interventions may be common among adolescents. We compared intervention completion rates among adolescent and young adult women with and without a prior pregnancy or sexually transmitted infection (STI) participating in a program to improve contraceptive continuation.
Design:
Secondary data analysis from a feasibility study of a health coaching intervention to improve contraceptive continuation.
Setting:
Three urban pediatric clinics in Philadelphia.
Participants:
Women ages 14–22 years who were English-speaking, sexually active in the past year, not desiring pregnancy in the next year, and starting a new contraceptive method.
Interventions:
At baseline, participants completed a sociodemographic questionnaire and semi-structured interview, followed by five monthly coaching sessions. Interviews and coaching sessions were audio-recorded, transcribed, and coded for thematic content.
Main Outcome Measures:
Intervention completion was defined as the number of completed coaching sessions.
Secondary Outcomes:
Qualitatively explored group differences in reproductive knowledge, attitudes, and risk perception.
Results:
Participants with a prior adverse outcome (a prior STI and/or a prior pregnancy) completed fewer coaching sessions than those without such history (median: 2 vs. 4, p=0.03). Both groups had low HIV/STI knowledge, negative attitudes towards pregnancy, and low HIV/STI risk perception. Those with a prior adverse reproductive outcome held more negative attitudes towards condoms.
Conclusion:
Despite similar reproductive knowledge, attitudes, and risk perception, young women who have experienced an adverse reproductive outcome may be less likely to fully engage in sexual risk reduction interventions. Future studies should confirm these findings and consider strategies to optimize interventions reach for vulnerable youth.
Keywords: reproductive health, adolescent, sexually transmitted diseases, sexual behavior, unintended pregnancy
INTRODUCTION
Adverse reproductive health outcomes are events that occur at a less than ideal time (unintended or mistimed pregnancy) or are associated with negative health effects (like sexually transmitted infections, or STIs). Adverse reproductive health outcomes disproportionally occur among young women under age 25 years.1 One in four youth aged 15–24 have been diagnosed with a sexually transmitted infection (STI)2 and unintended pregnancies are highest in this age group.3 Young women who previously experienced an STI are at a higher risk of subsequent reinfections and sequelae such as pelvic inflammatory disease and altered fertility.4,5 Additionally, unintended pregnancy places young women at greater risk of economic hardship and other negative social outcomes.6 These young women are therefore important to reach with sexual risk reduction interventions.
Among youth targeted by sexual risk reduction interventions, adherence (defined as the number of intervention sessions completed) varies among subpopulations.7,8 Nonadherence may result from attrition (drop-out or losses to follow-up) or inconsistent engagement. A review of 142 adolescent pregnancy prevention interventions found that while attrition is often reported and varies widely from 0.5% to 58%, few studies report variations in the intervention dose received. Moreover, little is known about the characteristics of those who drop out or engage inconsistently.9 In a meta-analysis of HIV prevention interventions, Noguchi and colleagues noted that participants who reported engaging in higher risk sexual behaviors at enrollment (e.g., condom non-use) were more likely to drop out.7 Whether nonadherence is higher among women who previously experienced an adverse reproductive outcome remains under-explored. Such information is valuable as it can inform efforts to tailor programs to reach those/women most vulnerable to experiencing preventable outcomes.
We compared intervention completion rates among young women with and without a prior adverse reproductive outcome participating in an intervention to improve contraceptive continuation rates. We explored whether the two groups differed in behavioral determinants known to predict intervention engagement, including reproductive health knowledge, attitudes, and risk perception.
MATERIALS AND METHODS
Study Design, Eligibility, Recruitment and Setting
This is a secondary analysis of data from the feasibility study of the Health Coaching for Contraceptive Continuation (HC3) intervention, which was conducted between March and December 2017. Eligible Participants were ages 14–22 years, sexually active with a male in the prior 12-months, not desiring pregnancy in the next 12-months, English-speaking, able to provide informed consent, and started a new contraceptive method within 14 days prior to enrolling. Eligible contraceptives included the hormonal pill, transdermal patch, vaginal ring, intramuscular shot, intrauterine device (IUD), and implant. Recruitment occurred at three urban clinics in Philadelphia affiliated with the Children’s Hospital of Philadelphia (CHOP), including an adolescent specialty clinic and two Title X adolescent family planning clinics at pediatric primary care sites. Annually, these sites care for 2,000 age-eligible youth residing in city zip codes with high rates of early sexual debut and teen pregnancy.10,11 The study was reviewed and approved by the Institutional Review Board at the CHOP and AccessMatters, the local Title X program funder. All participants provided informed consent, regardless of age; consistent with Federal law and Pennsylvania law, parental permission was waived for minors seeking confidential contraceptive services.
HC3 Intervention Description
Study visits occurred via phone or in-person at the clinic or a research office, based on participant preference. At baseline, participants completed a questionnaire, semi-structured contraceptive needs assessment interview, and created a ranked list prioritizing nine reproductive health topics they wanted to learn about during coaching. The questionnaire assessed demographic characteristics (age, race, education level, and insurance), medical history (chronic conditions and medications), and reproductive history (menarche, pregnancies and STIs). We included age of menarche as it is associated with early pregnancy and some STIs.12 Interviews were conducted by coaches, audio-recorded, lasted approximately 40 minutes and explored pregnancy attitudes; influences on contraceptive use decisions; and barriers and facilitators of contraceptive adherence. Coaches used the interview data to collaboratively develop a structured coaching plan that identified participants’ unique barriers and supports for contraceptive use. The reproductive topics participants ranked included contraceptive methods, contraceptive adherence strategies, condoms, STIs, healthy relationships, bleeding and weight change side effects, and the physiology of pregnancy.
During the six months following enrollment, participants completed up to five monthly coaching sessions. Each was audio-recorded and lasted approximately 30 minutes. Coaches reassessed attitudes towards pregnancy and barriers to contraceptive continuation, reinforced contraceptive knowledge, provided tailored reproductive health education, screened for side effects, and referred participants for clinical care, when appropriate. At each session, participants completed follow-up questionnaires assessing self-reported contraceptive continuation, adherence to short-acting methods, and updated participants’ medical and reproductive histories. To maximize retention, at recruitment, each participant provided contact information for at least three people who could be contacted should the participant’s contact information change. Participants were contacted up to five times for each study visit (via email, text, phone calls, or in clinic reminders, where applicable), between visits participants received holiday and birthday cards to maintain a connection with the study team, and at each study visit contact information was updated for participants and their additional contact.
Measures
The primary outcome, intervention completion, was defined as the number of coaching sessions completed. The main predictor, prior adverse reproductive outcome, was defined based on self-reported history of a prior pregnancy or STI on the baseline questionnaire or interview. STIs included chlamydia, gonorrhea, syphilis, trichomonas, genital herpes, genital warts, human papillomavirus (HPV), HIV, and/or pelvic inflammatory disease (PID). Covariates included age, race, education, income, marital status, housing, age at contraceptive initiation, and current contraceptive method.
Quantitative Analysis
Descriptive statistics were calculated for questionnaire variables using STATA 15.13 Bivariate analyses were done using χ2 and Wilcoxon rank-sum tests to determine associations between demographic and reproductive characteristics and the number of coaching sessions completed. Wilcoxon rank-sum tests were also used to determine differences in the median number of intervention sessions completed between the groups. Two multivariate linear regression models were used to assess the association between a prior adverse reproductive outcome and the number of completed coaching sessions. The first adjusted for age and race – demographic characteristics associated with having an adverse reproductive outcome and differential engagement in health service utilization.14,15 The second model adjusted for age at contraceptive initiation and type of contraceptive initiated at baseline, with methods grouped into long acting reversible contraceptives (LARC: implant and IUD) and short acting reversible contraceptive (SARC: pill, patch, ring, and shot). We adjusted for these two reproductive characteristics as they could be influenced by having a prior adverse reproductive outcome16,17. We ran two models rather than one due to the small sample size, which limited the number of covariates allowable in the models without sacrificing accuracy.18
Qualitative Analysis
Qualitative data were included from participants completing ≥4 coaching sessions to avoid misclassifying themes as absent when they might be due to missing data. Baseline interviews and coaching sessions were transcribed, and content analysis performed. Directed content analysis, which uses predetermined coding domains, was employed to identify similarities and differences in three behavioral determinants: knowledge, attitudes, and risk perception. Transcripts were deductively coded for STI/HIV knowledge; pregnancy attitudes and condom use; and STI/HIV risk perception. STI/HIV knowledge was categorized as low if participants divulged erroneous beliefs or were unable to answer STI/HIV-related questions. Coders included two medical students (C.A. and N.F.) and an MPH-level experienced coder (D.P.) who trained the students. Coding proceeded iteratively: each coder independently coded three transcripts then met to compare coded passages; coding discrepancies were resolved via consensus with the MPH-level coder serving as the final arbitrator; and, kappa reliability statistics were calculated to ensure agreement regarding the words, phrases, and passages that reflected key coding domains. Coders repeated this process to develop a final codebook with definitions, rules, and examples for each code. This codebook was used to recode all transcripts. The study team then met to review all coded passages and reach consensus regarding the key findings. Coding was performed using Nvivo.19
RESULTS
Sample Characteristics
Descriptive statistics for the parent sample (n=33) and those included in the qualitative analysis (n=21) are displayed in Table 1. All participants were cisgender females. In the parent sample, participants’ mean age was 17.4±2.1 years. Most were Black, privately insured, in high school, and previously used contraception. Thirteen (39%) endorsed a prior adverse reproductive outcome, with nine reporting a prior STI alone, two reporting a prior pregnancy alone, and two reporting both a prior pregnancy and an STI. The 21 participants included in the qualitative analysis were similar to the parent sample but had a smaller proportion of participants with a prior adverse reproductive outcome (n=5, 23.8%).
Table 1:
Demographic Characteristics of the Full Study Sample and of those Included in the Qualitative Analysis
| Characteristic | Total Sample (n=33) n (%) |
Completed ≥4 coaching sessions (n=21) n (%) |
|---|---|---|
| Mean age, years (SD) | 17.4 (2.1) | 17.5 (2.3) |
| Gender Identity | ||
| Woman | 33 (100) | 21 (100) |
| Race | ||
| Non-Hispanic White | 5 (15.2) | 3 (14.3) |
| Non-Hispanic Black | 24 (72.7) | 17 (81.0) |
| Other | 4 (12.1) | 1 (4.8) |
| Hispanic | 5 (15.2) | 3 (14.3) |
| Education level | ||
| Less than HS degree | 23 (69.7) | 14 (66.7) |
| Graduated HS or equivalent | 2 (6.1) | 1 (4.8) |
| Some college but no degree | 8 (24.2) | 6 (28.6) |
| Insurance | ||
| Public/Government | 12 (36.4) | 7 (33.3) |
| Private | 21 (63.6) | 14 (66.7) |
| Marital Status | ||
| Never married/not living with partner | 31 (93.9) | 19 (90.5) |
| Never married/living with partner | 1 (3.0) | 1 (4.8) |
| Not Reported | 1 (3.0) | 1 (4.8) |
| Housing | ||
| Living with parents or other family members | 28 (84.8) | 18 (85.7) |
| Living with romantic partner | 1 (3.0) | 1 (4.8) |
| Living on own | 1 (3.0) | 1 (4.8) |
| Other (dormitory & foster care) | 2 (6.1) | 0 (0.0) |
| Not reported | 1 (3.0) | 1 (4.8) |
| Reproductive History | ||
| Ever used contraception | 23 (69.7) | 17 (81.0) |
| Prior adverse reproductive outcome | 13 (39.4) | 5 (23.8) |
| Mean age at menarche (SD) | 11.8 (2.0) | 11.9 (2.1) |
| Baseline contraceptive method | ||
| Short acting reversible contraceptive (SARC) | 21 (63.6) | 12 (57.1) |
| Long acting reversible contraceptive (LARC) | 12 (36.4) | 9 (42.9) |
SD = standard deviation
Intervention Completion Rates
Among the 33 participants, nine attended all five coaching sessions, 12 attended four sessions, one attended three sessions, four attended two sessions, two attended one session, and five attended no coaching sessions after enrollment. Compared to those without a prior adverse reproductive outcome, Participants with a prior adverse outcome completed fewer coaching sessions (median: 2 vs. 4, respectively, p=0.03). As shown in Figure 1, when participants were divided into those who completed ≤2 sessions, 3 or 4 sessions, or all 5 sessions, those with a prior adverse reproductive health experience were represented in all three groups. However, a larger proportion of participants with a prior adverse reproductive outcome completed ≤2 sessions compared to those without such a history (53.8% vs 20.0%, p=0.04), and few completed all 5 sessions (7.7% vs 40.0%, p=0.04).
Figure 1:
Number of coaching sessions attended by women with or without a history of an adverse reproductive outcome (n=33)
Predictors of Intervention Completion
In the bivariate analyses (Table 2), there were no significant associations between the demographic or reproductive variables and having a prior adverse reproductive outcome. After controlling for age and race (Table 3, model 1), those with a prior adverse reproductive outcome still attended fewer coaching sessions than those without such a history. White participants attended fewer coaching sessions compared to Black participants. No differences were noted in the number of sessions completed across age groups. Controlling for reproductive health covariates (Table 3, model 2), those with a prior adverse reproductive outcome attended fewer coaching sessions compared to those without such a history.
Table 2:
Bivariate analyses of demographic and sexual/reproductive health characteristics associations with adverse reproductive health outcomes
| Characteristic | Prior adverse reproductive outcome (n=13) N (%) | No Prior adverse reproductive outcome (n=20) N (%) | p-value |
|---|---|---|---|
| Demographics | |||
| Mean age, years (SD) | 18.0 (1.8) | 17.0 (2.2) | 0.18 |
| Race | 0.60 | ||
| Non-Hispanic White | 1 (20.0) | 4 (80.0) | |
| Non-Hispanic Black | 10 (41.7) | 14 (58.3) | |
| Other | 2 (50.0) | 2 (50.0) | |
| Hispanic | 2 (40.0) | 3 (60.0) | 0.98 |
| Education level | 0.19 | ||
| Less than HS degree | 8 (34.8) | 15 (65.2) | |
| Graduated HS or equivalent | 2 (100.0) | 0 (0.0) | |
| Some college but no degree | 3 (37.5) | 5 (62.5) | |
| Insurance | 0.09 | ||
| Public/Government | 7 (58.3) | 5 (41.7) | |
| Private | 6 (28.6) | 15 (71.4) | |
| Marital Status | 0.40 | ||
| Never married/not living with partner | 13 (41.9) | 18 (58.1) | |
| Never married/living with partner | 0 (0.0) | 1 (100.0) | |
| Not Reported | 1 (100.0) | 0 (0.0) | |
| Housing | 0.71 | ||
| Living with parents or other family members | 11 (39.3) | 17 (60.7) | |
| Living with romantic partner | 0 (0.0) | 1 (100.0) | |
| Living on own | 0 (0.0) | 1 (100.0) | |
| Other (dormitory & foster care) | 1 (50.0) | 1 (50.0) | |
| Not reported | 0 (0.0) | 1 (100.0) | |
| Sexual and reproductive health | |||
| Prior history of contraceptive use | 9 (39.1) | 14 (60.9) | 0.96 |
| Baseline contraceptive method | 0.84 | ||
| Short acting reversible contraceptive (SARC) | 8 (38.1) | 13 (61.9) | |
| Long acting reversible contraceptive (LARC) | 5 (41.7) | 7 (58.3) | |
| Mean age of menarche | 12.0 (2.1) | 11.7 (2.0) | 0.63 |
| Mean age of contraceptive method initiation | 15.8 (1.8) | 15.5 (2.2) | 0.69 |
Table 3:
Likelihood coefficient of intervention session completion among those with and without a prior adverse reproductive outcome derived from multivariate linear regression models
| Covariate | Likelihood Coefficient (95% CI) | p-value |
|---|---|---|
| Model 1: Demographics Model | ||
| Prior adverse reproductive outcome | −1.64 (−2.92, −0.37) | 0.01 |
| Age | 0.25 (−0.07, 0.57) | 0.13 |
| Race | ||
| Black | Reference | 1.0 |
| White | −1.87 (−3.73, −0.02) | 0.05 |
| Other | −1.51 (−3.31, 0.29) | 0.10 |
| Model 2: Reproductive Health Model | ||
| Prior adverse reproductive outcome | −1.31 (−2.58, −0.04) | 0.04 |
| Age at contraceptive initiation | 0.25 (−0.09, 0.59) | 0.14 |
| Birth control type | ||
| Long acting reversible method | Reference | 1.0 |
| Short acting reversible method | 0.02 (−1.33, 1.36) | 0.98 |
Qualitative Differences in Reproductive Knowledge, Attitudes, and Risk perception
Participants with and without a prior adverse reproductive outcome were similar in their reproductive health knowledge, which was universally low. Both groups also held negative attitudes towards pregnancy and low HIV/STI risk perceptions. Subtle differences were noted in the social contexts that influenced their pregnancy attitudes and condom-use attitudes.
STI/HIV knowledge.
Participants ranked STI/HIV among the top three topics they wanted to learn more about. When asked why, most initially stated, ‘I don’t know anything.’ Among the 15 participants whose coaching sessions included STI/HIV knowledge-related discussions, seven shared an incorrect fact or myth about STIs after being asked or prompted to demonstrate their knowledge. A 15-year old with a prior adverse reproductive outcome thought all STIs could be transmitted vertically: “Somebody can be clean because they’re a virgin…but their parents probably have STDs and they probably got it.”
Attitudes towards pregnancy.
Participants with and without a prior adverse reproductive outcome were similar in their pregnancy attitudes. Both expressed positive attitudes towards future childbearing but wanted to avoid childbearing now. Both wanted to wait until they were in their 20’s or 30’s before getting pregnant, saying they would feel “ready to be a parent” once they achieved their career goals, felt financially stable, or obtained assets (e.g., house, car) deemed essential for parenting. A 15-year old without a prior adverse reproductive outcome explained, “In my twenty-year-old days, I’ll be able to go out and live my life…explore places [and] not have to worry about a child…but then when I’m at 28, that’s the end…That’s the time to sit at home with your child.” A 19-year old with a prior adverse reproductive outcome expressed a similar sentiment this way: “I don’t want a child right now…I’m not prepared. I don’t have the finances nor the resources to care for a child…I would just at least want wait until my career’s on track…I would be still in my twenties, so I would be still young, so it would be the right time to have a child…I just want to make sure that I’m happy where with where my life is before I have a child. Maybe [when] my student loans [are] half-way paid, I have a husband, I have travelled the world, studied abroad, I did all my research, and [then] I’m just ready to settle down and raise a family.”
Participants with and without a prior adverse reproductive experience identified people in their social networks who experienced an adolescent pregnancy. They explained how these experiences shaped their attitudes towards pregnancy. Participants with a prior adverse reproductive outcome commonly identified parents or siblings who experienced an unintended adolescent pregnancy. A 22-year old with a prior adverse outcome remarked, “My one sister just had a baby, I’ve had two friends who just had baby, and a cousin. My other sister is pregnant…My brother’s girlfriend is pregnant too. So, a lot of babies [are] coming.” On the other hand, participants without a prior adverse reproductive outcome were more likely to identify extended family (e.g., cousin) or someone in their social network as having experienced an adolescent pregnancy. A 15-year old without a prior adverse outcome reflected on a parenting classmate: “I know a girl, she’s 16 with a baby…You only [in] 10th, 11th grade with a baby? How are you supposed to take care of a baby, and then do your work? I think it would be stressful.” Witnessing others’ experiences with teen parenthood and the associated challenges influenced participants’ desire to delay childbearing.
Attitudes towards condom use.
Twelve participants expressed negative attitudes towards condom use, including all five Participants with a prior adverse reproductive outcome and seven of the 16 participants without such history. The five participants with a prior adverse reproductive outcome described strongly held attitudes against condom use, as exemplified in the dialogue below from a 19-year old:
Coach: Do you plan on using condoms in the future?
Participant: Nhhhn [indicates no] [laughs]
Coach: No. Why do you laugh?
Participant: [laughs] Yeah. Yeah, I’m not using them.
Coach: Mhhhm. And how come? Tell me more.
Participant: Well, I like [sex] better without it.
Coach: I’ve talked to a lot of people about sex…[That’s] certainly something a lot of people say. What are your thoughts or feelings about STDs?
Participant: They’re curable you just have to be careful. And you have to catch them early.
HIV/STI risk perception.
STI/HIV risk perception was low among those with and without a prior adverse reproductive experience. A 17-year old without a prior adverse outcome expressed concern that she would “get stupid” and make a mistake that would result in contracting an STI. The rest believed their risk was low or non-existent and attributed their lack of concern to trust in their partner, the exclusivity of their relationship, or both. A 17-year old with a prior adverse reproductive outcome said she knew her partner had other sexual partners, but trusted that he used condoms with the others, even though he did not use them with her. She felt she was unlikely to contract an STI, “because I grew up next to him… he isn’t into drugs…He was really clean.”
DISCUSSION
Few studies have examined whether completion rates for sexual risk reduction interventions differ among young women based on their history of adverse reproductive outcomes. Using data from a feasibility study of a health coaching intervention to improve contraceptive use among young women, we explored whether intervention completion rates varied between those with and without a prior pregnancy or STI. Despite high overall intervention completion rates, Participants who previously experienced a pregnancy or STI completed fewer intervention sessions than those who never experienced these outcomes. Age was not associated with the number of completed coaching sessions, but race/ethnicity was. Reproductive knowledge, attitudes, and risk perception were similar among Participants with and without a prior adverse reproductive outcome.
These findings fill an important gap in the literature by identifying Participants who may be at highest risk of nonadherence to sexual risk reduction intervention protocols. While prior research has identified socio-demographic predictors of attrition – such as being a member of a racial minority group and low socioeconomic status – published studies of sexual risk reduction programs rarely describe their success in reaching those who may benefit most from participation. This is particularly true regarding teen pregnancy prevention interventions.7,20 This is important because with behavioral interventions, receipt of a lower intervention dose generally translates into a lower intervention effect on outcomes.21 We found that those most at risk for the outcomes the program sought to prevent completed half as many intervention sessions. Recognition of this issue is invaluable for guiding researchers in identifying strategies that might mitigate this problem thereby reducing disparities in reproductive outcomes.
Our study was unable to fully explore why Participants with prior adverse reproductive outcomes completed fewer sessions. Neither sociodemographic characteristics nor traditional behavioral predictors accounted for the difference. It was particularly surprising that traditional determinants of health behaviors addressed in sexual risk reduction interventions – reproductive knowledge, attitudes, and risk perception – did not differ between those with and without a prior adverse reproductive outcome. If they had, the lessons for intervention research would have been far more straightforward.
Prior research has noted associations between sexual risk reduction intervention retention and intervention strategies. Participants with lower baseline knowledge showed greater retention in interventions that used information-based strategies. Similarly, participants with low baseline motivation to engage in targeted behaviors have higher retention rates in interventions that use motivational strategies.7 Results are mixed for interventions focused on behavioral strategies, with low retention among those with the lowest and highest rates of engagement in the target behavior at enrollment, suggesting a need to focus on perceived self-efficacy among the former and behavioral maintenance for the latter. In our study, we found no clear behavioral determinant to focus on to drive retention efforts. This may be due to the fact that participants were recruited from an area with disproportionately high rates of poverty and STIs, resulting in a population enriched for similarities in their socio-cognitive profiles. Other barriers that may explain these differences - such as lack of time, cost, and work, school, or other competing priorities - were not examined. To maximize intervention impact, future studies should confirm our findings and explore the factors that underlie the observed differences. By paying greater attention to documenting and exploring differences in intervention reach and retention among Participants with and without an adverse reproductive outcome, we may identify predictors amenable to being addressed through intervention tailoring.
Our findings provide insights for implementing adolescent sexual health interventions for vulnerable youth. By identifying a subpopulation of youth at increased risk for intervention noncompletion, our findings can inform sample size estimates and plans to reduce clinical trial attrition. For example, using a history of a prior adverse reproductive event as a stratifying variable during recruitment or in analyses ensures representation of this population in outcome data. A second insight is the importance of thoughtful analytic approaches that take differential rates of intervention engagement into consideration. Using an intention-to-treat approach maximizes external validity by including less adherent patients.22 This approach is far better at minimizing bias than excluding those who drop out early or who receive fewer intervention doses. Examining outcomes based on the dose received provides additional insights by helping to determine the number of intervention doses that achieve maximal effect. Such information can translate into protocol changes that reduce expenses without compromising impact.
This study has several limitations and strengths. Data are from a small feasibility study that was not explicitly designed to compare outcomes across our groups of interest. Relatively few participants in the full or qualitative samples had a prior adverse reproductive outcome, which may have limited our ability to identify group differences. Although STIs/HIV were frequently discussed, their prevention was not the primary focus of coaching sessions. Study strengths include our use of qualitative data, which allowed us to explore and compare several determinants of sexual risk behaviors with greater nuance than a survey would have allowed. In addition, our inclusion of participants who completed ≥4 coaching sessions maximized our ability to identify content related to the behavioral determinants.
Conclusions
Adolescent and young adult women who previously experienced a pregnancy or STI may complete fewer sexual risk reduction intervention sessions than those who never experienced these outcomes. Neither differences in socio-demographics nor in reproductive knowledge, attitudes, and risk perception explain this. These findings are important to consider when estimating intervention sample sizes and in planning outcome analyses to maximize external validity in clinical trial data.
Acknowledgements:
We wish to thank Alanna Butler and Stephanie Richardson, our coaching experts, as well as Nadia Dowshen, Jen Harding and Anne Teitelman for their invaluable support and assistance with study design, interventions monitoring, and the process evaluation. Financial support was provided by the University of Pennsylvania Center for Public Health Initiatives, the University of Pennsylvania Center for AIDS Research, the Children’s Hospital of Philadelphia, the U.S. Office of Adolescent Health and the University of Texas’ Innovative Teen Pregnancy Prevention Program, the National Institute of Mental Health (1K23MH119976), the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (HHS) under T71MC30798 Leadership Education in Adolescent Health (LEAH). This information or content and conclusions are those of the author and should not be construed as the official position or policy of, nor should any endorsements be inferred by HRSA, HHS or the U.S. Government.
Financial Disclosures:
Dr. Akers has received funding from Bayer Healthcare for an investigator-initiated grant, from Merck Inc. for their HPV Advisory Board, and from Mylan Pharmaceuticals for their Women’s Health Advisory Board.
Contributor Information
Christina Amutah, Perelman School of Medicine, 3400 Civic Center Boulevard, Philadelphia, PA 19104.
Danielle Petsis, Children’s Hospital of Philadelphia, PolicyLab, 2716 South St., Philadelphia, PA 19146, 2158177047.
Naomi F. Fields, Perelman School of Medicine at the University of Pennsylvania, Jordan Medical Education Center, 6th Floor, 3400 Civic Center Blvd Building 421, Philadelphia, PA 19104
Sarah Wood, Perelman School of Medicine at the University of Pennsylvania, Children’s Hospital of Philadelphia, PolicyLab, 2716 South Street, Philadelphia, PA 19146.
Alix Timko, Children’s Hospital of Philadelphia, PolicyLab, 2716 South Street, Philadelphia, PA 19146.
Aletha Y. Akers, University of Pennsylvania School of Medicine, Children’s Hospital of Philadelphia, Division of Adolescent Medicine, PolicyLab, 2716 South Street, Philadelphia, PA 19146.
REFERENCES
- 1.Fraser AM, Brockert JE, Ward RH. Association of Young Maternal Age with Adverse Reproductive Outcomes. N Engl J Med. 1995;332(17):1113–1118. doi: 10.1056/NEJM199504273321701 [DOI] [PubMed] [Google Scholar]
- 2.Forhan SE, Gottlieb SL, Sternberg MR, et al. Prevalence of sexually transmitted infections among female adolescents aged 14 to 19 in the United States. Pediatrics. 2009;124(6):1505–1512. doi: 10.1542/peds.2009-0674 [DOI] [PubMed] [Google Scholar]
- 3.Finer LB, Zolna MR. Declines in Unintended Pregnancy in the United States, 2008–2011. N Engl J Med. 2016;374(9):843–852. doi: 10.1056/NEJMsa1506575 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hosenfeld CB, Workowski KA, Berman S, et al. Repeat infection with Chlamydia and gonorrhea among females: a systematic review of the literature. Sex Transm Dis. 2009;36(8):478–489. doi: 10.1097/OLQ.0b013e3181a2a933 [DOI] [PubMed] [Google Scholar]
- 5.Weström L, Joesoef R, Reynolds G, Hagdu A, Thompson SE. Pelvic inflammatory disease and fertility. A cohort study of 1,844 women with laparoscopically verified disease and 657 control women with normal laparoscopic results. Sex Transm Dis. 1992;19(4):185–192. [PubMed] [Google Scholar]
- 6.Brown SS, Eisenberg L. Unintended Pregnancy and the Well-Being of Children and Families. JAMA. 1995;274(17):1332–1332. doi: 10.1001/jama.1995.03530170012005 [DOI] [PubMed] [Google Scholar]
- 7.Noguchi K, Albarracín D, Durantini MR, Glasman LR. Who participates in which health promotion programs? A meta-analysis of motivations underlying enrollment and retention in HIV-prevention interventions. Psychol Bull. 2007;133(6):955–975. doi: 10.1037/0033-2909.133.6.955 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McKenzie M, Tulsky JP, Long HL, Chesney M, Moss A. Tracking and follow-up of marginalized populations: a review. J Health Care Poor Underserved. 1999;10(4):409–429. doi: 10.1353/hpu.2010.0697 [DOI] [PubMed] [Google Scholar]
- 9.Oringanje C, Meremikwu MM, Eko H, Esu E, Meremikwu A, Ehiri JE. Interventions for preventing unintended pregnancies among adolescents. Cochrane Database Syst Rev. 2016;(2). doi: 10.1002/14651858.CD005215.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.CHOP-Care-Network-Information-for-grants-2017.pdf. Accessed December 9, 2020. https://perc.research.chop.edu/sites/default/files/CHOP-Care-Network-Information-for-grants-2017.pdf
- 11.Kann L, McManus T, Harris WA, et al. Youth Risk Behavior Surveillance —United States, 2017. 2018;67(8):479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ibitoye M, Choi C, Tai H, Lee G, Sommer M. Early menarche: A systematic review of its effect on sexual and reproductive health in low- and middle-income countries. PloS One. 2017;12(6):e0178884. doi: 10.1371/journal.pone.0178884 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.StataCorp. StataCorp LLC; 2017. [Google Scholar]
- 14.Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance 2018. U.S. Health and Human Services; 2019. Accessed November 15, 2019. https://www.cdc.gov/std/stats18/adolescents.htm [Google Scholar]
- 15.Champion JD, Harlin B, Collins JL. Sexual risk behavior and STI health literacy among ethnic minority adolescent women. Appl Nurs Res. 2013;26(4):204–209. doi: 10.1016/j.apnr.2013.06.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chacko MR, Wiemann CM, Buzi RS, Kozinetz CA, Peskin M, Smith PB. Choice of Postpartum Contraception: Factors Predisposing Pregnant Adolescents to Choose Less Effective Methods Over Long-Acting Reversible Contraception. J Adolesc Health. 2016;58(6):628–635. doi: 10.1016/j.jadohealth.2015.12.002 [DOI] [PubMed] [Google Scholar]
- 17.Madden T, Secura GM, Allsworth JE, Peipert JF. Comparison of contraceptive method chosen by women with and without a recent history of induced abortion. Contraception. 2011;84(6):571–577. doi: 10.1016/j.contraception.2011.03.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Knofczynski GT, Mundfrom D. Sample Sizes When Using Multiple Linear Regression for Prediction. Educ Psychol Meas. 2008;68(3):431–442. doi: 10.1177/0013164407310131 [DOI] [Google Scholar]
- 19.QSR International Pty Ltd. (2018) NVivo (Version 12).
- 20.Albarracín D, Wilson K, Durantini MR, Livingood W. When is Retention in Health Promotion Interventions Intentional? Predicting Return to Health Promotion Interventions as a Function of Busyness. Acta Investig Psicol. 2013;3(3):1311–1321. doi: 10.1016/S2007-4719(13)70969-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Voils CI, King HA, Maciejewski ML, Allen KD, Yancy WS, Shaffer JA. Approaches for informing optimal dose of behavioral interventions. Ann Behav Med Publ Soc Behav Med. 2014;48(3):392–401. doi: 10.1007/s12160-014-9618-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Pedlow CT, Carey MP. HIV sexual risk-reduction interventions for youth: a review and methodological critique of randomized controlled trials. Behav Modif. 2003;27(2):135–190. doi: 10.1177/0145445503251562 [DOI] [PMC free article] [PubMed] [Google Scholar]