Abstract
Purpose:
Youth living in group home settings are at significantly greater risk for sexual risk behaviors; however, there are no sexual health programs designed specifically for these youth. The study’s purpose was to assess the effectiveness of a teen pregnancy-prevention program for youth living in group home foster care settings and other out-of-home placements.
Methods:
The study design was a cluster randomized controlled trial involving youth (N = 1,037) recruited from 44 residential group homes located in California, Maryland, and Oklahoma. Within each state, youth (mean age = 16.2 years; 82% male; 37% Hispanic, 20% African-American, 20% white, and 17% multiracial) in half the group homes were randomly assigned to the intervention group (n = 40 clusters) and the other half were randomly assigned to a control group that offered “usual care” (n = 40 clusters). The intervention (i.e., Power Through Choices [PTC]) was a 10-session, age-appropriate, and medically accurate sexual health education program.
Results:
Compared to the control group, youth in the PTC intervention showed significantly greater improvements (p < .05) from preintervention to postintervention in all three knowledge areas, one of two attitude areas, all three self-efficacy areas, and two of three behavioral intention areas.
Conclusions:
This is the first published randomized controlled trial of a teen pregnancy-prevention program designed for youth living in foster care settings and other out-of-home placements. The numerous significant improvements in short-term outcomes are encouraging and provide preliminary evidence that the PTC program is an effective pregnancy-prevention program.
Keywords: Teen pregnancy prevention, Sex education, Reproductive health, Youth sexual risk behavior, Youth in foster care, Group homes, Intervention, Psychosocial change
Youth living in foster care or other out-of-home settings are substantially more likely to participate in sexual risk behaviors. For example, youth in foster care are more likely to report ever having had sex, to be currently sexually active, and are more likely to report initiating sex at an early age (before age 13) [1]. Among sexually active youth in foster care, 23.4% of girls and 8.5% of boys did not use any method of contraception at their last sexual encounter compared to 14.4% of girls and 7.5% of boys nationally. Nationally, a greater proportion of sexually active youth reported female partners use of hormonal contraception (48.3% of males and 42.7% of females) compared to youth in foster care (7.7% of males and 23.5% of females) [1].
Such behavioral differences result in disproportionately high pregnancy rates among youth in foster care. Nearly one in three young women in the foster care system are pregnant at least once by age 17 or 18 and by age 19 over half experience a pregnancy [2,3]. In comparison, one in four girls nationally become pregnant before age 20 [4]. Youth in foster care also experience a greater number of repeat pregnancies. Before age 20, 46.4% of females in foster care experience more than one pregnancy [2].
These data suggest the strong need for sexual risk behavior prevention programming for youth living in foster care or other out-of-home settings. However, there are no evidence-based pregnancy-prevention programs specifically for these youth.
A number of randomized controlled trials (RCTs) of teen pregnancy/sexually transmitted infection (STI)-prevention programs have been conducted. The results have generally found that sexuality education programs can have a positive impact on youths’ knowledge, attitudes, self-efficacy, and intentions to engage in sexual behaviors. For example, Jemmott et al. [5] evaluated the effectiveness of an HIV/STI-prevention program using an RCT design in a study that included 659 African-American youth (mean age = 11.8 years, 53% female). They found that youth in the safer-sex intervention were significantly more likely to report greater knowledge regarding condom use and HIV risk reduction, stronger beliefs that condoms prevent pregnancy, and greater self-efficacy for using condoms at 12-month follow-up. Markham et al. [6] conducted an RCT with 907 seventh-grade youth (48.4% Hispanic, mean age = 12.6, 59.8% female). Results at ninth-grade follow-up indicated that youth in the risk-reduction intervention group had greater knowledge about condoms, greater intentions to use condoms, stronger self-efficacy to use condoms, and greater intentions to remain abstinent until the end of high school [6]. Finally, Lawrence et al. [7] conducted an RCT involving 246 African-American youth (mean age = 15.3 years, 72% female). At 12-month follow-up, youth in the intervention group indicated greater increases in knowledge about HIV risk behavior, and greater increases in self-efficacy to prevent HIV risk behavior were noted post intervention although the differences were not maintained at 12 months [7]. These results suggest that sexual health education programs can significantly improve knowledge, attitudes, self-efficacy, and intentions toward sexual behaviors.
The results of research also suggests that effective programs can take place in a variety of settings (community based, schools, and health clinic) and can have positive effects on males and females as well as on racial and ethnic minority youth [7–9]. To date, however, no RCTs have been conducted to evaluate a comprehensive sexual health education curriculum designed specifically for youth in foster care and other out-of-home care settings.
The purpose of this RCT was to test the effectiveness of the Power Through Choices (PTC) program which is an age-appropriate and medically accurate sexual health education intervention for youth living in group home foster care settings and other out-of-home placements. The PTC program was delivered to youth living in group homes operated or contracted by child welfare (foster care) or the juvenile justice group care settings. Specific goals of the PTC intervention included delaying the initiation of sexual intercourse and reducing the incidence of unprotected sexual intercourse, STIs, and teen pregnancy among youth ages 13–18 who were living in group home settings. The PTC intervention focused on improving knowledge, attitudes, self-efficacy, behavioral intentions, and behaviors related to sexuality and reducing sexual risk behaviors of this understudied and underserved population. The specific purpose of this article is to present the preintervention to postintervention results in regard to the participants’ knowledge, attitudes, self-efficacy, and behavioral intentions.
Methods
The study was reviewed and approved by the institutional review board at the University of Oklahoma Health Sciences Center. The study design was a cluster RCT involving youth (N = 1,037) recruited from 44 residential group homes located in California (n = 19), Maryland (n = 10), and Oklahoma (n = 15) [10]. All youth living in the same group home were assigned to the same research condition to eliminate the possibility of contamination between treatment and control groups. Within each state, half the group homes were randomly assigned to a treatment group that offered the PTC program (n = 40 clusters) and the other half were assigned to a control group that offered “usual care” (n = 40 clusters).
Homes were approached to participate in the study if they had the capacity and commitment to support the study; therefore, sampling was purposive rather than random. Within each site’s catchment area, every group home that was willing to participate and that had youth residents between the ages of 13–18 was recruited. Exclusion criteria were group homes specifically for pregnant and parenting teens (maternity homes); group homes for adolescent sexual offenders; and group homes providing therapeutic services to youth with significant mental, emotional, or behavioral issues [10]. Based on these criteria, 72 homes were eligible to participate in the study yielding a response rate of 61%. Each participating group home completed a memorandum of agreement prior to randomization agreeing to participate regardless of randomization assignment.
Each state had its own team of intervention facilitators as well as two data collectors. Annual in-person trainings for all data collection personnel were conducted to standardize the data collection protocol and maintain the standardization across all sites for years two to five of the study. In year 1, a pilot study was conducted in two homes in each state. The evaluation protocols and instruments (approximately five questionnaires and 15 forms) were revised as a result of the pilot study and feedback during the PTC facilitator and evaluator trainings.
Sampling
PTC is designed and appropriate for youth living in many types of out-of-home care settings; however, the implementation of PTC described in this study is exclusive to youth living in group homes overseen by the child welfare (CW) (foster care) and/or Juvenile Justice (JJ) systems. A “group home” is considered a congregate care residential facility operated or contracted by a state child welfare agency, a state juvenile justice agency, or by a private care provider. There is evidence to suggest that youth involved in the CW systems are also at risk for involvement in the JJ system [10–12]. The inverse is also true, as youth first involved in the JJ system are at risk for involvement in the CW system [10–12]. Group homes served in the study included: (1) youth in the CW system; (2) youth in the JJ system; or (3) a mixture of youth from both systems. Eight homes in the study were CW homes (all in Oklahoma); nine homes were JJ homes (two in California, seven in Oklahoma); and 17 homes were a mixture of CW and JJ (17 in California, 10 in Maryland).
The plan was to enroll 20 intervention homes and 20 control homes resulting in 540 youth per group for a total of 1,080 youth and ultimately 360 evaluable youth per group for a total of 720 youth. For our primary outcome of condom or contraceptive use (or abstinence), 720 evaluable youth in 20 homes would provide over 95% power to detect a difference of 15% assuming an intracluster correlation (ICC) = .010. Power remained at over 80% for differences as small as 12.5% and an ICC as large as .03, and power increased if the number of youth remained the same but the number of homes increased. All youth, male and female, ages 13–18 who resided in the recruited group homes were eligible for participation. The study required consent for each youth from a legally authorized representative as well as youth assent. Data collectors obtained consent and assent for at least 80% of youth in a matched groups homes, before the homes were randomized.
Randomization
The group homes were the unit of randomization. Homes were stratified and clustered according to location (California, Maryland, or Oklahoma), recruitment date, number of youth served, and gender of youth served. The first four homes recruited in California were randomly assigned as a stratum of four clusters. All other clusters were grouped into matched pairs of two clusters for random assignment. An equal number of homes were assigned to the treatment and control groups, and the randomized homes contained a nearly equal number of youth. Homes were randomized by the biostatistician (S.K.V.) using a computer program (SAS version 9.4) [13].
The average group home consisted of 10.7 youth. Twenty homes were randomized once, 13 homes were rerandomized twice, nine homes were rerandomized three times, and two homes were randomized four times over the course of the trial (Figure 1).
Figure 1.
Power Through Choices participant flow diagram.
PTC intervention
Oklahoma Institute for Child Advocacy substantially updated and revised the original PTC program content using feedback from focus groups of youth in out-of-home care and experts in the field of sexuality education and teen pregnancy prevention [14]. Youth in out-of-home care often lack the skills and resources necessary to avoid risky sexual behaviors and teen pregnancy [14]. The PTC curriculum addresses specific characteristics which may motivate system-involved youth to become pregnant or engage in sexual risk-taking behavior. Some of these reasons may include an intense need for affection or belonging; absence of a dependable family or social support network; exposure to sexual abuse or violence; and limited skills in identifying and securing resources to support their present and future needs. The PTC intervention is sensitive to the issues of abuse and other trauma which may be part of the life story of the participants and provides opportunities for youth to examine how those experiences might influence feelings and behaviors related to sexual decision-making.
The PTC intervention consists of ten 90-minute sessions delivered twice per week to groups of 6–20 youth. The intervention program sessions are gender specific and feature an interactive approach that engages youth through role playing, group discussion, and other hands-on activities. Each session also includes time for questions and answers.
The PTC intervention was developed using Social Cognitive Theory and the Health Belief Model as guiding behavioral theories [15,16]. Theoretical constructs that are operationalized in the intervention to address these needs are skills building, role modeling, identification and reduction of barriers to change, goal setting, self-efficacy regarding postponing initiation of sexual intercourse, and self-efficacy regarding contraceptive and condom use for those who are sexually active. The intervention also focuses on self-empowerment and the impact of choices.
Youth residing in homes randomized to the control condition received “usual care” which was not programming related to reproductive health, but in some instances, they may have received educational information on topics such as healthy eating.
Data collection
Data were collected at pre-, post-, 6-month, and 12-month follow-up from the intervention and control group homes from 2012 to 2014. The period from preintervention to postintervention was typically about 6 weeks. Therefore, this study presents the results of the pre-to-post changes in participants’ knowledge, attitudes, self-efficacy, and behavioral intentions and does not include behavioral outcomes.
Data were collected on site in the group homes by trained data collectors using self-administered paper-and-pencil questionnaires. The questions and possible responses were read aloud by the data collectors to minimize any problems with reading comprehension or missing data due to skipped questions. The baseline survey was administered in the paired treatment and control homes approximately 1 week before the program began in the treatment group homes. The post survey was administered in groups or individually depending on whether the study participants were still residing in the participating group homes. The response rate was 98% and the pre-to-post retention rate was 92% [17].
Process evaluation was conducted to assess intervention implementation and fidelity. Data collectors observed one randomly selected intervention session each time the intervention was implemented. The purpose was to monitor and record curriculum fidelity, evaluate facilitator characteristics, provide corrective oversight when appropriate, and document any corrective actions. In addition, program facilitators completed program implementation checklists and youth attendance records as a measure of dosage.
Instrumentation
Measures reported in this study include knowledge, attitudes, self-efficacy, and intentions regarding sexuality, condom use, contraceptives, and sexual behaviors. Many items were from the Youth Risk Behavior Surveillance System or from the Prevention Minimum Evaluation Data Set [18,19].
Constructs were formed using exploratory factor analysis on a polychoric matrix using principal component analysis extraction and varimax rotation (Table 1) [20]. Items with a factor loading of .4 or higher were included in the construct. Responses for items representing each attitude and self-efficacy domain were summed and divided by the total number of items to create constructs with higher values indicating a more positive attitude or stronger self-efficacy. Internal consistency of the constructs was assessed with Cronbach’s alpha using polychoric correlations.
Table 1.
Scale information for knowledge, attitudes, self-efficacy, and behavioral intentions
| Construct/score | Number of items | Sample item | Item response format | Cronbach’s α | Scale response format | ICCa |
|---|---|---|---|---|---|---|
|
| ||||||
| Knowledge | ||||||
| Anatomy and fertility | 4 | The part of the female body where a baby grows during pregnancy | Multiple choice, true, false, or do not know | NA | Percent indicating correct responses | .0651 |
| HIV and STIs | 7 | All sexually active individuals are at risk for getting HIV | True, false, or do not know | NA | Percent indicating correct responses | .0474 |
| Methods of protection | 10 | Condoms are 100% effective in preventing pregnancy and STIs | True, false, or do not know | NA | Percent indicating correct responses | .0856 |
| Attitudes | ||||||
| Support for methods of protection | 6 | Condoms should always be used if a person your age has sexual intercourse | Four-point scale (strongly disagree to stronglyagree) | .84 | Mean of 6 items (range 1–4) | .0710 |
| Barriers to methods of protection | 5 | Condoms are a hassle to use | Four-point scale (strongly disagree to stronglyagree) | .84 | Mean of 5 items (range 1–4) | .1192 |
| Self-efficacy | ||||||
| Ability to communicate with partner | 3 | Tell your partner your feelings about what you do and do not want to do sexually | Four-point scale (very sure to very unsure) | .83 | Mean of 3 items (range 1 –4) | .0545 |
| Plan for protected sex and avoid unprotected sex | 3 | Plan ahead to have some method of protection available? | Four-point scale (very sure to very unsure) | .81 | Mean of 3 items (range 1 –4) | .0666 |
| Where to get methods of birth control | 1 | Find a place in your community to obtain methods of protection from pregnancy and STIs | Four-point scale (very sure to very unsure) | NA | “very sure” versus “somewhat sure,” “somewhat unsure,” and “very unsure” | .0336 |
| Behavioral intentions | ||||||
| Intention to have sex in the next year | 1 | Do you intend to have sexual intercourse in the next year? | Four-point scale (yes, definitely to no, definitely not) |
NA | “no, definitely not” and “no probably not” versus “yes, definitely” and “yes probably” | .3409 |
| Intention to use a condom in the next year | 1 | If you have sexual intercourse in the next year, do you intend to use (or have your partner use) a condom? | Four-point scale (yes, definitely to no, definitely not) | NA | “yes, definitely” and “yes probably” versus “no, definitely not” and “no, probably not” | .0639 |
| Intention to use birth control in the next year | 1 | If you have sexual intercourse in the next year, do you intend to use (or have your partner use) any of these methods of birth control: Condoms, birth control pill, the shot, the patch, the ring, intrauterine device, implant? | Four-point scale (yes, definitely to no, definitely not) | NA | “yes, definitely” and “yes probably” versus “no, definitely not” and “no, probably not” | .0514 |
ICC = intracluster correlation; NA = not applicable; STI = sexually transmitted infection.
Unadjusted baseline ICC.
Knowledge
Youth knowledge in the areas of Reproductive Anatomy and Fertility (four items), HIV and STIs (seven items), and Methods of Protection (10 items) was assessed (Table 1) [10,14,15]. The number of correct responses for the items representing each knowledge area were summed and divided by the total number of items to create a knowledge score for each domain with a higher value indicating greater knowledge.
Attitudes
The survey included 11 items that measured youth attitudes toward various methods of protection and using protection [19,21]. Two attitudes constructs were created: Support for Methods of Protection (α = .84) and Barriers to Methods of Protection (α = .84) [10].
Self-efficacy
Seven items were included to assess self-efficacy [19,21]. Two constructs were created: Ability to Communicate with Your Partner (items 1–3, α = .83) and Plan for Protected Sex and Avoid Unprotected Sex (items 4–6, α = .81) [10]. One single-item measure was included to assess self-efficacy regarding finding a place in the community to obtain a method of protection.
Behavioral intentions regarding sexual activity
Three items assessed intentions toward sexual activity [22]. The items determined the participants’ behavioral intentions regarding having sexual intercourse in the next year; using a condom (or their partner using a condom) if they had sexual intercourse; and using other methods of protection, such as birth control pills, the shot (Depo-Provera), or intrauterine devices. A binary measure was constructed for each item comparing youth who indicated they intended to “definitely not/probably not” or “definitely/probably” engage in the activity to youth who reported less positive behavioral intentions.
Analysis
All available data were used in the analyses. Generalized linear mixed model were utilized for both continuous (SAS PROC MIXED) and dichotomous outcomes (SAS PROC GLIMMIX, using Newton Raphson estimation). Measurement time (pre vs. post) and condition (control vs. intervention) represented, respectively, within- and between-cluster factors. Age, race, and gender were included as covariates in all analyses of mean outcomes. Random intercept statements were used to take into account the sampling strategy, whereby group homes were randomly assigned to treatment, and the same adolescents within a home were observed at two time points. Interactions with time × condition were evaluated based on clinical relevance and an alpha of .010. ICC was estimated from covariance parameters that were estimated from baseline models without covariates. Statistical analysis was performed using SAS version 9.4 [13].
Participants
Of the 1,183 youth assessed for eligibility, 146 youth were excluded because they were leaving the intervention or control group home before the conclusion of the intervention (n = 77), a parent or guardian did not consent (n = 22), or other reasons (n = 47). Other reasons for participation included age, previous participation in the program, youth employment that would make attendance sporadic, and ongoing therapy related to prior abuse. A total of 1,037 youth were randomized. After randomization, one youth did not assent to complete the baseline survey. Forty clusters each were assigned to the PTC intervention and control groups, 517 and 519 youth, respectively. Post-survey data were obtained from 952 youth (n = 468 youth in PTC intervention, n = 484 youth in control group). No group homes withdrew from the study (Figure 1).
Demographic data shown in Table 2 indicate the mean age of the participants was 16.2 years and a majority were male and in the 9th to 11th grades. More than one-third of the sample was Hispanic, approximately 20% was white or African-American, and approximately 17% was multiracial.
Table 2.
Demographic data by randomization status
| Measure | Treatment | Control | Difference | p value |
|---|---|---|---|---|
|
| ||||
| Sample size (clusters) | 468 (40) | 484 (40) | ||
| Age (mean) Gender | 16.2 | 16.1 | .0866 | .2689 |
| Male | 81.5 | 77.9 | .1727 | |
| Female | 18.6 | 22.1 | ||
| Race/ethnicity | .5547 | |||
| Hispanic | 37.1 | 37.4 | ||
| White, non-Hispanic | 19.6 | 20.9 | ||
| Black, non-Hispanic | 20.6 | 19.5 | ||
| American Indian/Alaska | 3.7 | 3.9 | ||
| Native, non-Hispanica Asian and Pacific Islander, | 1.4 | 1.8 | ||
| non-Hispanic Multiracial, non-Hispanic | 17.7 | 16.5 | ||
| Age entering foster care (mean) | 12.8 | 12.3 | .5 | .0795 |
| Last grade completed | .7030 | |||
| 8th grade or less | 26.7 | 29.3 | ||
| 9th–11th grade | 68.7 | 66.5 | ||
| 12th grade | 4.5 | 3.7 | ||
Values are percentages except where noted.
Analysis completed at the individual youth level.
p values are based on a chi-square test for all categorical data and a t test for all other measures.
7.5% of youth in the intervention group and 8.9% of youth the control group identified American Indian/Alaska Native as at least one of their races.
Results
The PTC program implementation rate was 100% and youth attended 87% of the sessions. There were no significant differences at baseline between the intervention and control groups in regard to demographic, knowledge, attitudes, or behavioral intentions outcomes data (Tables 2 and 3). As shown in Table 3, compared to changes in the control group, youth in the PTC intervention showed significantly greater improvements (p < .05) from preintervention to postintervention in all three knowledge areas, one of two attitude areas, all three self-efficacy areas, and two of three behavioral intention areas.
Table 3.
Preintervention to postintervention changes in youth knowledge, attitudes, self-efficacy, and behavioral intentions
| Measure | PTC intervention (n = 468, 40 clusters) |
Control group (n = 484, 40 clusters) |
Differencea | p value | ||
|---|---|---|---|---|---|---|
| Pre | Post | Pre | Post | |||
|
| ||||||
| Knowledge and awareness | ||||||
| Anatomy and fertilityb,c,d | 61.5% | 69.3% | 60.2% | 60.2% | 7.8% | <.0001 |
| HIV and STIsb,c,d | 69.2% | 77.9% | 67.0% | 64.9% | 10.8% | <.0001 |
| Methods of protectionb,c,d | 65.9% | 80.2% | 64.7% | 62.3% | 16.7% | <.0001 |
| Attitudes | ||||||
| Support for methods of protectionc,e | 3.37 | 3.48 | 3.39 | 3.33 | .17 | <.0001 |
| Barriers to methods of protectione | 2.48 | 2.63 | 2.47 | 2.57 | .06 | .1163 |
| Self-efficacy | ||||||
| Ability to communicate with partnerc,e | 3.35 | 3.55 | 3.33 | 3.29 | .23 | <.0001 |
| Plan for protected sex and avoid unprotected sexc,e | 2.99 | 3.27 | 3.02 | 3.03 | .27 | <.0001 |
| Where to get birth control (% very sure)c,e | 48.0% | 66.2% | 47.3% | 50.4% | 15.0% | .0017 |
| Behavioral intentions | ||||||
| % Definitely not and probably not have sexc,e | 8.2% | 6.9% | 7.6% | 7.6% | -1.4% | .5504 |
| % Definitely and probably use a condomd | 75.8% | 88.4% | 73.7% | 77.3% | 8.9% | .0052 |
| % Definitely and probably use a method of birth controle | 70.4% | 80.0% | 71.4% | 72.9% | 8.1% | .0422 |
Missing values and “do not know” coded as incorrect responses. Missing values for individual items and constructs ranged from 0 to 20 for pre and 0 to 38 for post. Knowledge scores were calculated for youth that answered over half of the items in the score.
Analysis was completed using mixed models to take into account cluster level randomization, matching of homes, and repeated data on individuals. Adjusted for baseline age, gender, and race/ethnicity. Values are means except where noted.
PTC = Power Through Choices; STIs = sexually transmitted infections.
Difference = (PTCpost – PTCpre) – (Controlpost – Controlpre).
Mean percentage correct responses.
Age p value <.05.
Race/ethnicity p value <.05.
Gender p value <.05
The greatest mean percentage difference in increased knowledge between the intervention and control groups was in the area of methods of protection (16%). Mean scores for attitudes regarding support for methods of protection significantly improved from pre to post by .17 for youth in the intervention group compared to youth in the control group. Mean scores for self-efficacy regarding planning for protected sex and avoiding unprotected sex significantly improved from pre to post by approximately .30 for youth in the intervention group compared to youth in the control group. Self-efficacy regarding where to get birth control significantly improved from pre to post by 15% for youth in the intervention group compared to youth in the control group. The greatest percentage difference in improved behavioral intentions between the intervention and control groups was in the area of condom use (8.9%).
Discussion
This is the first published study using an RCT design to evaluate a comprehensive pregnancy-prevention program for high risk youth living in foster care settings and other out-of-home placements. Our results demonstrate the feasibility of conducting an age appropriate, medically accurate intervention in these unique settings and found significant short-term program effects in regard to increases and improvements in youths’ knowledge, awareness, attitudes, self-efficacy, and behavioral intentions regarding sexuality and sexual behavior.
Although our results demonstrate short-term program effects (e.g., 6 weeks), the results are in agreement with other RCTs of sexuality education programs that focused on teen pregnancy prevention or HIV/AIDs risk reduction and that found significant improvements in knowledge, attitudes, self-efficacy, and intentions that were sustained over 12 months and as long as 24 months [5–7]. An important aspect for our future analyses will be to determine whether the magnitude of the difference between the intervention and control groups in regard to the pre-to-post changes in outcomes scores is sustained at 6- and 12-months post intervention.
The significant improvements in several short-term outcomes are promising nonetheless. The PTC program provided factual information on reproductive health, HIV, STIs, and methods of protection as well as increased the youths’ awareness of available health resources. Other goals of the intervention were to teach the youth to make informed decisions about their sexual risk behaviors and to recognize the potential consequences of these decisions for their future goals. In regard to immediate impact, the results indicate that the PTC intervention is effective in regard to improving pregnancy prevention relevant knowledge, attitudes, and self-efficacy of a racially/ethnically diverse majority male population that was generally sexually experienced.
Two areas that did not significantly improve were attitudes regarding barriers to methods of protection and behavioral intentions to not have sex in the next year. Nearly 90% of the youth reported having had sexual intercourse in the past, and therefore, it was not surprising that the intervention failed to significantly change their long-term intentions toward engaging in sexual intercourse [10]. Similarly, 82% of the participants were male and research literature suggests that male attitudes toward condoms (which were specified in several of the methods of protection questions) may be more difficult to change due to issues such as embarrassment and perceptions about reduced sexual pleasure [23]. Thus, although youth in the intervention group supported the idea of condom use and other methods of protection and reported significantly greater intentions to use a condom, it was difficult to significantly improve their attitudes related to the perceived barriers to using methods of protection that included condom use.
A study limitation is the short-term nature of the evaluation. It is unclear if the short-term program effects will be maintained or significantly associated with changes in youth sexual behavior. Future research will evaluate the PTC intervention’s possible long-term effects on psychosocial variables and their potential associations with youth risk behaviors. Another study limitation is that the control group received “usual care” which was unlikely to be equivalent to the PTC intervention group in regard to the amount of attention received. Therefore, although unlikely, the significant improvements noted in the intervention group compared to the control group may have been due to the greater amount of attention received by the intervention group rather than due to the PTC intervention components.
In summary, this is the first published RCT of a teen pregnancy prevention program designed for youth living in foster care settings and other out-of-home placements. The numerous significant improvements in short-term psychosocial outcomes suggest the PTC program is an effective intervention. However, the most informative results will come from analyses of longer term outcomes (6 and 12 months) that include behavior change data.
IMPLICATIONS AND CONTRIBUTION.
This study is a randomized controlled trial of a teen pregnancy-prevention program designed for youth living in foster care settings and other out-of-home placements. The numerous significant improvements in short-term outcomes are encouraging and provide preliminary evidence that the Power Through Choices program is an effective intervention. However, the most informative results will come from analyses of longer term outcomes (6 and 12 months) that include behavior change data.
Acknowledgments
The authors thank them for their support but acknowledge that the findings and conclusions presented in this publication are those of the authors alone, and do not necessarily reflect the opinions of the Annie E. Casey Foundation. The authors acknowledge the following sites—Oklahoma Institute for Child Advocacy, Kern County Superintendent of Schools, and Planned Parenthood of Maryland—for their outstanding work in regard to data collection and participant retention. The authors also acknowledge the efforts of Dr. Brian Goesling of Mathematica Policy Research who provided valuable consultation regarding the research design and conduct of the study. Finally, the authors thank Dr. Kristen Clements-Nolle for her comments on drafts of the manuscript.
Funding Sources
This publication was made possible by grant number 90AP2665 from the Department of Health and Human Services, Administration for Children and Families. This research was funded in part by the Annie E. Casey Foundation.
Footnotes
Conflicts of Interest: None of the authors have any conflicts, real or perceived, in regard to the submitted article and its content.
Disclaimer: Its contents are solely the responsibility of the Oklahoma Institute for Child Advocacy, University of Oklahoma Health Sciences Center, and the University of Nevada, Reno and do not necessarily represent the official views of the Department of Health and Human Services, Administration for Children and Families.
ClinicalTrials.gov Identifier: NCT01565304.
References
- [1].Casanueva C, Wilson E, Smith K, et al. NSCAW II wave 2 report: Child well-being final report. Washington, DC: US Department of Health and Human Services; 2012. [Google Scholar]
- [2].Dworsky A, Courtney ME. The risk of teenage pregnancy among transitioning foster youth: Implications for extending state care beyond age 18. Child Youth Serv Rev 2010;32:1351–6. [Google Scholar]
- [3].Courtney ME, Dworsky AL, Cusick GR, et al. Midwest evaluation of the adult functioning of former foster youth: Outcomes at age 26. Chicago, IL: Chapin Hall at the University of Chicago; 2011. [Google Scholar]
- [4].The National Campaign to Prevent Teen and Unplanned Pregnancy. Briefly: How is the ’roughly 1 in 4’ statistic calculated? Washington, DC: Author; 2015. [Google Scholar]
- [5].Jemmott JB III, Jemmott LS, Fong GT. Abstinence and safer sex HIV risk-reduction interventions for African American adolescents: A randomized controlled trial. JAMA 1998;279(19):1529–36. [DOI] [PubMed] [Google Scholar]
- [6].Markham CM, Tortolero SR, Peskin MF, et al. Sexual risk avoidance and sexual risk reduction interventions for middle school youth: A randomized controlled trial. J Adolesc Health 2012;50:279–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Lawrence JS, Brasfield TL, Jefferson KW, et al. Cognitive-behavioral intervention to reduce African American adolescents’ risk for HIV infection. J Consult Clin Psychol 1995;63:221. [DOI] [PubMed] [Google Scholar]
- [8].Villarruel AM, Jemmott JB, Jemmott LS. A randomized controlled trial testing an HIV prevention intervention for Latino youth. Arch Pediatr Adolesc Med 2006;160:772–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Jemmott JB III, Jemmott LS, Fong GT, Morales KH. Effectiveness of an HIV/STD risk-reduction intervention for adolescents when implemented by community-based organizations: A cluster-randomized controlled trial. Am J Public Health 2010;100:720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Oman RF, Vesely SK, Green J, et al. Knowledge, attitudes and behavior related to sexual risk behaviors in youth residing in group homes. Am J Public Health. In press. [Google Scholar]
- [11].Herz D, Lee P, Lutz L, et al. Addressing the needs of multi-system youth: Strengthening the connection between child welfare and juvenile justice. Washington, DC: Center for Juvenile Justice Reform, Georgtown Public Policy Institute; 2012. [Google Scholar]
- [12].Herz DC, Ryan JP, Bilchik S. Challenges facing crossover youth: An examination of juvenile-justice decision making and recidivism. Fam Court Rev 2010;48:305–21. [Google Scholar]
- [13].SAS Institute Inc. SAS, version 9.4. Cary, NC, USA: SAS Institute; 2014. [Google Scholar]
- [14].Becker MG, Barth RP. Power through choices: The development of a sexuality education curriculum for youths in out-of-home care. Child Welfare 2000;79:269. [PubMed] [Google Scholar]
- [15].Bandura A Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall, Inc; 1986. [Google Scholar]
- [16].Becker MH. The health belief model and personal health behavior, Vol. 2. Thorofare, NJ: Slack; 1974. [Google Scholar]
- [17].The American Association for Public Opinion Research. Standard definitions: Final dispositions of case codes and outcome rates for surveys. 9th ed. AAPOR; 2016. Available at: http://www.aapor.org/AAPOR_Main/media/publications/Standard-Definitions20169theditionfinal.pdf. Accessed August 10, 2016. [Google Scholar]
- [18].Kann L, Kinchen S, Shanklin SL, et al. Youth risk behavior surveillance—United States, 2013. MMWR Surveill Summ 2014;63(Suppl. 4): 1–168. [PubMed] [Google Scholar]
- [19].Brindis C, Mallari A. Prevention minimum evaluation data set (PMEDS): A minimum data set for evaluating programs aimed at preventing adolescent pregnancy and STD/HIV/AIDS. Los Altos, CA: Sociometrics Corporation; 1996. [Google Scholar]
- [20].Gadermann AM, Guhn M, Zumbo BD. Estimating ordinal reliability for Likert-type and ordinal item response data: A conceptual, empirical, and practical guide. Pract Assess Res Eval 2012;17:1–13. [Google Scholar]
- [21].Harris KM, Halpern CT, Whitsel E, et al. The national longitudinal study of adolescent health: Research design. University of North Carolina. Chapel Hill, NC: Carolina Population Center; 2009. [Google Scholar]
- [22].Bureau of Labor Statistics; U.S Department of Labor. National Longitudinal Survey of Youth 1997 cohort, 1997–2011 (rounds 1–15). Columbus, Ohio: Ohio State University; 2013. [Google Scholar]
- [23].Pleck JH, Sonenstein FL, Ku LC. Adolescent males’ condom use: Relationships between perceived cost-benefits and consistency. J Marriage Fam 1991;53:733–45. [Google Scholar]