Abstract
Purpose:
Foster youth have high rates of unintended teen pregnancies and other negative sexual health outcomes. Foster and kinship caregivers (FKC) are an untapped resource to reduce risks. We conducted a two-phase pilot study to evaluate feasibility, acceptability and assess preliminary efficacy of a training designed to improve caregiver communication, monitoring and conflict behaviors and improve sexual health outcomes for youth in foster care. Our study included a Randomized Controlled Trial (RCT) component.
Method:
Phase 1: We recruited 49 FKC and assessed feasibility quantitatively, and acceptability both qualitatively and quantitatively of our intervention (Heart to Heart). Phase 2: We conducted an RCT with 71 participants and evaluated caregiver communication, monitoring, and conflict behaviors as well as the psychological determinants thereof in intervention and control groups at 1, 3, and 6 months.
Results:
Phase 1: Facilitators delivered all intervention content; >90% participants received the entire training. The intervention was highly acceptable (mean score 4.9/5 on two questions). Phase 2: 71 participants were eligible and completed baseline (68 completed at least one follow up survey). We found significant improvements in the intervention group in knowledge, communication expectations, and caregiver-youth conflict behaviors in one or more waves; the control group demonstrated no significant improvements. When groups were compared, we found significant differences in knowledge, communication frequency, and conflict behaviors at 6 months.
Conclusions:
Heart to Heart is feasible, acceptable, and preliminary outcomes data is promising. More research is needed to better establish evidence of efficacy for long-term behavior change in caregivers and youth.
ClinicalTrials.gov identifer:
1. Introduction
1.1. Youth in Foster Care and Sexual Health
Current/former foster youth are at high risk of early unintended pregnancies, sexually transmitted infections (STIs), transactional sex, and intimate partner violence.[1–7] These outcomes can be associated with long-term health effects for them and their offspring, especially when combined with other risk factors (homelessness, economic insecurity, educational disparities).[8–10]
Multiple factors contribute to these outcomes, many of which are out of the youths’ control. Youth in care report earlier sexual debut, decreased use of contraception, higher rates of multiple partners, and sex with infected partners compared with general population youth.[1, 4, 6] Abuse, neglect and other adversities during childhood increases risk of these behaviors.[5, 11] Frequent school and placement changes can reduce access to sexual health information received via sexual education curricula and at home.[2, 12] Further, qualitative work suggests that school-based curricula may not resonate with youth in foster care due to different life experiences compared with some non-fostered peers, and/or may be insufficient.[13] Indeed, while some studies have shown short-term effects on behavior or long-term effects on psychosocial correlates thereof,[14–16] to date there are no sexual health interventions focused on foster youth that have been shown in high quality, longitudinal studies to change long-term rates of sexual risk behaviors among youth in foster care.
Current and former foster youth need and deserve tools to mitigate risks and gain control over their sexual health. Interventions to address systemic and family-level contributors to risk are needed to provide them with the support necessary to achieve healthy reproductive health outcomes.
1.2. Caregiver Interventions
Systematic reviews and meta-analyses indicate that theory-driven interventions targeting parent-adolescent communication, monitoring, and overall relationship quality can improve parent behaviors and adolescent sexual health outcomes.[17–20] Intervention dose may matter; however, even brief interventions can be effective.
Adolescents in foster care are likely to benefit from interventions targeting access to accurate sexual health information from the adults in their lives, including from foster and kinship caregivers (FKC). Prior studies with FKC, child welfare stakeholders, and foster youth support this assertion.[13, 21–26] Qualitative studies specifically suggest that caregivers encounter potentially modifiable barriers to effective communication and monitoring which, if mitigated, could significantly increase FKC’s impact on sexual health risks for youth in foster care. Reported caregiver barriers include: lack of knowledge of normal development and sexual health, lack of skills to effectively communicate and monitor youth, perceptions that talking about sex is against agency policy or puts the caregiver at risk of abuse allegations, and difficulties with effectively managing strong emotions during adolescent-caregiver interactions.[22, 23] Studies have also suggested that caregiver’s own difficulties with emotion regulation may be linked with caregiver-youth conflict, relationship quality and longevity, and the quality of caregiver-youth sexual health communications.[13, 23] In contrast, both caregiver-youth perceived closeness and non-judgmental caregiver communication style within dyads facilitate reduced pregnancy and STI risks for youth in foster care.[11, 24] Similarly, social support, self-care skills, and training opportunities may make it easier for caregivers to support healthy sexual development for teens in care.[24]
1.3. Intervention Development
We sought to develop a caregiver-focused intervention to reduce sexual health risks and improve healthy relationship outcomes among youth in foster care. In informal formative discussions, child welfare stakeholders felt that a caregiver-only training could be more easily implemented into existing training and certification maintenance programs than a combined intervention targeting caregiver-youth dyads. Some reasons cited include logistical (child care), placement instability, and participant retention. Additionally, most FKC care for multiple foster youth over time; thus, we hypothesized that an effective intervention directed towards caregivers could potentially improve outcomes for multiple youth in care.
An adolescent medicine physician, two psychologists, and two research team members developed the intervention, which we named “Heart to Heart”. Table 1 provide details on intervention content, including reason for inclusion and sources. Content was developed iteratively, with the team meeting regularly throughout the process to review and refine content. Some content was developed de novo and some was adapted from existing interventions. Intervention content areas were identified based on three sources: 1) prior adolescent sexual health intervention research suggesting that caregiver training on communication and monitoring can improve youth outcomes;[17–20] 2) mixed methods formative work indicating a need for both information (STI/pregnancy risks, birth control and condoms, normal adolescent development) and skill development (communicating, monitoring, regulating emotions during difficult interactions with foster youth, tailoring skills to the developmental level of the youth);[13, 22–24] and 3) extant theory. We used principles of Social Cognitive Theory to develop the intervention structure (e.g., use of interactive exercises to allow practice of skills).[27] We used principles of Emotion Self-Regulation theory to develop specific intervention content.[28] Specifically, given that both foster youth and FKC viewed caregiver emotion regulation as critical to effective sexual health communications,[13, 23] we included content that targeted emotion regulation both broadly and in the context of caregiver-adolescent discussions around sexual health. For these, we selected skills from Dialectical Behavior Therapy (DBT).[29, 30] We used DBT as the intervention from which to derive this content because it can be used with both adolescents and caregivers,[31] and because there is mounting evidence to suggest it improves caregiver self-regulation and communication effectiveness and improves mental health and reduces sexual and other risk behaviors in adolescents.[32–34]
Table 1:
Intervention Components and Hypothesized Caregiver Outcomes
| Intervention Component | Evidence-Base for Component | Caregiver Outcomes |
|---|---|---|
| Background and Importance | ||
| Youth in Foster Care and Sexual Health • Risks of pregnancy and STIs • Effective communication with youth • Effects of trauma on sexual health |
• Existing evidence • Stakeholder Interviews |
• Knowledge of sexual health and normal development • Sexual health communication expectations, perceived helpfulness, barriers, frequency, number of topics discussed |
| Low-literacy Health Information | ||
| Developmental Stages of Adolescence • Three stages (early, middle, and late) of adolescence described in physical, emotional/social, and cognitive frameworks • Distinguishing typical and atypical adolescent behavior • Identifying and understanding different forms of youth media and social media platforms • Understanding potential influences/impact (both positive and negative) of media and social media use among youth |
• Existing evidence • FKC Focus Groups • Stakeholder Interviews |
|
| Prevention of Pregnancy and STIs • Overview of birth control methods (effectiveness, proper use, benefits, side effects, ability to also protect against STIs, etc.) • Overview of STIs (prevention, treatment, symptoms, etc.) • Transmission of STIs (incidence among adolescents, causes, prevention methods) |
• Existing evidence • FKC Focus Groups • Stakeholder Interviews |
|
| Skill-building and Training | ||
| Communication and Monitoring Strategies related to Sexual Health and Media Use • Developing a tailored communication and monitoring approach based on youth’s developmental stage • Communicating with adolescents – tips and role plays • Addressing common barriers to communication |
• FKC Focus Groups • Stakeholder Interviews |
• Sexual health communication expectations, perceived helpfulness, barriers, frequency, number of topics discussed. • Monitoring behaviors |
| Emotion Regulation and Conflict Reduction Strategies and Techniques • Exercising identifying emotional, reasonable, and wise states of mind • Behaviorism and the importance of positive reinforcement • The importance of validation • Communicating mindfully and effectively with adolescents • Identifying social support sources in your community, family, or professionals • Building a social support system • Developing a personal action plan for self-care |
• FKC Focus Groups • Stakeholder Interviews • Existing evidence |
• Caregiver-youth conflict |
Heart to Heart contains 6 hours of content that can be delivered in one full day or two half days. Caregivers receive a manual accessible to a range of literacy levels containing intervention content designed to be used as a resource and as a tool to use with youth in their home. They receive links to two brief (10–20 minute) videos that reinforce DBT skills taught during the intervention, and that can be watched with youth to teach them skills to navigate sexual health situations and engage in effective communication when setting boundaries with partners. The facilitator manual clearly outlines core exercises and provides scripting for each content area. Content is delivered with a developmental lens; facilitators teach caregivers how to adapt their communication, monitoring, and other aspects of parenting to the youth’s cognitive and social-emotional developmental age rather than chronological age. Specifically, caregivers are taught to use straightforward, non-judgmental communication that emphasize shorter consequences, to use a combination of direct and indirect monitoring techniques and use positive reinforcements particularly in cognitively younger youth as well as when any youth is under stress. Thus, content is applicable for caregivers of youth ages 11–21 years. This broad age range was selected because in formative work and in informal conversations with public and private child welfare partners, stakeholders felt the content would be broadly desirable/applicable for FKC caring for youth across the entire adolescent age spectrum, and because over time FKC are likely to be caring for multiple youth within this age range.
1.4. The Present Study
We present data from a two-phase pilot study. In Phase 1, we used an uncontrolled, single group design to evaluate intervention feasibility and acceptability and collect feedback to refine the intervention. We also conducted an exploratory analysis of short-term changes in outcomes from baseline to one month after intervention delivery. In Phase 2, we conducted a randomized controlled trial with waitlist controls to collect longitudinal data over a 6-month period on caregiver knowledge and psychological and behavioral outcomes. Although content was designed to be suitable for delivery by lay persons, for both phases of this pilot content was taught by the two main intervention developers ([blinded for review]). Participants were consented prior to participation, and materials and procedures were approved by relevant institutional review boards and agencies.
2. Phase 1 Study (Single Group Feasibility & Acceptability)
2.1. Methods
2.1.1. Participants and Recruitment.
We recruited FKC from three large public and private child welfare agencies in New York, NY; Los Angeles, CA; and Seattle, WA who had been caring for at least one foster youth aged 11–18 for at least 3 months in the past year. The study was limited to English speaking caregivers due to funding limitations. We partnered with child welfare agencies who used a combination of flyers, emails, and phone/text to recruit and remind participants of the training.
Data Collection.
Surveys were administered in person or via phone (per agency preference) at baseline and 1-month post-training. Participants also completed a brief acceptability survey immediately after the intervention, and received up to $100 for participation.
2.1.2. Feasibility and Acceptability.
We collected data on number of participants recruited, proportion completing the entire intervention and at least one follow up survey, and proportion of key intervention components delivered. Acceptability was assessed quantitatively and qualitatively immediately after the intervention using two 5-point Likert items: How satisfied were you with this training? How likely would you be to recommend this training to another foster caregiver? (response choices ranged from not at all to very). Additionally, participants were asked three open-ended questions (What would you change about this training? What was most helpful about this training? What was least helpful about this training?). At the 1-month timepoint, participants were asked additional open-ended questions regarding what they remembered about the training, what was most helpful, and what skills, if any, they had used since the training. We also collected informal (verbal) feedback from agency personnel who attended the training with their caregivers regarding perceptions of the training and suggestions for improvement.
2.1.3. Demographic and Caregiving Experiences.
At baseline, we collected data on caregiver gender, age, race/ethnicity, years of experience fostering children/teens, annual income, highest level of education, and current and past caregiving experiences. We also asked about the foster youth in their home including age, gender, type and length of placement, and supervising agency.
2.1.4. Outcomes.
Outcome variables correspond to main intervention components described in Table 1. We used valid and reliable existing instruments for most outcomes; some nonbehavioral communication-related survey tools were shortened due to concerns about survey fatigue. For some existing questions designed to assess a construct related to a specific child (see below), caregivers were asked to answer with respect to the age-relevant youth who had been most recently placed in their home (designated as “Youth 1”).
Knowledge scale. We used eight true/false statements about sexual health and development created by lead researchers to specifically reflect unique intervention knowledge targets such as knowledge of normal development (e.g., “True or false: Youth who start puberty earlier than their peers are more likely to engage in sexual risk behaviors during their adolescence.” A summary score was calculated indicating the percent of correct answers.
Communication variables. In Phase 1 we assessed caregiver sexual health communication expectations regarding talking about sex with Youth 1 using five items from the Parenting Outcome Expectancy Scale.[35, 36] We included three positive expectations items about talking with Youth 1 about sex and two negative expectation items. Items were rated on a 5-point scale (Strongly Disagree/Disagree/Neutral/Agree/Strongly Agree). In Phase 2 we split a single question about whether talking about sex topics would make a youth less likely to “have a pregnancy or a sexually transmitted infection/disease” into two questions. Negative items were recoded and a summary scale reflecting mean score across items was created such that a higher score indicated more positive expectations. To assess the perceived helpfulness of these communications (Not at all/Somewhat/Very) and the frequency of sexual health discussions in the past four weeks (Never/Once/A few times/A lot), we adapted questions from the Adolescent Risk-specific Parent Child Communication Survey,[37, 38]. We also used this tool to assess the frequency of communication on specific topics, including age of sexual initiation, birth control use, condom use, pregnancy, STIs, and/or how to handle peer pressure, with Youth 1 ever (at baseline) or since the last training or survey (follow up timepoints). To assess the number of topics discussed, we created a summary scale of the specific topics. We created a single item to assess barriers to communication previously described by FKC in qualitative work.[22, 23] Participants were asked to check off any of 11 items that could make it challenging to talk about sex topics with any of the teens in their care.
Monitoring behaviors. We used five items adapted from Stattin and Kerr’s monitoring and child disclosure measures (Never/Rarely/Sometimes/Most Times/Always).[39, 40] Items were coded such that a higher score was associated with more parent monitoring/youth disclosure. We created a summary score reflecting mean score across items.
Caregiver-youth conflict behaviors. We used the 20-item Conflict Behavior Questionnaire (Parent Version).[41, 42] We asked caregivers to think back over the past 2 weeks and rate whether statements are true or false with respect to Youth 1. Items were coded such that a higher score was associated with more parent-youth conflict. We created a summary score reflecting the percentage of total items endorsed.
2.1.5. Analysis.
We calculated descriptive statistics on demographic and foster care variables as well as on quantitative feasibility and acceptability variables. We then systematically coded the open-ended qualitative acceptability questions. Two authors reviewed question responses, created a list of common and critical responses for each them, and divided them into subthemes as indicated. For the exploratory outcomes analysis, we compared participants’ outcomes scores at baseline with their 1-month post-intervention outcomes and evaluated significance using paired, two-tailed t-tests.
2.2. Results
Forty-nine FKC from four child welfare agencies participated in one of four intervention sessions in Seattle (one), Los Angeles (two), and New York (one). FKC were primarily female and African American, and 59.6 years old on average. They endorsed a variety of caregiving experiences, although most were experienced caregivers (Table 2).
Table 2:
Participant Demographics and Caregiving Experiences for Both Phases
| Parameter | Proportion or Mean (SD) | |||
|---|---|---|---|---|
|
| ||||
| PHASE 1 (n=49) | PHASE 2 | |||
|
| ||||
| Whole Sample (n=71) | Intervention Group (n=27) | Control Group (n=44) | ||
| Gender | ||||
| Female | 87.8% | 95.8% | 96.3% | 95.5% |
| Male | 12.2% | 4.2% | 3.7% | 4.5% |
|
| ||||
| Ethnicity | ||||
| Latina/Latino/Hispanic | 13.0% | 15.5% | 7.4% | 20.5% |
|
| ||||
| Race | ||||
| Black/African American/African | 75.0% | 67.6% | 47.7% | 61.4% |
| White/Caucasian | 8.3% | 11.3% | 11.1% | 11.4% |
| American Indian or Alaskan Native | 2.1% | 1.4% | 0.0% | 2.3% |
| Other | 6.3% | 1.4% | 3.7% | 0.0% |
| More Than One Race | 8.3% | 9.9% | 7.4% | 11.4% |
|
| ||||
| Age (years) | 59.6 (2.0) | 53.8 (15.9) | 50.0 (19.5) | 56.2 (12.9) |
|
| ||||
| Highest Level of Education | ||||
| Some High School | 12.8% | 2.9% | 3.8% | 2.3% |
| High School Diploma or GED | 8.5% | 10.1% | 7.7% | 11.6% |
| 1 or More Years of College, But No Degree | 21.3% | 36.2% | 34.6% | 37.2% |
| Associates or 2-Year Degree | 12.8% | 24.6% | 34.6% | 18.6% |
| Bachelors or 4-Year Degree | 4.3% | 11.6% | 3.8% | 16.3% |
| Masters, Doctorate, or Other Professional Degree | 4.3% | 14.5% | 15.4% | 14.0% |
|
| ||||
| Income | ||||
| $25,000 or less | 36.8% | 44.1% | 34.8% | 50.0% |
| $25,001 to $50,000 | 39.5% | 25.4% | 26.1% | 25.0% |
| $50,001 to $100,000 | 18.4% | 23.7% | 30.4% | 19.4% |
| $100,001 to $250,000 | 5.3% | 6.8% | 8.7% | 5.6% |
|
| ||||
| Years of Experience with Foster Youth Aged 11–21 | 15.5 (10.4) | 9.5 (8.8) | 11.1 (9.2) | 8.5 (8.5) |
|
| ||||
| How many youth ages 11–21 currently live with you? | ||||
| 1 | 52.8% | 43.7% | 37.0% | 47.7% |
| 2 | 31.5% | 35.2% | 37.0% | 34.1% |
| 3 | 7.9% | 15.5% | 22.2% | 11.4% |
| 4 or more | 3.4% | 5.6% | 3.7% | 6.8% |
|
| ||||
| Reason Foster Youth Have Been Placed With FKC a | ||||
| Family Member of Foster Youth | 55.1% | 52.1% | 59.3% | 47.7% |
| No Special Reason for Foster Youth’s Placement | 38.8% | 18.3% | 22.2% | 15.9% |
| Specializes in Working with Foster Youth with Emotional or Behavioral Problems | 24.5% | 29.6% | 29.6% | 29.5% |
| FKC for a Sibling of Foster Youth | 18.4% | 22.5% | 37.0% | 13.6% |
| Specializes in Taking Adolescent Placements | 18.4% | 21.1% | 29.6% | 15.9% |
| Non-related friend of Foster Youth or Youth’s Family | 12.2% | 29.6% | 25.9% | 31.8% |
|
| ||||
| Descriptors Related to Foster Youth 1 | ||||
|
| ||||
| FKC Relationship to Foster Youth 1 | ||||
| Foster Caregiver (Not related to or close friends with youth’s family) | 45.8% | 36.6% | 44.4% | 31.8% |
| Kinship/Relative Caregiver (Relative or close friend of youth’s family) | 43.8% | 45.1% | 44.4% | 45.5% |
| Adoptive Parent | 10.4% | 18.3% | 11.1% | 22.7% |
|
| ||||
| Foster Youth 1 Age (years) | 15.6 (2.1) | 14.0 (3.8) | 13.2 (3.8) | 14.6 (3.6) |
|
| ||||
| Foster Youth 1 Gender | ||||
| Female | 57.1% | 52.1% | ||
| Male | 40.8% | 47.9% | 59.3% | 47.7% |
| Trans Female | 2.0% | 0.0% | 0.0% | 0.0% |
|
| ||||
| Length of Time Foster Youth 1 Lived with FKC (years) | 6.4 (6.4) | 5.9 (4.7) | 5.8 (4.1) | 5.9 (5.1) |
|
| ||||
| Foster Youth 1 Placement Type / Supervising Entity | ||||
| Private Agency | 25% | 7.4% | 11.1% | 4.9% |
| State Child Welfare Agency | 59.3% | 72.1% | 55.6% | 82.9% |
| Indian Health Services | n/ab | 1.5% | 3.7% | 0% |
| Other | 15.6% | 19.1% | 29.6% | 12.2% |
In Phase 1 race and ethnicity were asked as separate questions whereas in Phase 2 they were asked in a combined question; Phase 2 respondents who only marked Latino were not included under Race since their race was unknown.
In Phase 1 Indian Health Services was included in Other category; this response choice was added during Phase 2 due to agency feedback.
We sought to recruit 8–10 caregivers for each intervention session but did not limit the size of the groups; actual intervention group sizes ranged from 4 to 29 FKC. This suggested recruitment feasibility. The intervention was delivered in the full day format (two trainings) or two half-day format (two trainings) depending on partnering agency preference. Splitting the intervention into two days had minimal impact on retention: among the 33 individuals who attended the first day of 2-day intervention, 31 (93.9%) successfully completed the second day of the intervention. All but three (93.8%) participants completed the 1-month post-intervention survey. Participants also reported high satisfaction (4.9/5) and likelihood of recommending the intervention to another FKC (4.9/5). Interventionists completed 100% of key intervention components for all trainings.
Systematic analysis of responses to open-ended acceptability questions confirmed high acceptability of both information and skills-based content (Table 3). Suggestions from informal agency feedback included adding activities on talking about sex with youth affected by trauma as well as with youth who identify as lesbian, gay, bisexual, transgender, queer, or questioning (LGBTQ). A few participants suggested breaking the intervention up into three days. However, we decided to continue delivering the intervention over 1–2 days during Phase 2, due to high feasibility/acceptability of these formats and feedback from multiple agencies that these schedules were preferred over a 3-day format. Additional changes made based on participant feedback included adding the two videos described in the introduction and increasing the age range of youth that eligible participants could be caring for from 11–18 to 11–21 years, due to feedback from both participants and staff members at hosting agencies that FKC caring for young adults over 18 in extended foster care could also benefit from the intervention.
Table 3:
Sample Responses from Phase 1 Open-Ended Acceptability Questions.
| Immediately After Training | |
| What was most helpful about this training? | What was least helpful about this training? |
|
STD and pregnancy information
“Breaking down STDs/STIs and symptoms. I truly enjoyed all the information presented to me by [the trainers]. I actually learned some valuable information today. The information was easy to digest and retain. I’m so glad to be here today.” Communication and emotion regulation skills “Where kids are psychologically, trying to understand where they are cognitively, which helps know how to help the youth based on where they are. Sex education, monitoring social media - everything was connected and easy to follow.” “Why it is important to overcome discomfort and talk to your teen.” |
More interactive elements
“Not enough role play” No least helpful “Unable to think of anything” “Everything was informative” |
| What would you change about this training? | |
|
Resources to use with youth
“Extra books for teens” “Maybe consider adding a small pamphlet to take home to the teens.” No changes recommended “The training was very informational. There is nothing I would change.” | |
| At 1 Month Follow Up | |
| What do you remember learning about in the training? | |
|
STD and birth control information
“Some of the STDs, how to put the condom on” “Different types of birth control, how to deal with your teen, being patient.” Developmentally-framed communication “Communication skills with children as it relates to their ability to understand the problem.” | |
| What about the training was most helpful for your life? What skills from the training, if any, have you used? | |
|
Developmental framing for communicating with youth
“Most helpful for my life was looking at different ways of how a person understands things - learning to meet a person where they are, meaning a child’s chronological age might be 18, but if you’re speaking about STDs, birth control, pregnancy, etc. they might be 13 in the mind.” “When we were discussing being able to understand that, just because a kid is 17, but could be walking around the house acting like they’re 13, you know how to understand them a bit better.” Intervention manual is a resource “Loved the booklet! I’ve shared it with multiple people with the social worker and teen together.” Communication and emotion regulation skills “I’m trying to listen a bit more, I tend to do a lot of talking.” “How to diffuse a situation. It’s okay to take a time-out and come back with a cooler head, once each side has calmed down…I hear the trainers’ voices in my head when I’m interacting with him.” “DEAR MAN I’ve used in not quite the same [way as in the training]…it can be used with almost anything. I’ve used bits and pieces of it from the training.” | |
For baseline to-1-month post-intervention outcomes, effects were in the expected direction. There were statistically significant changes in knowledge (53% to 75% baseline-post; p<0.001), communication expectations (3.33 to 3.64; p=0.008), perceived communication helpfulness (2.31 to 2.55; p=0.03), and caregiver-youth conflict behaviors (35% to 24%; p=0.002).
3. Phase 2 Study (Randomized Controlled Trial)
3.1. Methods
3.1.1. Participants and Recruitment.
We recruited English-speaking FKC who cared for a youth aged 11–21 for at least three months in the past year in Los Angeles County. Waitlist control participants participated in four surveys first, then received the training. We anticipated a higher attrition rate in the control group due to the 6-month delay in receiving the training; thus we randomized at a rate of 1:1.25 intervention:control.
3.1.2. Data Collection.
Data were collected at baseline as well as 1, 3, and 6 months later, in person, via phone, or online (again depending on caregiver/agency preference). Participants received $25 for the training, and $25 for each survey completed. We used handwritten sheets completed by a trained researcher to track the number of participants who attended trainings as well as the proportion of participants who attended a 2-day training who completed the full intervention.
3.1.3. Measures.
Measures were identical to Phase 1.
3.1.4. Analysis.
We calculated descriptive statistics on demographics, caregiving experiences, and feasibility data. We calculated means and standard deviations for continuous variables such as age and years of experience with foster youth, and proportions for categorical variables such as gender, education and income level. We used mixed-effects linear regression models to assess the effect of the intervention on study outcomes at each follow-up timepoint (i.e. 1, 3, and 6 months after baseline). We calculated the change in each outcome since baseline for each participant, at each follow-up timepoint. A separate model was run for each outcome of interest, with repeated measures by individual modeled as a random effect. Group (intervention vs control), timepoint (modeled as a categorical fixed effect), and an interaction term between the two were included in the models. This enabled comparison of outcomes in the intervention group to (a) their own baseline scores, and (b) to outcomes in the control group. For most outcomes, we calculated the predicted mean change from baseline in each outcome score along with 95% confidence intervals, and conducted tests of simple effects of intervention within and between groups at each timepoint.
For three outcomes (perceived helpfulness of communication, frequency of sexual health communication, number of topics discussed) similar models were used, except that we analyzed mean value at each timepoint rather than change from baseline. For one outcome this was because the question wording differed slightly across timepoints (for number of topics discussed, participants were asked at baseline about topics they had ever discussed, versus whether they had discussed those topics “since your last survey” at follow up timepoints); thus change from baseline would not be meaningful. Two additional outcomes (perceived helpfulness, frequency of communication) consisted of a single ordinal variable rather than a summary score; thus they were analyzed in a similar manner to the number of topics discussed variable. Participants were only asked questions pertaining to Youth 1 at follow up if they currently had the same adolescent youth in their care at the time of that follow-up survey as they did at the baseline survey; over two thirds of participants had the same youth in their home at each follow up timepoint and thus contributed data to these outcomes. We calculated predicted mean outcome scores and confidence intervals. We conducted tests of simple effects of intervention at follow-up timepoints compared to baseline within groups, as well as timepoint-specific differences between groups. In all models, we adjusted for age and education level. There were low levels of missing data for most variables with fewer than 5% of participants missing baseline covariates (one was missing age and two were missing educational level); these three were excluded from adjusted analyses. Mixed effects models can accommodate unequal numbers of follow-up timepoints, thus participants were not excluded from all analyses based on outcomes data missingness, which was also low (less than 10% for each outcome at each timepoint).
3.2. Results
Seventy-one eligible FKC participated in the baseline and were included in analyses, 27 of whom were randomized to intervention and 44 to control (Figure 1). Imbalance in groups was due to the fact that the control group attrition we anticipated did not occur. Sixty-eight (95.75) of these participants participated in at least one follow-up survey. Participants were recruited from seven child welfare agencies. A total of eight trainings were completed with intervention and waitlist control participants (four intervention only, three control, and one that included both intervention and control participants from different cohorts/agencies). The initial training site was excluded from the analysis due to unexpected logistical and eligibility assessment challenges that resulted in only two eligible participants enroll in the training. Specifically, the training happened to coincide with a major natural disaster (wild fire). As an additional contributor, a social work trainee who was not a permanent employee was tasked with recruitment at this initial site; at subsequent sites we partnered with educators and current/former foster parents who were permanent team members to assist in recruitment. This was a far more effective strategy, as it allowed agencies to publicize the training in a manner similar to how they publicize other trainings via the use of a combination of email, flyers, and phone calls. The number of participants ranged from 4–12 participants in each group. The intervention was delivered in the full day format (5 instances) or two half-day format (3 instances) depending on partnering agency preference. Again, splitting the intervention into 2 days had minimal impact on retention; among the 31 individuals who attended day 1 of the 2-day intervention, 30 (96.8%) successfully completed the second day of the intervention.
Figure 1:
CONSORT Diagram for Phase 2 Pilot RCT.
FKC were again primarily female, Black/African American/African, and in their fifties on average (Table 2). Results from main analyses (Table 4) were consistent with Phase 1 results. In the intervention group, there were statistically significant changes from baseline in the expected direction at one or more timepoints in four outcomes: knowledge, communication expectations, perceived communication helpfulness, and caregiver-youth conflict behaviors. For three of these outcomes (knowledge, communication expectations, and caregiver-youth conflict behaviors) the change from baseline was significant at the 6-month timepoint, suggesting sustained change in these outcomes. In contrast, in the control group there were no significant changes from baseline for any outcome at any timepoint. For three outcomes (knowledge, caregiver-youth conflict behaviors, and communication frequency) the comparison between intervention and control groups was significant for at least 1 timepoint.
Table 4.
Phase 2 Outcomes timepoint, adjusted for age, education, and recruiter.
| Outcome: | Months since baseline: | Intervention Group (n=27) | Control Group (n=41) | P-value for difference between groups | ||
|---|---|---|---|---|---|---|
|
| ||||||
| Mean (95% CI) | P-value for difference from zero | Mean (95% CI) | P-value for difference from zero | |||
| CHANGE FROM BASELINE OUTCOMES | ||||||
|
| ||||||
| Knowledge | 1 | 26% (18–34%) | <0.001 | 1% (−4–7%) | 0.57 | <0.001 |
| 3 | 23% (15–31%) | <0.001 | 5% (0–11%) | 0.07 | <0.001 | |
| 6 | 27% (20–34%) | <0.001 | 4% (−2–1%) | 0.18 | <0.001 | |
|
| ||||||
| Communication Expectations | 1 | 0.29 (−0.01–0.60) | 0.06 | −0.08 (−0.29–0.14) | 0.48 | 0.07 |
| 3 | 0.30 (0.01–0.60) | 0.041 | −0.06 (−0.34–0.22) | 0.69 | 0.09 | |
| 6 | 0.36 (0.02–0.70) | 0.037 | −0.04 (−0.29–0.21) | 0.77 | 0.07 | |
|
| ||||||
| Communication Barriers | 1 | 0.3 (−0.70–1.30) | 0.55 | 0.12 (−0.3–0.54) | 0.56 | 0.76 |
| 3 | −0.3 (−0.9–0.29) | 0.32 | −0.16 (−0.66–0.34) | 0.54 | 0.72 | |
| 6 | 0.04 (−0.77–0.86) | 0.91 | 0.3 (−0.22–0.82) | 0.26 | 0.62 | |
|
| ||||||
| Caregiver-Youth Conflict Behaviors | 1 | −0.92 (−2.13–0.29) | 0.14 | −0.73 (−1.83–0.38) | 0.20 | 0.83 |
| 3 | −0.66 (−2.21–0.88) | 0.40 | 1.17 (−0.58–2.93) | 0.19 | 0.15 | |
| 6 | −1.76 (−3.03−−0.48) | 0.007 | 0.48 (−1.19–2.14) | 0.58 | 0.050 | |
|
| ||||||
| Monitoring Behaviors | 1 | 0.06 (−0.15–0.28) | 0.57 | −0.08 (−0.32–0.16) | 0.50 | 0.40 |
| 3 | 0.09 (−0.18–0.37) | 0.50 | −0.08 (−0.32–0.16) | 0.51 | 0.37 | |
| 6 | 0.07 (−0.16–0.31) | 0.53 | −0.02 (−0.25–0.22) | 0.90 | 0.61 | |
|
| ||||||
| MEAN ONLY OUTCOMES | ||||||
|
| ||||||
| Perceived Communication Helpfulness | 0 | 2.35 (2.12–2.58) | (n/a) | 2.52 (2.35–2.69) | (n/a) | 0.23 |
| 1 | 2.56 (2.36–2.75) | 0.10 | 2.46 (2.28–2.63) | 0.46 | 0.46 | |
| 3 | 2.67 (2.45–2.89) | 0.039 | 2.60 (2.42–2.78) | 0.50 | 0.60 | |
| 6 | 2.54 (2.34–2.74) | 0.17 | 2.53 (2.35–2.72) | 0.93 | 0.95 | |
|
| ||||||
| Communication Frequency | 0 | 2.73 (2.28–3.18) | (n/a) | 2.46 (2.17–2.76) | (n/a) | 0.33 |
| 1 | 2.98 (2.71–3.25) | 0.24 | 2.38 (2.13–2.63) | 0.57 | 0.002 | |
| 3 | 2.90 (2.65–3.15) | 0.45 | 2.62 (2.40–2.84) | 0.29 | 0.11 | |
| 6 | 2.75 (2.52–2.98) | 0.94 | 2.61 (2.33–2.88) | 0.35 | 0.45 | |
|
| ||||||
| Number of Topics Discusseda | 0 | 4.47 (4.01–4.93) | (n/a) | 4.43 (4.14–4.71) | (n/a) | 0.87 |
| 1 | 4.66 (4.08–5.24) | (n/a) | 4.21 (3.64–4.79) | (n/a) | 0.36 | |
| 3 | 4.63 (3.83–5.42) | (n/a) | 4.38 (3.82–4.94) | (n/a) | 0.63 | |
| 6 | 5.24 (4.55–5.93) | (n/a) | 4.48 (3.96–5.00) | (n/a) | 0.11 | |
Change from baseline not assessed for Number of Topics Discussed because the question referred to a different time window at each timepoint.
4. Discussion
Adolescent foster youth have higher rates of several challenging sexual health outcomes.[1–7] Caregiver-oriented interventions can reduce risks;[17–20] however, FKC experience unique barriers in acquiring and implementing skills.[13, 22, 23] To the authors’ knowledge, this is the first study to establish feasibility and acceptability and provide preliminary efficacy evidence for a sexual health intervention tailored for FKC.
We saw preliminary evidence that this intervention produced sustained change in caregiver knowledge, communication outcomes, and caregiver-youth conflict behaviors. Findings are consistent with prior research in non-FKC caregivers suggesting that even brief interventions can create sustained change in targeted outcomes.[17–20] Heart to Heart holds promise as a tool to increase foster youth access to accurate, non-judgmental sexual health information from caregivers. It may also improve the quality of parenting in other ways. For example, it may improve use of developmentally-matched parenting techniques (e.g., emphasizing short-term consequences and using more direct monitoring techniques for youth who are cognitively or social-emotionally younger compared with youth who are cognitively and socio-emotionally more mature).[43] Similarly, reducing caregiver-youth conflict is likely to increase trust and bond strength between youth and FKC, and therefore increase the likelihood that communication about sexual health is absorbed by the youth.[13] It could also conceivably impact broader outcomes such as placement stability. Placement stability, in turn, has been linked with a variety of improved young adult outcomes for foster youth.[44]
Main limitations include: 1) small sample sizes, 2) recruitment exclusively from large metropolitan areas, and 3) no outcomes directly measuring caregiver emotion regulation, self-care, or social support, and 4) lack of data collection from adolescents in foster care themselves. Social desirability bias may have occurred for some outcomes; however, this is likely to affect control and intervention groups equally. To address limitations findings should be replicated in a larger RCT that: 1) is powered to detect subtle changes in behavioral outcomes, 2) includes urban, suburban, and rural participants, 3) assess caregiver emotion regulation, self-care, and social support outcomes 4) includes data collection in both foster youth and caregivers, and 5) includes longer term follow up windows (i.e. 12–24 months). It may take time for improved caregiver skills and knowledge to result in improved youth outcomes.
5. Conclusions
Heart to Heart is feasible, acceptable, and promising with respect to outcomes (changes in knowledge, sexual health communication, caregiver-youth conflict). More research is needed to better establish evidence of efficacy for long-term behavior change in a broad array of FKC outcomes as well as youth pregnancy-related behaviors.
Highlights.
Heart to Heart training for foster/kinship caregivers was feasible/acceptable.
We also demonstrated promising behavioral outcomes data.
H2H may have impacts broader than sexual health, including caregiver-youth conflict.
Acknowledgements:
This study was supported by grants from the Conrad N. Hilton Foundation, The National Center for Youth Law, and through NIH grant # K23-MH090898. We would also like to thank the Los Angeles Department of Children and Family Services, New York City Adminstration for Children’s Services, Children’s Village NYC, Washington State Department of Children, Youth and Families, and Casey Foundation. Finally, we would like to thank our other partner in the Los Angeles Reproductive Health Equity Project for Foster Youth.
Abbreviations:
- FKC
foster and kinship caregivers
- RCT
randomized controlled trial
- STI
sexually transmitted infection
- NY
New York
- CA
California
- WA
Washngton
Footnotes
Conflict of Interest
The first author (Ahrens) has confirmed with all co-authors that no authors have conflicts of interest related to this publication.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Courtney ME, Okpych NJ, Charles P, et al. Findings from the California Youth Transitions to Adulthood Study (CalYOUTH): Conditions of foster youth at age 19. Chapin Hall at the University of Chicago: Chicago, IL, 2016. Available at: https://www.chapinhall.org/research/calyouth/. Accessed July 17, 2020. [Google Scholar]
- 2.Courtney ME, Charles P, Okpych NJ, et al. Findings from the California Youth Transitions to Adulthood Study (CalYOUTH): Conditions of foster youth at age 17. Chapin Hall at the University of Chicago: Chicago, IL, 2014. Available at: https://www.chapinhall.org/research/calyouth/. Accessed July 17, 2020. [Google Scholar]
- 3.Courtney ME, Dworsky A, Brown A, et al. Midwest evaluation of the adult functioning of former foster youth: Outcomes at age 26. Chapin Hall at the University of Chicago: Chicago, IL, 2011. Available at: https://www.chapinhall.org/research/midwest-evaluation-of-the-adult-functioning-of-former-foster-youth/. Accessed July 17, 2020. [Google Scholar]
- 4.Courtney ME, Dworsky A, Ruth GR, et al. Midwest evaluation of the adult functioning of former foster youth: Outcomes at age 19. Chapin Hall at the University of Chicago: Chicago, IL, 2005. Available at: https://www.chapinhall.org/research/midwest-evaluation-of-the-adult-functioning-of-former-foster-youth/. Accessed July 17, 2020. [Google Scholar]
- 5.Ahrens KR, Katon W, McCarty C, et al. Association between childhood sexual abuse and transactional sex in youth aging out of foster care. Child Abuse Negl, 2012. 36(1):75–80. DOI: 10.1542/peds.2009-2424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ahrens KR, Richardson LP, Courtney ME, et al. Laboratory-diagnosed sexually transmitted infections in former foster youth compared with peers. Pediatrics, 2010. 126(1):e97–e103. DOI: 10.1542/peds.2009-2424. Epub 2010 Jun 14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Katz CC, Courtney ME, and Sapiro B, Emancipated Foster Youth and Intimate Partner Violence: An Exploration of Risk and Protective Factors. J Interpers Violence, 2017:886260517720735. doi. 10.1177/0886260517720735. [DOI] [PubMed] [Google Scholar]
- 8.Scott ME, Wildsmith E, Welti K, et al. Risky adolescent sexual behaviors and reproductive health in young adulthood. Perspect Sex Reprod Health, 2011. 43(2):110–8. DOI: 10.1363/4311011. Epub 2011 May 17. [DOI] [PubMed] [Google Scholar]
- 9.Zer S, Wainstock T, Walfisch A, and Sheiner E. Perinatal Outcomes and Long-term Health in Offspring of Teenage Mothers. J Pediatr Adolesc Gynecol, 2019. 32(6):622–627. DOI: 10.1016/j.jpag.2019.08.003. [DOI] [PubMed] [Google Scholar]
- 10.Patel PH and Sen B. Teen motherhood and long-term health consequences. Matern Child Health J, 2012. 16(5):1063–71. DOI: 10.1007/s10995-011-0829-2. [DOI] [PubMed] [Google Scholar]
- 11.Ahrens KR, McCarty C, Simoni J, et al. Psychosocial pathways to sexually transmitted infection risk among youth transitioning out of foster care: evidence from a longitudinal cohort study. J Adolesc Health, 2013. 53(4):478–85. DOI: 10.1016/j.jadohealth.2013.05.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Courtney ME, Terao S, and Bost N. Midwest evaluation of the adult functioning of former foster youth: Conditions of youth preparing to leave state care. Chapin Hall at the University of Chicago: Chicago, IL, 2004. Available at: https://www.chapinhall.org/research/midwestevaluation-of-the-adult-functioning-of-former-foster-youth/. Accessed July 17, 2020. [Google Scholar]
- 13.Ahrens KR, Spencer R, Bonnar M, et al. Qualitative Evaluation of Historical and Relational Factors Influencing Pregnancy and Sexually Transmitted Infection Risks in Foster Youth. Children and Youth Services Review, 2016. 61:245–252. DOI: 10.1016/j.childyouth.2015.12.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Green J, Oman RF, Lu M, et al. Long-Term Improvements in Knowledge and Psychosocial Factors of a Teen Pregnancy Prevention Intervention Implemented in Group Homes. J Adolesc Health, 2017. 60(6):698–705. DOI: 10.1016/j.jadohealth.2017.01.004. Epub 2017 Mar 1. [DOI] [PubMed] [Google Scholar]
- 15.Oman RF, Vesely SK, Green J, et al. Short-Term Impact of a Teen Pregnancy-Prevention Intervention Implemented in Group Homes. J Adolesc Health, 2016. 59(5):584–591. 10.1016/j.jadohealth.2016.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Taylor RJ, Shade K, Lowry SJ, and Ahrens KR. Evaluation of reproductive health education in transition-age youth Children and Youth Services Review, 2020. 108(104530). DOI: 10.1016/j.childyouth.2019.104530. [DOI] [Google Scholar]
- 17.Stanton B, Cole M, Galbraith J, et al. Randomized trial of a parent intervention: parents can make a difference in long-term adolescent risk behaviors, perceptions, and knowledge. Archives of Pediatrics & Adolescent Medicine, 2004. 158(10):947–955. DOI: 10.1001/archpedi.158.10.947. [DOI] [PubMed] [Google Scholar]
- 18.Sutton MY, Lasswell SM, Lanier Y, and Miller KS, Impact of parent-child communication interventions on sex behaviors and cognitive outcomes for black/African-American and Hispanic/Latino youth: a systematic review, 1988–2012. Journal of Adolescent Health, 2014. 54(4):369–84. DOI: 10.1016/j.jadohealth.2013.11.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Romer D, Stanton B, Galbraith J, et al. Parental influence on adolescent sexual behavior in high-poverty settings. Archives of Pediatrics & Adolescent Medicine, 1999. 153(10):10551062. DOI: 10.1001/archpedi.153.10.1055. [DOI] [PubMed] [Google Scholar]
- 20.Ethier KA, Harper CR, Hoo E, and Dittus PJ. The Longitudinal Impact of Perceptions of Parental Monitoring on Adolescent Initiation of Sexual Activity. Journal of Adolescent Health, 2016. 59(5):570–576. DOI: 10.1016/j.jadohealth.2016.06.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Svoboda DV, Shaw TV, Barth RP, and Bright CL. Pregnancy and parenting among youth in foster care: A review. Children and Youth Services Review, 2012. 34:867–875. DOI: 10.1016/j.childyouth.2012.01.023. [DOI] [Google Scholar]
- 22.Serrano J, Crouch J, Albertson K, and Ahrens KR. Stakeholder perceptions of barriers and strategies to sexual health discussions between foster/kinship caregivers and youth. Children and Youth Services Review, 2018. 88:434–440. DOI: 10.1016/j.childyouth.2018.03.020. [DOI] [Google Scholar]
- 23.Albertson KM, Crouch JM, Udell W, et al. Caregiver perceived barriers to preventing unintended pregnancies and sexually transmitted infections among youth in foster care. 2018. 94:82–87. DOI: 10.1016/j.childyouth.2018.09.034. [DOI] [Google Scholar]
- 24.Albertson KM, Crouch JM, Udell W, et al. Caregiver-endorsed strategies and facilitators to improving sexual health outcomes among foster youth. Child and Family Social Work, 2020. 10.1111/cfs.12726. Epub 2020 Jan 13. [DOI] [Google Scholar]
- 25.Dworsky A and Dasgupta S. Preventing Pregnancy and Promoting Sexual Health among Youth in Care: Results from the Evaluation of a Training for Caregivers and Child Welfare Workers. Chicago: Chapin Hall at the University of Chicago, 2014. [Google Scholar]
- 26.Love L, McIntosh J, Rosst M, and Tertzakian K. Fostering hope: Preventing teen pregnancy among youth in foster care. Washington D.C. The National Campaign to Prevent Teen Pregnancy, 2005. [Google Scholar]
- 27.McAlister A, Perry C, and Parcel G. How individuals, environments, and health behaviors interact: Social Cognitive Theory. In: Glanz K, RImer B, Viswanath, eds. Health Behavior and Health Education: Theory, Research and Practice. San Francisco, CA: Jossey-Bass, 2008:169–188. [Google Scholar]
- 28.Bariola E, Gullone E, and Hughes EK. Child and Adolescent Emotion Regulation: The Role of Parental Emotion Regulation and Expression. Clin Child Fam Psychol Rev, 2011. 14: 198–212. DOI: 10.1007/s10567-011-0092-5. [DOI] [PubMed] [Google Scholar]
- 29.Linehan MM. DBT Skills Training Handouts and Worksheets, 2nd edition. New York, NY: The Guilford Press, 2015. [Google Scholar]
- 30.Linehan MM, ed. DBT Skills Training Manual. New York, NY: The Guilford Press, 2015. [Google Scholar]
- 31.Miller AL, Rathus JH, and Linehan MM, Dialectical Behavior Therapy with Suicidal Adolescents. New York, NY: Guilford Publications, Inc, 2007. [Google Scholar]
- 32.McCauley E, Berk MS, Asarnow JR, et al. Efficacy of Dialectical Behavior Therapy for Adolescents at High Risk for Suicide: A Randomized Clinical Trial. JAMA Psychiatry, 2018. s75(8):777–785. DOI: 10.1001/jamapsychiatry.2018.1109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Zapolski TCB and Smith GT, Pilot Study. J Sch Nurs, 2017. 33(3):198–204. DOI: 10.1177/1059840516673188. Epub 2016 Oct 14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Miller AL. Introduction to a Special Issue Dialectical Behavior Therapy: Evolution and Adaptations in the 21(st) Century. Am J Psychother, 2015. 69(2):91–95. [DOI] [PubMed] [Google Scholar]
- 35.Diiorio C. Parenting Outcome Expectancy Scale. In: Fisher TD, Davis CM, Yarber DL, Davis SL, eds. Handbook of sexuality-related measures, 3rd ed. New York, NY: Routledge, 2011:255–257. [Google Scholar]
- 36.DiIorio C, Dudley WN, Wang DT, et al. Measurement of parenting self-efficacy and outcome expectancy related to discussions about sex. J Nurs Meas, 2001. 9(2): p. 135–49. [PubMed] [Google Scholar]
- 37.Dutra R, Miller KS, and Forehand R. The process and content of sexual communication with adolescents in two-parent families: Associations with sexual risk-taking behavior. AIDS and Behavior, 1999. 3(1):59–66. [Google Scholar]
- 38.Sales JM, Milhausen RR, Wingood GM, et al. Validation of a Parent-Adolescent Communication Scale for use in STD/HIV prevention interventions. Health Educ Behav, 2008. 35(3): p. 332–45. DOI: 10.1177/1090198106293524. [DOI] [PubMed] [Google Scholar]
- 39.Stattin H and Kerr M. Parental monitoring: a reinterpretation. Child Dev, 2000. 71(4):1072–85. DOI: 10.1111/1467-8624.00210. [DOI] [PubMed] [Google Scholar]
- 40.Lionetti F, Keijsers L, Dellagiulia A, and Pastore M. Evidence of Factorial Validity of Parental Knowledge, Control and Solicitation, and Adolescent Disclosure Scales: When the Ordered Nature of Likert Scales Matters. Front Psychol, 2016. 7: p. 941. DOI: 10.3389/fpsyg.2016.00941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Robin AL and Foster SL. Negotiating parent-adolescent conflict: A behavioral-family systems approach. The Guilford family therapy series. New York, NY: Guilford Press; US, 1989. [Google Scholar]
- 42.Prinz RJ, Foster S, Kent RN, and O’Leary KD. Multivariate assessment of conflict in distressed and nondistressed mother-adolescent dyads. J Appl Behav Anal, 1979. 12(4): p. 691–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Meschke LL, Peter CR, and Bartholomae S. Developmentally appropriate practice to promote healthy adolescent development: Integrating research and practice. Child & Youth Care Forum, 2012. 41(1): p. 89–108. DOI: 10.1007/s10566-011-9153-7 [DOI] [Google Scholar]
- 44.McGuire A, Cho B, Huffhines L, et al. The relation between dimensions of maltreatment, placement instability, and mental health among youth in foster care. Child Abuse & Neglect, 2018. 86:10–21. DOI: 10.1016/j.chiabu.2018.08.012 [DOI] [PMC free article] [PubMed] [Google Scholar]