Power to Decide, The National Campaign to Prevent Teen and Unplanned Pregnancy, partnered with the Center for Latino Adolescent and Family Health at New York University's Silver School of Social Work, Healthy Futures of Texas, and the Rio Grande Valley Council to implement Families Talking Together (FTT), an evidence-based teen pregnancy prevention program in Texas. This was a randomized clinical trial of a parent-based sexual and reproductive health intervention (SRH) to foster parent-adolescent communication about sex among Latino adolescents. The FTT intervention focuses on the parenting practices that are important to supporting healthy adolescent sexual behavior. As part of the intervention, families also received a module on the Affordable Care Act (ACA) to influence greater linkages to health care.The study was conducted with 634 parent-adolescent dyads in Willacy, Starr, Hidalgo, and Cameron counties in South Texas. Parent-adolescent dyads completed a baseline survey and were allocated to either an experimental or control group. Participating dyads completed follow-up surveys 3 and 9 months' post-baseline assessment. The proposed outcomes of the study were increased adolescent report of parent-adolescent communication about sex, decrease in sexual risk behaviors (e.g., sexual debut) as well as increased health care insurance enrollment among Latino families.

The study was a two-arm parallel RCT designed to test the efficacy of a teen pregnancy prevention program conducted in the Rio Grande Valley from 2013-2018. Parent-adolescent dyads were randomly assigned to either a passive control group that received no intervention or an experimental group which received the promotor delivered family-based SRH prevention program Families Talking Together (FTT) with an added Affordable Care Act (ACA) module. Parent-adolescent dyads completed baseline, immediate post-baseline (3 months) and delayed (9-month) follow-up assessments.

A combination of venue-based and area sampling methods was used in 4 Rio Grande Valley counties to recruit Latino adolescents and their parents residing in colonias for study participation. The 4 counties (Willacy, Starr, Hidalgo, Cameron) were chosen due to well documented SRH disparities among youth.Trained recruitment promotores attended local community health fairs, shopping malls, supermarkets, etc. where large numbers of resident families could be engaged. During these events, recruitment promotores made initial contact with potentially eligible youth and their families in order to screen for study inclusion criteria. In addition, recruiters went door-to-door in targeted communities conducting active outreach and engagement of families directly within their homes. Eligible families were informed that they were being asked to participate in a research project seeking to improve access to SRH programming for adolescents and that participation would require, at minimum, participating in a survey now as well as after 3 and 9 months. We obtained informed assent and parental consent for all adolescents participating in the study. Parents provided their consent to participate.Refusal bias data was collected as part of the screening process. Refusal data suggested no significant differences between those families declining to participate and those agreeing to be part of the study.

Adolescents and their parents completed baseline, immediate post-baseline (3 months), and delayed follow-up (9 months) assessments using self-administered surveys in either English or Spanish, based on preference. Prior to initiation of the RCT, surveys were pilot tested to ensure conceptual clarity and linguistic appropriateness. To ensure confidentiality, parents and adolescents completed questionnaires separately. A social desirability scale was included in the measurement protocol to assess the extent to which participant responses reflect social desirability bias.

A computer random number generation program was used to create a randomly permutated scheme that assigned specific subject identification numbers to either the experimental or passive control group in a one-to-one ratio. An allocation sequence was generated by the principal investigator. Each experimental/control group allocation sequence was placed in a sealed envelope which remained concealed until after participants had enrolled into the study. After the baseline was completed, the sealed envelope was opened with the family's allocation tied to their specific ID. Data collectors administering the immediate and delayed follow-up surveys were blinded to the condition of participants.

Parents randomized to the experimental group received the Families Talking Together (FTT) intervention, an evidence-based program designed to increase parent-adolescent communication about sex in order to delay sexual debut and prevent negative SRH outcomes in young adolescents (aged 10 to 14). The FTT intervention has been delivered with efficacy in multiple settings, including schools and clinics. FTT is also identified by the United States Department of Health and Human Services and the Office of Adolescent Health (OAH) as a high quality, effective prevention program. The FTT intervention consisted of two components. Intervention component 1 was comprised of two FTT intervention sessions between a parent and the bilingual and bicultural promotor trained to deliver FTT in either English or Spanish. Intervention sessions were delivered to parents in their home or a mutually agreed upon private location in the community. Intervention component 2 was comprised of written supplemental materials that promotores used to guide each intervention session. Written materials included a family FTT workbook and a short story entitled 'Victor and Maria.'

During each intervention session, promotores reviewed the intervention materials and instructed parents how to structure conversations about delaying sexual debut and the use of contraception with their adolescent. At three and 9 months post-baseline, adolescents in the experimental and control groups completed follow-up surveys administered by data collectors.

To examine group differences in parent-adolescent communication variables, ordinary least squares (OLS) regression was used. To ensure baseline equivalence between the two arms of the RCT, demographic and outcome variables were compared at baseline between the intervention and control groups. In order to assess the generalizability of results across traditional and robust methods of analysis, all analyses were repeated with bootstrapping. Additionally, outlier and specification-error for each analysis was conducted.

Inclusion Criteria

• Self-identify as Latino
• At least one a parent present in home, defined as a biological or non-biological primary caregiver
• Resident of target geographic community (Willacy, Starr, Hidalgo, or Cameron) in the Rio Grande Valley

Exclusion Criteria:

• Self-identify as non-Latino
• Lack of parent or caregiver in home during recruitment
• Resident of non-target community in the Rio Grande Valley

A computer random number generation program was used to create a randomly permutated scheme that assigned specific subject identification numbers to either the experimental or passive control group in a one-to-one ratio. An allocation sequence was generated by the principal investigator. Each experimental/control group allocation sequence was placed in a sealed envelope which remained concealed until after participants had enrolled into the study. After the baseline was completed, the sealed envelope was opened with the family's allocation tied to their specific ID. Data collectors administering the immediate and delayed follow-up surveys were blinded to the condition of participants.