HIV Risk Reduction in Youth in the Bahamas

HIV infection in the Bahamas is the leading cause of death among Bahamians between the ages of 15 and 44. It is estimated that 4.13% of adults in the Bahamas are HIV infected, and the Bahamas has the highest annual incidence of AIDS in the English-speaking Caribbean. One cause for these high rates of infection is the prevalence of risky behaviors among Bahamian youth. Among youth between the ages of 13 and 15, 32% are sexually experienced, with that rate rising to 57% among youth age 16 or older. Approximately half of these adolescents have never used birth control, and only one third use birth control consistently.

This study will examine the benefits of combining a youth HIV prevention education program with a parental program focused on communication and monitoring. The youth education program, developed in the United States but modified to address Caribbean culture, is called Focus on Youth in the Caribbean (FOYC). FOYC has already been shown to be effective in increasing condom use, knowledge of HIV, and knowledge of risks associated with unprotected sex and drug use. Including parents in this intervention effort may broaden and sustain its effects. A study of seventh to eleventh graders found that teens were six times less likely to engage in sexual activity if they thought their mothers disapproved of it. The study also found that teens often did not have a clear idea of what their parents approved of or disapproved of, indicating a need for better communication between parents and children. Other studies have shown that parents who monitor their children reduce the likelihood that those children have unprotected sex or use drugs. The parental intervention program, called Caribbean Informed Parents and Children Together (CImPACT), has been shown to increase communication between parents and children and, when paired with a face-to-face intervention, enhance condom skills and lower rates of sex, unprotected sex, cigarette use, and alcohol use. This study will examine whether pairing the FOYC and CImPACT interventions will reduce prevalence of behaviors that present high risk of HIV infection in youth in the Bahamas.

Parents will be randomly assigned to receive either the CImPACT program or a control program called Goal for It (GFI), both of which will be implemented in a 1.5-hour training session. Schools will be randomly assigned to receive either the FOYC program or a control program entitled Wonderous Wetlands. Both programs will be implemented as part of the curriculum in 15 Bahamian elementary schools. Children, therefore, will be part of 1 of 3 groups:

1. Parent receiving the CImPACT program; school receiving the FOYC program
2. Parent receiving the GFI program; school receiving the FOYC program
3. Parent receiving the GFI program; school receiving the Wonderous Wetlands program

Parents and children will be assessed at the beginning of the study and after 6, 12, 18, 24, and 36 months. These assessments will include questionnaires measuring high-risk behaviors, conceptions of risks and rewards involved in high-risk behaviors, and parent and adolescent communication and monitoring.

In the continuation of this study we will:

1. Recruit 2600 youth (and their parents) from 172 Grade 10 FHLE classrooms over two years in all seven government high schools in New Providence, The Bahamas
2. Randomly assign at the level of the classroom 2400 youth (and their parents) from 160 classes to one of four intervention conditions: 1) HFLE only (no parent intervention); 2) BFOOY only (no parent intervention); 3) BFOOY plus ImPACT (delivered to parent-youth dyads); or 4) BFOOY plus GFI (delivered to parent-youth dyads);
3. Randomly assign one of the seven high schools (with ~12 HFLE classes and ~200 students) to a "control school" condition to permit exploration of possible class-to-class contamination within schools
4. Assess intervention effects on the entire youth sample at 6, 12, 18 and 24 months post-intervention and among parents at 6 and 12 months
5. Conduct sub-group analyses to explore: i) the impact of related and intensive interventions received at two critical junctures during adolescence; ii) the effect of being part of a longitudinal trial on risk/protective behaviors; and iii) the extent of intervention contamination when randomization is conducted at the level of the class rather than the school. Subgroups will include: Exposed youth (youth who received FOYC in Grade 6) and Naïve youth (youth who did not receive FOYC in Grade 6); Enrolled youth (youth who were enrolled in the ongoing evaluation of FOYC whether as controls or intervention (FOYC) youth) and Not-enrolled youth (youth who were never part of the ongoing study); and Control School youth (those randomly assigned to the control condition at the level of the school in Year 2) and Control Class youth (those assigned to the control condition at the level of the class)

Intervention effect (primary hypotheses):

1. Youth randomized to HFLE only will exhibit significantly higher rates of HIV risk behavior, intentions and perceptions and lower knowledge and condom skills through 24 months post-intervention than each of the other three groups;
2. Youth randomized to BFOOY plus GFI will exhibit higher rates of abstinence behavior, intentions and perceptions than youth randomized to BFOOY alone through 24 months follow-up
3. Youth randomized to BFOOY plus ImPACT will exhibit lower rates of HIV risk behavior, intentions and perceptions and higher rates of knowledge, condom skills, and STD testing than each of the other three groups through 24 months follow-up; Parents receiving ImPACT will exhibit higher knowledge and condom use skills than GFI or control parents; ImPACT parents and youth will report higher rates of discussion about HIV prevention and safe sex than parents and youth in the other three intervention groups
4. Intervention behavioral effects will be consistent with changes in perceptions and intentions as posited by the PMT, the theoretic model of behavioral change on which BFOOY is based

Secondary/exploratory hypotheses will include:

1. At baseline, Exposed Youth will demonstrate higher rates of condom-use and condom-use intentions and perceptions than Naive youth
2. Exposed youth will demonstrate higher protective behaviors, intentions and perceptions compared to Naive Youth within an intervention category (BFOOY or HFLE) throughout the follow-up period
3. Enrolled youth who received FOYC in Grade 6 will not differ from Not-enrolled youth who received FOYC in Grade 6 with respect to intentions, perceptions and knowledge
4. Youth in the Control Classes will not differ from youth in the Control school with respect to HIV knowledge, perceptions and intentions As a result of this study we shall be able to identify of one or more interventions or intervention combinations targeting mid-adolescents to reduce HIV risk behavior and address questions regarding possible interactions between intervention receipt in both the pre-adolescent and mid-adolescent years. With this knowledge, health policy officials globally, including those in The Bahamas, will be able to make evidence-based decisions appropriate to their resources and the epidemiology of the epidemic in their settings.