Mediator / Moderator Analysis of Latino STYLE: Parent-child communication and other factors influencing the efficacy of a randomized, controlled trial of an adolescent HIV Intervention
Introduction
Latino adolescents face higher odds of HIV infection than their non-Hispanic counterparts in the U.S. The Latino STYLE (Strengthening Today's Youth Life Experiences) intervention is an HIV intervention targeted at Hispanic adolescents and their Spanish-speaking parents. Initial analysis of Latino STYLE data does not show a reduction in unsafe sex practices, but many factors could explain these results, including mediation of parent-adolescent communication, and moderating co-factors like gender, age at first sex, and acculturation. This study will explore these factors as they apply to the findings of the Latino STYLE intervention.
Methods
Latino adolescents were randomized into the Latino STYLE or the general health promotion treatment arm, along with one parent. They participated in an audio computer-assisted self-interview (ACASI) prior to the intervention, and a second ACASI 3-months post-intervention. Differences in their risky sexual behavior were estimated by exposure (an aggregate variable including proportion of condom use and sex-avoidance) for each treatment arm. These differences were assessed using SPSS Statistics version 24 for mediation effects of parent-adolescent communication, and moderation effects of gender, age at first sex, and acculturation.
Results
The sample consisted of 200 adolescents between ages 14 and 17 (mean=15.42, 95%CI=[15.26,15.57]), with 105 females. The experimental group had 102 adolescents, and baseline characteristics of the two groups showed no significant differences at a significance level of 0.1. Mediation analysis using change in parent-adolescent communication scores, showed that neither treatment nor communication were predictors of exposure, and treatment was not predictive of communication, suggesting no mediation effect. Moderator analysis found that only ethnic culture acculturation was predictive of exposure, but this did not significantly interact with treatment to predict exposure.
Conclusion
The two interventions did not differ in their level of communication change between Latino adolescents and their parents, and between the two intervention groups there was no major difference in exposure. Higher ethnic-culture acculturation reduced risky sexual behavior, which may reflect the tendency of more recent immigrants to have better health outcomes known as the "Immigrant Paradox." A study with more sexually-active adolescents could elucidate the insignificant differences in risky sexual behavior found in this study.
Latino adolescents face higher odds of HIV infection than their non-Hispanic counterparts in the U.S. The Latino STYLE (Strengthening Today's Youth Life Experiences) intervention is an HIV intervention targeted at Hispanic adolescents and their Spanish-speaking parents. Initial analysis of Latino STYLE data does not show a reduction in unsafe sex practices, but many factors could explain these results, including mediation of parent-adolescent communication, and moderating co-factors like gender, age at first sex, and acculturation. This study will explore these factors as they apply to the findings of the Latino STYLE intervention.
Methods
Latino adolescents were randomized into the Latino STYLE or the general health promotion treatment arm, along with one parent. They participated in an audio computer-assisted self-interview (ACASI) prior to the intervention, and a second ACASI 3-months post-intervention. Differences in their risky sexual behavior were estimated by exposure (an aggregate variable including proportion of condom use and sex-avoidance) for each treatment arm. These differences were assessed using SPSS Statistics version 24 for mediation effects of parent-adolescent communication, and moderation effects of gender, age at first sex, and acculturation.
Results
The sample consisted of 200 adolescents between ages 14 and 17 (mean=15.42, 95%CI=[15.26,15.57]), with 105 females. The experimental group had 102 adolescents, and baseline characteristics of the two groups showed no significant differences at a significance level of 0.1. Mediation analysis using change in parent-adolescent communication scores, showed that neither treatment nor communication were predictors of exposure, and treatment was not predictive of communication, suggesting no mediation effect. Moderator analysis found that only ethnic culture acculturation was predictive of exposure, but this did not significantly interact with treatment to predict exposure.
Conclusion
The two interventions did not differ in their level of communication change between Latino adolescents and their parents, and between the two intervention groups there was no major difference in exposure. Higher ethnic-culture acculturation reduced risky sexual behavior, which may reflect the tendency of more recent immigrants to have better health outcomes known as the "Immigrant Paradox." A study with more sexually-active adolescents could elucidate the insignificant differences in risky sexual behavior found in this study.
