Bridging Disciplinary Boundaries (January 11 - 14, 2007)

Background: Inner-city adolescents with HIV-positive (HIV+) parents are considered by many to represent one of the highest risk groups for their own future HIV infection (Mellins, et al., 2003; Mellins, Brackis-Cott, Dolezal, Richards, & Abrams, 2002). Unfortunately knowledge about contributions to positive sexual behavioral health outcomes in HIV-affected children and adolescents is limited as most research has been based on deficit models in the field of health promotion. The primary goal of the proposed study was to empirically validate a theoretical model to account for the strengths within the ecosystem of HIV affected adolescents that will promote positive behavioral health outcomes.

Method: Structural equation modeling of secondary longitudinal data generated from the “Risk and Resilience in Youth with HIV+ Mothers” study (PI: Mellins) was employed. Guided by a relevant theoretical model of health behavior, Social Action Theory (SAT) (Ewart, 1991), data from 144 families was used to evaluate the usefulness of this model in predicting resilient behavioral health outcomes in HIV-affected children. The specific aims of this study were to: 1) Modify Social Action Theory (Ewart, 1991), to depict the determinants of sexual health specific to HIV-affected adolescents; 2) Chart the resilience trajectories of sexual health protective behaviors, over time, in an urban sample of youths in mid to late adolescents.

Results: The findings from estimating the SAT measurement models at baseline and follow-up indicate that even though the characteristics of the sample seem stable over time, their impact on estimating SAT changes as teens age through adolescence. Hypothesis testing further indicated that SAT can not be applied in the same manner for younger adolescents versus older adolescents. Furthermore, the model predicts very different behavioral health outcomes (drug use) than first hypothesized (sexual risk taking).

Conclusion: The findings from this study point to the lack of empirically validated models that assess the causal factors related to adolescent sexual and drug risk. Indeed, one of the shortcomings of this field is the attempt to apply theoretical models designed for adults to adolescents. Furthermore, the appropriateness of theories varies as teens age through adolescents. This is a highly volatile time in human development and thus requires very specialized attention. This study, nevertheless, offers the opportunity to examine, and potentially address, a number of critical issues within the field of HIV and mental health. Foremost, this study has the potential to significantly contribute to the refinement of a theoretical model for predicting the strengths that children affected by HIV can employ to promote positive behavioral health outcomes.