Nonparametric/functional contributable-risk model with the application to trend decomposition for HIV/AIDS incidence rate in rakai teenagers, Uganda

Background: The HIV incidence rate declined over past 10 years among the teens in Rakai, Uganda. Meanwhile, social environment, individual sexual behaviors and demographic composition also changed over time. It is desirable to understand the dynamic contributions of those factors towards the decline of HIV incidence. Most of existing decomposition methods measure relative contributions over two time points. Methods: We propose to quantitatively decompose the declining trend to various contributors based a set of nonparametric models. Specifically, we model the association between the main outcome (HIV incidence) and potential contributors using nonparametric varying-coefficient poison regression; and then model the changes of the contributors over time using separate nonparametric regressions. Combining those models, we develop a new index that visualizes the relative contributions in proportions, and allow the decomposition to vary over time.

Results: We applied the methods to the Rakai Community Cohort Study from 1999-2011 for adolescent of 15 to 19 years old. Overall, 89% of the decline in HIV incidence rate in 1999-2011 is due to the decreasing proportion of sexual experienced teens, and the rest 11% is due to “safer sex” among sexual experienced teens. Safer sexual behavior has stronger contributions towards the decline of HIV incidence during 2000-2002 and during 2007-2008, while in other years, fewer sexually experienced teens dominates the contribution to the change of HIV incidence.

Conclusions: A new statistical method is proposed to attribute the change of an outcome to various contributors, and is promising by applying to a HIV study among Ugandan youth.