Aware, a large (n=5012) randomized comparative effectiveness trial, found that HIV risk reduction counseling for HIV negative individuals at the time of an HIV test did not have an impact on cumulative incidence of However, the question remains as to whether there are subgroups that would benefit from counseling. Further, understanding how counseling paradoxically increased STIs in MSM, the group most at risk for HIV in the US, and whether there are other subgroups who increased STIs is of public health importance. In recent innovations machine learning techniques have been used specifically to uncover subgroups with differential treatment responses in a fashion that is replicable and does not suffer model over-fitting associated with multiple testing. We will extend methods to explore treatment subgroups and differences across minority groups based on two of these approaches-Random Forests (RF), and Virtual Twins (VT). The VT approach uses random forests as a first step to create separate forest-based predictions of outcomes under both treatment and control conditions for each trial participant. Then a person-specific treatment effect is created for each individual and a tree-based prediction is made. Whereas this procedure has been shown to be very promising, there is a tendency for the procedure to have relatively low sensitivity, and low positive predictive value. These problems may be alleviated by replacing the single tree predictor of the individual-specific treatment effect by the random forest procedure and/or reweighting of the classification problem to equalize the number of treatment successes (STI incidence is far from affecting 50% of the sample). We will use simulations to uncover the optimal strategy and then use the optimal strategy to create a model that predicts likelihood of future STIs based on behaviors and likelihood of the individual having a positive or negative impact of counseling. We will also use an extension of this model to find the factors associated with the observed MSM by counseling interaction observed in Project Aware. The significance of this research lies in 1) determining if there are subgroups of STD clinic patients who would benefit from or be harmed by short HIV risk reduction counseling 2) providing a model to target these individuals or understanding why they may show increased STIs, and 3) providing the groundwork for use of these approaches to understanding the heterogeneous response to other HIV prevention interventions.

Public Health Relevance

A recent large (n=5012) national study, Project Aware1, showed that on average there was no impact of brief risk reduction counseling on the likelihood of getting a new sexually transmitted infection (STI). This project will determine if there are subgroups of individuals who actually did benefit from brief risk reduction counseling and decreased their likelihood of a future STI as a result or subgroups that showed increases in STIs when counseled. If there are any subgroups with benefit of counseling evident in the data, this project will create a predictive model which could be used to identify individuals prospectively who would benefit from brief sexual risk reduction counseling when visiting an STD clinic;if there are subgroups with negative impact the project will help to understand why.

National Institute of Health (NIH)
National Institute on Drug Abuse (NIDA)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Jenkins, Richard A
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Miami School of Medicine
Public Health & Prev Medicine
Schools of Medicine
Coral Gables
United States
Zip Code
Traynor, Sharleen M; Rosen-Metsch, Lisa; Feaster, Daniel J (2018) Missed Opportunities for HIV Testing Among STD Clinic Patients. J Community Health 43:1128-1136
Lu, Min; Sadiq, Saad; Feaster, Daniel J et al. (2018) Estimating Individual Treatment Effect in Observational Data Using Random Forest Methods. J Comput Graph Stat 27:209-219
Pan, Yue; Liu, Hongmei; Metsch, Lisa R et al. (2017) Factors Associated with HIV Testing Among Participants from Substance Use Disorder Treatment Programs in the US: A Machine Learning Approach. AIDS Behav 21:534-546
Kim, Minjung; Lamont, Andrea E; Jaki, Thomas et al. (2016) Impact of an equality constraint on the class-specific residual variances in regression mixtures: A Monte Carlo simulation study. Behav Res Methods 48:813-26
Van Horn, M Lee; Feng, Yuling; Kim, Minjung et al. (2016) Using Multilevel Regression Mixture Models to Identify Level-1 Heterogeneity in Level-2 Effects. Struct Equ Modeling 23:259-269
Feaster, Daniel J; Parish, Carrigan L; Gooden, Lauren et al. (2016) Substance use and STI acquisition: Secondary analysis from the AWARE study. Drug Alcohol Depend 169:171-179
Alcaide, Maria L; Feaster, Daniel J; Duan, Rui et al. (2016) The incidence of Trichomonas vaginalis infection in women attending nine sexually transmitted diseases clinics in the USA. Sex Transm Infect 92:58-62
Gooden, Lauren; Metsch, Lisa R; Pereyra, Margaret R et al. (2016) Examining the Efficacy of HIV Risk-Reduction Counseling on the Sexual Risk Behaviors of a National Sample of Drug Abuse Treatment Clients: Analysis of Subgroups. AIDS Behav 20:1893-906
Van Horn, M Lee; Jaki, Thomas; Masyn, Katherine et al. (2015) Evaluating differential effects using regression interactions and regression mixture models. Educ Psychol Meas 75:677-714
Tross, Susan; Feaster, Daniel J; Thorens, Gabriel et al. (2015) Substance Use, Depression and Sociodemographic Determinants of HIV Sexual Risk Behavior in Outpatient Substance Abuse Treatment Patients. J Addict Med 9:457-63