The HIV epidemic continues to disproportionately affect ethnic minority youth. These disconcerting health disparities indicate that although existing HIV preventive strategies for ethnic minority youth have been efficacious, they have not significantly reduced the impact of the epidemic in this population. Macro-level interventions, such as structural or policy interventions, have the potential to impact the HIV epidemic at a population level, and thus reduce the HIV health disparities that exist among ethnic minority youth and other segments of the U.S. population. This article calls for a paradigm shift to develop, evaluate, and disseminate interventions that target upstream/macro-level factors or that, at a minimum, integrate both a macro and individual level perspective. The article also discusses the challenges in developing and evaluating such interventions. Psychologists and other behavioral scientists can play a critical role in reducing the impact of HIV on ethnic minority youth by integrating macro-level approaches to future HIV prevention strategies.

College drinking is a serious health concern. Few studies have examined screening measures and methods of administration. This study compares two alcohol screens (NIAAA 5/4 binge drinking question or Alcohol Use Disorders Identification Test (AUDIT)) in a college student health clinic waiting room and two modes of administration (self-administered either on a computer kiosk or on a tablet computer). Participants were 259 undergraduates from the University of Miami. Most (78-98%) students completed screening. More students were identified with risky alcohol use with the 5/4 (49%) than AUDIT (14%). On the 5/4, administration method was not linked to completion, 93% kiosk vs. 95% tablet, p = .554, but was related to identification as a risky alcohol user, 42% kiosk vs. 56% tablet, p = .033. On the AUDIT, administration method was significantly related to completion, 73% kiosk vs. 98% tablet, p < .001, and identification, 8% kiosk vs. 23% tablet, p = .003. Method of administration of the single item 5/4 binge drinking question was related to the a higher proportion of students identified with risky alcohol use when screened by a computer tablet, but not completion rates; the AUDIT method of administration was related to both completion and identification rates (higher rates with the tablet in both cases). Education of student health providers who make decisions about what screening tools to use in their centers and who interpret the results of alcohol screening in college health centers should consider the potential influence of administration method. Future research should examine the reasons that method of administration might influence screening results.

Many preventive trials randomize individuals to intervention condition which is then delivered in a group setting. Other trials randomize higher levels, say organizations, and then use learning collaboratives comprised of multiple organizations to support improved implementation or sustainment. Other trials randomize or expand existing social networks and use key opinion leaders to deliver interventions through these networks. We use the term contextually driven to refer generally to such trials (traditionally referred to as clustering, where groups are formed either pre-randomization or post-randomization - i.e., a cluster-randomized trial), as these groupings or networks provide fixed or time-varying contexts that matter both theoretically and practically in the delivery of interventions. While such contextually driven trials can provide efficient and effective ways to deliver and evaluate prevention programs, they all require analytical procedures that take appropriate account of non-independence, something not always appreciated. Published analyses of many prevention trials have failed to take this into account. We discuss different types of contextually driven designs and then show that even small amounts of non-independence can inflate actual Type I error rates. This inflation leads to rejecting the null hypotheses too often, and erroneously leading us to conclude that there are significant differences between interventions when they do not exist. We describe a procedure to account for non-independence in the important case of a two-arm trial that randomizes units of individuals or organizations in both arms and then provides the active treatment in one arm through groups formed after assignment. We provide sample code in multiple programming languages to guide the analyst, distinguish diverse contextually driven designs, and summarize implications for multiple audiences.