Objective: The aim of this study was to evaluate HIV testing yield under several candidate strategies for outreach testing at venues (i.e. places where people socialize and meet new sex partners) in East Africa cross-border areas.
Design: Population-based cross-sectional biobehavioural survey of people who had not been previously diagnosed with HIV found in venues.
Methods: We identified participants who would have been tested for HIV under each of 10 hypothetical outreach testing strategies and calculated the proportion who would have newly tested positive for HIV under each strategy. On the basis of this proportion, we calculated the 'number needed to test' (NNT) to identify one new case of HIV under each strategy. All estimates were obtained by applying survey sampling weights to account for the complex sampling design.
Results: If testing was performed at a random sample of venues, 35 people would need to be tested to identify one new case of HIV, but higher yield could be found by limiting testing to venues with specific characteristics. Strategies focusing on women had higher testing yield. Testing women employed by venues would result in highest yield of all strategies examined (NNT = 15), while testing men under age 24 would result in the lowest yield (NNT = 99).
Conclusion: Quantitatively evaluating HIV testing strategies prior to implementation using survey data presents a new opportunity to refine and prioritize outreach testing strategies for the people and places most likely to result in high HIV testing yield.