Estimating duration in sex work from cross-sectional studies: a Bayesian global synthesis suggests shorter durations

Bayesian global synthesis of cross-sectional sex-work survey durations estimates true average selling durations that are significantly shorter than prior estimates, e.g., Africa 2.1 vs 5.5 years.

• Why it matters: Cross-sectional surveys can over-represent longer-selling individuals and right-censor reported durations, creating competing biases that prior methods did not jointly correct.
• What they did: The authors derived how reported and true duration distributions relate under censoring, then built a parametric Bayesian model and re-analyzed aggregate data from a global systematic review.
• The result: Model-estimated average true durations are shorter across all regions, including Asia 1.4 (1.2–1.5) vs 2.8–4.0 years, implying HIV incidence in sex work may be higher to fit the same prevalence.

AI summary of the abstract · may contain errors · not medical advice · read the paper.