To characterize adherence to post-exposure prophylaxis after non-forcible sexual exposure to HIV, we conducted a review of the literature and meta-analysis. Articles were considered if they contained primary adherence data following non-forcible sexual exposure. Random-effects meta-analysis was used to create pooled point estimates for adherence. Of 1,257 abstracts identified through our search algorithm, 17 were eligible for inclusion in this review, representing 3,634 patients enrolled in 3 randomized controlled trials (RCTs), 9 prospective and 5 retrospective observational studies. Pooled adherence, primarily assessed by self-report, was 77 % [95 % confidence interval (CI) 68-87] in prospective observational studies, 81 % (95 % CI 65-96) in retrospective studies, 78 % (95 % CI 65-91) in RCTs, and 78 % (95 % CI 72-85) overall. Overall adherence was moderately high, with high variability between studies. Assessment of adherence could be enhanced by the use of objective measurements.