Objective      To understand the cohort retention and its predictors in pre-exposure prophylaxis (PrEP) clinical trial among men who have sex with men (MSM).       Methods      Using non-probability sampling methods

167 eligible subjects were recruited and then assigned to a daily dose arm or an intermittent dose arm randomly at the proportion of 1∶1. Follow-up and questionnaire were carried out in weeks 0 (baseline)

4

8

12

16

20

24

and 28. Survival curve based on the Kaplan-Meier estimation was conducted to describe the cohort retention rates of different trial arms. Univariate and multivariate logistic regression procedures were conducted to analyze the predictors of cohort retention.       Results      Overall

80.2% of the subjects completed at least one follow-up after taking medicine

and 59.9% completed a 28-week follow-up. Intermittent dose arm (ARR=0.60

95% CI:0.36-0.99)

age of 24 years and older (ARR=0.46

95% CI:0.25-0.83)

middle income (ARR=0.45

95% CI:0.21-0.96)

and participants who attended the experiment wishing to spread a kind of medication to prevent the infection of human immunodeficiency virus (HIV) (ARR=0.59

95% CI:0.36～0.99) showed lower attrition

namely higher cohort retention. Those who attended the experiment for free medication (ARR=1.90

95% CI:1.13-3.21) had higher attrition

namely lower cohort retention.       Conclusion      The cohort retention of PrEP clinical trial is not high. The cohort retention rate is higher in the intermittent dose arm than in the daily dose arm. Appropriate strategies are needed to promote the cohort retention in the further PrEP clinic trial.