Background Recommendations for the frequency of STI screening in high-risk populations are limited by lack of data about when infections occur following a negative screening test.

Methods Participants (18–29 years of age; women [N = 192] and men [N = 156]) in a 12-week study of incident STI were recruited from a county STI clinic. Self-collected vaginal samples (women) and urine samples (men) were obtained weekly at participant’s home. Nucleic acid amplification tests (NAAT) were used for diagnosis of C. trachomatis, N. gonorrhoeae, and T. vaginalis infections. Infections diagnosed at enrollment were treated immediately. Based on cumulative diary reports of partner-specific sexual behaviours, an exposure variable was created to indicate vaginal exposure with only one partner, or with more than one partner. Time to infection was modelled using Kaplan-Meier curves; group differences were assessed via logrank test. Logistic regression was used to assess factors associated with infection status at the study completion.

Results 146/192 (76%) women and 126/156 (81%) men were uninfected at enrollment. At 4 weeks post-enrollment, 16% of women and 8% of men had at least one STI. At 8 weeks post-enrollment, about 23% of women and 10% of men were infected. Overall, 54/192 (28%) women and 19/156 (12%) men had at least one STI during the 12-week followup. Survival curves differed significantly by gender (p = 0.0003). In a multivariable logistic model, women had significantly higher odds of STI (odds = 3.3, p = 0.004), controlling for infection status at enrollment and monogamy status. Gender remained significant in models including number of partners during the follow-up period.

Conclusion In high STI-risk populations, rescreening of initially negative persons identifies high subsequent infection rates, especially among women. Early rescreening (e.g. within 3 months) of any tested person would significantly increase identification of reinfections among initially positive and treated as well as new infections among those previously testing negative.