P3.314 Can Chlamydia Prevalence Monitoring Data Be Used to Evaluate Impact of Screening? The US CDC Infertility Prevention Project Experience

Background Chlamydia testing data are often used for prevalence monitoring to evaluate screening programmes; however, trends in positivity are impacted by changes in screening coverage and criteria, independent of changes in population prevalence, and thus are difficult to interpret. Given limited resources, many chlamydia control programmes in the US target screening to improve cost-effectiveness. We explored the potential impact of focusing screening on high prevalence (i.e., > 3%) clinics on trends in chlamydia positivity. Methods We analysed line-listed data on women tested for chlamydia in family planning clinics participating in the Infertility Prevention Project (IPP) during 2000–2011. Trends in annual positivity in family planning clinics participating in IPP who reported at least 120 tests restricted to patients aged 15–24 years were examined among two cohorts: (1) all clinics and (2) all clinics, with testing data removed from subsequent years for clinics where positivity fell below 3%. Results Positivity trends for both cohorts are shown in the table, along with overall percent change in positivity over the period. All trend lines increased over time; however, trend lines with low prevalence clinics removed had a higher positivity at each year. Similar patterns were seen for both cohorts of clinics, as well as when stratified by geographic region. Percent change in positivity over the 12 year period was 52.2% for cohort 1 and 64.1% in cohort 2. Conclusion Our analysis of chlamydia data suggests that individual point estimates of chlamydia positivity are likely overestimated when chlamydia screening was targeted to high prevalence clinics; however trends over time were similar in the two analytic groups. Data used for programme monitoring and evaluation may bias point estimates of prevalence. Caution should be used when using prevalence monitoring data to evaluate impact of screening without considering clinic-level confounders. YEAR All Clinics Below 3% Removed