Background: EPT’s effect on Chlamydia trachomatis (CT) population prevalence is unknown.
Objectives: To determine if a program that includes population-wide provision of free patient delivered partner therapy (PDPT) and targeted partner services (PS) decreases CT prevalence.
Methods: From 2007-10 we conducted a stepped-wedge community-level randomized trial of EPT in Washington State. Twenty-four local health jurisdictions were randomly assigned into four study waves. Waves initiated the intervention at 6-8 month intervals. The intervention included universal access to free PDPT and provision of PS based on diagnosing clinician assessment which was defined and recorded using a case report form. PDPT use was estimated by interviewing randomly selected cases. The study outcome was chlamydial prevalence in women age 14-25 tested in Infertility Prevention Project (IPP) clinics. Analyses used generalized linear models with random effects for area and clinic, controlling for time.
Results: The intervention increased the percentage of persons receiving PDPT from diagnosing clinicians (16 to 28%) and the percentage receiving PS (26 to 46%). The prevalence of chlamydial infection in women decreased from 8.1 to 6.4% during the study period (p<.001). By intention-to-treat analysis, the intervention did not significantly decrease CT prevalence (RR=.89, 95% CI .76-1.04). CT prevalence was significantly associated with a combined measure of the proportion of persons receiving PDPT or PS (RR=.63, 95% CI .42-.95), but not with either intervention alone.
Conclusions: Although the prevalence of CT declined during the study period, that decline was not clearly associated with our EPT intervention. The decrease in CT may reflect an intervention effect undetectable by our design or analysis, or could be a consequence of other, unidentified factors.
Implications for Programs, Policy, and Research: A population-based intervention that makes free PDPT widely available significantly increased PDPT use. EPT’s population-level impact remains uncertain.