Population effectiveness of opportunistic chlamydia testing in primary care in Australia: a cluster-randomised controlled trial

Abstract

Screening young adults who are sexually active for genital Chlamydia trachomatis infection is promoted in several high-income countries, but its effectiveness at the population level is highly debated. We aimed to investigate the effects of opportunistic chlamydia testing in primary care on the estimated chlamydia prevalence in the population aged 16-29 years in Australia. We did a cluster-randomised controlled trial. Clusters were rural towns with a minimum of 500 women and men aged 16-29 years and no more than six primary care clinics. We randomly allocated each cluster using a computer-generated minimisation algorithm to receive a multifaceted, clinic-based chlamydia testing intervention or to continue usual care. The intervention included computerised reminders to test patients, an education package, payments for chlamydia testing, and feedback on testing rates. The primary outcome was chlamydia prevalence, estimated before randomisation (survey 1) and at the end of the trial (survey 2) in patients aged 16-29 years who attended the clinics. Analyses were done by intention to treat. General practitioners and clinic staff were aware of group allocation, whereas patients and laboratory staff who performed the chlamydia tests were not. This trial was completed on Dec 31, 2015, and is registered (ACTRN12610000297022). Between Dec 14, 2010, and Sept 14, 2015, 26 clusters (63 clinics) received the chlamydia testing intervention and 26 (67 clinics) continued usual care. Over a mean duration of 3·1 years (SD 0·3), 93 828 young adults attended intervention clinics and 86 527 attended control clinics. The estimated chlamydia prevalence decreased from 5·0% (95% CI 3·8 to 6·2) at survey 1 to 3·4% (2·7 to 4·1) at survey 2 in the intervention clusters (difference -1·6%, 95% CI -2·9 to -0·3) and from 4·6% (95% CI 3·5 to 5·7) at survey 1 to 3·4% (2·4 to 4·5) at survey 2 in the control clusters (difference -1·1%, -2·7 to 0·5). The unadjusted odds ratio for the difference between intervention and control clusters was 0·9 (95% CI 0·5 to 1·5). These findings, in conjunction with evidence about the feasibility of sustained uptake of opportunistic testing in primary care, indicate that sizeable reductions in chlamydia prevalence might not be achievable. Australian Government Department of Health, National Health and Medical Research Council, Victorian Department of Health and Human Services, and New South Wales Ministry of Health.