Screening for Chlamydial and Gonococcal Infections: A Systematic Review Update for the U.S. Preventive Services Task Force [Internet]

Background In 2014, the U.S. Preventive Services Task Force (USPSTF) recommended screening for chlamydia and gonorrhea in asymptomatic, sexually active women, aged 24 years or younger and in older women who are at increased risk for infection. There was insufficient evidence to recommend for or against screening in men. Purpose To update the 2014 systematic review and synthesize evidence for the USPSTF on the effectiveness of screening for chlamydial and gonococcal infection in adults and adolescents, including those who are pregnant. Data Sources Cochrane Central Register of Controlled Trials (January 2014 through May 2020) and Cochrane Database of Systematic Reviews (January 2014 through May 2020), and MEDLINE (January 2014 through May 2020), and manually reviewed reference lists); with surveillance through May 21, 2021. Study Selection English-language trials and observational studies on screening effectiveness, accuracy of risk stratification and alternative screening methods, test accuracy, and screening harms. Data Extraction One investigator abstracted data and a second investigator checked data abstraction for accuracy. Two investigators independently assessed study quality using methods developed by the USPSTF. Data Synthesis (Results) 20 studies met inclusion criteria (N=179,515); seven studies were carried forward from the prior report. Three previously included trials of women found chlamydia screening associated with decreased risk of PID versus no screening, though effects were not statistically significant in two trials. One new, good-quality trial of young women and men in primary care clinics found screening for chlamydia associated with decreased risk of hospital diagnosed primary pelvic inflammatory disease (PID) (Relative Risk [RR] 0.6; 95% Confidence Interval [CI], 0.4 to 1.0), though absolute effects were small (absolute difference −0.137%). Screening was not associated with decreased risk of clinic diagnosed PID (RR 1.1; 95% CI, 0.7 to 1.8) or epididymitis in men (RR 0.9; 95% CI, 0.6 to 1.4). Three studies on the accuracy of risk prediction instruments in asymptomatic persons reported an area under the receiver operating characteristic curve (AUC) that ranged from 0.64 to 0.73. Using age criteria alone ( 98%) followed by endocervical (>96%) and urine samples (>89%). The sensitivity of urine testing for gonococcal infection in males was 93 to 100 percent, while sensitivity ranged from 89 to 100 percent for other sites. Three studies demonstrated that self- and clinician-collected vaginal samples for chlamydia and gonorrhea diagnosis were highly sensitive (90 to 100%); no studies meeting inclusion criteria compared collection methods in males. False-positive and false-negative rates were low for testing across anatomic sites and for self- versus clinician- collection of samples. No studies evaluated screening intervals or accuracy of concurrent testing for other infections. Data was lacking for effects of screening on psychosocial harms or effect on risk behaviors or risk perception. Limitations English language articles only; methodological limitations in the trials; most studies conducted in higher risk populations or settings; meta-analysis not performed; unable to assess for publication bias; studies lacking for pregnant individuals. Conclusions Screening for chlamydial infection may reduce the incidence of PID in young women. Risk prediction criteria to identify persons with asymptomatic chlamydial or gonococcal infection are associated with limited accuracy and require further validation. Testing for asymptomatic chlamydial and gonococcal infections is accurate at most anatomic sites, using urine sampling, and using self-collected specimens. Research is needed to determine the effectiveness of screening in men, optimal screening intervals, and adverse effects of screening.