Screening pregnant women for chlamydia: what are the predictors of infection?

OBJECTIVES: To determine the risk factors associated with chlamydial infection in pregnancy and the sensitivity and specificity of these when used for selective screening. METHODS: A prospective, cross-sectional study of pregnant women aged 16-25 years attending four major public antenatal services across Melbourne, Australia. Between October 2006 and July 2007, women were approached consecutively and asked to complete a questionnaire and to provide a first-pass urine specimen for Chlamydia trachomatis testing using PCR. RESULTS: Of 1180 eligible women, 1087 were approached and 1044 (88%) consented to participate. Among the 987 women for whom a questionnaire and a definitive diagnostic assay were available, the prevalence of chlamydia was 3.2% (95% CI 1.8 to 5.9). In a multiple logistic regression model, more than one sexual partner in the past year (AOR 11.5; 95% CI 7.1 to 18.5) was associated with chlamydia infection. The use of any antibiotic within 3 months (AOR 0.2; 95% CI 0.1 to 0.6) was associated with a decreased risk of infection. Screening restricted to women who reported more than one sexual partner in the past year would have detected 44% of infections in women aged 16-25 years and would have required only 7% of women to be screened. The addition of those women aged 20 years and under would have required 27% of women to be screened and detection of 72% of infections. CONCLUSIONS: Selective chlamydia screening of pregnant women based on risk factors can improve the yield from screening. However, the potential harm of missed infections among excluded women would need to be considered.