Background Accurate information about the prevalence of Chlamydia trachomatis infections is needed to assess prevention and control measures at national and regional level.
Methods Systematic review up to August 2012 of population-based cross-sectional studies that estimated chlamydia prevalence in high income countries, focussing on EU/EEA Member States. Data were extracted about study findings; the risk of bias was assessed and meta-analysis was performed where appropriate. Meta-regression was used to examine the relationship between chlamydia prevalence estimates and study response rates.
Results 25 population-based studies were included from 10 EU/EEA countries and 14 studies from 6 other high income countries. Four EU/EEA Member States reported on nationally representative surveys of sexually experienced adults ≤ 26 years, with response rates from 52–71%. Chlamydia point prevalence estimates in 18–24 year olds (3 studies) ranged from 3.0–4.7% in women and from 2.4–4.7% in men. Chlamydia prevalence estimates in EU/EEA countries and other high income countries were statistically consistent. The combined estimate from 5 studies in 18–26 year olds in Europe and the USA was 4.3% (95% CI 3.7, 5.0%) in women and 3.6% (95% CI 2.9, 4.3%) in men. In most studies there was either a high risk of selection bias in the methods used or insufficient information to judge. Estimates of chlamydia prevalence were inversely associated with response rate (p = 0.005 in women, 0.011 in men).
Conclusions Selection bias in chlamydia prevalence surveys is likely, with over-estimation of prevalence being more likely than under-estimation. Cross-sectional surveys with lower response rates are associated with higher estimates of chlamydia prevalence. In studies with low response rates the percentage of chlamydia positive tests should not be interpreted as an estimate of population prevalence. Applying standards for the reporting of prevalence surveys might help to improve consistency in future.