What is the impact of change in diagnostic test method on surveillance data trends in Chlamydia trachomatis infection?

Objective: To describe the impact of change from culture to more sensitive nucleic acid amplification testing (NAAT) tests on the detection of Chlamydia trachomatis in a genitourinary medicine (GUM) clinic population. Methods: Data were collected between January 1992 and December 2003 on results of C trachomatis tests on male and female attenders at the Lothian GUM clinic (n = 81 590). Routine diagnosis switched from culture to NAAT methods in September 1998. Association of test result with age, sex, year of test, and test type was analysed using logistic regression. Results: 6.1% (95% CI: 5.7% to 6.5%) of women and 7.1% of men (95% CI: 6.7% to 7.5%) tested positive with culture and 9.9% of women (95% CI: 9.4% to 10.3%) and 11.1% of men (95% CI: 10.7% to 11.5%) tested positive with NAATs. This corresponds to a 56% increase for men (95% CI: 47% to 66%) and 62% for women (95% CI: 50% to 67%). Logistic regression showed that a positive test result was strongly associated with test type with or without adjustment for year of test, sex, and young age. Conclusions: The significant increase in chlamydial infections detected following a change from culture to NAATs has important implications for interpretation of trends ascertained from surveillance data. Not all of this can be a direct effect of enhanced sensitivity and there may be indirect effects that improve ascertainment of existing infections. As more laboratories switch to NAATs similar patterns of stepwise increases in positive results are expected and trend analysis based on such surveillance data might thus show an artefactual rise in chlamydia infection rates. Accumulated surveillance data should therefore include timing of introduction of NAAT, so as to take account of under-ascertainment by previous methods.