Nucleic acid amplification tests are more accurate and cost-effective than the Chlamydia rapid test for diagnosing genital chlamydia
- Correspondence to Thomas L Gift
Centers for Disease Control and Prevention, 1600 Clifton Road NE, MS E-80, Atlanta, GA 30333, USA;
Chlamydia is the most prevalent bacterial sexually transmitted infection in the world and has been associated with reproductive sequelae in women, though the frequency with which sequelae develop is not precisely known.1 2 As most cases of chlamydia are asymptomatic, screening – primarily of women – has been the primary intervention used to control chlamydia.2 Screening programmes typically involve collecting a specimen from a patient, sending it to a lab for processing and notifying the patient of results and arranging treatment if positive. As Hislop and colleagues note, rapid tests have been developed to reduce the delay between specimen collection and treatment and to improve treatment rates in some settings where follow-up with patients is problematic. Rapid tests available at present have improved sensitivity over those available 10–15 years ago, up to 80% compared with nucleic acid amplification tests (NAATs). Some can use urine or vaginal swab specimens. Optimal test selection (lab based vs rapid) depends on case detection, test cost, treatment delay, loss to follow-up and consequences of untreated chlamydial infection (both sequelae and further transmission).
Cost-effectiveness of rapid testing
Hislop and colleagues provide a thorough literature review of the recent studies evaluating the patient acceptability and diagnostic accuracy of rapid tests versus NAATs. Findings from this review are applied to a decision model to compare the costs of three tests: two rapid tests and a lab-based NAAT. The authors use a relatively high baseline treatment rate for the rapid tests and the NAAT (95%) and similar test costs. They find that the NAAT is lower in overall cost and treats more cases of chlamydia than the two rapid tests both at baseline and in most sensitivity analyses. The authors did not consider sequelae costs that might result from untreated cases of chlamydia, but this would not have impacted their finding that the NAAT was the best-performing test because the rapid tests led to treatment of fewer cases of chlamydia and had higher overall costs. Incorporating the impact of rapid treatment on transmission may have changed their findings, however.
The authors' findings highlight an important clinical consideration for rapid tests: in settings where a high proportion of patients who test positive are rapidly treated, the advantages of rapid tests over lab-based tests are diminished. Rapid tests are most obviously suited to settings where a substantial proportion of patients may be lost to follow-up. Emergency departments are one such place; studies in the USA have shown that there may be no documented treatment for up to 32% of patients not empirically treated at the time of testing.3 4 There are a number of other settings in which follow-up could be problematic.
Rapid tests could also serve an important role in disease prevention when used in settings in which treatment after testing is delayed. Although the authors did not consider treatment delay in their model, some studies have shown that a small percentage (<5%) of women may develop signs and symptoms of pelvic inflammatory disease by the time of treatment that were not present at the time of testing.1 One study in the USA showed that even in settings where >90% of patients were treated eventually, up to 40% remained untreated 30 days after testing.5 Coupled with expedited partner therapy (where legal), rapid testing could favourably impact the speed of index patient and partner treatment, thus potentially having a substantial impact on transmission.
The decision to utilise rapid tests must be considered carefully; their relatively low sensitivity (and possibly inferior specificity) compared with NAATs relegates them to a niche in which patients not treated at the time of testing may be lost to follow-up or may only be treated after a substantial delay. Within that niche, however, they may have a role.
Disclaimer The findings and conclusions in this manuscript are those of the author and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Competing interests None.