Article Text

Download PDFPDF

Can we climb out of the “pit” of poorly performing rapid diagnostic tests for chlamydia?
  1. Charlotte A Gaydos
  1. Dr Charlotte A Gaydos, Division of Infectious Diseases, Johns Hopkins University School of Medicine, 530 Rangos Building, 855 North Wolfe Street, Baltimore, MD 21205, USA; cgaydos{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Michel et al (see page 187) present documentation in this issue that a Conformitée Européenne (CE) marked home-use test for Chlamydia trachomatis that is available over the internet performs very poorly.1 The authors demonstrated a lack of accuracy of this assay compared with Amplicor PCR with sensitivities of 12.5% and 20% in both a lower (8% prevalence) and high prevalence (22.9% prevalence) population, respectively. False-positive results were equally as problematic, since the specificities were only 93.5% and 88.1% in the two populations studied. Since this test could be used by asymptomatic women, taking advantage of easy purchasing through the internet, the demonstrated positive predictive value of 28% is also unacceptable. With an overall negative predictive value of 85%, 15% of women would assume they were uninfected, when actually they were infected. This assay is reported to be CE-marked, a status which is based on strict performance assessment. The reported poor performance of the home test should generate regulatory action to have this CE mark removed. Another study supported the poor performance reported by Michel et al.1 2

In order to deal with this problem, educators and healthcare providers must inform at-risk women about internet assays marked for “self-administered tests” and for “home use” that do not provide reliable results. The proliferation of such poor testing merchandise available from internet sites could damage the credibility of internet sites that offer well-documented and reliable testing for sexually transmitted infections. The internet currently offers women the opportunity for home collection of samples for chlamydia screening that are then mailed to a laboratory for nucleic acid amplification tests (NAAT).3 4 Hopefully, we will someday have such tests that can be performed by the user at home, but we are not there yet.

The type of assay studied by Michel et al falls under the assay category called rapid, “point of care” (POC), or “near patient” test, which can quickly provide results while the patient is still in the clinic or, at least in theory, immediately “at home” if the patient could perform the assay accurately.1 As a group, these types of assays are currently poor performers, with sensitivities in the range of 50–60% compared with the gold standard of NAAT.510 Many of the POC tests for chlamydia are only US Food and Drug Administration (FDA) cleared for cervical samples and not vaginal samples, which will be required for any home test. Many of these assays have obtained clearance by regulatory bodies because they were compared with the old gold standard of culture, which is no longer the gold standard but is rather “tarnished.” Because of their poor performance compared with NAATs and package inserts that may be overstated, clinicians who use them are likely unaware of the risk they pose to obtaining a correct diagnosis.6 10 In order to escape the perils and “pitfalls” of poorly performing POC assays, we must require new platforms with new technologies to meet or exceed the test sensitivities and specificities currently available. Better tests are being developed, and the diagnostic community awaits their evaluation and clearance by regulatory bodies such as the FDA.11 12 Internet-available diagnostic assays are expected to become more, not less, accessible. Therefore, consideration must be given to both national and international regulation to prevent substandard tests being obtained and used by unsuspecting consumers via the internet.



  • Competing interests: None.

Linked Articles