|
|||||||||||||||
|
|||||||||||||||
 |
|||||||||||||||
Sexually
Transmitted Diseases > Screening
Tests > Appendix C
Appendix CConducting Studies To Evaluate Chlamydia trachomatis and Neisseria gonorrhoeae Tests, Including Studies Required by the Regulations of the Clinical Laboratory Improvement Amendments of 1988 for Verifying or Establishing Test Performance Characteristics When a test is a candidate to replace an established test, the sensitivities and specificities of the candidate and established tests should be compared. Recommendations for the conduct of test evaluation studies have been reviewed elsewhere (C-1–C-4). The imperfect sensitivity and specificity of reference standards complicate evaluation of C. trachomatis and N. gonorrhoeae tests. Discrepant analysis has been used universally to address this problem in the past. However, this approach has been criticized because of the likelihood of producing biased estimates of test characteristics (C-5–C-8). Key to advancing the evaluation of C. trachomatis and N. gonorrhoeae tests is the introduction of study designs that compare candidate and established tests in the same study by using an independent reference standard or by using approaches that do not require designation of a reference standard. Investigation of such designs for evaluation of C. trachomatis tests has begun (C-6, C-9–C-15). The Clinical Laboratory Improvement Amendments of 1988 (CLIA) requires that a clinical laboratory verify or establish the performance of a test to be used in the medical care of persons when it is introduced into that laboratory (C-16). Verification of performance under CLIA applies to the introduction of tests that are Food and Drug Administration (FDA)-cleared, and the CLIA requirement is that the laboratory demonstrate that test performance (i.e., accuracy, precision, reportable range of results, and appropriate normal values) in that laboratory be equivalent to the performance claimed by the manufacturer in the FDA-cleared package insert. Establishing performance under CLIA applies to tests or modifications of tests that have not been FDA-cleared for a particular use. When a modification or use of a test has not been FDA-cleared, CLIA requires that the laboratory establish the performance characteristics of the test for that use. In addition to the performance specifications noted previously, establishing performance must include determinations of analytical sensitivity and specificity, to include interfering substances. Such verification or establishment studies are relatively straightforward when the candidate test performs similarly to the established test. In such cases, the laboratory can verify or establish performance by using the established test as the reference standard if discrepancies between the candidate and established test are uncommon. If the candidate test performs differently from the established test, the laboratory is faced with the economic and logistical challenge of verifying or establishing the performance of the candidate test by introducing an independent reference standard. If the laboratory lacks an independent reference standard, a reference laboratory with an independent reference standard might be enlisted to 1) blindly test a suitable sample of verification or establishment study specimens with the reference standard tests or 2) supply a panel of specimens previously characterized by the reference laboratory. The reader is referred to the American Society for Microbiology CUMITECH series for detailed guidance regarding conduct of verification and establishment studies (C-17). References
Date last reviewed: October 18, 2002 Content Source: MMWR October 18, 2002 / Vol. 51 / No. RR--15 Content Owner: Division of STD Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention |
|||||||||||||||
|
|||||||||||||||
|
Screening
Tests To Detect Chlamydia trachomatis and Neisseria gonorrhoeae Infections
- 2002
