Advances in molecular disease analysis and directed national funding of efforts such as the Human Genome Project have begun to bring potentially valuable information about many diseases through molecular genetic testing. While molecular genetic testing in the clinical laboratory enables identification of patients at risk for a disease process, many clinicians do not fully appreciate the complexity implied in these tests. To begin with, the ease with which preliminary research data have turned into presumed valid genetic testing is of concern. What few understand is that availability of a test that identifies a disease should be only the beginning of a carefully thought out process. However, it is not difficult to set up a genetic testing service. Clinical Laboratories Improvement Act registration and Health Care Financing Administration certification are usually not difficult to obtain. A recent National Institutes of Health (NIH) task force raised concerns about the quality of the laboratories performing genetic tests. Although not required to, most laboratories voluntarily participate in inspections or other programs to assess quality of testing. The NIH group also was concerned that laboratory requirements that do exist for accreditation insufficiently emphasize the preanalytic and postanalytic aspects of the tests.1 While we are justly proceeding with molecular genetic testing, many clinicians and even more patients do not completely understand the data the testing provides. With numerous laboratories performing this technically demanding analysis, enough time has elapsed to begin asking questions about the reliability of molecular genetic testing. In this issue of The Journal, Hofgartner and Tait 2 review the frequency of problems in clinical molecular genetic testing, a relatively new specialty of laboratory testing. As evidenced in their survey, the mean age of a molecular genetics testing laboratory is 6.5 years. This study surveyed laboratories on site (n = 2) and by questionnaire (n = 42 respondents) to determine the incidence of problems in the tests performed and where problems occurred, that is, in the pretest, laboratory, or posttest phase, and attempted to determine the level of actual harm resulting from errors. Approximately 60% of the institutions were academic, 20% private not-for-profit, 12% private forprofit laboratories, and 7% classified as "other." The overall frequency of significant problems was 33% in the questionnaire-surveyed laboratories and 38% in the on-site inspected laboratories. At this reported rate, the frequency of errors in genetic testing is somewhat comparable to that in other areas within the laboratory and in medicine in general. A majority of significant problems occurred in the pretest phase, with inappropriate test selection being the most common in the category. The authors concluded that significant problems during genetic testing occur infrequently, and problems resulting in significant harm to the patient are relatively rare.Genetic tests can carry more inherent...