Recent technologic developments have introduced a number of improvements in the ability of clinical laboratories to cultivate and identify Mycobacterium tuberculosis complex more quickly than previously. These developments include more rapid detection of growth and tests to identify RNA or DNA of M. tuberculosis complex directly in clinical samples. United States Food and Drug Administration (FDA) panels have recently recommended approval of two direct amplification tests (DAT), the Gen-Probe MTD (San Diego, CA) and the AMPLICOR M. tuberculosis test (Roche Diagnostic Systems, Inc., Branchburg, NJ). The FDA has approved the MTD for identification of M. tuberculosis complex in respiratory specimens that are smear-positive for acid-fast bacilli (AFB). In addition, the specimen must be from a patient who has not received antituberculous medication for seven or more days or within the last 12 months. From the data reviewed by the FDA, the specificity (100%) and sensitivity (95%/96% in the two studies) of these two tests in AFB smear-positive specimens were found to be comparable to the Accuprobe (Gen-Probe) for identification of M. tuberculosis complex in culture, with the advantage that the DAT results are available much sooner. The DAT are significantly more sensitive than the AFB smear. However, in AFB smear-negative samples, the specificity, sensitivity, and positive predictive value were 96/99%, 48/53% and 24/58%, respectively, in the two studies. For some results, the Gen-Probe assay had the higher value and for others, the Roche assay was higher. The DAT result, particularly when discordant with the AFB smear, must be used in conjunction with clinical assessment. While both the MTD and the AMPLICOR M. tuberculosis test have undergone extensive testing in clinical laboratories, neither test has been examined for its utility in routine clinical use or public health settings in the United States. An American Thoracic Society Workshop was convened to examine the data and technology available to date, to develop a consensus addressing the appropriate use of the rapid diagnostic tests (in particular, DAT's) for tuberculosis, and to identify future research needs and directions. The consensus among three focus groups, clinical, laboratory, and public health, was that, while these tests are a major improvement over standard techniques, there is currently insufficient information on their clinical and public health utility. When the AFB smear and DAT are both positive, the diagnosis of tuberculosis can be considered to be established. Furthermore, when the AFB smear is negative and the DAT is also negative, it is unlikely that M. tuberculosis will be grown from that sample. When there is discordance between the AFB smear and the DAT, additional consideration must be given to the overall clinical picture and repeat testing should be done. It is recommended that the currently available DAT's always be performed in conjunction with microscopy and culture, and each test result must be interpreted within the overall clinical setting in which it is used.