A renewed interest in the disease has resulted. Tuberculosis has become a featured topic in many continuing medical education programs, conferences, and grand rounds. Younger physicians have been reviewing the literature of yesterday for insights into the epidemiology, infectiousness, pathogenesis, control, and treatment of tuberculosis. Indeed, in these archives there is much to be relearned, but in the disease itself, there is much yet to be discovered.

Three articles about this ancient plague in this issue demand the attention of the reader [1-3]. As an introduction, it should be recalled that in 1985, although confirming that drug-resistant mycobacteria were just as contagious as drug-sensitive organisms, Snider and colleagues [4] found that there was enormous variability in the infectivity of individual patients with clinically similar disease status. These investigators noted that infectiousness tended to be an "all or nothing" phenomenon, even after their results were adjusted for sputum smear status (a "positive" acid-fast smear indicates that the patient's sputum carries many tuberculosis bacilli and therefore that the patient is a significant contagious threat to others) [5]. They concluded that a threshold exists for infectiousness in the air that, once exceeded, results in the transmission of the organism to others.

Two of the articles in this issue focus on health care workers and address this risk for tuberculosis transmission [1, 2]. They present strong evidence to support the long-held belief that new exposure poses little risk to those previously infected; past infection is documented by a previous positive tuberculin skin test result. This confirms once again the value of knowing one's tuberculin reaction status.

Cutting through the myriad end points that have been used to establish a positive tuberculin reaction, Stead [1] compares five tuberculosis outbreaks from past eras with six outbreaks from the present. New infection rates range from 16% to 67% after exposure. Although most of these exposures extended over time (lasting 2 weeks to months), an infection rate of 31% was seen after several persons were exposed to Mycobacterium tuberculosis during one bronchoscopy. In the second paper on tuberculosis in this issue, Templeton and colleagues [2] further emphasize this point; they report a new infection rate of 100% among persons exposed to M. tuberculosis during an autopsy. Even more important are the high rates of clinical disease that rapidly followed these new infections: 14% to 59%. Of particular note, two of the five persons (40%) whose tuberculin test results converted after the autopsy exposure progressed to culture positivity.

Confirming the findings of decades past, Stead clearly shows the therapeutic benefit of an appropriate course of preventive isoniazid treatment. He documents that 19% of untreated nursing home residents who became tuberculin-positive progressed to active disease, whereas no disease developed in a similar group who received preventive therapy.

It is hoped that Stead's results will put to final rest the confusion raised by the long debate over the risk for hepatitis versus the benefit of preventive isoniazid therapy. This debate has centered on the recommendation that patients whose only known risk factor for tuberculosis is a positive tuberculin test result, but who are not known to be recent converters, be offered a course of preventive isoniazid therapy. The likelihood of disease reactivation in this low-risk group ranges only from 30 to 80 per 100 000 patient-years [6].

Obviously, the more years the patient is likely to live, the greater the benefit of preventive therapy. However, because the likelihood of subsequent development of clinically active tuberculosis is relatively small, the cut-off age at which the risk for hepatitis was thought to exceed the benefit of preventive therapy for this low-risk group was set at 35 years.

To the dismay of knowledgeable tuberculosis experts, physicians often apply this age-related decision rule to high-risk patients. This occurred in the recent converters described by Stead. There has never been any question that these and other high-risk tuberculin-positive groups should be considered for preventive isoniazid treatment at any age. In addition to recent converters, high-risk patients include tuberculin-positive patients with human immunodeficiency virus infection or other chronic medical conditions that might impair the patient's immune system. It is hoped these two papers regenerate an awareness of the value of preventive antituberculous therapy.

It is most appropriate that both of these articles include William W. Stead as an author. His career in tuberculosis spans its prechemotherapy era, the early treatment trials that led to our understanding of the principles underlying effective therapy, the development of short-course chemotherapy, the premature decline in government commitment and funding support for tuberculosis control programs, and finally the resultant re-emergence of the disease. Stead's unending commitment to the battle against tuberculosis is documented by a seemingly endless series of papers of the highest caliber: epidemiologic investigations, treatment trials, and observations—all of which substantially contributed to the advancement of our knowledge. He recently brought national attention to the problem of tuberculosis within both nursing homes and correctional institutions. In addition, only a professional of his stature and known concern for the poor and disadvantaged could report differences in genetic susceptibility to tuberculous infection without fear of misinterpretation.

It is also fitting that the third of the papers on tuberculosis in this issue [3] is by three physicians, who, like William Stead, belong to that small group of tuberculosis experts who have committed their entire professional lives to the fight against the disease. They now look to the future.

Thirty years after high-dose intermittent tuberculosis treatment was first initiated in the United States, directly administered therapy has finally become the standard of treatment of clinically active tuberculosis. Physicians deviating from this standard put their patients and society at risk, should the patient fail to comply with a self-administered treatment regimen. Indeed, the increase in multidrug-resistant tuberculosis can be largely laid at the feet of physicians and public health professionals who failed to recognize or to respond to the danger associated with patient nonadherence to therapy.

However, there are some circumstances in which directly observed therapy is not possible. For these situations, fixed-dose combination medications, composed of two or three antituberculous medications, provide a realistic and welcome option that minimizes the opportunity for a patient to selectively take only a single medication. In this issue, Moulding and colleagues [3] carefully address the potential problems associated with such fixed combinations but do not emphasize that all of a day's medication should be taken at the same time to achieve the highest peak serum level. The greater the ratio between the peak serum level and the minimum inhibitory concentration of the drug, the greater is the drug's bacterial effect [7].

Fixed-dose medications bode well for tomorrow. Fortunately for all concerned, they are already available today.