© 2004 American Thoracic Society doi: 10.1164/rccm.2406009
How Much Directly Observed Therapy Is Enough?Denver Metro Tuberculosis Clinic, Denver Public Health, and the Department of Medicine (Division of Infectious Disease) University of Colorado Health Sciences Center Denver, Colorado In the mid-1980s tuberculosis control was converted from a quiet backwater of public health to a front-page issue in major U.S. cities. After decades of steadily falling case rates, there was a substantial increase in cases and large outbreaks of highly drug-resistant strains. Although other factors (the HIV pandemic, continued immigration from tuberculosis-endemic areas) played a role, perhaps the most important cause of the increase in case rates and drug resistance was the erosion of the public health infrastructure to assure completion of tuberculosis treatment. An influential paper marking the nadir of tuberculosis control in the United States was a study from Harlem in which 89% of persons diagnosed with tuberculosis were lost from follow-up before completing a course of therapy (1). The nation embarked on a major rebuilding of the public health infrastructure for tuberculosis that included a dramatic expansion in the use of directly observed therapy (DOT), a form of tuberculosis treatment in which the ingestion of medications is observed by a healthcare worker. The expansion in the use of DOT in the United States was followed by dramatic decreases in multidrug-resistant tuberculosis and the return of overall trends in case rates to the steady decline of previous decades. Despite these successes, controversies remain regarding the use of DOT. Some have argued that these recent successes in tuberculosis control are due to increased funding and a management strategy of focusing on outcomes of tuberculosis treatment, rather than being due to the coincident expansion of DOT. Concerns remain about the ethics of DOT; it is intrusive to require that a healthcare worker watch the ingestion of each dose of medication. A paper by Jasmer and colleagues in this issue of the Journal (pp. 561–566) provides an opportunity to review the role of DOT in U.S. tuberculosis control efforts (2). The gold standard for evidence-based public health is the randomized clinical trial. By this standard, the evidence base for DOT is somewhat tarnished; two randomized clinical trials in high-burden countries showed no benefit over self-administered therapy (3, 4), whereas a third trial showed a higher treatment completion rate among persons randomized to DOT (5). There are three important issues relating to these previously published trials. First, the trials were done in clinical settings very different from urban America. Second, the randomized trials did not include standardized corrective responses to nonadherence. In particular, timely corrective responses and the use of incentives and enablers to promote adherence (enhanced DOT) are associated with high rates of successful treatment completion. Finally, enhanced DOT is not a standard approach that can be easily tested in a randomized trial, because there are likely to be differences in the ability to institute timely interventions to maintain adherence. For U.S. tuberculosis control programs, the study by Jasmer and colleagues (2) highlights the effectiveness of DOT in the management of the challenging patients encountered in many urban settings. Despite much higher rates of HIV coinfection, drug and alcohol abuse, and homelessness, patients treated with enhanced DOT had better outcomes than patients treated with self-administered treatment. It is important to note that this was an observational study, and there are a number of differences between the two patient populations that may affect comparisons. The higher death rate among patients treated with self-administered therapy may have been due, in part, to the older age distribution in this patient population. However, it is instructive to focus just on the DOT group in this study—to have a cure rate of more than 95% and no cases of acquired drug resistance in a patient population with so many risk factors for a poor outcome of tuberculosis treatment is outstanding. This study shows that enhanced DOT given through a well-organized tuberculosis control program is remarkably effective for "high-risk" patients. Previous analyses also suggest that DOT is cost-effective among such high-risk patients, by decreasing expensive events such as treatment failure, relapse, and acquired drug resistance (6). What does this study say about patients not meeting the criteria for DOT suggested by the San Francisco program, such as presence of sputum smear positivity, drug resistance, HIV infection, substance abuse, or psychiatric illness? In a subanalysis of 126 patients, Jasmer and colleagues argue that treatment outcomes among tuberculous patients appropriately selected for self-administered therapy are equivalent to those with DOT (2). The question of whether selective use of DOT works as well as near-universal DOT is important to U.S. tuberculosis control programs. In the interest of full disclosure, we work in a tuberculosis control program that pioneered DOT and has used near-universal DOT for several decades. In an unfortunate turn of phrase, Jasmer and colleagues label the group of patients without homelessness, substance abuse, and other risk factors as "patients not at risk for nonadherence" (2), suggesting that providers can accurately predict adherence using simple clinical criteria. We think that a better description would be "patients at lower risk for complications of nonadherence," because poor adherence is common and providers are poor predictors, generally overestimating adherence (7, 8). In tuberculosis treatment, the risks of overestimating adherence are failure of treatment, relapse, acquired drug resistance, and transmission to others. The results of the study by Jasmer and colleagues show trends toward these poor outcomes being more common among patients "appropriately treated" with self-administered therapy. In addition to eliminating concerns about undetected nonadherence, use of near-universal DOT has the advantage of sending a consistent and nonjudgmental message to all patients with active tuberculosis (i.e., "We use this form of treatment for all patients"). In our view, all patients with pulmonary tuberculosis should be started on DOT, with perhaps limited use of self-administered therapy after the critical first 2 months of treatment. A final point raised by this study is the important role of the tuberculosis control program in ensuring the appropriate treatment of tuberculosis by medical care providers outside the health department. Similar to the results of other recent observational studies (9–11), tuberculosis treatment by providers outside of the tuberculosis control program was associated with inferior treatment outcomes (2). Jasmer and colleagues call for programs to educate private providers about the importance of DOT, but educational programs have little effect on provider behavior (12). We urge public health officials to consider the regulation adopted by the state of Colorado requiring DOT for all patients with respiratory tract tuberculosis unless a waiver is granted by the tuberculosis control program (13). DOT works, and we should assure that its benefits are extended to all patients and their communities. FOOTNOTES Conflict of Interest Statement: W.J.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; R.R.R. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
