No evidence for multiple-drug prophylaxis for tuberculosis compared with isoniazid alone in southeast Asian refugees and migrants: Completion and compliance are major determinants of effectiveness

Background. The use of multiple-drug prophylaxis for tuberculosis (TB) has not been shown to be more effective than prophylaxis with isoniazid alone. The boundary between inactive pulmonary TB (class 4 TB) and culture-negativ e "active" pulmonary TB (class 3 TB) is often unclear, as is the intention to treat such patients as a preventive measure or as a curative measure. Methods. We compared the effectiveness of single drug preventive therapy wi th isoniazid to the effectiveness of multiple drug preventive therapy for p atients with asymptomatic, inactive TB, in a retrospective cohort study of 984 Southeast (SE) Asian migrants and refugees who received prophylaxis bet ween 1978 and 1980. Results. The rate of TB developing in this cohort was 122 per 100,000 perso n-years. There was no significant difference in development of TB between p eople who received isoniazid only and those who received multiple drugs. Th e only significant predictor of TB was noncompletion of prophylaxis [relati ve risk (RR) = 62, 95% confidence interval (CI) = 20-194]. Subgroup analysi s on people who had completed therapy showed noncompliance as a significant predictor of TB (RR = 16, 95% CI = 1.4-179). The risk of noncompletion (RR = 4.7, 95% CI = 2.37-9.39, P < 0.0001) and noncompliance (RR = 2.2, 95% CI = 1.03-4.7, P = 0.03) was higher for patients who received multiple drugs compared with isoniazid alone. Multiple-drug therapy cost 30 times more tha n isoniazid alone. Conclusions. We did not find evidence in support of the empirical practice of giving multiple drugs for prevention of TB. This practice is also more c ostly and more likely to result in noncompliance and adverse drug reactions . (C) 2000 American Health Foundation and Academic Press.