Twelve months of isoniazid compared with four months of isoniazid and rifampin for persons with radiographic evidence of previous tuberculosis: an outcome and cost-effectiveness analysis.

Isoniazid taken daily for 12 mo and isoniazid and rifampin taken daily for 4 mo are both recommended options for patients with radiographic evidence of previous tuberculosis and positive tuberculin skin tests who have not had prior treatment. We compared the completion rates, number of adverse effects, and cost effectiveness of these two regimens. Patients were treated at the San Francisco Tuberculosis Clinic from 1993 through 1996. A Markov model was developed to assess impact on life expectancy and costs. One thousand twenty-two patients, with a mean age of 52 yr, and > 90% foreign born, were treated; 545 received isoniazid and 477 received isoniazid and rifampin. For isoniazid, 79.8% completed 12 mo of therapy and 4.9% had adverse effects versus 83.6% completion, 6.1% adverse effects for isoniazid and rifampin (p > 0.05 for all between-group comparisons). Both regimens increased life expectancy by 1.4-1.5 yr. Compared with isoniazid, isoniazid and rifampin produced net incremental savings of $135 per patient treated. In patients with radiographic evidence of prior tuberculosis who have not been previously treated, isoniazid for 12 mo and isoniazid and rifampin for 4 mo have similar rates of completion and adverse effects, and both increase life expectancy compared with no treatment. Isoniazid and rifampin for 4 mo is cost saving compared with isoniazid alone. This advantage was maintained even when compared with 9 mo of isoniazid, the new American Thoracic Society/Centers for Disease Control (ATS/CDC) recommendation for treatment with isoniazid alone.