Benefit-risk considerations in preventive treatment for tuberculosis in elderly persons.

Of 2135 elderly residents of nursing homes in Arkansas (mean age, 79.4 years) who have been treated with isoniazid for prevention of tuberculosis, data from 1935 were suitable for analysis. About 12 months of therapy was successfully completed in 1600 persons. Therapy could not be completed in 84 persons (4.4%) because of incipient hepatic toxicity and in 116 (6.0%) because of other types of drug intolerance. Although 135 persons (7.0%) died during the course of therapy, no evidence was found that isoniazid contributed to any death. The ratio of benefit (reduction of risk for tuberculosis) to risk (for nonfatal isoniazid-related hepatitis) was clearly favorable in persons who had definite conversions (1.6 for women, 3.4 for men) but less so for persons who had tuberculin reactions of unknown duration and for persons with minor increases in size of tuberculin reaction (less than 12 mm increase from an initially negative reaction).