The effectiveness of isoniazid (INH) monotherapy for the treatment of Mycobacterium tuberculosis infection to prevent active tuberculosis (TB) was first seen in a U.S. Public Health Service trial in 1962 (1). After that sentinel trial, over 20 randomized trials with more than 100 000 participants have shown a 25% to 90% reduction in active TB with 6 to 12 months of INH therapy compared with no treatment or placebo (1). In 1970, the regimen with optimal efficacy, 12 months of INH, was recommended as standard therapy for latent TB infection by the American Thoracic Society (2). Subsequently, a large Eastern European trial of older persons with fibrotic lung lesions showed that 6 months of INH had similar effectiveness to 12 months of INH because of better treatment completion rates (3). As a result, 6 months of INH was recommended by many authoritative agencies, including the World Health Organization. When a post hoc analysis of randomized trials in the Eskimo population of Alaska showed that 9 months of INH offered optimal effectiveness, this regimen became the standard in North America (4). However, completion rates under program conditions are very low for both 6 and 9 months of INH, and INH-associated hepatotoxicity, which can be fatal, is a major concern (4).