After an outbreak of hepatitis in Washington, D.C. in 1970 among a group of persons taking isoniazid to prevent tuberculosis, an isoniazid surveillance study was conducted among 13,838 persons in 21 participating health departments. Age appeared to be the predominant factor influencing the risk of developing isoniazid-related hepatitis, i.e., increasing age was associated with an increasing risk. Drinking alcohol, especially on a daily basis, also seemed to enhance the risk of hepatitis among persons concurrently taking isoniazid. In general, case rates among males and females of the same race, and rates among different races, were not markedly different; however, there were striking differences in the case rates among males of different races. The incidence of hepatitis varied greatly among the 21 cities, but was not unique to any geographic region, nor was it related to a specific manufacturer of isoniazid. The onset, in most cases, occurred within the first few months of treatment. Eight fatalities were reported by the 21 participating health departments, 7 occurring in one city. Black females accounted for 5 of the 8 deaths. This information provides a basis for weighing the benefits of isoniazid in preventing tuberculosis against the risk of its causing hepatitis. Close monitoring for overt signs or symptoms of hepatitis among persons receiving isoniazid preventive therapy is indicated, especially for persons greater than or equal to 35 years of age and those who drink alcoholic beverages on a daily basis.