Extreme statements regarding the prevalence of problems with alcohol and other drugs among physicians have often been made without firm empirical support. The data that are available can be categorized as follows: license and disciplinary actions, known or registered addicts, mortality rates, hospital admissions and treatment populations, and surveys of selected groups of physicians. Because of difficulties in interpreting each type of data, it must be concluded that the prevalence of drug problems among physicians is unknown. A previously unpublished survey of Ontario physicians found that the proportion who have been treated for problems with alcohol and other drugs is not greater than that in the general population. Indeed, when alcohol and other drugs are considered together, physicians may not be unusually likely to have such problems.