Psychoactive Drug Use and AIDS

To the Editor.— Kaslow et al 1 reported that they found no evidence that alcohol may accelerate immunodeficiency in human immunodeficiency virus—seropositive men during an 18-month follow-up. These results are surprising in light of the wealth of literature documenting alcohol's detrimental effects on immune defense. We suggest several reasons why the authors were unable to detect a clear trend between alcohol consumption and immunocompromise despite their large number of study participants. First, their estimates of alcohol intake are based solely on the individual's report of daily alcohol consumption. Reporting of actual consumption by alcoholics is notoriously unreliable and has inspired the development of more reliable devices such as the Michigan Alcoholism Screening Test to aid clinicians in diagnosing alcoholism. 2 By relying on individual reporting, many alcoholics may minimize their drinking and be misclassified as minimal or moderate drinkers. Second, the difference in number of drinks per day in each