OBJECTIVE: To measure any difference in the utilization of hospital resources between alcoholic patients and nonalcoholic patients (controls) in a department of internal medicine. DESIGN: Prospective comparative study. Alcoholics were identified as patients with Michigan Alcoholism Screening Test (MAST) scores of greater than or equal to 8. Controls were defined as patients with MAST scores of less than or equal to 4, and matched with alcoholics for sex, age, and time of admission. The length of stay, as well as several indicators of utilization of diagnostic and therapeutic procedures, was used for the comparison of resource utilization. SETTING: General wards of internal medicine of a 1,000-bed city and teaching hospital in Lausanne, Switzerland. PARTICIPANTS: One hundred and three alcoholic patients and 103 controls aged 20-75 years, admitted from September 1, 1988, to March 18, 1989. RESULTS: Alcoholics had the same lengths of stay (16 days), durations of intravenous infusions (six days), and durations of bladder catheterization (one day). Statistically nonsignificant differences were found between alcoholics and nonalcoholics regarding the charges for routine laboratory examinations [693 vs. 734 Swiss francs (Sfrs)], antibiotic therapies (218 vs. 145 Sfrs), and x-ray procedures (568 vs. 774 Sfrs; p = 0.06). The average number of electrocardiograms (two vs. five; p less than 0.005) and the duration of intensive care unit (ICU) stay (one vs. two days; p less than 0.05) were significantly lower for alcoholics than for controls. A total hospital charges index was also lower for alcoholics than for controls (11,900 Sfrs vs. 12,800 Sfrs), but not significantly. CONCLUSION: The authors' results suggest that alcoholics do not use more hospital resources per admission than do nonalcoholics. Moreover, alcoholics tend to use less frequently some procedures, such as the ICU, electrocardiography, and x-ray examinations. Several hypotheses are developed to explain these results in relation to those of previous studies, which showed more use of medical care by alcoholics than by nonalcoholics.