Markers for Patients with Alcohol Problems in an Outpatient General Medicine Clinic

Our purpose was to determine the correlation among alcohol problems, certain health habits, and hypertension in a general internal medicine outpatient setting. Two-hundred eighty-seven patients who were undergoing complete examinations in a general internal medicine department of a large multispecialty clinic were asked to participate in a "lifestyle survey" interview, which included questions regarding social history, gastrointestinal complaints, anxiety/depression, sexual dysfunction, and the Alcohol Use Disorders Identification Test (AUDIT). Charts were screened for documentation of alcohol problems and the presence of hypertension. Correlations between alcohol problems (defined as an AUDIT score >/=8) and the presence of hypertension were made using chi-square tests for categorical variables and the t test for continuous variables. Of the 287 patients contacted, 12 refused, 3 were ineligible, and 29 had incomplete data precluding analysis, yielding a total of 233 patients (81%) who were included in the analysis. Sixty percent were female, 76% were white, and the mean age was 53 years. The overall prevalence of alcohol problems was 7.3% and less than one-quarter of those patients had chart documentation of an alcohol problem. Thirty percent of those patients had a diagnosis of hypertension. No statistically significant correlation between a positive AUDIT and hypertension was found. Failure to use seat belts (P = 0.020), history of smoking (P < 0.001), alcohol use within 24 hr of the office visit (P < 0.001), and self-reported family history of alcoholism (P = 0.012) were significantly more likely in patients with active alcohol problems. There was no difference in somatic complaints between patients with and patients without alcohol problems. We identified certain health behaviors which served as markers for patients with active alcohol problems in an outpatient setting. Physician diagnosis of alcohol problems (sensitivity of 24% in this study) may be improved with attention to such markers. Despite the known relationship of excess alcohol use and elevations in blood pressure, we found no statistically significant correlation between these variables in our study. Symptomatic concerns of the patient appear to be too nonspecific to distinguish patients with and without alcohol problems.