Can Doctors Be Divided Into Groups Based on Their Attitudes to Addressing Alcohol Issues in Their Patients? Analyses From a Survey of Portuguese General Practitioners.

Background: The effectiveness of interventions to increase general practitioners' management of alcohol problems is affected by their attitudes toward at-risk drinkers. Tailoring training programs to general practitioners' attitudes may be useful in increasing alcohol screening and brief advice.

Objectives: to determine whether general practitioners could be divided into distinct groups based on their attitudes toward at-risk drinkers. Another aim of this study was to develop and validate a model for classifying general practitioners into distinct groups.

Methods: A total of 234 general practitioners answered the questionnaire. Physicians' attitudes toward patients with hazardous or harmful alcohol use were measured with the Short Alcohol and Alcohol Problems Perception Questionnaire. Cluster analysis was performed to identify distinct general practitioner groups based on their attitudes toward at-risk drinkers. Logistic regression analysis was used to develop a model for predicting group membership.

Results: Cluster analysis identified two distinct groups of general practitioners, one with more positive attitudes (adequacy = 10.8 ± 1.6, legitimacy = 11.8 ± 1.7, motivation = 9.8 ± 1.7, satisfaction = 8.1 ± 1.9, and self-esteem = 9.7 ± 2.1), the other with more negative attitudes (adequacy = 8.9 ± 1.8, legitimacy = 11.0 ± 1.8, motivation = 7.8 ± 1.6, satisfaction = 5.7 ± 2.0, and self-esteem = 6.8 ± 1.7). The predictors in the final model were self-esteem, motivation, and adequacy. The model predicted general practitioner groups on the training set with 90.4% accuracy (area under receiver operating characteristic [ROC] curve = 0.96), and maintained its predictive performance when applied to the test set (accuracy 93.6%, area under ROC curve = 0.97).

Conclusions: General practitioners could be divided into distinct groups based on their attitudes toward at-risk drinkers. These findings may prove useful in designing alcohol-specific training programs for general practitioners.