The significance of random breath alcohol sampling in the accident and emergency department.

Using a detailed questionnaire, 126 patients attending an accident and emergency department were screened for drinking problems. Eleven per cent (14) were identified as having established drinking problems, but only two of these had received help from professional agencies. In all, 39% (50) were identified as being adversely affected by their drinking habits. A breath alcohol test (BT) alone, on the same patients failed as a screening device for hidden drinking problems in these circumstances, and we do not recommend its use. The simple 'CAGE' questionnaire was a little more sensitive, but asking more questions identified more problems. A positive BT in the presence of a positive CAGE occurred in three patients and although insensitive it was absolutely specific for a serious drinking problem. A significant number of patients who attend an accident and emergency department have a drinking problem. The most effective method of detecting this is to ask patients about their drinking habits.