Studies carried out specifically in the general hospital setting suggest that brief interventions can lead to significant reductions in alcohol use or related problems, or both.¹ Reported barriers to the implementation of brief intervention centre upon a failure to identify hazardous drinkers.²

The work of Huntley et al³ showed that selective screening of just 10 presenting conditions could account for up to 77% of hazardous drinkers, who could then be offered access to interventions designed to reduce their consumption. No matter how effective an intervention may be, it is reliant upon the willingness of a patient to accept it. The identification of underlying characteristics associated with the rejection of brief interventions could help clinicians to shape the way in which screening results are presented to patients, and thus increase the proportion of hazardous drinkers willing to accept help. We have found that by emphasising the potential connection between the hazardous pattern of alcohol consumption and the possible future consequences to health, patients are more likely to accept the offer of an intervention.⁴

PARTICIPANTS/METHODS/RESULTS

As part of a wider randomised controlled trial investigating screening and brief intervention in the A&E department of a large inner London teaching hospital, 1120 non-treatment seeking hazardous drinkers (aged 18+) were identified using the Paddington Alcohol Test⁵ over a one year period (21.4% of all patients screened). Of these 717 accepted an intervention (64.0%). Screening occurred as per routine practice,^3,4 with senior house officers first dealing with the patients’ primary presenting condition, and then asking questions about their drinking. Male patients consuming eight or more units of alcohol on any single occasion at least once per week, or female patients consuming six or more units, or any patient who declared that their visit to the A&E department was alcohol related were recognised as hazardous drinkers, were told that their level of alcohol consumption may be harmful, and offered help. Patients who did not accept the offer of help were provided with contact details of local alcohol agencies, should they reconsider.

Tables 1A and 1B shows differences in proportions among the demographics and presenting circumstances of patients who did or did not accept our offer of help. Binary logistic regression showed that a patient’s presenting condition, together with the total number of units consumed on a single occasion, predicts uptake of the intervention. There was a significant positive association between presentation to the A&E department complaining of abdominal pain/discomfort and accepting advice (77.7%). A significant negative association was detected for head injury (52.9%), accident (55.4%), or after a fall (59.8%).

COMMENT

A patient’s age or sex alone does not predict the uptake of an offer of brief intervention. However, presenting condition and current level of alcohol consumption are associated with the acceptance of an offer of help.

Underlying reasons for the negative association between falls, head injuries, accidents, and the acceptance of alcohol related advice are unclear. These patients display insight as to the relation between their injuries and alcohol, yet they remain resistant to advice. One possible explanation might be that such events such as falls and accidents are believed to be attributable in part to misfortune or other factors over which the patient has no control.

Most patients identified as hazardous drinkers (64%) accepted the offer of advice, however all hazardous drinkers may benefit from an intervention that aims to reduce their alcohol consumption. It may be possible to increase the proportion of patients who present after a fall, head injury, or accident that accept advice by further emphasising the relation between their drinking and possible health consequences, as well as the potential link between alcohol and their presentation to the A&E department.