Objectives: To examine the associations of self-reported alcohol consumption in a community based sample of elderly subjects.

Methods: A total of 518 non-institutionalised community dwelling elderly identified from general practice registers were interviewed in their own homes using the Geriatric Mental State (GMS-AGECAT), the Mini-Mental State Examination and the sociodemographic questionnaire. Physical health was rated on a six-point scale. Self reported alcohol consumption was recorded in units per week. The group was then divided according to the presence or absence of excessive alcohol consumption (ie. consumption of over 14 units per week for females and 21 units per week for males). Results were analysed using multivariate regression analysis with excessive alcohol consumption as the dependent variable.

Results: The mean age of the sample was 73 (range 65-95), 63% (n = 329) were female. Thirty-six per cent of the population were abstinent and 7% reported excessive alcohol consumption. Analysis of the data revealed no association between excessive alcohol consumption and diagnosis, age, cognitive function or poor physical health. Excess consumption was found to be significantly associated with gender (male) and widowed status (p < 0.001, p = 0.013 respectively).

Conclusions: As alcohol misuse is commonly missed in the elderly identifying high risk groups is important for the development of intervention strategies. Our results suggest that elderly widowers may be more at risk than their peers of alcohol misuse