Abstract

OBJECTIVE: To determine whether the frequency and pattern of use of the accident and emergency (A&E) department by individuals with diabetes is different from that of the general population. METHODS: A historical cohort of 696 individuals with diabetes from six randomly selected general practices and a non-diabetic comparison cohort matched on age, sex, and general practice were identified. The use of an urban A&E department by the two cohorts was compared for number of visits between 1984 and 1996 for injuries, diabetes related and non-diabetes related illness, proportion referred by a general practitioner, proportion arriving by ambulance, and proportion admitted. RESULTS: More visits were made by the diabetic cohort (1002 v 706, P = 0.0001); 121 visits were directly related to diabetes, including 52 for hypoglycaemia. The diabetic cohort also had more visits for medical illness unrelated to diabetes (357 v 231, P = 0.0001). The number of visits for injuries was similar (524 v 475, P = 0.3). Individuals with diabetes who attended A&E were not significantly more likely to be referred by a general practitioner (14% v 16%) or admitted (20% v 17%). CONCLUSIONS: Individuals with diabetes made more frequent visits than the general population to the A&E department. Since there was no excess of visits for injuries and the proportion requiring admission was similar, the hypothesis that they have a different threshold for attending is not supported.