First steps in building ACME--an admission case-mix system for the elderly.

In the United Kingdom, specialists in Geriatric Medicine usually have a major role in treating acute medical problems of elderly people, in addition to running rehabilitation services and continuing care. The proportion of the different types of patients varies widely from service to service, however, making it difficult for clinicians to compare their performance with that of colleagues. Existing casemix measurement systems are usually designed to deal with a more homogeneous patient group (e.g. rehabilitation, long-stay care) and/or are too detailed for day-to-day use. We describe our attempts to devise a simple casemix system which would be of practical day-to-day use for individual specialists in Geriatric Medicine. We have classified patients according to (1) the acuteness and potential for recovery of their presenting illness and (2) their functional status (based on simple measures of mobility and cognitive impairment). These factors have been incorporated into a three-point score, CMIX, which was capable of explaining 19.5% of the variability in duration of stay in a prospective study of 400 new admissions in two centres. In contrast, age and sex explained only 1% of variability in these patients. The pattern of patient outcome also differed significantly between the three CMIX categories. We also propose a simple graphical method of classifying outcome which should prove useful for audit purposes even when our casemix system is not employed.