BACKGROUND: In the UK the replacement of long-term in-patient care with community-based support has been part of central government health policy for many years. One of the challenges of implementing such a policy is the prediction of support and service needs in the community and the associated costs. METHOD: Using research data from north London analyses were undertaken to examine the associations between service use and costs in the community and the characteristics of hospital in-patients. RESULTS: Although clinical diagnosis was not a useful predictor of either service utilisation or costs, more than a third of the variation in community care costs could be explained by symptoms, behaviour and personal characteristics at least one year earlier.
BACKGROUND: In the UK the replacement of long-term in-patient care with community-based support has been part of central government health policy for many years. One of the challenges of implementing such a policy is the prediction of support and service needs in the community and the associated costs. METHOD: Using research data from north London analyses were undertaken to examine the associations between service use and costs in the community and the characteristics of hospital in-patients. RESULTS: Although clinical diagnosis was not a useful predictor of either service utilisation or costs, more than a third of the variation in community care costs could be explained by symptoms, behaviour and personal characteristics at least one year earlier.