The intent of this proposal is to develop a conceptually simple but clinically relevant measure of illness that is useful for predicting the extent of and need for health services. The need for such a measure is becoming increasingly evident as the aged and other populations, differing in their burdens of morbidity, become enrolled in ambulatory care organizations with prospective payment and pre-budgeting. In order to avoid penalizing facilities or organizations that contract to care for populations with higher-than-average need for services, it is necessary to develop a measure of need for services that is objective, valid, and not susceptible to manipulation either by intent or inadvertently. The proposed method is designed to require no special collection instruments or methods. The data are regularly available in clinical facilities, in the form of medical records, clinical information systems, and administrative systems used in billing. The method is particularly suited to the planning and evaluation of primary health services as it is neither targeted at, nor dependent on, the presence of specific diagnoses whose relative demand on medical resources may change over time, but instead clusters diagnoses into morbidity groupings. The approach in this research is both methodologic and analytic. Preliminary studies done in one HMO on conceptualization of morbidity will be extended to other age groups and types of populations. Subsequently, the usefulness of the conceptual scheme in predicting use of resources will be tested in several HMOs across the country and in a Medicaid population in one state. It will also be tested against other existing developmental approaches to ambulatory case mix. If successful, this research would have applicability in the development of approaches to plan and pay for services. It should also be useful to researchers and managers interested in exploring the reasons for variations in resource use for populations with similar morbidity burdens.
