The specific aims of the proposed study are to elucidate factors which explain the observed decline in total mortality and cause- specific mortality among elderly persons in a well-defined population between 1967 and 1981 and to ascertain whether there has been an increase or decrease in medical care utilization associated with mortality decline. The relative contributions of the following factors to the decline in mortality will be assessed: a) decline in incidence of stroke and pneumonia, common underlying or contributing cause of death among the elderly; b) decline in case fatality rates for stroke and pneumonia due to less frequent or severe occurrence of life- threatening complications, co-morbidities or risk factors. These research questions are of critical importance for predicting the future health status and health services needs of the rapidly increasing numbers of elderly in the population. The study will consist of retrospective analyses of secular trends of age-specific rates of mortality, morbidity and medical care utilization among elderly Kaiser Permanente members during the period 1967-1981 in the Northwest Region. Study data will be from existing computerized membership, hospital discharge and outpatient files, and from medical record abstraction and follow-up. Poisson regression and log-linear models for extra-Poisson variation will be used to test for period effects in the mortality data. The proposed study involves the collaboration of two established investigators a medical epidemiologist and a biostatistian--who have successfully completed several federally funded studies using the Kaiser Permanente data systems and medical records.
