| The ability of Geriatric Medicine inpatient services to modify mortality rates is well established in studies of duration of 1 year or less. Experience in real world effectiveness is lacking. A cohort model was used to explore mortality among all Manitobans aged 65 or more by March 31, 1990, who used hospital-based health care services in the interval April 1, 1988 through March 31, 1990. Using the Manitoba Health data set a prospective analysis of service utilization for the index contact was compared between users and non-users of Geriatric Medicine inpatient services in Winnipeg Manitoba. This resulted in a sample size of 23,568 people. The index contact was defined as the first inpatient stay on a Geriatric unit during the study interval. Where such did not exist, the initial inpatient admission was used. Survival analysis demonstrated that mortality rates are higher for users of Geriatric Medicine inpatient services, having accounted for age at admission, income decile, interaction of marital status and gender, Charlson index of comorbidity, marital status and gender. Most of the effect is due to a difference in early survival times. These results expand upon the position statements of the American Geriatrics Society that Geriatric services decreased mortality rates among older people involved in Randomized Controlled Trials. |
