Search

Objective To determine the yield of standardized targeting criteria for identifying hospitalized HMO patients who are at high risk for mortality and nursing home placement and, therefore, may be appropriate for comprehensive geriatric assessment. Design Prospective cohort study. Setting Four HMO (Southern California Kaiser-Permanente) hospitals. Subjects 3,697 patients 65 years of age or older admitted to any inpatient service from July 1 through September 30, 1990. Main Outcome Measures Screening measures for standardized targeting criteria, administered 24–48 hours after admission. Results Overall, 35.1% of patients screened met at least one inclusionary criterion, which is considerably higher than other published reports. The most commonly met criteria were incontinence, impairment in performing activities of daily living, malnutrition, and immobility. The percentage of patients meeting at least one criterion and meeting each criterion varied considerably among hospitals. The vast majority of patients (83%) who were judged appropriate for assessment met at least three criteria. Conclusions The yield of standardized criteria differs when applied to patients in different health care systems and to patients at different hospitals within a health care system, which may reflect differences in the populations being served and the practice styles of physicians and the delivery system. Because no single criterion identified more than approximately half of patients who were judged appropriate for assessment, it is likely that a battery of criteria will be necessary to identify these patients.

Objective: To describe the evaluation of an interdisciplinary comprehensive geriatric assessment (CGA) consultation program for targeted hospitalized patients. Design: Multi-center randomized clinical trial (RCT) at four hospitals where patients were randomly assigned to CGA consultation or usual care by the attending physician, and a non-equivalent control group (NCG) at two hospitals. Setting: Six hospitals in a multi-specialty group practice model health maintenance organization (HMO). Participants: 3593 patients age 65 years or older meeting at least one of 13 inclusionary criteria at admission. Intervention: Screening by hospital staff and standardized CGA consultation conducted by a nurse practitioner, social worker, and geriatrician at the four RCT hospitals. Main Outcome Measures: Functional and health status, mortality, rehospitalization, and cost-effectiveness of the CGA program at 1 year post-randomization; validation of targeting (inclusionary) criteria that identify subgroups of patients deriving benefit from CGA; and physician contamination (learning from CGA and changing treatment provided to control patients). Conclusions: A number of methodological issues need to be considered when conducting effectiveness trials of CGA. The concurrent design of a multi-center RCT, coupled with the NCG to determine physician contamination, is an innovative approach intended to determine more precisely the cost-effectiveness of CGA for frail hospitalized elderly persons. The large and heterogeneous patient population and the broad array of inclusionary criteria will permit the evaluation of the benefit of CGA for subgroups. All these features are intended to enhance the generalizability of study results.