A controlled trial of outpatient geriatric evaluation and management

To evaluate the effects of targeted outpatient geriatric evaluation and management (GEM).

Controlled clinical trial. Elderly persons were identified as being at high risk for hospital admission on the basis of their responses to a short, mailed, self-administered questionnaire. The high-risk elders who chose to participate in the GEM program were compared to those who continued to receive usual care.

Outpatient GEM clinic at an urban university hospital.

Elderly Medicaid recipients whose probability of repeated hospital admission (Pra) within 4 years was calculated as 40% or greater (Pra > or = 0.40).

Mortality, use of institutional services, satisfaction.

Of the 1210 persons who were sent questionnaires, 624 responded (response rate = 51.6%), of whom 154 (24.7%) were deemed to be a high risk for hospitalization. Of these, 43 received GEM (experimental subjects); 111 received usual care (controls). At baseline, the experimental and control groups' demographic and health-related characteristics did not differ significantly. The average experimental subject was 76.5 years old, had 9.6 significant medical problems, and took 6.7 significant long-term prescription medications. During the program's first 17 months of follow-up, the experimental subjects had lower annual rates of mortality (2.9% vs 19.2%, P = 0.03) and emergency room use (0.6 vs 1.0 visits, P = 0.01) than did the controls. The experimental subjects also tended to use nursing homes, but not hospitals, at a lower rate than the controls. All of the experimental subjects rated the program as either excellent (81.0%) or good (19.0%); 100% said they would recommend it to others. Their established primary physicians rated the GEM services as appropriate and helpful.

Targeted outpatient GEM was associated with reduced mortality, reduced use of emergency rooms, and a trend toward reduced use of nursing homes.