Illness and hospitalization often trigger functional decline among older persons. Home care services implemented for functional decline provide an opportunity to intervene to improve outcomes. To compare functional status and the likelihood of remaining at home for persons receiving restorative care vs usual home care. Intervention using prospective individual matching conducted between November 1, 1998, and April 30, 2000. Six offices of a home care agency in Connecticut were used. One branch office served as the restorative care unit and the other 5 served as usual care offices. Patients receiving home care through the restorative care office who were 65 years or older; in receipt of Medicare-covered home care lasting at least 7 days; with absence of severe cognitive impairment; and not terminal, bedridden, or requiring total care were matched with patients from 1 of the usual care offices. The matching factors included age, sex, race, baseline self-care function, cognitive status, whether hospitalization preceded the home care episode, and date of the home care episode. Of the 712 eligible restorative care patients, 691 (97%) were matched with a usual care patient. Restorative care, provided by the home care agency nursing, therapy, and home health aide staff, was based on principles from geriatric medicine, nursing, rehabilitation, and goal attainment. Remaining at home, functional status at completion of the home care episode, and duration and intensity of home care episode. Compared with usual care, and after adjusting for baseline characteristics and other factors, restorative care was associated with a greater likelihood of remaining at home (82% vs 71%; odds ratio [OR], 1.99; 95% confidence interval [CI], 1.47-2.69) and a reduced likelihood of visiting an emergency department (10% vs 20%; OR, 0.44; 95% CI, 0.32-0.61). Home care episodes were shorter (mean [SD], 24.8 [26.8] days vs 34.3 [44.2] days; S = -17 821; P<.001). Restorative care patients had better mean (SD) scores than usual care patients in self-care (11.0 [2.1] vs 10.7 [2.5]; P =.07 after adjustment), home management (9.5 [2.9] vs 9.2 [3.0]; P =.05 after adjustment), and mobility (3.3 [0.8] vs 3.2 [0.9]; P =.02 after adjustment). This trial suggests that reorganizing the structure and goals of home care can enhance health outcomes of older patients without increasing health care utilization.