Use of physician and acute care services by persons with and without Alzheimer's disease: a population-based...

To estimate differences in use of acute care services between persons with and without Alzheimer's disease (AD).

Population-based historical cohort study.

All Rochester, Minnesota, residents with AD onset between January 1, 1980, and December 31, 1984 (n = 301), plus one age- and sex-matched nondemented control per case, were identified with a retrospective review of community-based medical records.

Cases and controls were followed in their medical records for number of acute care encounters in the year before January 1 of the index year (year of onset for AD case and their matched control) and in the 4 years following December 31 of the index year. Encounters included clinician visits (office or nursing home), emergency room (ER) visits, hospitalizations (inpatient and outpatient), and inpatient days. Multivariate regression analyses were adjusted for age, sex, pre-index level of illness, and follow-up time.

In the pre-index period, cases and controls were similar with respect to level of illness, number of office visits, ER visits, and hospitalizations. In the year before AD onset, 17 cases (7%) had a clinician visit in the nursing home compared with no controls. In the 4 years after the index year, mean length of follow-up was 3.4 years for both cases and controls. The numbers of ER visits, hospitalizations, and inpatient days were similar for cases and controls. Sixty-four percent of AD cases had a clinician visit in a nursing home versus 1% of controls. Controls experienced more office visits than cases (median = 16 vs 10, P < .001).

The onset of AD is not associated with greater use of acute care services. However, neither is the high use of nursing home care offset by fewer ER or hospital encounters.