Cognitive impairment and length of hospital stay in older persons

We looked at performance on the Folstein Mini-Mental State Exam (MMSE) as a predictor of hospitalization and length of stay in the coming year in community-dwelling older persons from the National Institute of Mental Health Epidemiologic Catchment Area program. They had been assessed with the MMSE at the outset and were re-evaluated with the MMSE and a Health Services Questionnaire 1 year later. Subjects were more likely to be hospitalized in the subsequent year if they were older than 75 years or if they scored poorly on the MMSE. Severe cognitive impairment increased the risk more than mild impairment. Multivariate analyses that controlled for demographic variables demonstrate that MMSE performance is a significant predictor of any subsequent hospitalization (medical or psychiatric) among whites and among those at both high and low educational levels. This effect was not explained solely by the increased rate of psychiatric hospitalizations. We also determined that a decline in MMSE score over 1 year was associated with an increased risk of hospitalization, more hospital days, longer average length of stay, and a prolonged (greater than 20 days) hospital stay. We conclude that both initial poor performance on the MMSE and deterioration increase the risk of hospital use and lead to more extended hospital stays. However, even with poor MMSE performance, most older persons remain out of the hospital and most of those hospitalized do not have prolonged stays; thus, MMSE score alone is insufficient as a predictor of impending hospitalization. Further studies are needed to add other measures of risk for hospitalization and prolonged hospital stays.