Social isolation predicts re-hospitalization in a group of older American veterans enrolled in the UPBEAT Program

Objectives Does social isolation predict re‐hospitalization in a group of older men enrolled in Unified Psychogeriatric Biopsychosocial Evaluation and Treatment (UPBEAT), a mental health care‐coordination project at nine Veterans Affairs Healthcare Centers nationwide? Methods The current study examined 123 UPBEAT patients located at West Los Angeles, whose ratings were available on the Lubben Social Network Scale (LSNS), the SF‐36 scale, the Cumulative Illness Rating Scale (CIRS) and the Mental Health Index (MHI‐38) Depression and Anxiety subscales. Within one year of enrollment, 55% of patients were re‐hospitalized. Odds of re‐hospitalization were calculated using two logistic regression models. Social isolation risk (LSNS) and demographic covariates were included. In addition, Model 1 contained depression and anxiety measures (MHI‐38) and physician‐rated medical burden (CIRS), while in Model 2, patient‐perceived physical (PCS) and mental health (MCS) subscales from the SF‐36 were included. Results The group of patients who were socially isolated or at high or moderate risk for isolation, were 4–5 times more likely to be re‐hospitalized within the year, than low isolation risk patients. In both Models 1 (chi‐square = 19.86; p = 0.031) and 2 (chi‐square = 26.42; p = 0.002) demographic characteristics were not significant predictors of re‐hospitalization, but social isolation risk was a significant predictor (Model 1: odds ratio (OR) = 5.31; 95% confidence intervals (CI) = 1.81–15.53; and Model 2: OR = 3.86; 95% CI = 1.39–10.73). In addition, MHI‐Anxiety was a significant predictor (OR = 1.22; 95% CI = 1.05–1.43) in Model 1 and in Model 2, patient‐perceived physical health significantly predicted re‐hospitalization (OR = 0.91; 95% CI = 0.86–0.96). Conclusion When controlling for other covariates, social isolation, physical health and mental health were significant risk factors for re‐hospitalization. These findings underline the importance of assessing and addressing lack of social support, along with other factors, in the health care of older male veterans. Copyright © 2001 John Wiley & Sons, Ltd.