Abstract
Background: Diabetes mellitus (DM) may be a risk factor for frailty and individuals with DM are more likely to be frail than individuals without DM, however, there is limited research in hospitalized older adults. Objectives: To determine the prevalence and impact of frailty in hospitalized older adults with and without DM using a Frailty Risk Score (FRS) and assess the role of frailty in predicting 30-day rehospitalization, discharge to an institution, and in-hospital mortality. Methods: The study was a retrospective, correlational design and secondary analysis of electronic health record data for a sample of hospitalized older adults. Logistic regression was performed for 30-day rehospitalization and discharge location. Cox proportional hazards regression was used to analyze time to in-hospital death and weighted using propensity scores. Results: Of 278 hospitalized older adults, 49% had DM, and the mean FRS was not significantly different by DM status (9.6 versus 9.1, p = .07). Increased FRS was associated with significantly increased odds of rehospitalization (AOR = 1.24, p = .04) and increased odds of discharge to an institution (AOR = 1.48, p < .001). The FRS was not significantly associated with increased risk of in-hospital death (p = .17), but DM was associated with a 484% increase in the instantaneous risk of death (AHR=5.84, p = .005). Conclusion: Both DM and frailty were prevalent, and frailty was significantly associated with rehospitalization and discharge to an institution. Frailty assessment may augment conventional clinical assessment and facilitate tailoring more intensive care for frail patients with DM.