Abstract
OBJECTIVE: The objective of this study was to examine the effects of physical activity during hospital stay on falls, emergency room visits, and rehospitalizations among older patients with dementia at 1 and 6 months post discharge. DESIGN: This is a secondary data analysis. Bidirectional analysis using panel data from a cluster randomized clinical trial was utilized. SETTING AND PARTICIPANTS: The sample comprised 455 older adults with dementia from a clustered randomized clinical trial among 12 hospitals in 2 east coast states. METHODS: The bidirectional relationships between physical activity at 1 and 6 months and falls, emergency room visits, and rehospitalizations at 1 and 6 months post discharge and their relations with physical activity at discharge were assessed using cross-lagged panel models, accounting for the autoregression of physical activity and adverse events over time. Age, race, gender, cognitive status, admission location, comorbidity, and intervention groups were controlled in the modeling. RESULTS: Physical activity at hospitalization discharge showed autoregressive associations for the 3 time points. Patients with a higher level of physical activity at discharge had a lower chance of falling (β = -0.250, SE = 0.094, OR = 0.779, P = .008) and rehospitalizations (β = -0.228, SE = 0.092, OR = 0.796, P = .013) within 1 month post discharge. Falls, emergency room visits, or rehospitalizations at 1 month were not significantly associated with physical activity, or any adverse events at 6 months. CONCLUSIONS AND IMPLICATIONS: The findings from this study highlighted the significance and value of engaging older adults, particularly those living with dementia, in physical activity when hospitalized, as this reduced the likelihood of adverse events such as falls and rehospitalization at least in the first month post discharge.