Objectives: To explore how neuropsychiatric symptoms (NPS) are associated with number of falls and how exercise modifies the risk of falling in community-dwelling people with Alzheimer's disease (AD) and NPS.
Design: Secondary analysis of a randomized controlled trial.
Participants: Community-dwelling individuals with AD (N=210) who completed the Neuropsychiatric Inventory (NPI) (N = 179).
Intervention: Participants were randomized into 3 groups: group-based exercise (4-hour sessions with approximately 1 hour of training) and tailored home-based exercise (1 hour of training) twice a week for 1 year and a control group receiving usual community care. In this secondary analysis, we merged the home-based and group-based exercise groups and compared this group with the control group.
Measurements: NPS were measured using the NPI at baseline, and spousal caregivers recorded falls in daily fall diaries during 1 year of follow-up.
Results: The number of falls increased linearly with NPI score in the control group. Fall rates were 1.48 (95% confidence interval (CI)=1.26-1.73) per person-year in the intervention group and 2.87 (95% CI=2.43-3.35) in the control group. Adjusted for age, sex, Mini-Mental State Examination (MMSE) score, and Short Physical Performance Battery (SPPB) score, incidence rate ratio (IRR) was 0.48 (95% Cl=0.39-0.60, p < .001). Main effects for fall rate were significant for group (p < .001) and NPI total (p < .02); the interaction effect was also significant (p = .009) (adjusted for sex, age, MMSE score, SPPB score, and psychotropic medication use).
Conclusion: Exercise may decrease the risk of falling in community-dwelling individuals with AD and NPS. Future exercise trials should confirm this finding in participants with significant NPS.
Trial registration: ACTRN12608000037303. J Am Geriatr Soc 66:2377-2381, 2018.
Keywords: Alzheimer's disease; NPI; exercise; falls; neuropsychiatric symptoms.
© 2018, Copyright the Authors Journal compilation © 2018, The American Geriatrics Society.