SENSORIMOTOR, COGNITIVE AND AFFECTIVE FUNCTIONS CONTRIBUTE TO THE PREDICTION OF FALLS IN OLD AGE AND NEUROLOGICAL DISORDERS: AN OBSERVATIONAL STUDY

ABSTRACT Objective To determine whether impairments across cognitive and affective domains provide additional information to sensorimotor deficits for fall prediction amongst various populations. Design We pooled data from five studies for this observational analysis of prospective falls. Setting Community or low-level care facility. Participants A total of 1,090 older people (74.0±9.4 years;579♀); 500 neurologically intact (NI) older people and three groups with neurological disorders (cognitive impairment, N=174; multiple sclerosis (MS), N=111; Parkinson’s disease, N=305). Interventions None. Main Outcome Measure(s) Sensorimotor function was assessed with the Physiological Profile Assessment, cognitive function with tests of executive function, affect with questionnaires of depression, and concern about falling with falls efficacy questionnaires. These variables were associated with fall incidence rates, obtained prospectively over 6-12 months. Results Poorer sensorimotor function was associated with falls (incidence rate ratio[95% confidence interval]: 1.46[1.28-1.66]). Impaired executive function was the strongest predictor of falls overall (2.91[2.27-3.73]), followed by depressive symptoms (2.07[1.56-2.75)] and concern about falling (2.02[1.61-2.55]). Associations were similar among groups, except for a weaker relationship with executive impairment in NI and a stronger relationship with concern about falling in MS. Multivariable analyses showed that executive impairment, poorer sensorimotor performance, depressive symptoms and concern about falling were independently associated with falls. Conclusions Deficits in cognition (executive function) and affect (depressive symptoms) and concern about falling are as important as sensorimotor function for fall prediction. These domains should be included in fall risk assessments for older people and clinical groups.