Proportion of falls reported in persons with Parkinson's disease: A meta-analysis

Abstract

Objective: Falls can be highly debilitating and have an important negative impact on the quality of life of patients with Parkinson's disease (PD). The aim of this systematic review and meta-analysis is to provide an up-to-date overview of the prevalence of ≥1 fall and ≥2 falls in idiopathic PD.

Data sources: MEDLINE, Web of Science, Embase and Cinahl databases were systematically searched until 04 July 2022 for prospective studies reporting fall prevalence in persons with idiopathic PD.

Methods: Pooled prevalence rates with 95% confidence intervals (CIs) were computed using random-effects models. Heterogeneity among studies was assessed using the I² statistic.

Results: A total of 54 studies (7546 participants) were included, and random-effects meta-analysis yielded a pooled proportional fall rate of 0.48 (95% CI [0.43-0.52], I² = 93%, 46 studies, 6874 participants) for classification 1 (≥1 fall) and a pooled proportional fall rate of 0.32 (95% CI [0.27-0.37], I² = 78%, 31 studies, 5672 participants) for classification 2 (≥2 falls). Subgroup analysis on the classification of falls, and length and method of monitoring falls did not reveal significant differences and did not reduce between-study variability.

Conclusion: Pooled estimates suggest that one in two persons with PD fall at least once, and one in three fall at least twice within the registered time period. Substantial variability remains after pooling fall prevalence rates according to the length and method of monitoring. Therefore, no recommendations can be made concerning these methodological aspects. Future research on falls in PD is encouraged to implement best practice recommendations to monitor and report fall data.

Keywords: Falls; Parkinson's disease; prevalence.