Relationship Between the 25-Question Geriatric Locomotive Function Scale and Falls: A One-Year Longitudinal Observational Study of 1,173 Healthy Community-Dwelling Residents Aged 65 and Older

Introduction This study aimed to explore the relationship between the 25-question Geriatric Locomotive Function Scale (GLFS-25) score (i.e., total score and domain scores) and falls (i.e., history with or without falls and frequency of falls). Methodology We conducted a one-year longitudinal observational study involving 1,173 healthy community-dwelling residents aged ≥65 years who attended a basic health checkup in Minami-Aizu Town and Tadami Town, Fukushima, Japan, from 2016 to 2017. The following clinical information was collected: age, sex, body mass index, smoking status, alcohol consumption, living situation, metabolic syndrome, physical activity, and GLFS-25 score during the participants' health check in 2016. The GLFS-25 measures various domains, including body pain, movement-related difficulties, usual care, social activities, and anxiety. Participants were diagnosed with locomotive syndrome (LS) based on their GLFS-25 total scores: Non-LS (0-6 points), LS-1 (7-15 points), LS-2 (16-23 points), and LS-3 (24-100 points). We assessed the annual occurrence of falls during the participants' health check in 2017 and the monthly frequency of falls. Student's t-test, Mann-Whitney's U test, and Fisher's exact test were performed to compare parameters between fallers and non-fallers. To examine the association between the annual occurrence of falls and the diagnosis of LS, a multivariate logistic regression analysis was performed to calculate adjusted odds ratios (ORs), controlled based on the clinical information. To assess the association between the monthly frequency of falls and GLFS-25 scores, a multivariate regression analysis was performed to calculate the adjusted standardized partial regression coefficient (β), controlled based on the clinical information. Results Fallers were significantly older (p < 0.001), had a higher body mass index (p = 0.034), and had higher GLFS-25 total scores (p < 0.001) than non-fallers. In the multiple logistic regression analysis, falls were significantly associated with LS-1 or more (OR = 2.32, p < 0.001), LS-2 or more (OR = 2.72, p < 0.001), and LS-3 or more (OR = 2.99, p < 0.001). Furthermore, the annual occurrence of falls was significantly associated with GLFS-25 body pain (OR = 1.94, p = 0.012) and anxiety scores (OR = 2.09, p = 0.021). In the multiple regression analysis, the monthly frequency of falls was significantly associated with the GLFS-25 total score (β = 0.29, p < 0.001). The monthly frequency of falls was also significantly associated with GLFS-25 domain scores, including body pain score (β = 0.23, p < 0.001), movement-related difficulty score (β = 0.21, p < 0.001), usual care score (β = 0.18, p < 0.001), social activity score (β = 0.26, p < 0.001), and anxiety score (β = 0.22, p < 0.001). Conclusion Our findings emphasize the importance of fall prevention in individuals with LS-1 and suggest that the GLFS-25 total score may predict recurrent falls. Our study first provides valuable evidence regarding the relationship between the GLFS-25 (total score and domain scores) and falls. The monthly frequency of falls was correlated with the total GLFS-25 score and all GLFS-25 domain scores. However, the annual occurrence of falls was found to have no correlation with anything other than the GLFS-25 domain scores regarding physical pain and anxiety. Therefore, further investigations are needed.