2023/01/31 | Research | Biomechanics

History of falls as best fall rate predictor

Although age, sex, physical performance, or comorbidities are associated with falling, these variables are not good indicators to determine the expected number of falls in older persons. A study by the Musculoskeletal Biomechanics group in collaboration with the School of Health Sciences of Physiotherapy of the HES-SO Valais has shown that out of 10 parameters believed to be related to falls in older adults, only the number of previous falls within 12 months and associated fear of falling could clearly predict future falls. These findings are relevant in the context of targeted prevention of fall-induced hip fractures which the group studies in a big multidisciplinary project.

Incidence rate ratios (IRR) and corresponding 95% confidence intervals in the model generated with variable selection via backward elimination, in numerical form (left) and as forest plot (right). (https://doi.org/10.3389/fragi.2022.1056779)

Falls are a very complex phenomenon yet too little understood. Since most hip fractures in older adults are related to falls, the risk of a fracture increases with the number of falls. A randomized controlled trial with 350 people around 80 years of age found that the number of falls within a year and the fear of falling are the best predictors to identify persons at risk of falling. The statistical model applied revealed a strong increase in the likelihood of falls from 3 or more falls in the previous 12 months. Contrary to the current formula for assessing fall risk, other factors previously believed to be highly relevant – such as balance or physical fitness – were not predictive of future falls.