Predictive Models for Falls in Older Adults with Dementia: A Brief Review

Falls among older adults with dementia are common events with serious consequences for their health and quality of life. Predicting falls may enable the development of interventions to prevent them, especially in institutional settings.

Objectives: To identify and describe existing models for predicting falls among older adults with dementia, including both clinical models and those based on technological solutions.

Methodology: A rapid exploratory literature review was conducted in the Scopus, Ovid Medline, and PubMed databases, including studies published through 2024. Primary studies that developed fall prediction models in older adults with dementia were included. Reviews, studies on other populations, or studies without explicit predictive models were excluded.

Results: A total of 2,293 records were identified; after removing duplicates and applying the inclusion and exclusion criteria, 22 studies were included. Of these, 54.5% were conducted in Europe, 18.2% in Japan, 18.2% in North America, and 9.1% in other settings. The models were developed in residential, home, and hospital settings and utilized sociodemographic, clinical, functional, cognitive, and mobility-related predictors. 22.7% used artificial intelligence algorithms. Only three studies applied internal validation methods. Conclusions: There are multiple models for predicting falls in older adults with dementia, but their clinical applicability is limited by insufficient validation and heterogeneity among the predictors. Additional studies are needed to develop and validate robust models that are applicable in clinical practice.