Tecnologias para predição de quedas em idosos: Uma revisão de escopo.

Abstract

The study addresses the growth of population aging and the consequent increase in the incidence of falls among older adults, highlighting the severe physical, social, and economic repercussions of these events. In response to this scenario, the work aimed to map technologies focused on fall prediction, whether they consist of an electronic or computational nature.

The research was conducted as a scoping review, following the PRISMA 2020 guidelines to ensure methodological rigor and transparency in the selection of studies. Searches were performed in relevant scientific databases, applying predefined inclusion and exclusion criteria, resulting in the analysis of 17 published over the past decade.

The results showed that wearable sensors and machine learning algorithms have made significant advances in fall detection and prediction, although they face limitations related to user acceptance, adaptation to prolonged use, and implementation costs. Force platforms proved to be affordable and reliable options in clinical settings for balance assessment but less practical for continuous monitoring.

The study concludes that preventing and predicting falls is more effective than reacting after their occurrence, both in the physical aspect, by preserving older adults’ autonomy, and economically by reducing costs associated with hospitalizations and rehabilitation. The need for further research with larger samples and clinical validation is emphasized, aiming for technologies that are simultaneously accurate, affordable, and acceptable to the elderly population.