OBJECTIVES: This study evaluates the effectiveness of a predictive & preventative intervention aimed to reduce the number of falls based on data from elderly patients in community care settings collected by Cera Care. The goal is to evaluate whether the implementation of fall-specific care tasks and timely care plan reviews based on patient risk can significantly lower the incidence of falls in the domiciliary care setting.

METHODS: The intervention employed an AI-powered Falls Risk tool that used comprehensive patient data to stratify clients into different fall risk levels and generate daily alerts for branch staff. Based on these predictions, personalised fall-specific care tasks and timely care plan reviews were implemented to reduce fall incidence. The study, conducted over six weeks with 614 patients across ten Cera Care branches, utilised a pre/post analysis to evaluate fall incidence and associated healthcare outcomes.

RESULTS: The implementation of the predictive model where users in moderate and high-risk categories have up to a 6% chance of falling on any given week and the new fall specific, personalised care tasks and timely care plan reviews resulted in a 20% (p-value 0.00003) reduction in fall rates among the 614 patients involved in the pilot study.

CONCLUSIONS: The integration of a AI-powered Falls Risk assessment tool using data from elderly patients in community care settings, with personalised fall-specific care tasks and timely care plan reviews, highlights the effectiveness of advanced predictive analytics and real-time interventions. The observed 20% reduction in fall incidents could, when applied to the whole Cera Care's population of 16,000 patients, prevent an estimated 602 hospitalizations and save approximately £5.5 million in healthcare costs annually. These findings emphasise the potential of AI technology, leveraging comprehensive patient monitoring data, to enhance patient outcomes, reduce healthcare costs, and revolutionise elderly care.