Answers are grounded in this paper and cite their sources.
Retrospective matched cohort study observes nonsignificant reductions in inpatient falls with AI-assisted video monitoring, highlighting the need to optimize clinical implementation.
Background/Objectives: Effective strategies to prevent inpatient falls are essential for reducing fall-related injuries, mortality, length of hospital stay, and healthcare costs. Although advanced technologies have increasingly been adopted for fall prevention, evidence regarding effectiveness in real-world clinical settings remains limited. This study evaluated the effect of implementing an artificial intelligence-assisted video monitoring system on the incidence of inpatient falls and fall-related injuries. Methods: This retrospective matched cohort study used electronic medical record data from a tertiary hospital in C city, South Korea. The system was implemented in January 2022. Patients admitted between 2020 and 2021 comprised the non-exposed group, whereas those admitted between 2023 and 2024 comprised the exposed group. Nearest neighbor propensity score matching based on age, sex, the number of diagnoses, and the number of ward days was performed to create comparable groups. Fall incidence rates per 1000 patient-days were calculated, and Firth’s penalized likelihood logistic regression and Cox proportional hazards regression with robust errors were conducted. Results: Propensity score matching yielded a 1:1 matched sample of 3002 cases per group. The fall incidence rate was 1.017 per 1000 patient-days in the exposed group, lower than 1.286 in the non-exposed group. However, penalized likelihood logistic regression and Cox proportional hazards regression revealed no statistically significant effect of artificial intelligence-assisted video monitoring on fall reduction. Conclusions: Artificial intelligence-assisted video monitoring was associated with a lower fall incidence, but no statistically significant effect was identified. These findings highlight the potential and limitations of artificial intelligence-assisted video monitoring for inpatient fall prevention and underscore the need for further research to enhance its clinical utility.
No takes yet. Share an insight, caveat, or question.
Lee et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: