Feature ArticleCan the Electronic Health Record Predict Risk of Falls in Hospitalized Patients by Using Artificial Intelligence? A Meta-analysisHsu, Yen MD; Kao, Yung-Shuo MD Author Information Author Affiliations: Department of Family Medicine, Changhua Christian Hospital (Dr Hsu), Changhua, Taiwan; and Department of Radiation Oncology, China Medical University Hospital (Dr Kao), Taichung, Taiwan. The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article. Corresponding author: Yung-Shuo Kao, MD, China Medical University Hospital, No. 2, Yude Rd, North District, Taichung City 404332, Taiwan, ROC ([email protected]). Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.cinjournal.com). CIN: Computers, Informatics, Nursing ():10.1097/CIN.0000000000000952, October 14, 2022. | DOI: 10.1097/CIN.0000000000000952 Buy PAP Metrics Abstract Because of an aging population worldwide, the increasing prevalence of falls and their consequent injuries are becoming a safety, health, and social-care issue among elderly people. We conducted a meta-analysis to investigate the benchmark of prediction power when using the EHR with artificial intelligence to predict risk of falls in hospitalized patients. The CHARMS guideline was used in this meta-analysis. We searched PubMed, Cochrane, and EMBASE. The pooled sensitivity and specificity were calculated, and the summary receiver operating curve was formed to investigate the predictive power of artificial intelligence models. The PROBAST table was used to assess the quality of the selected studies. A total of 132 846 patients were included in this meta-analysis. The pooled area under the curve of the collected research was estimated to be 0.78. The pooled sensitivity was 0.63 (95% confidence interval, 0.52–0.72), whereas the pooled specificity was 0.82 (95% confidence interval, 0.73–0.88). The quality of our selected studies was high, with most of them being evaluated with low risk of bias and low concern for applicability. Our study demonstrates that using the EHR with artificial intelligence to predict the risk of falls among hospitalized patients is feasible. Future clinical applications are anticipated. Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.