Rural Hospital AI Use and Evaluation
Josiah A. Cruikshank, Brian E. Whitacre, Alison F. DavisABSTRACT
Purpose
Hospital use of predictive artificial intelligence (AI) is increasing. This study examines rural–urban differences in the adoption, use, and evaluation of predictive AI by US hospitals in 2023. It then assesses the characteristics associated with AI adoption and evaluation among all rural hospitals.
Methods
We analyze the 2023 American Hospital Association's Information Technology Supplement, which was the first to include questions on predictive model use. Adjusted Wald tests assess differences in AI adoption between rural and urban hospitals, and in categories of use and evaluation of such models for those that do adopt. We then use weighted multivariate logistic regressions on a subset of rural‐only hospitals to document associations between predictive model adoption/evaluation and hospital characteristics.
Findings
Rural hospitals adopted and used predictive AI at substantially lower rates than urban hospitals (51% vs. 81%) in 2023. Among hospitals that adopted AI, rural facilities were less likely to evaluate their models for accuracy (59% vs. 65%) and bias (39% vs. 47%). They also relied more heavily on AI tools developed by their electronic health record vendor. Regression analyses show that system membership and stronger financial performance were positively associated with AI adoption, while Critical Access Hospital and for‐profit status were negatively associated.
Conclusions
Rural hospitals remain constrained in their ability to adopt, use, and evaluate AI tools, driven largely by resource limitations. Targeted policies may be needed to ensure rural hospitals appropriately implement and evaluate AI systems as they become increasingly integral to hospital operation and clinical care.