Uptake of Clinical Decision Support Systems Among Health Care Professionals in Six European Countries and the United States: Cross-Sectional Survey Within the I-CARE4OLD Project.

BACKGROUND: The use of Clinical Decision Support Systems (CDSS), such as clinical decision rules, algorithms, or machine learning-based applications, has gained attention in recent years. However, their adoption and effectiveness may vary across different health care systems and settings. For a CDSS to be adopted, it must effectively address the practical issues encountered by professionals; however, little research has been done to identify these needs and requirements. OBJECTIVE: This study aims to describe and compare the current use of various decision-support and prediction tools in long-term care for older people across health professionals from 6 European countries and the United States. METHODS: This study analyzed survey data from a CDSS pilot study in a purposive sample of health professionals working with older adults with complex chronic conditions from six European countries and the United States. The survey included participants' general background information, their current use of decision support tools, and their attitudes on the potential benefits of CDSS. About 20 participants were sampled per country. Closed responses were analyzed using correlation coefficients and regression models, while open-ended responses were clustered in a qualitative manner, categorizing each response. RESULTS: A total of 151 professionals (mean age 45.5, SD 11.6 years, 71.5%, 108/151 female) participated in the pilot study. Most participants were physicians (85/151, 56.3%) or nurses (57/151, 37.7%). About 51% (78/151) of the participants reported using CDSS, while 22.4% (34/151) used predictive CDSS, showing important variation across samples from the seven countries. The regression model for comfort with technology showed a positive association for openness to new technologies (β=0.622; P<.001), although an inverse significant association was found for age (β=-0.022; P<.001). No significant associations were found for the actual use of CDSS. Participants reported using CDSS mainly for diagnostic purposes or for guideline implementation, not aimed at prognostic information. In contrast, examples of prognostic tools were most frequently mentioned by respondents as being valuable improvements to clinical practice. CONCLUSIONS: While some countries' samples reported well-integrated digital health infrastructures and higher CDSS adoption rates, others still face challenges in implementing these. However, we found multiple examples of emerging tools, and at the same time, an important demand for predictive CDSS. Our findings highlight the need for improvement of current CDSS implementation and both development and implementation of particularly predictive CDSS.