133 AI-based risk predictions: Evaluation of pilot DRAAI 5 Quantitative data from questionnaires and field notes were analyzed using descriptive statistics. For continuous variables, data are presented as mean (standard deviation) for normally distributed variables or median (interquartile range) for skewed data. Categorical variables are summarized as percentages. Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp. For the qualitative analysis of open-ended responses and field notes, deductive content analysis was conducted. The implementation plan and the NASSS domains served as a framework for interpreting the data. The research team reviewed and discussed the field notes and open-ended responses, identifying patterns, reflecting on the findings, and assigning meaning to the observations. A narrative summary of the main findings was incorporated into the results section where relevant [41]. Ethical Considerations Ethical considerations for retrieving data to generate and show the risk predictions adhered to hospital protocols, applicable privacy laws and regulations (Data Protection Impact Assessment) and principles for the responsible use of AI in healthcare [42]. The Medical Ethics Review Committee of the Erasmus University Medical Center reviewed the study and determined that it was not subject to the Dutch Medical Research Involving Human Subjects Act. As a result, obtaining informed consent for this process evaluation study was deemed unnecessary (MEC-2023-0752). Results Overall, it was feasible to provide daily predictions for every patient throughout the entire pilot period. On two occasions, errors occurred during the preparation and display of the predictions; however, these issues were resolved within the same day. Participants A total of 55 nurses completed the questionnaire, resulting in a response rate of 39% (55/140). Of these, 25 nurses (43%; 25/58) were from the pulmonary ward, 21 (38%; 21/56) from the internal medicine ward, and nine (35%; 9/26) from the COVID-19 ward.
RkJQdWJsaXNoZXIy MTk4NDMw