Usability and Quality of UCISAB, a Clinical Decision Support System for Optimization of Analgesia and Sedation Infusions in the ICU: A Multicenter, Cross-Sectional, Mixed-Methods Usability Evaluation in Colombia
TL;DR
UCISAB showed favorable perceived quality and a high stated willingness to use, while usability remained marginal; further interface and pharmacological refinements, together with prospective clinical validation in real-world settings, are required before routine implementation.
Abstract
Background: The management of sedative and analgesic infusions in intensive care units (ICUs) represents a global clinical challenge. Clinical decision support systems (CDSSs) have emerged as an approach to facilitate protocol standardization and mitigate adverse effects associated with prolonged sedation and analgesia. Objective: To evaluate the usability and technical quality of UCISAB, a CDSS designed to optimize sedoanalgesia in the ICU. Methods: A multicenter, cross-sectional, mixed-methods usability evaluation was conducted. A prototype was evaluated using the System Usability Scale (SUS) and the User Mobile Application Rating Scale (uMARS), complemented by a qualitative analysis of open-ended comments. Results: A total of 31 physicians participated (attending physicians and residents). The mean SUS usability score was 66.13 (SD ±21.26), categorized as “Marginal High.” On the uMARS, the median scores were 4.28 for objective quality (good), 3.75 for subjective quality (acceptable), and 4.33 for impact (good). The stated willingness to apply the tool’s recommendations in clinical practice reached 93.5%, with a median intention-to-use score (uMARS) of 4.0 (good). Exploratory analysis revealed no significant difference in SUS scores between residents and attending physicians (p = 0.072). In these exploratory comparisons, statistically significant differences by professional role were found only in the uMARS dimensions: objective quality (p < 0.001), subjective quality (p = 0.023), and impact (p = 0.036). Qualitative feedback suggested that the tab-based architecture may increase cognitive load among expert users, highlighting a need for interface simplification. Conclusions: UCISAB showed favorable perceived quality and a high stated willingness to use, while usability remained marginal; further interface and pharmacological refinements, together with prospective clinical validation in real-world settings, are required before routine implementation. Professional role may influence perceived quality, suggesting the potential value of adaptive interface designs; however, this association requires confirmation in future studies. Developing systems that optimize sedation and analgesia in critical care remains an unmet need in ICUs.