Implementation and maintenance of an established clinical decision support tool algorithm: challenges and recommendations from a prospective cohort study

Abstract

Background: Computerized clinical decision support systems (CDSS) offer predictive statistical measurements of patient outcomes to clinicians using amalgamated data. Methods: The focus of this study was to investigate the usability and reliability of an established CDSS at a pilot site in the management of surgical patients with lumbar degenerative spondylolisthesis, and to assemble recommendations in order to implement a CDSS at other organizations. Results: Physicians reported that in 100% of patients, the clinical decision support tool algorithm (CDSTA) could help discuss surgical options. In 78% of patients, physicians agreed that it could help make the decision to recommend surgery. Physicians agreed with the CDSTA outcome predictions in 96% of patients. Conclusions: Various challenges were identified concerning CDSTA usability, and the main stumbling point was communication between the clinical and information technology staff who developed the CDSTA, demonstrated through technical language barriers, CDSTA backend updating delays, and reluctance of clinicians and staff to adapt to the CDSTA in addition to the typical daily workflow. The most significant lesson was maintaining open lines of dialogue between the developer and pilot site to allow for seamless CDSS integration into a practice and minimal technical downtime.