Understanding Clinicians’ Needs, Requirements, and Perception of a Sepsis Clinical Decision Support System to Support Successful Implementation
Event Type
Poster Presentation
TimeThursday, April 152:42pm - 2:43pm EDT
LocationDigital Health
DescriptionClinicians are faced with ever increasing patient data as well as medical evidence which are all required for them to make the best possible decisions [1]. Clinical Decision Support Systems (CDSS) are computer technology that analyse patient data using the information to make diagnoses and provide patient-specific recommendations to assist clinicians in their decision making [1] [2]. CDSS can be in form of solicited information, unsolicited information, physician order, disease management systems and integrated information systems [1]. They have been proven by research to improve patient safety, clinical management, cost containment, administrative functions, and diagnostics support [3]. They are commonly integrated into EMRs/EHRs in the healthcare organisation [3] [4][8].
According to the Centers for Medicare & Medicaid Services (CMS), a CDSS should ‘deliver the right information, to the right people, through right channels, in right intervention formats and at the right points in the workflow’[4]. However, many healthcare organisations encounter significant challenges regarding implementing user-friendly CDSS that fit into clinician’s workflow[4]. Research has shown that CDSS are more likely to be used if they are properly integrated into work processes [6]. CDSS can disrupt workflow if the design does not follow human cognition and behaviours. Disrupting clinicians’ workflows can cause increased cognitive workload, longer time to complete tasks and reduced interaction with patients. Some studies have shown that as a result, clinicians with more experiential knowledge are less receptive to the use of CDSS [3][6]. Poor implementation of CDSS also results in alert fatigue and clinical burnout [3] [4][6][8]. In the nearest future, it is expected that machine learning, and AI will be heavily applied to make these tools smarter [9]. It is therefore important for healthcare organisations to deal with these foundational human factors challenges by focusing on the end user’s requirements in the design and implementation process for decision support tools.
Research and literature have shown that mimicking cognitive process, properly investigating user’s needs, prioritizing usability, monitoring, and getting feedback are all strategies that can help with the successful implementation of CDSS [5][7]. Usability score by end-users and time to completion for tasks are amongst the key performance indicators outlined by CMS to understand the impact of a CDSS [4].
This project is looking to investigate the needs and requirements of SickKids clinicians regarding a Sepsis clinical decision support tool to be implemented. The project is in two phases, a semi-structured one-on-one interview and an observational phase. The project aims to use human factors study to drive the successful implementation of the Sepsis BPA in the organization. The objectives of the study include eliciting clinicians’ requirements, perception of the Sepsis Best Practice Alert (BPA) tool, and discovering possible usability issues, acceptance, usage behaviour, and cognitive workload. We anticipate that the study would result in a better understanding of how to fit CDSS into clinicians' workflow, while simultaneously learning about the perception and acceptance of decision support tools in the healthcare community.
Although this current research is focused on the Sepsis tool at SickKids, the knowledge gained and results obtained can inform developers, researchers and health systems designers on the development of decision support.

[1] M. Pusic and M. Ansermino, "Clinical Decision Support Systems," BCMJ, vol. 46, pp. 236-239, 2004.
[2] Agency for Healthcare Research and Quality, "Clinical Decision Support," 2019.
[3] R. T. Sutton, D. Pincock, D. C. Baumgart, D. C. Sadowski, R. N. Fedorak and K. I. Kroeker, "An overview of clinical decision support systems: benefits, risks, and strategies for success," npj Digital Medicine, vol. 17, no. 3, 2020.
[4] J. Bresnick, "Understanding the Basics of Clinical Decision Support Systems," Health IT Analytics, 2017.
[5] A. G. Fiks, "Designing Computerized Decision Support That Works for Clinicians and Families," Curr Probl Pediatr Adolesc Health Care, pp. 60-88, 2011.
[6] E. S. Berner, "Clinical Decision Support Systems: State of the Art," Agency for Healthcare Research and Quality, 2009.
[7] Z. Dimitrios, "A Framework to Design Successful Clinical Decision Support Systems," 2017.
[8] J. Kent, "What Are the Top Challenges of Clinical Decision Support Tools?," Quality and Governance news, 2020.
[9] J. Kent, "How Machine Learning is Transforming Clinical Decision Support Tools," Health IT Analytics, 2020.