Clinical Decision Support Overview

Principles

Definition

Clinical decision support is an academic concept and methodological approach concerned with providing clinicians, patients, and staff with knowledge and person-specific information, intelligently filtered or presented at appropriate times, to enhance health and health care. As a research field, it investigates the principles, design, implementation, and evaluation of systems and interventions that leverage data, knowledge bases, and technology to aid clinical decision-making processes. Its significance lies in its potential to improve patient safety, reduce errors, promote evidence-based practice, and optimize healthcare delivery.

Ontological type

Core Functions

Evaluation Strategies

Knowledge Representation

Key developments

Key Figures

Computable Probabilistic CDS era

David F. Lobach [1], Duke University [3], emerged as a leading figure in computable probabilistic clinical decision support during the era. His work on computerized decision support based on a clinical practice guideline [6] improved compliance with care standards, and structuring clinical practice guidelines in a relational database model for decision support on the Internet [7] demonstrated scalable representations of guidelines for point-of-care use. Jeremy C Wyatt [2], Stanford University [4], and University College London [5], emerged as a leading voice in the philosophies for the design and development of clinical decision-support systems during this era. Philosophies for the design and development of clinical decision-support systems [8] outlined design principles and evaluation considerations that shaped CDS development and practice.

Guideline and Standards Integration era

Amiram Gafni [1] is associated with Albert Einstein College of Medicine [3] and the University of Toronto [4] during the Guideline and Standards Integration era. In this era, his key contribution centers on the paper Decision-making in the physician–patient encounter: revisiting the shared treatment decision-making model [7], which framed patient preferences as central to guideline-concordant care and the feasibility of embedding shared decision processes into clinical decision support. Timothy J. Whelan [2] is associated with Harvard University [5] and the University of California, San Francisco [6] in this period. His key contribution in this era, as reflected in the paper Decision-making in the physician–patient encounter: revisiting the shared treatment decision-making model [7], helped advance shared decision-making as a mechanism to operationalize guidelines within interoperable CDS and to emphasize safety, modifiability, and workflow integration.

Explainable AI in Practice era

Gary S. Collins [1] is a leading figure in Explainable AI in Practice, with affiliations spanning University of California, San Francisco [3] and Stanford University [4] in this era. In 2018, Collins [1] contributed to PROBAST [5], a tool to assess the risk of bias and applicability of prediction model studies, a contribution that strengthened methodological rigor for evaluating explainable AI in clinical decision support and fostered clinician trust. Johannes B. Reitsma [2] is a recognized figure connected to University of California, San Francisco [3] and Stanford University [4] in this era, with his work on PROBAST [5] advancing systematic bias assessment in prediction models. This emphasis on validation and bias assessment through PROBAST [5] helped align explainable AI outputs with interpretable, trustworthy decision support in real-world clinical workflows.