Clinical Decision Support for Disseminating and Implementing Patient-Centered Outcomes Research (PCOR) Clinical Evidence - Clinical decision support (CDS) systems can enhance the diagnostic process by minimizing diagnostic errors
and delays, which are major components of patient safety. Recent federal regulations, including the Protecting
Access to Medicare Act (PAMA), encourage CDS adoption and use by requiring that ambulatory providers
consult with evidence-based Appropriate Use Criteria through qualified CDS mechanisms for ordering certain
advanced imaging exams (e.g., CT, MRI). Optimally implemented, PAMA is a unique opportunity to improve
the quality of diagnostic imaging care. However, if poorly executed, it could create interruptive electronic health
record (EHR) workflows and low value clinical alerts, with little or no impact on diagnostic imaging utilization.
The Harvard Medical School Library of Evidence (HMS-LOE) led a multi-disciplinary team of clinicians and
medical informaticists to represent and grade clinical evidence in an effort to create CDS-consumable
knowledge artifacts, which are publicly available at the HMS Countway Medical Library. Public domain sources
of clinical evidence are represented as a Clinical Evidence Logic Statement (CELS) of “If-Then” form and
graded using a system developed by the Oxford Centre for Evidence-Based Medicine for strength of evidence.
18% of the CELS in the HMS-LOE are also fully represented in standardized syntax and formats (e.g.,
Nomenclature of Medicine-Clinical Terms [SNOMED CT], Fast HealthCare Interoperability Resources [FHIR]
and Clinical Quality Language [CQL]). CELS are specifically designed to be CDS-consumable and will greatly
facilitate representation and public dissemination and sharing of artifacts.
In this study, we will: 1) Assess the technical capability of FHIR, CQL and SNOMED CT to represent three
patient-centered outcomes research (PCOR) guidelines from the HMS-LOE in a higher-level, structured
representation, specifically a) Using Breast MRI (instead of or in addition to mammography) to diagnose a
second breast cancer in women with previous breast cancer, b) Using CT scan for Pulmonary Embolism, and
(c) Physician education for those at Risk for Pulmonary Embolism. In addition, we will: 2) Validate PCOR CELS
represented in a structured representation at our healthcare system (Brigham and Women’s Hospital), and (3)
Scale previously developed CDS artifacts to a different healthcare system (Covenant Health Inc.) with a
different EHR implementation and health information technology (HIT) platform. Implementing transparently
graded and represented knowledge artifacts will promote widespread dissemination and interoperability across
varied HIT platform, end-users and workflows. In addition, validation and evaluation of the CDS artifacts,
including usability and impact on clinical practice, will enhance widespread adoption. Ultimately, the goal of
CDS for diagnostic imaging is to enhance evidence-based, effective care delivery.
