Project Summary/Abstract
Phrase Health is a clinical decision support (CDS) analytics company that empowers health systems to deliver
high value clinical care through data-driven improvements of CDS. CDS enhances health-related decisions
and actions with pertinent, organized clinical knowledge, and patient information. For example, physicians may
have trouble remembering to order all guideline-recommended care for sepsis. CDS delivered via an order set
in the electronic health record (EHR) can simplify this process and reduce mortality by bundling the
recommended diagnostic and therapeutic orders together. However, CDS may fail to improve outcomes
because: (1) the CDS tool is underutilized; (2) the user may not follow the recommended action from the CDS;
(3) the recommended action may not lead to the appropriate evidence-based practice (EBP); and/or (4) the
EBP may not translate to the expected outcome in a novel population. Healthcare organizations need an
efficient, rigorous, and scalable process evaluation method to diagnose when and why CDS is not leading to
the intended improvements. In Phase 1, our team demonstrated the technical feasibility and usability of a new
software product, Quality Decisions, that guides quality improvement (QI) advocates of all experience levels to
(1) rigorously evaluate the impact of CDS on clinical outcomes and (2) convert data-driven insights into action.
In this Phase 2 proposal, we will use the RE-AIM framework to evaluate implementation of Quality Decisions at
three diverse health systems (Children’s Healthcare of Atlanta, Children’s Hospital of Philadelphia, and The
University of Vermont Health Network).
In Aim 1 of this proposal, we will evaluate the implementation effectiveness of Quality Decisions. Our primary
outcome will be how often the software leads to (1) new QI or CDS intervention(s), (2) increased confidence in
CDS effectiveness, or (3) changes in cohort or measure definitions. Using mixed methods including surveys,
log data, and focus groups, we will evaluate implementation efficacy, adoption, and fidelity. We will also assess
implementation barriers and facilitators using the Consolidated Framework for Implementation Research.
In Aim 2, we will determine the customer resources required to implement Quality Decisions at scale. In a pilot
phase, each health system will implement the software for a limited number of CPGs and collect hours of work
required to get the software up and running. In an expansion phase, we will estimate the number of institutional
CPGs that could feasibly be incorporated into the software and the number actually built in one year.
At the end of this project, we will have collected the primary data required to commercialize our software to
potential customers including (1) a description of the customer resource requirements to use the software
across CPGs in a health system and (2) implementation outcomes demonstrating that users use the software
frequently as intended and gain meaningful insights. These data will demonstrate value for future cust omers
and investors, allowing us accelerate the translation of knowledge into better health outcomes.
