Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website

PPO 15-178 – HSR Study

PPO 15-178
Co-designing Clinical Decision Support to Advance Value-based Shoulder Imaging
Michael Cary Brunner, MD
William S. Middleton Memorial Veterans Hospital, Madison, WI
Madison, WI
Funding Period: January 2017 - December 2017
Medicine's demand for advanced imaging continues to increase faster than available healthcare dollars, often diverging from evidence-based practice. Provider ordering habits are difficult to alter, particularly given barriers to following computerized clinical reminders and provider preference for technology-laden diagnostics. Partnering with providers to "co-design" better clinical decision support (CDS) in substituting for MRI with clinically effective, less costly technology like shoulder ultrasound, would improve appropriate patient MRI access by reducing unnecessary exams. This study outlined a strategy to co-design CDS to encourage an evidence-based approach to transition imaging for rotator cuff disease from MRI to musculoskeletal ultrasound as a clinically-equivalent, efficacious and less costly alternative.

This pilot grant's successful use of co-designed CDS to promote ultrasound over MRI will inform longer term, funded efforts to maximize this CDS maintenance over time (beyond the scope of a one-year pilot) and quantification of resultant improved clinical and economic outcomes. If this collaborative design process to CDS proves durable, its implementation should be disseminated to other facilities and studied in promoting this and other evidence-based provider test-ordering behaviors.

Our goal was to demonstrate that co-designing CDS can exploit provider input to reduce local barriers to its adoption and can transition to more efficient alternatives informed by comparative effectiveness research. While clearly informative CDS facilitates the practice of evidence-based medicine, CDS unconventionally co-designed in collaboration with targeted providers, has potential to influence practice change long-term. The Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) implementation and Sociotechnical model frameworks served to guide the following aims:

Aim 1: Identify providers' reasons for selecting MRI over ultrasound in diagnosing rotator cuff tear, and distinguish their preferences and barriers to the use of ultrasound.

Aim 2: Co-design CDS informed by primary care provider interviews, implement it and assess CDS impact on utilization of ultrasound for diagnosing rotator cuff tears.

To perform these studies, we used a convergent mixed-methods approach organized by the RE-AIM and Sociotechnical model frameworks. Formative evaluation during provider education and roll-out of musculoskeletal ultrasound to diagnose rotator cuff tear informed CDS design to prompt the appropriate transition from MRI to ultrasound. Quantitative data obtained throughout the course of the study was used to track CDS implementation with sufficient power to detect significant differences in ordering patterns over the life of the project. Semi-structured interviews of 10-15 participants were conducted during the first phase of the study and repeated after initiating co-designed CDS. The CDS construction team included two referring providers, an interviewer, a shoulder surgeon, musculoskeletal and chief radiologists, CPRS and CDS experts.

Co-designed clinical decision support (CDS) was remarkably effective in changing provider ordering habits, informing Aim 2. CDS was co-designed to include concise education and require a minimal number of provider added keystrokes. It concretely emphasized to the Veteran the need for plain film evaluation prior to further imaging and the benefits of shoulder ultrasound over MRI.

Informing Aim 1, there was a statistically significant shift in (particularly non-musculoskeletal specialty) provider ordering from MRI to ultrasound post CDS, with appropriateness of ordering more than doubling (both p<.001). The rate of MRIs ordered after ultrasound was 11%, with the large majority ordered by musculoskeletal specialists for "operative planning". Higher than the literature standard 5%, this could reflect orthopedic surgeon greater comfort interpreting MRI over ultrasound or reflect clinical concern for intra-articular pathologies that could alter surgical management.

Provider ordering habits can prove difficult to alter, especially when attempting to reduce unnecessary use of advanced diagnostic imaging. While CDS through an electronic health record can help, alert fatigue and other barriers limit success; these were defined through a formative evaluation. By partnering to co-design CDS with medical providers to reduce these defined barriers, transition to ultrasound of the shoulder from MRI for rotator cuff diagnosis was accomplished and should improve health care for these Veterans and MRI access for others in a quantifiable way.

External Links for this Project

NIH Reporter

Grant Number: I21HX001852-01A1

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


Journal Articles

  1. Brunner MC, Sheehan SE, Yanke EM, Sittig DF, Safdar N, Hill B, Lee KS, Orwin JF, Vanness DJ, Hildebrand CJ, Bruno MA, Erickson TJ, Zea R, Moberg DP. Joint Design with Providers of Clinical Decision Support for Value-Based Advanced Shoulder Imaging. Applied clinical informatics. 2020 Feb 19; 11(1):142-152. [view]

DRA: Health Systems, Other Conditions
DRE: Technology Development and Assessment
Keywords: none
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.