CT and MRI Protocol Variation and Optimization at an Academic Medical Center.
J Am Coll Radiol. 2018 Sep;15(9):1254-1258. doi: 10.1016/j.jacr.2018.06.002. Epub 2018 Aug 3.
Glazer DI1, DiPiro PJ2, Shinagare AB3, Huang RY4, Wang A5, Boland GW4, Khorasani R6.
PURPOSE: To reduce CT and MRI protocol variation across a multisite radiology practice at an academic medical center so that patients with similar clinical presentations are examined the same way. MATERIALS AND METHODS: This study was performed at a large academic radiology practice performing ∼800,000 radiology examinations annually. To diminish variability across the enterprise (2 general radiology divisions; 10 subspecialty imaging divisions), a Harmonization Oversight Committee was created and tasked with ensuring patients with similar clinical presentations undergo the same CT or MRI protocol, regardless of where they are imaged. A process for decision making and conflict resolution was established, supported by the department chair. Primary outcome measure was standardization of CT and MRI protocols across all sites. Secondary outcome was percent reduction of CT and MRI protocols postharmonization. RESULTS: Over the 5-month harmonization process, most conflicts arose for abdominal imaging protocols because they are performed in four distinct subspecialty divisions, but all were addressed effectively through the conflict resolution process. Overall, there was a 31% reduction in the total number of CT and MRI protocols (before harmonization 481, after harmonization 331). There was significant variation in reduction of protocols per workgroup (multiple P values; range <.0001 to .9) with largest reduction in workgroups that overlapped multiple divisions. CONCLUSION: A structured, organ system- and consensus-based quality improvement process with unambiguous decision-making and conflict resolution processes can be used to harmonize imaging protocols across complex, matrixed, multisite radiology practices so that patients with similar clinical presentations are imaged with the same imaging protocol.