Lead/Presenter: Andrea Nevedal, COIN - Palo Alto
All Authors: Nevedal AL (Center for Innovation to Implementation, VA Palo Alto Health Care System), Lewis E (Program Evaluation Resource Center; Center for Innovation to Implementation, VA Palo Alto Health Care System), Wu J (Center for Innovation to Implementation, VA Palo Alto Health Care System) Jacobs J (Health Economics Resource Center; Center for Innovation to Implementation, VA Palo Alto Health Care System) Lo J (Health Economics Resource Center, VA Palo Alto Health Care System) Illarmo S (Health Economics Resource Center, VA Palo Alto Health Care System) Barnett P (Health Economics Resource Center; Center for Innovation to Implementation, VA Palo Alto Health Care System)
Clinical practice guidelines suggest that magnetic resonance imaging of the lumbar spine (LS-MRI) is unneeded during the first 6 weeks of new onset low-back pain without red-flag conditions. Despite recommendations, approximately 30% of LS-MRI are unnecessary and this type of advanced imaging is not associated with improved pain, back function, quality of life or mental health. Unneeded LS-MRI costs the U.S. healthcare system about $300 million dollars per year, including $14 million to Veterans Affairs (VA). The factors that lead to unneeded LS-MRI are not known. The purpose of this study was to conduct a qualitative assessment among VA primary care providers (PCPs) to identify potential modifiable factors associated with unneeded ordering which could subsequently be targeted for de-implementation.
VA administrative data were used to identify primary care providers (physicians, physician assistants, nurse practitioners) with higher and lower rates of guideline concordant LS-MRI for new onset low back pain without red-flag conditions in 2016. We conducted 55 semi-structured telephone interviews with 22 higher and 33 lower guideline concordant VA primary care providers and recruited from all 18 VA regional service networks. Open-ended questions focused on environmental, guideline, patient, and provider factors contributing to unneeded LS-MRI. Data were analyzed using directed content analysis and summarized in a matrix to compare differences between high and low guideline concordant provider groups.
Lower guideline concordant PCPs described environmental factors (lenient radiologist review, patient travel burden, increased time constraints); guideline factors (less familiarity, more disagreement); and provider factors (giving in to patients, seeing value in early scans) that contributed to unneeded LS-MRI. Higher guideline concordant PCPs described environmental factors (diligent radiologist review); guideline factors (greater familiarity, more agreement); and provider factors (not giving in to patients, no value in early scans) that contributed to appropriate LS-MRI. Both PCP groups reported frequent patient pressure for LS-MRI, but they differed in their response to patient pressure.
Higher and lower guideline concordant PCPs differed in environmental, guideline, and provider factors, relevant to their decision to order unneeded LS-MRI.
Learning which factors contribute to unneeded LS-MRI will help researchers and administrators develop targeted de-implementation strategies to reduce these scans.