In a survey of 431 headache clinicians at the VA health system, we found that a majority of clinicians believed neuroimaging was overused for patients with headaches; however, many would utilize neuroimaging in low-risk scenarios that have a small probability for change in patient management. Utilization of guideline-discordant neuroimaging may have been driven by clinicians' beliefs that the benefits outweigh the harms of headache neuroimaging even in low-risk scenarios, and their preference towards errors of commission rather than omission. Because potential red flags greatly influence decision making, improved evidence on which red flags should influence neuroimaging decisions is desperately needed. This information would allow future guidelines to be more precise in describing the specific clinical scenarios that should prompt neuroimaging. Additionally, CDSS that include details of headache neuroimaging guidelines and patient-specific assessments of neuroimaging benefits/risks may reduce unnecessary neuroimaging, especially for APCs who had high utilization and were the most accepting of such interventions.
ANOVA, analysis of variance; APCs, advanced practice clinicians; CDSS, clinical decision support systems; EHR, electronic health record; MRI, magnetic resonance imaging; NP, nurse practitioner; PA, physician assistant; REDCap, Research Electronic Data Capture; SD, standard deviation; TDF, Theoretical Domains Framework; VA, Veterans Affairs.
Dr. Reynolds is supported by the National Institutes of Health (T32NS0007222). Dr. Burke, Ms. Evans, and Dr. Callaghan are supported by VA CSRD Merit CX001504. Dr. Syed, Dr. Liao, Dr. Lobo, Dr. Cooper, and Dr. Charleston have no relevant funding to disclose
Headache. 2022;62(1):36-56. © 2022 Blackwell Publishing