Implementation Outcomes of Strategies to Promote Short-Course Radiation for Nonspine Bone Metastases in an Academic-Community Partnership: Survey Results from the ALIGNMENT Trial

Abstract

Local treatment of nonspine bone metastases has become increasingly complex, resulting in physician practice variability nationwide. The purpose of this study was to assess physician perceptions of 3 implementation strategies to promote adoption of short course radiotherapy (RT) for nonspine bone metastases. ALIGNMENT ("Alliance Group for Bone Metastasis") was a multi-institutional stepped wedge cluster randomized implementation trial testing strategies to increase use of ≤5 fractions for nonspine bone metastases conducted across 3 clinical sites in an academic-community partnership. Strategies included a) multidisciplinary consensus guidelines, b) e-Consults, an email-based consultation platform, and c) personalized audit and feedback (A&F) reports with peer comparison. Using the Proctor et al. framework and validated questions from Weiner et al., physician surveys were used to assess each strategy's usefulness, acceptability (i.e., "I welcome [strategy]"), appropriateness (i.e., "[strategy] seems like a good match"), and feasibility (i.e., "[strategy] seems implementable" or "easy to use"). Survey responses were anonymized, so Fisher's Exact test was used to compare proportions with significance set at p<0.05. Overall, 29 of 38 and 30 of 38 physicians participated in the pre- and post-implementation surveys, respectively, with 80% completing both. Pre-implementation, guidelines was most often ranked 1st in terms of usefulness (61%), followed by eConsults (38%) and A&F (3%). Post-implementation, guidelines and eConsults had the most and least favorable acceptability, appropriateness, and feasibility scores, respectively (Table), with 77% of physicians being likely to recommend the guidelines to other oncologists. In contrast, while 43% of physicians reported having at least 1 difficult clinical question regarding bone metastases during the study, only 33% of physicians preferred eConsults, while 50% preferred reaching out to a friend/colleague. Lastly, although A&F had the lowest perceived usefulness pre-implementation, A&F had the greatest increase in acceptability (72%→90%; p = 0.10), appropriateness (66%→90%; p = 0.03) feasibility ("implementable": 59%→93%, p = 0.002; "easy to use": 45%→93%, p<0.001). In this multicenter trial, all strategies were acceptable, appropriate, and feasible, with guidelines and A&F showing the most favorable outcomes post-implementation. While guidelines were assessed as the most useful, A&F had significant increases in appropriateness and feasibility.