A Prospective Analysis of Radiation Oncologist Compliance With Early Peer Review Recommendations

Abstract

We conducted a prospective observational cohort study of physician compliance with daily early pretreatment planning peer review recommendations and quantified factors associated with compliance. All patient cases in our department are presented at 2 peer review conferences: (1) "early" preplanning, occurring daily for patients who have undergone simulation review, and (2) "late" (chart rounds), occurring weekly for patients who have started treatment. Peer review recommendations were prospectively recorded during early review, and compliance with recommendations was determined at chart rounds. Recommendations were assigned magnitude scores (minor, moderate, or major). We analyzed the association of patient, physician, and recommendation characteristics and compliance (scored as a binary variable) with early peer review recommendations, using logistic regression with a mixed effects model. From February 2017 to May 2018, 1271 patient cases underwent early peer review, and 326 (26%) received peer-based recommendations. Of 356 recommendations, 37% were minor, 36% were moderate, and 27% were major. Overall compliance was 59% (95% confidence interval, 54%-64%). On univariate analysis, compliance decreased as the recommendation magnitude increased (minor, 65%; moderate, 60%; major, 47%; P=.019; odds ratio, 0.71 per increase in magnitude). Compliance also differed among different treating physicians (range, 38%-73%, χ2 test, P=.003) but was not associated with other physician characteristics. Disease group and treatment technique were not associated with compliance. On multivariable analysis, increasing recommendation magnitude remained significantly associated with decreased compliance (multivariate P=.042; odds ratio, 0.74). Daily early peer review resulted in a substantial proportion of recommended changes. Compliance with early peer review recommendations was fair but varied among physicians. Compliance declined with increasing recommendation magnitude, suggesting that physicians may be reluctant to adopt major changes. These results highlight the potential importance of peer review timing.