Identifying benchmarks for ASCO's “Choosing Wisely” measures that address low-value imaging in early-stage prostate and breast cancer.

Abstract

194 Background: ASCO has recently selected “Choosing Wisely” measures that identify physician health behaviors of low value in cancer care. Two of these measures address the use of advanced imaging for early stage disease. These measures aim to reduce imaging in patients at low risk for metastatic disease unless there is a clinical indication. To date, no evidence-based benchmarks for meeting these measures have been reported. Methods: We analyzed all patients from January 2010 to June 2012 at the Duke Cancer Institute (DCI), an NCI-designated, comprehensive cancer center. We investigated conformance to two Choosing Wisely measures: avoiding imaging for staging in early stage (ES) prostate cancer (PC) and ES breast cancer (BC). ES was defined by Choosing Wisely as Stage IIb or less for BC and Gleason <7 or PSA <10 ng/mL for PC. Advanced imaging was defined as bone scan, computed tomography (CT) scan, or positron emission tomography (PET) performed at the DCI within 60 days of cancer diagnosis. Descriptive statistics and chi-square were performed. Results: Total 1143 BC and 29 PC were identified. Median age was 58 and 61 years, respectively. 0% (0/29) of PC cases had advanced imaging. Within BC, 20.6% (235/1143) had at least one imaging procedure performed; 16.5% had two or more. Patients with imaging were more likely hormone receptor negative, triple negative, younger (<50), and higher stage (Stage IIb), (all p<0.0001). Of imaging performed in BC, 41% were CT only, 22% were PET or PET/CT, and 36% were bone scans. Ongoing chart abstractions are identifying clinical indications for the ordering of imaging and associated clinical consequences. Conclusions: We have established internal benchmarks for conformance to two Choosing Wisely measures. The lack of advanced imaging for low-risk PC patients demonstrates the feasibility of this measure even in a multi-disciplinary setting. Imaging in <25% low-risk BC patients suggests that these are performed for clinical rather than staging purposes. Further data sharing and comparisons are needed to establish oncology-wide benchmarks.