MP11-06 REGIONAL-LEVEL INAPPROPRIATE IMAGING RATES FOR PROSTATE AND BREAST CANCERS ARE CORRELATED: POTENTIAL LESSONS FOR THE CHOOSING WISELY CAMPAIGN

Abstract

You have accessJournal of UrologyGeneral & Epidemiological Trends & Socioeconomics: Practice Patterns, Cost Effectiveness II1 Apr 2014MP11-06 REGIONAL-LEVEL INAPPROPRIATE IMAGING RATES FOR PROSTATE AND BREAST CANCERS ARE CORRELATED: POTENTIAL LESSONS FOR THE CHOOSING WISELY CAMPAIGN Danil Makarov, Sounok Sen, Pamela Soulos, Heather Gold, James Yu, Joseph Ross, and Cary Gross Danil MakarovDanil Makarov More articles by this author , Sounok SenSounok Sen More articles by this author , Pamela SoulosPamela Soulos More articles by this author , Heather GoldHeather Gold More articles by this author , James YuJames Yu More articles by this author , Joseph RossJoseph Ross More articles by this author , and Cary GrossCary Gross More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2014.02.421AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Signal detection theory suggests the criterion for medical decision making varies between individuals. It is unknown whether hospital referral regions (HRRs) similarly have characteristic criteria. To answer this question, we sought to determine whether rates of inappropriate imaging among incident prostate and breast cancer patients were associated at the HRR-level. METHODS We identified low risk prostate and breast cancer patients from the SEER-Medicare database as defined in the Choosing Wisely Campaign. Our dependent variable of interest was the HRR-level rates of advanced imaging among low risk prostate cancer patients. Our independent variable of interest was the HRR-level rates of breast cancer imaging among low risk patients. We calculated the Pearson correlation between HRR-level imaging rates, weighted by the inverse variance of the breast cancer imaging rates. We also used logistic regression to model prostate cancer imaging rates as a function of breast cancer imaging rates, adjusted for patient- and environmental-level covariates. RESULTS We identified 9,219 male Medicare beneficiaries with prostate cancer, and 30,398 female beneficiaries with breast cancer. Prostate cancer imaging rates were associated with breast cancer imaging rates, with a Pearson correlation coefficient of 0.35 (p<0.01). Adjusted for age and comorbidity, a man with low risk prostate cancer had odds ratios of 1.87 (1.21-2.88), 1.17 (0.77-1.80) and 1.73 (1.12-2.69) to undergo inappropriate imaging if he were living in the fourth, third and second quartiles of inappropriate breast cancer imaging, as compared to living in the lowest quartile of breast cancer imaging. CONCLUSIONS At a regional-level, there is an association in imaging rates between prostate and breast cancers, a pair of diseases diagnosed in non-overlapping populations and cared for by non-overlapping providers. This finding confirms the predictions of signal detection theory that the criterion for medical decision making may be characteristic of a region. It also suggests that policy efforts should seek to improve quality of care and reduce healthcare spending in high-utilization areas, rather than focus on professional societies or specific-disease support groups. © 2014FiguresReferencesRelatedDetails Volume 191Issue 4SApril 2014Page: e97 Advertisement Copyright & Permissions© 2014MetricsAuthor Information Danil Makarov More articles by this author Sounok Sen More articles by this author Pamela Soulos More articles by this author Heather Gold More articles by this author James Yu More articles by this author Joseph Ross More articles by this author Cary Gross More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...