MP34-15 PROSTATE CANCER QUALITY OF CARE DISPARITIES AND THEIR IMPACT ON PATIENT OUTCOMES

Abstract

You have accessJournal of UrologyProstate Cancer: Epidemiology & Natural History II1 Apr 2018MP34-15 PROSTATE CANCER QUALITY OF CARE DISPARITIES AND THEIR IMPACT ON PATIENT OUTCOMES Keith Lawson, Katherine Daignault, Olli Saarela, Robert Abouassaly, and Antonio Finelli Keith LawsonKeith Lawson More articles by this author , Katherine DaignaultKatherine Daignault More articles by this author , Olli SaarelaOlli Saarela More articles by this author , Robert AbouassalyRobert Abouassaly More articles by this author , and Antonio FinelliAntonio Finelli More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2018.02.1107AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES The ability to rigorously measure disparities in care that drive poor patient outcomes is essential for developing effective strategies for quality improvement. As real-world case-mix adjusted quality variations have yet to be reported in prostate cancer, our objective was to apply a data-driven analytical framework to determine the validity of expert-defined quality indicators (QIs) in order to reveal true disparities in prostate cancer care and their impact on patient outcomes. METHODS Hospital-level quality of care was assessed according to 10 published, expert-defined QIs utilizing data derived from the United States National Cancer Database (NCDB). Case-mix adjusted hospital benchmarking was performed using indirect standardization methodology and multivariable regression models as previously described by our group [1]. In a training set of hospitals, a composite measure of hospital quality, the Prostate Cancer Quality Score (PC-QS), was derived from those individual QIs that discriminated hospital performance in a manner associated with poor patient outcomes; including the need for salvage therapy, initiation of androgen deprivation therapy, 30- and 90-day mortality, and overall mortality. Subsequently, associations between the PC-QS and hospital volume, academic affiliation, and patient outcomes (as above) were determined in a separate validation set of hospitals. RESULTS Collectively, data from over 1100 hospitals were analyzed, with widespread variation observed across all QIs. In the training set, between 3-36% of hospitals were identified as poor outliers compared to the national average level of care for a given QI. Based on associations between outlier status and the aforementioned patient outcomes, 5 individual QIs were incorporated into the PC-QS. In the validation set, lower PC-QS hospitals displayed smaller referral volumes and were less likely to be academic hospitals than those with higher PC-QS (p < 0.001). Higher PC-QS was associated with lower rates of salvage therapy, ADT initiation and 30-day mortality (adjusted hazard ratio [confidence interval]: 0.85 [0.79-0.91], 0.94 [0.91-0.98], adjusted odds ratio 0.92 [0.87-0.97] per PC-QS unit increase, respectively). CONCLUSIONS Data-driven benchmarking of hospital-level quality performance reveals the widespread disparities that exist in prostate cancer care. This data supports the use of our novel PC-QS composite metric as a benchmarking tool for quality improvement. 1. Lawson KA et. al. Eur Urol 2017; 72: 379-386. © 2018FiguresReferencesRelatedDetails Volume 199Issue 4SApril 2018Page: e443 Advertisement Copyright & Permissions© 2018MetricsAuthor Information Keith Lawson More articles by this author Katherine Daignault More articles by this author Olli Saarela More articles by this author Robert Abouassaly More articles by this author Antonio Finelli More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...