Addressing Critiques of the Proposed CMS Metric of Organ Procurement Organ Performance: More Data Isn't Better. Issue 8 (August 2020)

• Record Type: Journal Article
• Authors: Goldberg, David S.; Doby, Brianna; Lynch, Raymond
• Is Part Of: Transplantation. Volume 104:Issue 8(2020)
• Publication Date: 2020-08
• Subjects: Transplantation of organs, tissues, etc -- Periodicals; Transplantation immunology -- Periodicals; 617.95
• DOI: 10.1097/TP.0000000000003071
• Languages: English
• ISSNs: 0041-1337

Abstract

Abstract : Background: Centers for Medicare and Medicaid Services (CMS) has proposed a rule change to redefine the metric by which organ procurement organizations (OPOs) are evaluated. The metric relies on Centers for Disease Control and Prevention (CDC) data on inpatient deaths from causes consistent with donation among patients <75 years of age. Concerns have been raised that this metric does not account for rates of ventilation, and prevalence of cancer and severe sepsis, without objective data to substantiate or refute such concerns. Methods: We estimated OPO-level donation rates using CDC data, and used Agency for Healthcare Research and Quality/Healthcare Cost and Utilization Project data from 43 State Inpatient Databases to calculate "adjusted" donation rates. Results: The CMS metric and the ventilation-adjusted CMS metric were highly concordant in absolute terms (Spearman and Pearson correlation coefficients ≥0.95). In the Bland-Altman plot, 100% (48/48) of paired values (standard deviations [SDs] of the CMS and "ventilation adjusted" metrics) were within 1.96 SDs of the mean difference, with near-perfect correlation in Passing and Bablok regression (Lin's concordance correlation coefficient: 0.97). The CMS metric and the ventilation/cancer/sepsis-adjusted metric were highly concordant in absolute terms (Spearman and Pearson correlation coefficients ≥0.94). In the Bland-Altman plot, 97.9% (47/48) of paired values (SDs of the CMS and "ventilation/cancer/sepsis adjusted" metrics) were within 1.96 SDs of the mean difference, with near-perfect correlation in the Passing and Bablok regression (Lin's concordance correlation coefficient: 0.97). Conclusions: These conclusions should provide CMS, and the transplant community, with comfort that the proposed CMS metric using CDC inpatient death data as a tool to compare OPO is not compromised by its lack of inclusion of ventilation or other comorbidity data.