Centers for Medicare and Medicaid Services (CMS) is making small adjustments to physician reimbursement to improve the value of care delivered to Medicare beneficiaries under fee-for-service. Using localized prostate cancer as a model, we examined whether these small reimbursement changes were enough to increase the value of cancer care.
Small variations in fee-for-service reimbursement for costly physician-administered prostate cancer drugs in the early 2000s did not reduce low-value care, suggesting that small payment changes may not be enough to motivate improvements in care quality. Moreover, patient clinical characteristics and practice organization were associated with low-value care (Table 3).
We conducted a natural experiment exploiting unintended differences in regional Medicare carriers' payment of Part B drug claims. Using data from SEER registries that are linked to Medicare claims, we conducted multilevel analysis to assess whether the reimbursement variation affected urologists' inappropriate use of gonadotropin-releasing hormone (GnRH) agonists. These drugs are often used, and misused, in prostate cancer treatment.
We conducted our study among urologists. Much of cancer care is delivered by medical and radiation oncologists and our results may not be generalizable to these providers. Future studies should address these groups specifically to compare how reimbursement changes, both small and large, affect prescribing of GnRH agonists to complete our understanding of the clinical and practice contexts in which reimbursement does or does not change clinical decision making.
CMS' efforts to align reimbursement with high quality care should continue. However, our findings suggest that small adjustments to reimbursement on the order of 2% to 3% may be insufficient to improve the value of care delivered by urologists treating localized prostate cancer. Other efforts may also be needed to change physicians' behavior, such as CMS' identification of low-performing practices through publication of quality ratings and targeted, physician-centered interventions to support health care providers who are not adherent to guidelines.jop;JOP.2015.007344v1/T03T1T03Table 3.Multilevel Regression Predicting Primary GnRH Agonist UseFactorOdds Ratio95% CIReimbursement generosity index1.001.00 to 1.00Practice type Group practiceReference Solo practice1.271.07 to 1.52 Missing0.860.58 to 1.26Physician prostate panel size, prostate patients per year < 20 Reference 21-37 0.750.65 to 0.86 ≥ 38 0.810.68 to 0.96T stage T1Reference T21.911.72 to 2.13Grade Well differentiated, 2-4Reference Moderately differentiated, 5-72.261.78 to 2.88 Missing3.502.50 to 4.91No. of comorbidities 0Reference 11.401.24 to 1.57 21.431.18 to 1.72 ≥ 3 2.071.65 to 2.60Age, years1.781.38 to 2.30Age, squared1.001.00 to 1.00Race/ethnicity Non-Hispanic whiteReference Non-Hispanic black1.421.17 to 1.73 Hispanic1.251.00 to 1.56 Other1.300.94 to 1.79 Missing1.951.50 to 2.53Radiation oncology consultation NoReference Yes0.260.23 to 0.31Medical oncology consultation NoReference Yes0.880.65 to 1.19Urology consultation NoReference Yes5.623.19 to 9.90Primary care consultation NoReference Yes0.410.36 to 0.46Constant0.950.87 to 1.04NOTE. Sample includes only patients (n = 15,128) of urologists who prescribed GnRH agonists. The model also adjusts for urologists' length of time in practice, sex, training location, board certification status, medical school affiliation, proportion of minority patients in practice, patients' marital status, rural residence, community educational attainment, community income, prior primary care use, year treated, and SEER region.Abbreviation: GnRH, gonadotropin-releasing hormone.
Journal of oncology practice / American Society of Clinical Oncology. 2016 Mar 08 [Epub ahead of print]
Shellie D Ellis, Ronald C Chen, Stacie B Dusetzina, Stephanie B Wheeler, George L Jackson, Matthew E Nielsen, William R Carpenter, Morris Weinberger, Shellie D Ellis, Ronald C Chen, Stacie B Dusetzina, Stephanie B Wheeler, George L Jackson, Matthew E Nielsen, William R Carpenter, Morris Weinberger
University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC ., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC. ., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC., University of Kansas School of Medicine, Kansas City, KS; University of North Carolina at Chapel Hill, Chapel Hill; Durham Veterans Affairs Medical Center; and Duke University Medical Center, Durham, NC.