Impact of Centers for Medicare & Medicaid Services national coverage determination on erythropoiesis‐stimulating agent and transfusion use in chemotherapy‐treated cancer patients

Abstract

In July 2007, the Centers for Medicare & Medicaid Services released a national coverage determination (NCD) for erythropoiesis-stimulating agent (ESA) use in cancer patients, mandating payment restrictions likely to reduce ESA use and possibly increase red blood cell transfusions. We aimed to quantify ESA and transfusion use pre-NCD and post-NCD. Medicare 5% sample data, 2005-2007, were used. Patients were 66 years or older, had lung, breast, or colorectal cancer or lymphomas, and initiated chemotherapy in pre-NCD and post-NCD periods (September-November 2006, September-November 2007). ESA use and transfusions were identified from claims. Differences in proportions of patients using ESAs and receiving transfusions pre-NCD and post-NCD were evaluated using logistic regression; differences in transfusion event rates were evaluated using a Poisson model. The pre-NCD cohort included 1897 patients and the post-NCD cohort 1877. In the pre-NCD cohort, 31% of patients had lung cancer, 29% lymphoma, 20% colorectal cancer, and 20% breast cancer; distribution was similar in the post-NCD cohort. Overall, ESA use decreased from 35.0% pre-NCD to 15.2% post-NCD. Transfusion use increased from 9.3% to 10.4%, and transfusion event rates from 19.0 to 21.8 per 100 patient-quarters. Results adjusted for baseline characteristics and comorbid conditions were similar. ESA use reduction achieved statistical significance; transfusion use and rate increases did not. ESA use decreased sharply post-NCD. This was accompanied by an estimated 1.1% (95% confidence interval -0.8% to 3.0%) absolute increase in transfusion use.