Assessment of the current Medicare reimbursement system for breast cancer operations

Background: Medicare determines procedural reimbursement by means of formulas considering physician work, practice, and liability expenses. Since no mechanism exists to consider outcomes in calculating reimbursements, we hypothesized that Medicare reimbursements do not correlate with outcomes for different breast cancer operations. Methods: We prospectively studied 240 patients with T1, 2N0M0 breast cancer in three surgical treatment arms: segmental mastectomy with axillary node dissection (SM&ALND n = 42); SM with sentinel node dissection (SM&SLND n = 96); and mastectomy without reconstruction (MRM; n = 102). Outcome parameters of complications, hospital stay, analgesic usage, and days to return to work were correlated with procedure reimbursements. Results: Median follow-up was 26 months. SM&SLND patients rarely required hospital stays (14%) in comparison with either SM&ALND (96%) or MRM patients (99%) (P < 0.001). SM&ALND and MRM patients required 9 and 10 median days of narcotics, respectively, versus 1 day in the SLND group (P < 0.001). SM&SLND patients returned to work at a median of 3 days, in comparison with 19 for SM&ALND and 26 for MRM patients (P < 0.001). Complications were more common in the MRM group (67% numbness/10% pain) and the SM&ALND group (56%/9%) than in the SM&SLND group (0%/1%). Reimbursements were inversely correlated with outcomes. MRM was reimbursed the highest, at an average of $1,075.03, with SM&ALND at $882.72. SM&SLND was reimbursed at $642.00. Conclusions: Medicare reimbursements for breast cancer operations do not correlate with outcomes. Less-invasive procedures are paid for at lower rates despite better outcomes and fewer complications. The data from this study raise the question of the impact of reimbursement on breast procedure selection.