Diffusion of Laparoscopic Adjustable Gastric Banding after the Medicare National Coverage Decision for Bariatric Surgery
We used State Inpatient Data from 12 large, geographically diverse states to track diffusion of the LAP-BAND among Medicare patients and spillovers to non-Medicare patients. The sample includes 315,696 patients undergoing weight-loss surgery for morbid obesity at 522 hospitals between 2003 and 2009. We find that the NCD caused a 7.7 percentage point increase in the probability of LAP-BAND receipt relative to other bariatric procedures for Medicare beneficiaries. Use of the LAP-BAND spread most rapidly in hospitals that were just below the Center of Excellence volume threshold at the time of the NCD. Medicare patients induced to receive a LAP-BAND as a result of the NCD were less likely to die and experience a complication. However, these improvements in short-term safety came at the cost of lower probability of long-term success; weight loss was 35% greater with gastric bypass versus LAP-BAND. We will estimate the effects of this change in procedure mix on long-term health outcomes of bariatric surgery patients.
Hospitals responded to a volume-based selective referral policy by increasing use of a less effective procedure. Increased use of this lower-skill procedure was associated with improvements in short-term outcomes at the cost of limited weight loss in the long-run. These potentially unintended consequences of the coverage decision highlight the challenges payers and policymakers face in legislating quality.