Bariatric Centers of Excellence: The Effect of Centralization on Access to Care

Lindsay E Kuo, MD, Kristina D Simmons, PhD, Rachel R Kelz, MD, MSCE. Hospital of the University of Pennsylvania

?Background: In 2006, Centers for Medicare & Medicaid Services restricted coverage for bariatric procedures to designated high-volume Centers of Excellence (COEs) in an effort to improve outcomes. This policy has subsequently been reversed as the resultant improvement in outcomes was not substantiated. Regionalization remains a strategy recommended to improve outcomes for other complex surgical procedures. In an attempt to examine the effects of regionalization on access to care, we used the bariatric surgery experience as a model. We sought to examine the effect of the Medicare coverage restriction to COEs on access to bariatric surgery for all patients.

Methods: Inpatient claims data from three state databases (California, Florida and New York) from 2011 was used. All obese patients 18 years of age or older who underwent a bariatric surgical procedure were included. The three states were divided into hospital service areas (HSAs) and the per capita rates of bariatric surgery and minimally invasive bariatric surgery were calculated for each HSA with adjustment for age and race using the direct method. The number of designated bariatric COEs per 100,000 obese residents was used to divide the HSAs into quintiles. Standardized per capita rates of performance between HSAs were compared using ANOVA.

Results: In 2011, there were 132 COEs. Centers of Excellence were not widely distributed throughout the three states: out of 65 HSAs, only 37 contained one or more COEs. In HSAs without a COE, only 1.75 bariatric procedures were performed per 100,000 residents, compared to 11.20 in the HSAs in the lowest quintile of COEs and 38.37 in the highest-quintile HSAs (p<0.001). See Figure 1 below. In HSAs with no COEs, only 12.1% of bariatric surgical procedures were performed via minimally invasive techniques, compared to 65.5% in the lowest-quintile HSAs and 71.8% in the highest-quintile HSAs (p<0.001).

Conclusions: HSAs without COEs performed fewer bariatric surgery procedures than HSAs with COEs. Rates of MIS performance were relatively low in regions without COEs compared to regions with COEs. Patients residing in HSAs without COEs were less likely to undergo a bariatric surgical procedure and were less likely to receive the gold standard minimally invasive procedure than patients residing in HSAs with a COE. In this study using bariatric procedure policy as a model for the effects of centralization of surgical procedures by Medicare, we found that the coverage restriction resulted in centralization of bariatric surgery for all patients and diminished access to care for people residing in outlier regions. This pattern of care should be investigated further prior to the regionalization of other complex surgical procedures.


« Return to SAGES 2015 abstract archive