K. R. Chhabra1,2,3, J. Yang1, D. A. Telem1,4, J. R. Thumma1, J. B. Dimick1,4 1University Of Michigan,Center For Healthcare Outcomes And Policy,Ann Arbor, MI, USA 2University Of Michigan,IHPI Clinician Scholars Program,Ann Arbor, MI, USA 3Brigham And Women’s Hospital,Department Of Surgery,Boston, MA, USA 4University Of Michigan,Department Of Surgery,Ann Arbor, MI, USA
Introduction:
Sleeve gastrectomy has rapidly become the most common bariatric operation performed in the United States. Though sleeve gastrectomy has an excellent short-term safety profile, its longer-term safety is poorly described. Prior studies comparing sleeve gastrectomy to gastric bypass are limited by low sample size (in randomized trials) and selection bias (in observational studies). Traditional observational methods cannot fully account for unmeasured confounding, such as the increased disease severity and comorbidity burden of patients selected for gastric bypass.
Methods:
We compare the safety of sleeve gastrectomy and gastric bypass in the fee-for-service Medicare population and the commercially insured population of the IBM MarketScan claims database. We identified re-interventions and complications from 30 days to 2 years from surgery using CPT and ICD-9/10 codes. To overcome unmeasured confounding, we use the prior year’s sleeve gastrectomy utilization within each state as an instrumental variable. Instrumental variables are an econometric technique that uses external causes of variation as a natural experiment, to allow causal inferences from observational data. Thus, we exploited the variation in the timing of payers’ decisions to cover sleeve gastrectomy as a natural experiment.
Results:
Among 90,877 patients who underwent bariatric surgery, the share of sleeve gastrectomy rose from 6.08% to 61.49% (Medicare) and 52.60% to 72.07% (commercial) from 2012-2015. At 30 days, sleeve gastrectomy had a lower re-intervention rate relative to bypass (relative risk 0.6 in Medicare, 0.4 in commercial), readmission rate (RR 0.7 in Medicare, 0.6 in commercial), and lower ED visit rate in commercially insured patients only (RR 0.7, all p<.001). In Medicare patients, there was no difference in ED visit rates between sleeve and bypass at 30 days. At 2 years, sleeve continued to have lower rates of re-intervention, readmission, and ED visits across both payers—though the difference between sleeve and bypass narrowed at the 2-year mark. The rates of revisional surgery were higher in sleeve gastrectomy, though this did not achieve statistical significance. In the commercially insured, total healthcare spending was lower for sleeve gastrectomy at 2 years ($46,640 vs. bypass $54,070, p<.001), but in Medicare populations spending was statistically equivalent.
Conclusion:
In both the Medicare and commercially insured populations, for patients eligible for both sleeve and bypass, sleeve gastrectomy has a superior safety profile up to 2 years from surgery. Sleeve gastrectomy also causes lower postoperative health spending than bypass in commercially insured populations.
