Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jun 1.
Published in final edited form as: Ann Surg. 2022 Aug 29;277(6):979–987. doi: 10.1097/SLA.0000000000005692

Comparative Safety of Sleeve Gastrectomy and Gastric Bypass Up to 5 Years After Surgery in Patients with Medicaid

Ryan Howard 1,2, Jie Yang 2, Jyothi Thumma 2, Anne Ehlers 1,2, Sean O’Neill 1,2, David Arterburn 3, Andrew Ryan 2,4,5, Dana Telem 1,2,6, Justin B Dimick 1,2,6
PMCID: PMC9971342  NIHMSID: NIHMS1830941  PMID: 36036493

Structured Abstract

Objective:

Compare adverse outcomes up to 5 years after sleeve gastrectomy and gastric bypass in patients with Medicaid.

Summary Background Data:

Sleeve gastrectomy is the most common bariatric operation among patients with Medicaid, however its long-term safety in this population is unknown.

Methods:

Using Medicaid claims, we performed a retrospective cohort study of adult patients who underwent sleeve gastrectomy or gastric bypass from January 1, 2012 to December 31, 2018. Instrumental variables survival analysis was used to estimate the cumulative incidence and heterogeneity of outcomes up to 5 years after surgery.

Results:

Among 132,788 patients with Medicaid, 84,717 (63.8%) underwent sleeve gastrectomy and 48,071 (36.2%) underwent gastric bypass. 69,225 (52.1%) patients were White, 33,833 (25.5%) were Black, and 29,730 (22.4%) were Hispanic. Compared to gastric bypass, sleeve gastrectomy was associated with a lower 5-year cumulative incidence of mortality (1.29% vs. 2.15%), complications (11.5% vs. 16.2%), hospitalization (43.7% vs. 53.7%), ED use (61.6% vs. 68.2%), and reoperation (18.5% vs. 22.8%), but a higher cumulative incidence of revision (3.3% vs. 2.0%). Compared to White patients, the magnitude of the difference between sleeve and bypass was smaller among Black patients for ED use (5-year aHR 1.01 [95% CI 0.94−1.08] vs. 0.94 [95% CI 0.88-1.00], P<.001) and Hispanic patients for reoperation (5-year aHR 0.95 [95% CI 0.86−1.05] vs. 0.76 [95% CI 0.69−0.83], P<.001).

Conclusions:

Among patients with Medicaid undergoing bariatric surgery, sleeve gastrectomy was associated with a lower risk of mortality, complications, hospitalization, ED use, and reoperations, but a higher risk of revision compared to gastric bypass. Although the difference between sleeve and bypass was generally similar among White, Black, and Hispanic patients, the magnitude of this difference was smaller among Black patients for ED use and Hispanic patients for reoperation.

Mini Abstract

Among 132,788 patients with Medicaid, sleeve gastrectomy was associated with a lower risk of mortality, complications, hospitalization, emergency department use, and reoperations, but a higher risk of revision. These results were similar across racial and ethnic groups except for emergency department use in Black patients and reoperation in Hispanic patients.

Introduction

Over the past decade, sleeve gastrectomy has rapidly replaced gastric bypass as the most common surgical treatment of severe obesity in the United States.1 Comprising fewer than 10% of bariatric procedures just a decade ago, sleeve gastrectomy now accounts for nearly 70% of all bariatric procedures nationally.2 Sleeve gastrectomy has also overtaken gastric bypass as the most common bariatric procedure among the 75 million Americans enrolled in Medicaid, a trend accelerated by coverage of sleeve gastrectomy by the Centers for Medicare and Medicaid Services (CMS) in 2012 and the passage of the Affordable Care Act in 2014.3,4

Despite the rapid adoption of sleeve gastrectomy among patients with Medicaid, its long-term outcomes in this population have yet to be described. Previous claims-based studies comparing sleeve gastrectomy and gastric bypass have evaluated patients with commercial insurance or Medicare.5–8 However, patient characteristics and outcomes in these groups differ from patients with Medicaid. For example, a systematic review found that short-term mortality and health care utilization were higher after bariatric surgery among patients with Medicaid.9 Moreover, many of these studies are limited to short-term outcomes, small cohorts of patients, and administrative databases in which patients with Medicaid comprise less than 20% of the cohort.10–12 This lack of evaluation of patients with Medicaid also impedes our understanding of the comparative risks and benefits of these two procedures in minority and vulnerable populations. Although prior work suggests that Black and Hispanic patients may have higher short-term complication rates after bariatric surgery, to date no study has assessed these outcomes in a large sample of Medicaid patients with long-term follow-up.13,14

Therefore, we used national Medicaid claims to assess the long-term comparative outcomes of sleeve gastrectomy and gastric bypass up to 5 years after surgery. Given the demographic composition of patients with Medicaid, we were able to evaluate whether the previously described superior safety profile of sleeve gastrectomy differed among minority subgroups. We hypothesized that sleeve gastrectomy would be associated with a lower incidence of adverse events and healthcare utilization compared to gastric bypass, similar to what has been previously found in patients with private insurance and Medicare, but that the magnitude of this difference would vary between subgroups of minority patients.

Methods

Data Source and Study Cohort

We used claims from the Inpatient (IP), Other Services (OT), and Personal Summary (PS) files in the Medicaid Analytic Extract (MAX) and Transformed Medicaid Statistical Information System Analytic Files (TAF) to identify patients with severe obesity undergoing laparoscopic sleeve gastrectomy and gastric bypass between January 1, 2012 and December 31, 2018. Severe obesity was identified using International Classification of Diseases 9th and 10th Edition (ICD-9/10) diagnosis codes. Laparoscopic sleeve gastrectomy and gastric bypass were identified using Current Procedural Terminology (CPT) and ICD-9/10 codes. Patients with diagnoses of small bowel cancer and patients who did not have continuous enrollment in Medicaid for at least one year prior to surgery were excluded.

This secondary analysis of de-identified administrative data was exempted from regulation by the University of Michigan Institutional Review Board and the requirement for informed consent was waived. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Outcomes and Covariates

The outcomes of this study were chosen based upon bariatric surgical outcomes previously described in patients with commercial insurance and Medicare.5,7 Therefore, the outcomes of interest were mortality, complications, all-cause hospitalization, emergency department (ED) use, revision, and reoperation. Mortality was identified as death of a patient according to the Personal Summary File. Complications were identified using established methods for identifying splenic, hemorrhagic, anastomotic, wound-related, obstruction-related, pulmonary, cardiac, neurologic, genitourinary, thromboembolic, and shock-related complications (Supplemental Table 1).7 Hospitalization and ED use were identified with revenue center codes previously published by the Research Data Assistance Center.15 An emergency department visit that resulted in hospitalization was categorized as hospitalization only.

Postoperative revisions and reoperations were identified using previously published CPT codes (Supplemental Table 2). Revision is of particular interest in bariatric surgery given prior evidence that the incidence of revision is higher after sleeve gastrectomy than gastric bypass.7 Revisions included any subsequent operation that involved directly modifying the index bariatric procedure, such as gastrectomy, anastomotic revision, or conversion of a sleeve gastrectomy to a gastric bypass. Reoperations included any subsequent abdominal operation potentially related to the index bariatric procedure, such as abdominal wall hernia repair, biliary procedures, internal hernia repair, and paraesophageal hernia repair. These procedures were chosen as they have previously been investigated as related to index bariatric procedures in the literature, and we have used them extensively in prior evaluations of bariatric surgical outcomes in patients with Medicare and private insurance.5,7,8,16–18

Covariates included in this study were patient age, sex, race and ethnicity, year of surgery, type of operation, and 29 Elixhauser comorbidities. Race and ethnicity were defined in Medicaid claims using the following categories: White, Black, Hispanic (including Hispanic or Latino), American Indian or Alaskan Native, Asian or Pacific Islander, Native Hawaiian or other Pacific Islander, more than one race, and unknown. Given that a goal of this study was to compare differences between sleeve gastrectomy and gastric bypass by race and ethnicity, we excluded patients with unknown race. Additionally, in accordance with guidance on reporting and analyzing race and ethnicity from the American Medical Association, due to extremely small sample size we also excluded patients identified as American Indian or Alaskan Native, Asian or Pacific Islander, Native Hawaiian or other Pacific Islander, and more than one race, since each of these groups comprised less than 1% of the overall sample.19,20 Therefore, our study ultimately included patients identified as White, Black, and Hispanic.

Statistical Analysis

Descriptive statistics were calculated for all covariates. Baseline differences between patients undergoing sleeve gastrectomy and gastric bypass were calculated using the Chi-squared test or unpaired t-test.

This study used an instrumental variable to evaluate the independent association of each procedure with outcomes while mitigating selection bias inherent to observational study design. The instrumental variable, which we have used and described previously, was the state-level rate of sleeve gastrectomy in the prior year.5–7 We chose this instrument because in June 2012, CMS initiated coverage of sleeve gastrectomy for the treatment of severe obesity.21 Following this decision, adoption of this procedure varied significantly between states. This variation allowed for a natural experiment in which treatment choice was “pseudo-randomized”: a patient who just so happened to live in a region with high sleeve gastrectomy utilization in the prior year was more likely to undergo this operation, whereas a patient who just so happened to live in an area with low utilization was less likely to undergo this operation, independent of their individual patient characteristics. This instrument was strongly associated with treatment (F statistic=660.6). An F statistic greater than 10 is considered strong.22

The main analysis of this study used a 2-stage residual inclusion estimation method to estimate the adjusted hazard ratios (aHR) and cumulative incidence of outcomes after sleeve gastrectomy compared to gastric bypass. We have previously described and employed this method.7 In the first stage, multivariable logistic regression was performed to estimate the likelihood that a patient would undergo sleeve gastrectomy. These models included the following covariates: previous-year state-level sleeve gastrectomy rate, age, sex, race and ethnicity, comorbidities, and year of surgery. In the second stage, Cox proportional hazards regression models were constructed to estimate the aHR and cumulative incidence of outcomes. These models included the following covariates: treatment (sleeve gastrectomy vs. gastric bypass), age, sex, race and ethnicity, comorbidities, year of surgery, and residuals from the first-stage regression model which represent unobserved confounding associated with treatment choice. Both models included an interaction term between treatment and race/ethnicity to evaluate whether the magnitude of difference in outcomes between sleeve and bypass was significantly different between White, Black, and Hispanic patients (i.e., if sleeve is associated with decreased mortality compared to bypass, is this difference in mortality between the two procedures the same or different between White, Black, and Hispanic patients). In both models, the included covariates were selected a priori because they are factors known to be associated with outcomes after bariatric surgery.23–25 Both models accounted for clustering at the state level and robust standard errors were used to account for state-level heteroscedasticity.

Censoring criteria in the Cox model included death, disenrollment from Medicaid, or reaching the end of the follow-up period (December 31, 2019) before an outcome occurred. Assessment of Schoenfeld residuals revealed that the proportional hazards assumption was violated in all outcomes except revision, indicating that the treatment effect varies over time. Therefore, covariates in those models were interacted with time, and aHRs at 1, 3, and 5 years after surgery were estimated. The cumulative incidence of each outcome was calculated with covariates in the Cox proportional hazards model held at their mean. Finally, descriptive statistics were also calculated for individual complication and reoperation categories (Supplemental Table 3).

All statistical tests were performed using SAS version 9.4 (SAS Institute Inc) and Stata version 15.1 (StataCorp LLC). Tests were 2-sided and significance was set at P<.05. Data analyses were performed from August 2021 to May 2022.

Results

We identified 132,788 patients who underwent bariatric surgery during the study period, of whom 84,717 (63.8%) underwent sleeve gastrectomy and 48,071 (36.2%) underwent gastric bypass (Table 1). Mean age was 41.0 (11.0) years and 115,357 (86.9%) patients were female. 69,225 (52.1%) patients were White, 33,833 (25.5%) were Black, and 29,730 (22.4%) were Hispanic. Median (interquartile range) follow-up time was 540 (257–923) days. By 2018, Black patients underwent the highest proportion of sleeve gastrectomy (77.7%), followed by Hispanic patients (71.4%) and White patients (67.8%) (Figure 1).

Table 1 –

Cohort Characteristics

Characteristic All (N=132788) White (N=69225) Black (N=33833) Hispanic (N=29730)
Operation Sleeve Bypass P Sleeve Bypass P Sleeve Bypass P Sleeve Bypass P
Number of patients 84717 (63.8) 48071 (36.2) 41868 (60.5) 27357 (39.5) 23779 (70.3) 10054 (29.7) 19070 (64.1) 10660 (35.9)
Age (years, mean (SD)) 40.7 (11.0) 41.5 (10.8) <.001 41.6 (11.2) 42.0 (10.8) <.001 40.3 (10.4) 41.7 (10.4) <.001 39.3 (11.2) 39.8 (11.0) <.001
Female 73249 (86.5) 42108 (87.6) <.001 34975 (83.5) 23594 (86.2) <.001 21707 (91.3) 9193 (91.4) 0.655 16567 (86.9) 9321 (87.4) 0.164
Year of Surgery
 2012 3247 (3.8) 6504 (13.5) <.001 1340 (3.2) 3486 (12.7) <.001 881 (3.7) 1372 (13.7) <.001 1026 (5.4) 1646 (15.4) <.001
 2013 6759 (8.0) 6512 (13.6) 2968 (7.1) 3429 (12.5) 1895 (8.0) 1443 (14.4) 1896 (9.9) 1640 (15.4)
 2014 9750 (11.5) 6455 (13.4) 4759 (11.4) 3586 (13.1) 2609 (11.0) 1348 (13.4) 2382 (12.5) 1521 (14.3)
 2015 8665 (10.2) 4041 (8.4) 4419 (10.6) 2416 (8.8) 2236 (9.4) 762 (7.6) 2010 (10.5) 863 (8.1)
 2016 15957 (18.8) 7875 (16.4) 8181 (19.5) 4720 (17.3) 4501 (18.9) 1629 (16.2) 3275 (17.2) 1526 (14.3)
 2017 19861 (23.4) 8403 (17.5) 10056 (24.0) 4909 (17.9) 5705 (24.0) 1787 (17.8) 4100 (21.5) 1707 (16.0)
 2018 20478 (24.2) 8281 (17.2) 10145 (24.2) 4811 (17.6) 5952 (25.0) 1713 (17.0) 4381 (23.0) 1757 (16.5)
Elixhauser Comorbidities
 Hypertension 40087 (47.3) 23840 (49.6) <.001 19594 (46.8) 13224 (48.3) <.001 12887 (54.2) 5981 (59.5) <.001 7606 (39.9) 4635 (43.5) <.001
 Chronic pulmonary disease 19557 (23.1) 11899 (24.8) <.001 9390 (22.4) 6707 (24.5) <.001 5842 (24.6) 2676 (26.6) <.001 4325 (22.7) 2516 (23.6) 0.070
 Diabetes without chronic complications 18456 (21.8) 13093 (27.2) <.001 9055 (21.6) 7078 (25.9) <.001 5095 (21.4) 2969 (29.5) <.001 4306 (22.6) 3046 (28.6) <.001
 Depression 15220 (18.0) 10492 (21.8) <.001 9684 (23.1) 7337 (26.8) <.001 2872 (12.1) 1579 (15.7) <.001 2664 (14.0) 1576 (14.8) 0.054
 Liver disease 9476 (11.2) 7381 (15.4) <.001 4813 (11.5) 4526 (16.5) <.001 2194 (9.2) 1293 (12.9) <.001 2469 (13.0) 1562 (14.7) <.001
 Hypothyroidism 8085 (9.54) 5067 (10.5) <.001 5404 (12.9) 3640 (13.3) 0.128 1100 (4.6) 543 (5.4) 0.002 1581 (8.3) 884 (8.3) 0.995
 Deficiency anemias 5339 (6.3) 2856 (5.9) 0.009 1629 (3.9) 1173 (4.3) 0.010 2538 (10.7) 1060 (10.5) 0.723 1172 (6.2) 623 (5.8) 0.295
 Diabetes with chronic complications 3805 (4.5) 3108 (6.5) <.001 2028 (4.8) 1875 (6.9) <.001 1122 (4.7) 706 (7.0) <.001 655 (3.4) 527 (4.9) <.001
 Psychoses 3654 (4.3) 2597 (5.4) <.001 2377 (5.7) 1829 (6.7) <.001 740 (3.1) 428 (4.3) <.001 537 (2.8) 340 (3.2) 0.068
 Fluid and electrolyte disorders 2338 (2.8) 1494 (3.1) <.001 955 (2.3) 832 (3.0) <.001 912 (3.8) 376 (3.7) 0.675 471 (2.5) 286 (2.7) 0.263
 Congestive heart failure 2026 (2.4) 1141 (2.4) 0.837 961 (2.3) 654 (2.4) 0.417 830 (3.5) 382 (3.8) 0.162 235 (1.2) 105 (1.0) 0.054
 Other neurological disorders 1986 (2.3) 1281 (2.7) <.001 1332 (3.2) 943 (3.5) 0.055 394 (1.7) 216 (2.2) 0.002 260 (1.4) 122 (1.1) 0.108
 Renal failure 1752 (2.1) 857 (1.8) <.001 733 (1.8) 478 (1.8) 0.973 703 (3.0) 232 (2.3) 0.001 316 (1.7) 147 (1.4) 0.063
 Rheumatoid arthritis 1645 (1.9) 819 (1.7) 0.002 815 (2.0) 469 (1.7) 0.027 483 (2.0) 193 (1.9) 0.503 347 (1.8) 157 (1.5) 0.026
 Drug abuse 510 (0.6) 255 (0.5) 0.098 330 (0.8) 182 (0.7) 0.065 103 (0.4) 33 (0.3) 0.163 77 (0.4) 40 (0.4) 0.706
 Valvular disease 487 (0.6) 253 (0.5) 0.253 280 (0.7) 161 (0.6) 0.195 131 (0.6) 57 (0.6) 0.856 76 (0.4) 35 (0.3) 0.341
 Coagulopathy 396 (0.5) 251 (0.5) 0.169 222 (0.5) 164 (0.6) 0.232 97 (0.4) 42 (0.4) 0.897 77 (0.4) 45 (0.4) 0.812
 Peripheral vascular disease 349 (0.4) 231 (0.5) 0.069 214 (0.5) 159 (0.6) 0.218 83 (0.4) 41 (0.4) 0.414 52 (0.3) 31 (0.3) 0.776
 Paralysis 229 (0.3) 123 (0.3) 0.623 120 (0.3) 75 (0.3) 0.762 66 (0.3) 28 (0.3) 0.988 43 (0.2) 20 (0.2) 0.496
 Pulmonary circulation disease 204 (0.2) 192 (0.4) <.001 114 (0.3) 117 (0.4) 0.001 66 (0.3) 56 (0.6) <.001 24 (0.1) 19 (0.2) 0.254
 Alcohol abuse 182 (0.2) 83 (0.2) 0.098 88 (0.2) 54 (0.2) 0.716 55 (0.2) 19 (0.2) 0.446 39 (0.2) 10 (0.1) 0.024
 Chronic blood loss anemia 126 (0.2) 80 (0.2) 0.431 36 (0.1) 37 (0.1) 0.051 57 (0.2) 24 (0.2) 0.986 33 (0.2) 19 (0.2) 0.918
 Acquired immune deficiency syndrome 107 (0.1) 44 (0.1) 0.071 31 (0.1) 14 (0.1) 0.249 61 (0.3) 16 (0.2) 0.086 15 (0.1) 14 (0.1) 0.163
 Peptic ulcer disease 106 (0.1) 157 (0.3) <.001 61 (0.2) 106 (0.4) <.001 21 (0.1) 25 (0.3) <.001 24 (0.1) 26 (0.2) 0.017
 Weight loss 97 (0.1) 102 (0.2) <.001 54 (0.1) 61 (0.2) 0.003 27 (0.1) 23 (0.2) 0.012 16 (0.1) 18 (0.2) 0.038
 Solid tumor without metastasis 81 (0.1) 52 (0.1) 0.487 46 (0.1) 29 (0.1) 0.880 21 (0.1) 18 (0.2) 0.025 14 (0.1) --- 0.386
 Lymphoma 51 (0.1) 10 (<.01) 0.001 25 (0.1) --- 0.005 15 (0.1) --- 0.644 11 (0.1) --- 0.047
 Metastatic cancer --- --- 0.055 --- --- 0.553 --- --- 0.016 --- --- 0.103
Length of stay (days, mean (SD)) 2.05 (7.7) 2.49 (9.2) <.001 1.98 (7.6) 2.39 (7.5) <.001 2.15 (6.4) 2.69 (9.5) <.001 2.1 (9.1) 2.54 (12.2) 0.001
Open in a new tab

Covariates with fewer than 10 observations were censored according to Medicare/Medicaid policy.

Figure 1 –

Figure 1 –

Open in a new tab

Annual Proportion of Sleeve Gastrectomy in patients with Medicaid

Compared to gastric bypass, sleeve gastrectomy was associated with a lower 5-year cumulative incidence of mortality (1.29% [95% CI 1.27%−1.31%] vs. 2.15% [95% CI 2.14%−2.17%]), complications (11.5% [95% CI 11.4%−11.6%] vs. 16.2% [95% CI 16.2%−16.3%]), hospitalization (43.7% [95% CI 43.4%−44.1%] vs. 53.7% [95% CI 53.3%−54.1%]), ED use (61.6% [95% CI 61.5%−61.7%] vs. 68.2% [95% CI 68.1%−68.3%]), and reoperation (18.5% [95% CI 18.3%−18.6%] vs. 22.8% [95% CI 22.6%−23.1%]) (Figure 2, Table 2). However, sleeve gastrectomy was associated with a higher 5-year cumulative incidence of revisions (3.3% [95% CI 3.2%−3.3%] vs. 2.0% [95% CI 1.9%−2.0%]) compared to gastric bypass.

Figure 2 –

Figure 2 –

Open in a new tab

Adjusted Cumulative Incidence of Mortality, Complications, All-Cause Hospitalization, Emergency Department Use, Revision, and Reoperation After Sleeve Gastrectomy and Gastric Bypass

Cumulative incidence estimated using instrumental variable-adjusted Cox proportional hazards models with the following covariates: age, sex, race and ethnicity, Elixhauser comorbidities, and year of surgery. X-axis displays the cumulative number of patients with each outcome and the number of patients remaining at risk for each year of follow-up.

Table 2 –

Adjusted cumulative incidence of outcomes.

Outcome All White Black Hispanic
Sleeve Gastrectomy Gastric Bypass Sleeve Gastrectomy Gastric Bypass Sleeve Gastrectomy Gastric Bypass Sleeve Gastrectomy Gastric Bypass
Mortality
 Year 1 0.35 (0.34–0.35) 0.58 (0.58–0.58) 0.42 (0.42–0.42) 0.67 (0.67–0.67) 0.47 (0.47–0.47) 0.67 (0.66–0.67) 0.21 (0.21–0.22) 0.36 (0.35–0.35)
 Year 2 0.60 (0.59–0.60) 1.00 (0.99–1.00) 0.73 (0.72–0.73) 1.16 (1.15–1.16) 0.82 (0.81–0.82) 1.15 (1.14–1.15) 0.37 (0.36–0.37) 0.62 (0.61–0.62)
 Year 3 0.83 (0.82–0.84) 1.38 (1.38–1.39) 1.01 (1.00–1.01) 1.60 (1.59–1.61) 1.13 (1.12–1.13) 1.59 (1.58–1.60) 0.51 (0.50–0.52) 0.85 (0.84–0.86)
 Year 4 1.04 (1.03–1.06) 1.74 (1.73–1.76) 1.27 (1.26–1.28) 2.02 (2.01–2.03) 1.42 (1.42–1.43) 2.00 (1.99–2.01) 0.64 (0.63–0.66) 1.08 (1.07–1.09)
 Year 5 1.29 (1.27–1.31) 2.15 (2.14–2.17) 1.57 (1.55–1.58) 2.49 (2.47–2.51) 1.76 (1.74–1.77) 2.47 (2.45–2.49) 0.80 (0.78–0.81) 1.33 (1.31–1.35)
Complications
 Year 1 6.08 (6.06–6.10) 8.69 (8.67–8.71) 6.21 (6.20–6.22) 8.55 (8.53–8.57) 6.72 (6.71–6.73) 9.24 (9.21–9.27) 5.89 (5.86–5.91) 8.42 (8.40–8.45)
 Year 2 7.65 (7.62–7.68) 10.90 (10.87–10.92) 7.81 (7.79–7.83) 10.72 (10.70–10.74) 8.45 (8.43–8.46) 11.58 (11.54–11.62) 7.42 (7.39–7.44) 10.57 (10.54–10.60)
 Year 3 9.01 (8.96–9.05) 12.78 (12.76–12.82) 9.19 (9.17–9.21)) 12.58 (12.55–12.62) 9.94 (9.92–9.95) 13.58 (13.53–13.63) 8.73 (8.70–8.76) 12.40 (12.37–12.45)
 Year 4 10.30 (10.25–10.36) 14.58 (14.54–14.63) 10.52 (10.49–10.55) 14.36 (14.31–14.40) 11.36 (11.33–11.38) 15.48 (15.41–15.54) 9.99 (9.95–10.04) 14.16 (14.11–14.21)
 Year 5 11.48 (11.41–11.55) 16.21 (16.15–16.26) 11.72 (11.68–11.76) 15.96 (15.89–16.02) 12.65 (12.61–12.68) 17.19 (17.10–17.28) 11.14 (11.08–11.19) 15.73 (15.67–15.80)
Hospitalization
 Year 1 12.58 (12.53–12.62) 16.47 (16.43–16.52) 12.63 (12.59–12.67) 16.02 (15.96–16.08) 15.19 (15.10–15.27) 18.38 (18.27–18.49) 11.86 (11.80–11.92) 15.55 (15.46–15.65)
 Year 2 23.32 (23.21–23.44) 29.93 (29.80–30.06) 23.41 (23.31–23.52) 29.17 (29.02–29.32) 27.78 (27.64–27.91) 33.05 (32.87–33.23) 22.07 (21.92–22.22) 28.40 (28.19–28.60)
 Year 3 31.83 (31.64–32.03) 40.15 (39.94–40.36) 31.95 (31.78–32.12) 39.21 (38.97–39.45) 37.48 (37.30–37.66) 43.96 (43.72–44.20) 30.22 (29.98–30.47) 38.25 (37.9438.55)
 Year 4 38.63 (38.36–38.90) 48.00 (47.70–48.29) 38.76 (38.51–39.01) 46.96 (46.62–47.29) 45.02 (44.81–45.24) 52.18 (51.89–52.47) 36.77 (36.44–37.11) 45.88 (45.48–46.29)
 Year 5 43.73 (43.38–44.09) 53.71 (53.33–54.09) 43.88 (43.55–44.21) 52.61 (52.19–53.04) 50.58 (50.32–50.31) 58.06 (57.72–58.40) 41.73 (41.30–42.16) 51.48 (50.99–51.99)
ED Use
 Year 1 51.32 (51.26–51.38) 57.81 (57.76–57.87) 50.72 (50.68–50.76) 58.35 (58.27–58.44) 55.70 (55.66–55.75) 60.69 (60.60–60.79) 47.18 (47.07–47.29) 53.47 (53.35–53.60)
 Year 2 57.85 (57.78–57.92) 64.50 (64.44–64.56) 57.23 (57.18–57.27) 65.05 (64.95–65.14) 62.36 (62.30–62.41) 67.39 (67.29–67.49) 53.51 (53.39–53.63) 60.08 (59.95–60.21)
 Year 3 60.25 (60.18–60.32) 66.92 (66.86–66.98) 59.63 (59.57–59.67) 67.46 (67.36–67.55) 64.78 (64.72–64.84) 69.78 (69.68–69.89) 55.87 (55.74–56.00) 62.49 (62.36–62.63)
 Year 4 61.23 (61.16–61.30) 67.89 (67.82–67.95) 60.60 (60.55–60.65) 68.43 (68.33–68.53) 65.76 (65.70–65.82) 70.74 (70.64–70.85) 56.83 (56.70–56.97) 63.47 (63.33–63.61)
 Year 5 61.57 (61.49–61.64) 68.22 (68.15–68.29) 60.94 (60.88–61.00) 68.76 (68.67–68.86) 66.09 (66.03–66.16) 71.07 (70.96–71.18) 57.16 (57.03–57.30) 63.81 (63.67–63.95)
Revision
 Year 1 0.56 (0.56–0.57) 0.33 (0.33–0.33) 0.55 (0.55–0.55) 0.36 (0.35–0.36) 0.60 (0.60–0.60) 0.34 (0.33–0.34) 0.47 (0.47–0.47) 0.28 (0.27–0.28)
 Year 2 1.14 (1.12–1.15) 0.67 (0.67–0.68) 1.11 (1.11–1.12) 0.72 (0.72–0.73) 1.21 (1.21–1.22) 0.68 (0.67–0.69) 0.96 (0.95–0.96) 0.47 (0.56–0.58)
 Year 3 1.71 (1.69–1.73) 1.02 (1.01–1.02) 1.68 (1.67–1.69) 1.09 (1.08–1.10) 1.83 (1.82–1.85) 1.03 (1.014–1.05) 1.44 (1.43–1.45) 0.85 (0.84–0.87)
 Year 4 2.41 (2.37–2.44) 1.43 (1.41–1.44) 2.36 (2.34–2.37) 1.53 (1.52–1.55) 2.57 (2.54–2.60) 1.45 (1.422–1.48) 2.03 (2.01–2.05) 1.20 (1.17–1.23)
 Year 5 3.28 (3.23–3.34) 1.95 (1.93–1.98) 3.22 (3.20–3.25) 2.10 (2.08–2.12) 3.51 (3.46–3.56) 1.98 (1.938–2.03) 2.77 (2.73–2.80) 1.64 (1.56–1.69)
Reoperation
 Year 1 6.09 (6.07–6.11) 7.65 (7.62–7.68) 5.79 (5.76–5.81) 9.05 (9.00–9.10) 3.75 (3.74–3.77) 6.27 (6.24–6.30) 5.27 (5.23–5.31) 6.62 (6.58–6.67)
 Year 2 10.78 (10.74–10.83) 13.45 (13.38–13.53) 10.26 (10.19–10.32) 15.82 (15.70–15.94) 6.71 (6.68–6.74) 11.09 (11.02–11.63) 9.35 (9.29–9.42) 11.70 (11.61–11.78)
 Year 3 13.93 (13.85–14.00) 17.31 (17.18–17.43) 13.26 (13.16–13.36) 20.26 (20.07–20.44) 8.73 (8.67–8.78) 14.32 (14.20–14.43) 12.11 (12.02–12.20) 15.09 (14.97–15.21)
 Year 4 16.46 (16.35–16.57) 20.38 (20.20–20.56) 15.69 (15.54–15.83) 23.77 (23.53–24.03) 10.37 (10.30–10.45) 16.91 (16.76–17.07) 14.34 (14.23–14.46) 17.81 (17.66–17.97)
 Year 5 18.48 (18.33–18.63) 22.81 (22.57–23.05) 17.62 (17.43–17.81) 26.54 (26.21–26.87) 11.70 (11.59–11.81) 18.98 (18.77–19.19) 16.13 (15.99–16.27) 19.98 (19.77–20.18)
Open in a new tab

Cumulative incidence estimated using instrumental variable-adjusted Cox proportional hazards models with the following covariates: age, sex, race and ethnicity, Elixhauser comorbidities, and year of surgery.

Regarding differences in sleeve gastrectomy versus gastric bypass by race and ethnicity, compared to White patients, the adjusted hazard ratio of ED use differed among Black patients (5-year aHR 1.01 [95% CI 0.94–1.08] vs. 0.94 [95% CI 0.88–1.00], P<.001) (Table 3). This corresponded to a higher 5-year cumulative incidence and a smaller between-procedure difference in ED use among Black patients (sleeve 66.1% [95% CI 66.0%−66.2%] vs. bypass 71.1% [71.0%−71.2%]) compared to White patients (sleeve 60.9% [60.9%−61.0%) vs. bypass 68.8% [68.7%−68.9%]). Similarly, compared to White patients, the adjusted hazard ratio of reoperation differed among Hispanic patients (5-year aHR 0.95 [95% CI 0.86–1.05] vs. 0.76 [95% CI 0.69–0.83], P<.001). This corresponded to a lower 5-year cumulative incidence and a smaller between-procedure difference in reoperation among Hispanic patients (sleeve 16.1% [16.0%−16.3%] vs. bypass 20.0% [19.8%−20.2%]) compared to White patients (sleeve 17.6% [17.4%−17.8%] vs. bypass 26.5% [26.2%−26.9%]).

Table 3 –

Adjusted Hazard Ratios for Outcomes of Sleeve Gastrectomy vs. Gastric Bypass

Outcome Adjusted Hazard Ratio (95% CI) P value of Interaction
1 Year 3 Year 5 Year
Mortality
 All 0.63 (0.55–0.71) 0.73 (0.62–0.85) 0.79 (0.63–0.94) N/A
 White 0.61 (0.52–0.70) 0.71 (0.58–0.84) 0.76 (0.60–0.93) N/A
 Black 0.69 (0.57–0.81) 0.81 (0.66–0.95) 0.86 (0.68–1.04) 0.195
 Hispanic 0.58 (0.42–0.74) 0.68 (0.48–0.86) 0.73 (0.51–0.95) 0.724
Complications
 All 0.70 (0.64–0.76) 0.79 (0.70–0.88) 0.84 (0.73–0.94) N/A
 White 0.71 (0.63–0.78) 0.80 (0.70–0.89) 0.84 (0.74–0.95) N/A
 Black 0.71 (0.63–0.78) 0.80 (0.69–0.90) 0.84 (0.72–0.96) 0.985
 Hispanic 0.68 (0.59–0.77) 0.77 (0.66–0.87) 0.81 (0.69–0.93) 0.539
Hospitalization
 All 0.77 (0.74–0.81) 0.89 (0.83–0.95) 0.95 (0.88–1.03) N/A
 White 0.77 (0.72–0.82) 0.89 (0.82–0.96) 0.95 (0.87–1.03) N/A
 Black 0.80 (0.75–0.85) 0.92 (0.85–1.00) 0.98 (0.89–1.08) 0.267
 Hispanic 0.75 (0.71–0.78) 0.86 (0.80–0.91) 0.92 (0.84–0.99) 0.371
ED Utilization
 All 0.84 (0.80–0.88) 0.92 (0.88–0.97) 0.96 (0.91–1.02) N/A
 White 0.82 (0.77–0.87) 0.90 (0.84–0.96) 0.94 (0.88–1.00) N/A
 Black 0.89 (0.84–0.94) 0.97 (0.91–1.03) 1.01 (0.94–1.08) <.001
 Hispanic 0.85 (0.82–0.88) 0.93 (0.89–0.96) 0.97 (0.92–1.01) 0.264
Revision*
 All N/A N/A 1.86 (1.37–2.35) N/A
 White N/A N/A 1.73 (1.28–2.19) N/A
 Black N/A N/A 2.03 (1.32–2.73) 0.319
 Hispanic N/A N/A 1.98 (1.08–2.90) 0.467
Reoperation
 All 0.64 (0.60–0.68) 0.74 (0.68–0.79) 0.79 (0.72–0.86) N/A
 White 0.62 (0.58–0.65) 0.71 (0.66–0.77) 0.76 (0.69–0.83) N/A
 Black 0.58 (0.52–0.63) 0.67 (0.60–0.74) 0.71 (0.64–0.80) 0.235
 Hispanic 0.78 (0.71–0.84) 0.90 (0.82–0.97) 0.95 (0.86–1.05) <.001
Open in a new tab

Adjusted hazard ratios estimated using a Cox proportional hazards model including age, sex, race and ethnicity, year of surgery, and Elixhauser comorbidities, as well as an instrumental variable for prior-year state-level proportion of sleeve gastrectomy vs. gastric bypass. Interaction tested by P value is interaction of race/ethnicity and operation (sleeve gastrectomy vs. gastric bypass). In the case of ED use, there was a statistically significant interaction between Black and White patients despite overlapping 95% confidence intervals.

*

Results of the Cox proportional hazards model for revision did not violate the proportional hazards assumption, and therefore covariates in this model were not interacted with time and had a single 5-year adjusted hazard ratio for each subgroup.

Discussion

In this national study of patients with Medicaid undergoing bariatric surgery, sleeve gastrectomy was associated with a lower long-term risk of mortality, complications, hospitalization, ED use, and reoperation, but a higher long-term risk of revision compared to gastric bypass. Although the comparative safety of sleeve gastrectomy was generally similar among White, Black, and Hispanic patients, compared to White patients, the magnitude of the difference between sleeve gastrectomy and gastric bypass was smaller among Black patients for ED use and among Hispanic patients for reoperation. Taken together, these results suggest that the superior long-term safety of sleeve gastrectomy previously described in patients with private insurance and Medicare is maintained in patients with Medicaid, but that the comparative difference in safety between these two procedures differs across minority and potentially vulnerable subgroups.

To our knowledge, this is the largest study to date to evaluate long-term bariatric surgical outcomes in patients with Medicaid. Medicaid is the largest single payer for health care in the United States.26,27 What’s more, over 60% of Medicaid beneficiaries belong to minority or vulnerable populations where obesity is more prevalent than the general population.28–31 Prior work has demonstrated that use of bariatric surgery in these populations has risen rapidly, increasing by 15.8% per quarter in Medicaid expansion states since the passage of the Affordable Care Act.3 The current results not only corroborate the preferential adoption of sleeve gastrectomy in patients with Medicaid, but demonstrate that this procedure is even more common among Black and Hispanic patients compared to White patients. Overall, it is reassuring that the safety benefits of sleeve gastrectomy are maintained in this population, making this operation an excellent option for the treatment of severe obesity in patients with Medicaid. However, differential use of these two procedures has important implications. Existing evidence suggests that sleeve gastrectomy is associated with fewer complications whereas gastric bypass is associated with superior weight loss and comorbidity resolution.17 So while it is unsurprising that sleeve gastrectomy – the “safer” procedure – has been rapidly adopted among older populations with more comorbidities, the younger age of Medicaid patients may justify greater use of gastric bypass – the “more effective” procedure – insofar as these patients will live with the results of this operation for much longer.32

It is important to understand the results of the current study within the context of prior work comparing sleeve gastrectomy and gastric bypass in patients with private insurance and Medicare. Specifically, the cumulative incidence of adverse outcomes in the current study was higher than what has been previously reported among privately insured patients, but lower than what has been reported among patients with Medicare.5,7 This is consistent with studies that have directly compared patients with Medicaid to privately insured patients and found that Medicaid status was independently associated with increased mortality, complications, and health care utilization.9,10,33–35 These results also mirror previous work demonstrating that adverse outcomes are higher among Medicare beneficiaries compared to patients with Medicaid.36 These insurance-based differences are likely multifactorial. Again, patients with Medicaid are on average younger than Medicare beneficiaries, as was the case in the current study. However, compared to privately insured patients, Medicaid patients experience the longest wait times for bariatric surgery, as well as substantially higher barriers to care and socioeconomic strain.37,38 Importantly, studies suggest that bariatric surgery is just as effective for patients with Medicaid as it is for patients with private insurance and Medicare.39,40 Therefore, continued efforts to ensure the equitable delivery of timely, high-quality bariatric surgical care to patients with Medicaid may improve their outcomes.

This study also builds upon prior work evaluating bariatric surgical outcomes in minority and potentially vulnerable patient groups. Courcoulas et al.8 previously found evidence for heterogeneity in the safety of sleeve gastrectomy versus gastric bypass depending on race. In the current study, we found that these comparative differences were largely similar between White, Black, and Hispanic patients with two exceptions. Compared to White patients, the difference between sleeve and bypass was not as substantial among Black patients for ED use and Hispanic patients for reoperation. This appeared to be due to the fact that Black patients had higher rates of ED use overall and Hispanic patients had lower rates of reoperation overall. Higher ED use among Black patients, as well as the higher cumulative incidence of other adverse outcomes among Black patients observed in this study is consistent with prior literature demonstrating worse bariatric surgical outcomes among minority groups. For example, O’Neill et al.14 found that Black and Hispanic patients had a higher risk of 30-day readmission after bariatric surgery compared to White patients. Other studies also suggest that minority patients experience worse outcomes after bariatric surgery.13 Similarly, the lower rate of revision and reoperation among Hispanic patients is also consistent with prior literature, although the reasons for this remain unclear.41 On the one hand, some have suggested that this may be due to superior weight loss and comorbidity resolution among Hispanic patients, although other studies have reported no difference between racial and ethnic subgroups.42,43 On the other hand, this finding may reflect limited access to care, given that prior studies have shown that Hispanic patients undergo disproportionately fewer bariatric procedures despite a higher prevalence of obesity compared to White patients.44 Future work is needed to better understand the drivers of these differential outcomes by race.45

While the strengths of this study are its use of a unique population of patients with Medicaid, large sample size, and use of an instrumental variable, it has limitations. First, the observational nature of this study introduces the possibility of confounding due to selection bias and loss to follow-up. Moreover, the use of Medicaid claims to construct the study cohort may limit the generalizability of these findings to non-Medicaid populations. However, the specific goal of this study was to investigate bariatric surgical outcomes in a population of patients that has been historically understudied, and by doing so, these results enable important comparisons with other patient populations. Additionally, the criteria for Medicaid coverage of bariatric surgery differs from state to state, such that some states require patients to have more obesity-related comorbidities than others to qualify for Medicaid coverage of their operation.46 Nevertheless, to the extent that our analysis specifically compared two different bariatric procedures and established adequate balance between those groups, this likely had little effect on the current results. Second, although we found that the magnitude of the difference in outcome varied by race and ethnicity, the current study is unable to determine the clinical relevance of these differences, which as discussed above, may reflect differential outcomes or differences in access to care. Third, although this study evaluated a number of adverse outcomes that are critical to inform surgical decision making, it did not evaluate equally important outcomes such as weight loss, comorbidity resolution, and medication discontinuation. Although a number of short-term studies suggest that both sleeve gastrectomy and gastric bypass are highly effective in this population, future work should specifically evaluate long-term weight- and comorbidity-related outcomes in patients with Medicaid to better inform decision making. This study did not evaluate differences in endoscopic intervention after surgery, although Courcoulas et al. has previously demonstrated that sleeve gastrectomy was associated with a lower rate of endoscopic intervention compared to gastric bypass.8 Fourth, although this study used a powerful econometric technique to control for selection bias, this study is still limited by its non-randomized, observational design. Moreover, the use of administrative claims, which lack clinical granularity, may have resulted in the inclusion of patients who underwent partial gastrectomy for reasons unrelated to obesity, although our requirement of a concomitant diagnosis of severe obesity and exclusion of patients with gastric and small bowel cancer makes this less likely. Another limitation is that the reoperation outcome may include procedures that were unrelated to the index bariatric procedure and were simply incidentally performed during the follow-up period, although all individual procedures included in this outcome have been previously evaluated as potentially related to the index bariatric procedure. Finally, small sample size precluded this study from evaluating outcomes in other minority populations. Additional studies that focus specifically on these populations will be critical to better understand the heterogeneity of bariatric surgical outcomes between different patient populations.

Conclusion

Among patients with Medicaid undergoing bariatric surgery, sleeve gastrectomy was associated with a lower long-term risk of mortality, complications, hospitalization, ED use, and reoperations, but a higher long-term risk of revision compared to gastric bypass. Although the difference between sleeve and bypass was generally similar among White, Black, and Hispanic patients, the magnitude of this difference was smaller among Black patients for ED use and among Hispanic patients for reoperation. This suggests that the comparative safety of sleeve gastrectomy previously described in patients with private insurance and Medicare is maintained in patients with Medicaid, but that the difference between these two procedures may be slightly less among minority and potentially vulnerable subgroups.

Supplementary Material

Revised Supplemental Data File

Acknowledgments

Disclosures:

Funding for this study was provided by NIDDK grant 5R01DK115408-02. Dr. Howard receives unrelated funding from the Blue Cross Blue Shield of Michigan Foundation and the National Institute of Diabetes and Digestive and Kidney Diseases (5T32DK108740-05). Dr. Yang, Ms. Thumma, Dr. Ehlers, Dr. O’Neill, and Dr. Ryan have no disclosures. Dr. Arterburn receives unrelated funding from the National Institutes of Health and nonfinancial support from International Federation for the Surgery of Obesity and Metabolic Disorders Latin America Chapter. Dr. Telem receives unrelated funding from the Agency for Healthcare Research and Quality and receives consulting fees from Medtronic. Dr. Dimick receives unrelated grant funding from the National Institutes of Health, the Agency for Healthcare Research and Quality, Blue Cross Blue Shield of Michigan Foundation, and is a cofounder of ArborMetrix, Inc. No funder had any part in the design or execution of this study.

References

  • 1.Khorgami Z, Shoar S, Andalib A, Aminian A, Brethauer SA, Schauer PR. Trends in utilization of bariatric surgery, 2010–2014: sleeve gastrectomy dominates. Surg Obes Relat Dis. 2017;13(5):774–778. [DOI] [PubMed] [Google Scholar]
  • 2.English WJ, DeMaria EJ, Brethauer SA, Mattar SG, Rosenthal RJ, Morton JM. American Society for Metabolic and Bariatric Surgery estimation of metabolic and bariatric procedures performed in the United States in 2016. Surg Obes Relat Dis. 2018;14(3):259–263. [DOI] [PubMed] [Google Scholar]
  • 3.Gould KM, Zeymo A, Chan KS, et al. Bariatric surgery among vulnerable populations: The effect of the Affordable Care Act’s Medicaid expansion. Surgery. 2019;166(5):820–828. [DOI] [PubMed] [Google Scholar]
  • 4.Hanchate AD, Qi D, Paasche-Orlow MK, et al. Examination of Elective Bariatric Surgery Rates Before and After US Affordable Care Act Medicaid Expansion. JAMA Health Forum. 2021;2(10):e213083–e213083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Chhabra KR, Telem DA, Chao GF, et al. Comparative Safety of Sleeve Gastrectomy and Gastric Bypass: An Instrumental Variables Approach. Ann Surg. 2022;275(3):539–545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Chao GF, Chhabra KR, Yang J, et al. Bariatric Surgery in Medicare Patients: Examining Safety and Healthcare Utilization in the Disabled and Elderly. Ann Surg. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Howard R, Chao GF, Yang J, et al. Comparative Safety of Sleeve Gastrectomy and Gastric Bypass Up to 5 Years After Surgery in Patients With Severe Obesity. JAMA Surg. 2021;156(12):1160–1169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Courcoulas A, Coley RY, Clark JM, et al. Interventions and Operations 5 Years After Bariatric Surgery in a Cohort From the US National Patient-Centered Clinical Research Network Bariatric Study. JAMA Surg. 2020;155(3):194–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Takemoto E, Andrea SB, Wolfe BM, Nagel CL, Boone-Heinonen J. Weighing in on Bariatric Surgery: Effectiveness Among Medicaid Beneficiaries-Limited Evidence and Future Research Needs. Obesity (Silver Spring). 2018;26(3):463–473. [DOI] [PubMed] [Google Scholar]
  • 10.Funk LM, Suzo A, Mikami DJ, Needleman BJ. Two-year outcomes for medicaid patients undergoing laparoscopic Roux-en-Y gastric bypass: a case-control study. Obes Surg. 2014;24(10):1679–1685. [DOI] [PubMed] [Google Scholar]
  • 11.Takemoto E, Wolfe BM, Nagel CL, et al. Insurance status differences in weight loss and regain over 5 years following bariatric surgery. Int J Obes (Lond). 2018;42(6):1211–1220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Altieri MS, Yang J, Yin D, Talamini MA, Spaniolas K, Pryor AD. Patients insured by Medicare and Medicaid undergo lower rates of bariatric surgery. Surg Obes Relat Dis. 2019;15(12):2109–2114. [DOI] [PubMed] [Google Scholar]
  • 13.Acevedo E Jr., Lu X, Zhao H, Mazzei M, Sarvepalli S, Edwards MA. Outcomes in racial minorities after robotic Roux-en-Y gastric bypass and sleeve gastrectomy: a retrospective review of the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database. Surg Obes Relat Dis. 2021;17(3):595–605. [DOI] [PubMed] [Google Scholar]
  • 14.O’Neill SM, Needleman B, Narula V, Brethauer S, Noria SF. An analysis of readmission trends by urgency and race/ethnicity in the MBSAQIP registry, 2015–2018. Surg Obes Relat Dis. 2022;18(1):11–20. [DOI] [PubMed] [Google Scholar]
  • 15.Arterburn D, Wellman R, Emiliano A, et al. Comparative Effectiveness and Safety of Bariatric Procedures for Weight Loss: A PCORnet Cohort Study. Ann Intern Med. 2018;169(11):741–750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Verhoeff K, Mocanu V, Dang J, et al. Five Years of MBSAQIP Data: Characteristics, Outcomes, and Trends for Patients with Super-obesity. Obes Surg. 2022;32(2):406–415. [DOI] [PubMed] [Google Scholar]
  • 17.Howard R, Chao GF, Yang J, et al. Medication Use for Obesity-Related Comorbidities After Sleeve Gastrectomy or Gastric Bypass. JAMA Surg. 2022;157(3):248–256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Chao GF, Chhabra KR, Yang J, et al. Bariatric Surgery in Medicare Patients: Examining Safety and Healthcare Utilization in the Disabled and Elderly. Ann Surg. 2022;276(1):133–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Flanagin A, Frey T, Christiansen SL, Committee AMAMoS. Updated Guidance on the Reporting of Race and Ethnicity in Medical and Science Journals. JAMA. 2021;326(7):621–627. [DOI] [PubMed] [Google Scholar]
  • 20.Flanagin A, Frey T, Christiansen SL, Bauchner H. The Reporting of Race and Ethnicity in Medical and Science Journals: Comments Invited. JAMA. 2021;325(11):1049–1052. [DOI] [PubMed] [Google Scholar]
  • 21.Medicare Coverage Database. Center for Medicare and Medicaid Services. National Coverage Determination (NCD) for Bariatric Surgery for Treatment of Morbid Obesity (100.1). https://www.cms.gov/medicare-coverage-database/details/ncd-details.aspx?NCDId=57&ncdver=4&CoverageSelection=Both&ArticleType=All&PolicyType=Final&s=Iowa&KeyWord=bariatric%2Bsurgery&KeyWordLookUp=Title&KeyWordSearchType=And&bc=gAAAABAAAQAA&. Accessed February 7, 2021
  • 22.Rassen JA, Brookhart MA, Glynn RJ, Mittleman MA, Schneeweiss S. Instrumental variables II: instrumental variable application-in 25 variations, the physician prescribing preference generally was strong and reduced covariate imbalance. J Clin Epidemiol. 2009;62(12):1233–1241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Birkmeyer NJ, Dimick JB, Share D, et al. Hospital complication rates with bariatric surgery in Michigan. JAMA. 2010;304(4):435–442. [DOI] [PubMed] [Google Scholar]
  • 24.Finks JF, Kole KL, Yenumula PR, et al. Predicting risk for serious complications with bariatric surgery: results from the Michigan Bariatric Surgery Collaborative. Ann Surg. 2011;254(4):633–640. [DOI] [PubMed] [Google Scholar]
  • 25.Courcoulas AP, Yanovski SZ, Bonds D, et al. Long-term outcomes of bariatric surgery: a National Institutes of Health symposium. JAMA Surg. 2014;149(12):1323–1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Centers for Medicare & Medicaid Services. NHE Fact Sheet. Published December 16, 2020. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/NHE-Fact-Sheet.
  • 27.Gee RE, Shulkin D, Romm I. A Blueprint for Comprehensive Medicaid Reform. JAMA. 2021;325(7):619–620. [DOI] [PubMed] [Google Scholar]
  • 28.Report to Congress on Medicaid and CHIP Medicaid and CHIP Payment and Access Commission.; 2020. https://www.macpac.gov/wp-content/uploads/2020/06/June-2020-Report-to-Congress-on-Medicaid-and-CHIP.pdf.
  • 29.Hales CM, Carroll MD, Fryar CD, Ogden CL. Prevalence of obesity and severe obesity among adults: United States, 2017–2018. NCHS Data Brief, no 360. Hyattsville, MD: National Center for Health Statistics. 2020. [PubMed] [Google Scholar]
  • 30.Ogden CL, Lamb MM, Carroll MD, Flegal KM. Obesity and socioeconomic status in adults: United States 1988–1994 and 2005–2008. NCHS data brief no 50. Hyattsville, MD: National Center for Health Statistics. 2010. [PubMed] [Google Scholar]
  • 31.Mylona EK, Benitez G, Shehadeh F, et al. The association of obesity with health insurance coverage and demographic characteristics: a statewide cross-sectional study. Medicine (Baltimore). 2020;99(27):e21016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Carlin AM, Zeni TM, English WJ, et al. The comparative effectiveness of sleeve gastrectomy, gastric bypass, and adjustable gastric banding procedures for the treatment of morbid obesity. Ann Surg. 2013;257(5):791–797. [DOI] [PubMed] [Google Scholar]
  • 33.Dallal RM, Datta T, Braitman LE. Medicare and Medicaid status predicts prolonged length of stay after bariatric surgery. Surg Obes Relat Dis. 2007;3(6):592–596. [DOI] [PubMed] [Google Scholar]
  • 34.Alexander JW, Goodman HR, Martin Hawver LR, James L. The impact of medicaid status on outcome after gastric bypass. Obes Surg. 2008;18(10):1241–1245. [DOI] [PubMed] [Google Scholar]
  • 35.Chen EY, Fox BT, Suzo A, et al. One-year Surgical Outcomes and Costs for Medicaid Versus Non-Medicaid Patients Undergoing Laparoscopic Roux-en-Y Gastric Bypass: A Single-Center Study. Surg Laparosc Endosc Percutan Tech. 2016;26(1):38–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Hennings DL, Baimas-George M, Al-Quarayshi Z, Moore R, Kandil E, DuCoin CG. The Inequity of Bariatric Surgery: Publicly Insured Patients Undergo Lower Rates of Bariatric Surgery with Worse Outcomes. Obes Surg. 2018;28(1):44–51. [DOI] [PubMed] [Google Scholar]
  • 37.Alvarez R, Bonham AJ, Buda CM, Carlin AM, Ghaferi AA, Varban OA. Factors Associated With Long Wait Times for Bariatric Surgery. Ann Surg. 2019;270(6):1103–1109. [DOI] [PubMed] [Google Scholar]
  • 38.Fiscella K, Franks P, Gold MR, Clancy CM. Inequality in quality: addressing socioeconomic, racial, and ethnic disparities in health care. JAMA. 2000;283(19):2579–2584. [DOI] [PubMed] [Google Scholar]
  • 39.Liu N, Venkatesh M, Hanlon BM, et al. Association Between Medicaid Status, Social Determinants of Health, and Bariatric Surgery Outcomes. Ann Surg Open. 2021;2(1):e028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hayes S, Napolitano MA, Lent MR, et al. The effect of insurance status on pre- and post-operative bariatric surgery outcomes. Obes Surg. 2015;25(1):191–194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Sudan R, Nguyen NT, Hutter MM, Brethauer SA, Ponce J, Morton JM. Morbidity, mortality, and weight loss outcomes after reoperative bariatric surgery in the USA. J Gastrointest Surg. 2015;19(1):171–178; discussion 178–179. [DOI] [PubMed] [Google Scholar]
  • 42.Sudan R, Winegar D, Thomas S, Morton J. Influence of ethnicity on the efficacy and utilization of bariatric surgery in the USA. J Gastrointest Surg. 2014;18(1):130–136. [DOI] [PubMed] [Google Scholar]
  • 43.Khorgami Z, Arheart KL, Zhang C, Messiah SE, de la Cruz-Munoz N. Effect of ethnicity on weight loss after bariatric surgery. Obes Surg. 2015;25(5):769–776. [DOI] [PubMed] [Google Scholar]
  • 44.Sundaresan N, Roberts A, Thompson KJ, McKillop IH, Barbat S, Nimeri A. Examining the Hispanic paradox in bariatric surgery. Surg Obes Relat Dis. 2020;16(10):1392–1400. [DOI] [PubMed] [Google Scholar]
  • 45.Bhogal SK, Reddigan JI, Rotstein OD, et al. Inequity to the utilization of bariatric surgery: a systematic review and meta-analysis. Obes Surg. 2015;25(5):888–899. [DOI] [PubMed] [Google Scholar]
  • 46.Lee JS, Sheer JL, Lopez N, Rosenbaum S. Coverage of obesity treatment: a state-by-state analysis of Medicaid and state insurance laws. Public Health Rep. 2010;125(4):596–604. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Revised Supplemental Data File
NIHMS1830941-supplement-Revised_Supplemental_Data_File.docx (32.6KB, docx)

RESOURCES