Cohort studycited by the Tayebat archivefull text checked★ key study A-BAR-04 in-depth review
12 years after gastric bypass
Weight and Metabolic Outcomes 12 Years after Gastric Bypass.
Overview
Adams TD et al.. N Engl J Med 2017 · PMID 28930514 · DOI 10.1056/NEJMoa1700459
Half of the patients with diabetes were still in remission after 12 years; new diabetes was about 90% less frequent than in non-operated people.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Long-term outcomes
- How was the study done?
Prospective controlled cohort
- How many people, animals or samples were studied?
1156 adults
- What was actually found?
Remission declines but benefit persists
- How large was the observed effect?
Remission 51% at 12 y; incident T2D OR 0.08
- What do the results mean in everyday life?
Relapse happens, but many stay in remission
- What are the study's limitations?
Non-randomized; small diabetic subgroup; many controls later had surgery; mostly women in Utah
- What does it say about Elawady's specific claim?
Relapse claim true; 'benefit erodes away' overstated
- Do the results apply to Germany and Europe?
Applicable
- What can explicitly not be concluded from this study?
Exact causal effect; outcomes for sleeve gastrectomy
Scientific in-depth review
1. Full reference
- Authors
- Adams TD, Davidson LE, Litwin SE, Kim J, Kolotkin RL, Nanjee MN, Gutierrez JM, Frogley SJ, Ibele AR, Brinton EA, Hopkins PN, McKinlay R, Simper SC, Hunt SC
- Journal
- N Engl J Med
- Year
- 2017
- Identifier
- PMID 28930514 DOI 10.1056/NEJMoa1700459 · PMC5737957
- Verification
- verified – PubMed E-utilities efetch, 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
NIDDK grant DK-55006, NCRR grant MO1-RR00064, Weill Cornell Biomedical Research Program, Intermountain Healthcare. Disclosures: Kolotkin royalties for IWQOL-Lite questionnaire (Duke University); Ibele fees from Obalon Therapeutics trial. No other relevant COI reported.
2. Research question
12-year weight and metabolic outcomes after RYGB
3. Study design and population
- Study design
- Cohort study – Prospective observational controlled cohort (single bariatric centre, Salt Lake City, Utah) with two non-surgical comparison groups
- Population
Adults 18-72 years with severe obesity (mean baseline BMI about 45-47; 84% women in surgery group)
- Country
USA
- Period
Enrolment from July 2000; follow-up through March 2016 (12-year examination); vital status via National Death Index through 2014
- Sample size
1156 (418 RYGB; 417 sought but did not have surgery; 321 did not seek surgery)
- Inclusion and exclusion criteria
Age 18-72; severe obesity; no history of alcohol or narcotics abuse, no prior bariatric surgery, no gastric/duodenal ulcer, no myocardial infarction in previous 6 months, no active cancer in previous 5 years
4. Methods and endpoints
- Exposure / intervention
RYGB
- Comparator
Nonsurgery group 1: sought RYGB but did not have it (mainly insurance); nonsurgery group 2: population-based sample not seeking surgery. 147/417 (35%) of group 1 and 39/321 (12%) of group 2 later had bariatric surgery (excluded from incidence/remission tables)
- Endpoints
Primary: % weight lost; incidence and remission of T2D (FPG >=126 mg/dl, HbA1c >=6.5% or medication), hypertension, dyslipidemia at 12 years. Secondary: quality of life, mortality
- Statistical methods
Adjusted mean changes (baseline value plus sex, age, baseline BMI, marital status, income, education); logistic regression for incidence/remission; adjustment for multiple comparisons; medication-adjusted biochemical values
5. Actual results
- Results
Adjusted weight change in surgery group: -45.0 kg (95% CI -47.2 to -42.9; -35.0%) at 2 y, -36.3 kg (-39.0 to -33.5; -28.0%) at 6 y, -35.0 kg (-38.4 to -31.7; -26.9%) at 12 y vs -2.9 kg (-6.9 to 1.0) and 0 kg (-3.5 to 3.5) in nonsurgery groups 1 and 2. 93% maintained >=10% weight loss, 70% >=20%, 40% >=30%; 4/387 (1%) regained all weight. T2D remission in surgery group 66/88 (75%) at 2 y, 54/87 (62%) at 6 y, 43/84 (51%) at 12 y; of 62 patients in remission at 2 y with 12-y data, 69% remained free of T2D. 12-y remission vs controls: 51% vs 10% (5/52) and 5% (4/76); adjusted OR 8.9 (95% CI 2.0-40.0) and 14.8 (2.9-75.5). Remission by baseline treatment: 73% no medication, 56% oral only, 16% (3/19) insulin. Incident T2D 3% (8/303) vs 26% (42/164) and 26% (47/184); OR 0.08 (0.03-0.24) and 0.09 (0.03-0.29). Incident hypertension OR 0.23 (0.11-0.49) and 0.23 (0.11-0.51). Hypertension remission OR 5.1 (1.7-15.6) vs group 1, 2.4 (0.9-5.9) vs group 2. 7 suicides (5 in surgery group, 2 in group 1 after they had bariatric surgery).
- Absolute risks
12-year T2D incidence 3% (8/303) after RYGB vs 26% in both control groups; T2D remission 51% (43/84) vs 10% (5/52) and 5% (4/76) in controls; 1% regained all weight; 5 suicides in the surgery group (n=418).
- Authors' conclusion
Long-term durability of weight loss and effective remission and prevention of T2D, hypertension and dyslipidemia after RYGB.
6. Strengths
- >90% clinical follow-up and 96-99% with some 12-year data
- Two non-surgical control groups, one population-based
- 12 years, objective metabolic definitions
- Adjustment for multiple comparisons in adjusted analyses
7. Weaknesses
- Non-randomized; residual confounding acknowledged by authors
- Small diabetic subgroup (n=84-88) with wide CIs for remission ORs
- Substantial crossover to surgery in controls (35% and 12%), excluded from comparisons (possible selection)
- Single centre, 84% women, mostly white Utah population
- Exclusion of people with alcohol/narcotics abuse history limits generalisability for psychiatric outcomes
- Suicide counts small and not formally compared
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Structured single-reviewer judgement from full text; a formal assessment would require two independent reviewers and a protocol. The very large effects (OR about 0.08) are unlikely to be explained by confounding alone.
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | some concerns | Self-selection into surgery; group 1 (insurance denial) mitigates confounding by motivation; adjustment for six sociodemographic/BMI covariates; unmeasured confounding acknowledged. |
| Exposure measurement | low | Surgery status documented. |
| Selection of participants | some concerns | Controls who later had surgery (35%/12%) excluded from incidence/remission tables; analyses restricted to survivors. |
| Post-exposure interventions | some concerns | Controls free to seek weight-loss therapy; medication changes handled by adjustment and definitions. |
| Missing data | low | Some 12-year data in 96-99%; clinical measurements in 90-94%. |
| Outcome measurement | low | Objective laboratory/BP criteria, though medication use is part of definitions and outcome assessors not blinded. |
| Selection of reported result | low | Prespecified primary endpoints reported consistently with earlier waves. |
Full-text review (25 October 2026): Added remission comparison with controls (OR 8.9/14.8), persistence among 2-year remitters (69%), remission by baseline medication (insulin 16%), incidence counts (3% vs 26%), weight-maintenance distribution, crossover to surgery in controls, 7 suicides, full funding/COI; bias domains added, overall unchanged (some_concerns).
9. Assessment of the statistical interpretation
Remission declined from 75% to 51%, but 69% of 2-year remitters remained diabetes-free at 12 years and remission remained 5-10 times more frequent than in controls (adjusted OR 8.9 and 14.8, wide CIs). Remission depended strongly on pre-surgery diabetes stage (16% if on insulin). New-onset diabetes was reduced by 91-92%.
10. Relevance to the specific claim
Shows partial relapse of T2D remission; also strong long-term benefit.
11. Independent contrary findings
ARMMS-T2D (I-BAR-02), Mingrone (I-BAR-03).
12. Transferability to Germany and Europe
Generalizable with caution.
13. Overall methodological judgement
Correct numbers, selective framing.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | low |
| Directness | direct |
| Reproducibility | replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | moderate |
| Adequacy of the interpretation (by the archive) | partly correct but over-interpreted |
14. Open questions
Glycaemic control and complications in relapsers (only medication counts reported: mean change -0.3 vs +0.8/+1.1); causes of the excess suicides; applicability to sleeve gastrectomy.
15. Linked fact checks
- The diabetes remission achieved by bariatric surgery erodes over time (SOS: 72% at 2 years to 30.4% at 15 years; Adams: 75% to 51%). – Studies supporting the claim
- Gastric bypass is preferable to sleeve gastrectomy (less reflux) and can fully treat type 2 diabetes; for patients who fail dieting, bypass rather than sleeve should be chosen (2021 statement). – Studies contradicting or limiting the claim
Use by the Tayebat archive
Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.
- What the archive says about the study
Diabetes remission eroded from 75% at 2 years to 51% at 12 years after Roux-en-Y; framed as 'diabetes relapse is common'.
- A. What was actually studied?
12-year outcomes of RYGB vs non-surgical groups
- B. What did the authors find?
Remission 75%->62%->51%; 69% of 2-year remitters still in remission at 12 y; remission OR 8.9-14.8 vs controls; incident T2D 3% vs 26% (OR 0.08-0.09); 5 suicides in surgery group
- C. What conclusions do the authors themselves draw?
Durable weight loss and effective remission and prevention of T2D
- D. What does the archive derive from it?
Used to show that diabetes remission 'erodes'.
- E. Is this derivation scientifically justified?
The remission numbers are correct, but the archive omits that 69% of early remitters stayed in remission, that remission remained 5-10 times higher than in controls, and that new diabetes was 91-92% less frequent; the authors' main conclusion (durable benefit) is the opposite emphasis.
- Adequacy of the interpretation (by the archive)
- partly correct but over-interpreted