Cohort studycited by the Tayebat archivefull text checked A-BAR-16 in-depth review
Swedish cohorts: more self-harm after surgery
Risk of suicide and non-fatal self-harm after bariatric surgery: results from two matched cohort studies.
Overview
Neovius M et al.. Lancet Diabetes Endocrinol 2018 · PMID 29329975 · DOI 10.1016/S2213-8587(17)30437-0
Self-harm and suicide were more frequent after surgery, but absolute risks were low.
Does the study support the specific claim? Yes · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Self-harm
- How was the study done?
Matched cohorts
- How many people, animals or samples were studied?
>40,000
- What was actually found?
Higher risk
- How large was the observed effect?
aHR 1.78-3.16
- What do the results mean in everyday life?
Mental-health follow-up is needed
- What are the study's limitations?
Observational; few suicides; self-selection; no sleeve gastrectomy
- What does it say about Elawady's specific claim?
Supports risk; authors oppose general discouragement
- Do the results apply to Germany and Europe?
European
- What can explicitly not be concluded from this study?
Mechanism; that surgery shortens life overall; risk after sleeve
Scientific in-depth review
1. Full reference
- Authors
- Neovius M, Bruze G, Jacobson P, Sjöholm K, Johansson K, Granath F, Sundström J, Näslund I, Marcus C, Ottosson J, Peltonen M, Carlsson LMS
- Journal
- Lancet Diabetes Endocrinol
- Year
- 2018
- Identifier
- PMID 29329975 DOI 10.1016/S2213-8587(17)30437-0 · PMC5932484
- Verification
- verified – PubMed E-utilities efetch, 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
NIDDK R01DK105948; Swedish Research Council; Sahlgrenska ALF; Diabetesfonden. Funders had no role. Three authors received consulting fees from Itrim's scientific advisory committee; one received lecture fees from J&J, AstraZeneca, MSD; two are former/current SOReg directors.
2. Research question
Suicide and non-fatal self-harm after bariatric surgery vs non-surgical treatment
3. Study design and population
- Study design
- Cohort study – Two matched cohort studies: SOS (prospective non-randomised controlled intervention study, sequential treatment assignment matching on 18 variables; NCT01479452) and SOReg/Itrim (nationwide register linkage, coarsened exact matching); outcomes from national registers
- Population
Swedish adults with obesity attempting weight loss. SOS: age 37-60, BMI >=34 (men)/>=38 (women), excluding drug/alcohol abuse and psychiatric contraindications. SOReg/Itrim: age >=18, BMI 30-49.9
- Country
Sweden
- Period
SOS recruitment 1987-2001, follow-up to Dec 31, 2013 (median 18 years); SOReg surgery 2007-2012 and Itrim 2006-2013, follow-up to Dec 31, 2014 (median 3.9 years)
- Sample size
SOS: 2010 surgery (1369 VBG, 376 banding, 265 gastric bypass) vs 2037 usual-care controls. SOReg/Itrim: 20,256 gastric bypass vs 16,162 intensive lifestyle (matched from 26,388 and 18,365 eligible)
- Inclusion and exclusion criteria
SOS: identical criteria for both groups (see population); exclusions incl. prior bariatric/ulcer surgery, recent MI, malignancy, bulimia, drug/alcohol abuse. SOReg/Itrim: BMI 30-49.9, baseline weight recorded; missing BMI/education excluded (<0.5%). Crossovers to surgery censored (SOS 289, Itrim 335); SOS reversals censored (n=100)
4. Methods and endpoints
- Exposure / intervention
Bariatric surgery (SOS: VBG, banding, gastric bypass; SOReg: primary gastric bypass, 96% laparoscopic)
- Comparator
SOS: customary non-surgical treatment (not standardised). SOReg/Itrim: commercial intensive lifestyle programme (3-month LCD/VLCD plus 9-month maintenance)
- Endpoints
Death by suicide (incl. undetermined intent) and suicide or non-fatal self-harm (ICD9 E950-959, E980-989; ICD10 X60-84, Y10-34, Y870) from Causes of Death Register and National Patient Register
- Statistical methods
Cox regression; SOS adjusted for age, sex, history of self-harm, BMI; SOReg/Itrim weighted by matching strata and additionally adjusted for age, BMI, income, marital status, disability pension, unemployment; subgroup analyses (procedure, psychiatric history, education); proportional hazards violated for suicide in SOReg/Itrim (P=0.0497)
5. Actual results
- Results
SOS (68,528 person-years): suicide or non-fatal self-harm 87 vs 49 events, aHR 1.78 (95% CI 1.23-2.57); suicides 9 vs 3, aHR 3.06 (0.79-11.88). By procedure vs controls: gastric bypass 3.48 (1.65-7.31), banding 2.43 (1.23-4.82), VBG 2.25 (1.37-3.71). Additional adjustment for diabetes/CVD: 1.74 (1.20-2.52). SOReg/Itrim (149,582 person-years): 341 vs 84 events, aHR 3.16 (2.46-4.06); suicides 33 vs 5, aHR 5.17 (1.86-14.37). Elevated risk also in subgroups without psychiatric disorder or self-harm history. Surgery patients with events had similar or greater weight loss than those without. Substance abuse among those with non-fatal self-harm: SOS 48% vs 28%; SOReg 51% vs 29%. No between-group difference in SOS during the first 5 years.
- Absolute risks
SOS: 9 vs 3 suicides and 87 vs 49 suicide/self-harm events over 68,528 person-years in total; SOReg/Itrim: 33 vs 5 suicides and 341 vs 84 events over 149,582 person-years. Authors: 42 suicides in SOS and SOReg over about 117,000 person-years after surgery; absolute risk low. Cited Utah data: all-cause mortality 37.6 vs 57.1 per 10,000 person-years (surgery vs controls) vs suicide 2.6 vs 0.9.
- Authors' conclusion
Bariatric surgery was associated with suicide and non-fatal self-harm; absolute risks are low; findings indicate need for pre-operative information and post-operative psychiatric surveillance. The authors state the findings should not discourage bariatric surgery, at least not from a survival perspective, given documented survival benefits.
6. Strengths
- Two independent, complementary cohorts (very long follow-up; modern gastric bypass with intensive lifestyle comparator)
- Matching/adjustment for baseline psychiatric history, self-harm, substance abuse and psychotropic drugs
- Near-complete nationwide register outcome ascertainment
- Consistent findings in subgroups without psychiatric history
7. Weaknesses
- Non-randomised; selection bias and residual confounding acknowledged (self-harm history more common in SOS surgery group: 3.4% vs 1.9%)
- Itrim participants paid for treatment (socioeconomic selection), partly addressed by adjustment
- Non-fatal self-harm captured only when hospital care occurred; outpatient data only from 2001
- Few suicide events (wide CIs; SOS suicide HR not significant)
- Proportional hazards violated for suicide in SOReg/Itrim
- No sleeve gastrectomy; predominantly white population
- SOS by primary procedure may misattribute risk because of conversions
- Full text is the accepted, unedited manuscript
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.
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | some concerns | Extensive matching on psychiatric history and socioeconomic factors, but treatment choice is self-selected; residual confounding acknowledged. |
| Exposure measurement | low | Surgery documented in study/registry. |
| Selection of participants | some concerns | Matched subsets (77% and 88% of eligible); paying lifestyle participants differ socioeconomically. |
| Post-exposure interventions | low | Crossovers and reversals censored. |
| Missing data | low | Register follow-up near complete; minimal missing covariates. |
| Outcome measurement | some concerns | Register-based, objective for suicide; non-fatal self-harm requires hospital contact and surveillance intensity may differ (surgery patients have more healthcare contact). |
| Selection of reported result | low | Outcomes and subgroup analyses described a priori; consistent reporting. |
Full-text review (25 October 2026): Added procedure-specific HRs, follow-up periods, person-years per cohort, subgroup results, weight-loss finding, absolute framing (42 suicides/117,000 PY), authors' explicit statement on survival benefit, funding/COI; bias domains added, overall unchanged (some_concerns).
9. Assessment of the statistical interpretation
Relative risks are substantial and consistent across two cohorts and subgroups, but absolute suicide risk is low (42 suicides over about 117,000 post-surgery person-years). The study does not address total mortality; the authors explicitly cite survival benefit of surgery.
10. Relevance to the specific claim
Supports C-BAR-07; limits use as argument against surgery in general.
11. Independent contrary findings
12. Transferability to Germany and Europe
Directly European.
13. Overall methodological judgement
Strong, consistent evidence of an association between bariatric surgery (especially gastric bypass) and suicide/self-harm with low absolute risk; supports the self-harm part of C-BAR-07 but not the notion that surgery is lethal or shortens life overall.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Risk after sleeve gastrectomy; mechanisms (alcohol/substance abuse, pharmacokinetics); effective targeted surveillance.
15. Linked fact checks
- Sleeve gastrectomy (and bariatric surgery in general) is a lethal operation that kills patients / shortens life. – Studies with mixed or unclear results
- Bariatric surgery is followed by documented rises in self-harm, suicide-related deaths and alcohol use disorder. – Studies supporting 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
Listed in the archive bibliography only; the archive gives no specific interpretation.
- A. What was actually studied?
Suicide/self-harm risk
- B. What did the authors find?
aHR 1.78 and 3.16
- C. What conclusions do the authors themselves draw?
Association with suicide/self-harm; absolute risks low; need for information and surveillance; should not discourage surgery from a survival perspective
- D. What does the archive derive from it?
Supports narrative.
- E. Is this derivation scientifically justified?
Supports increased risk; the authors' explicit statement that this does not justify general discouragement is not reported by the archive.
- Adequacy of the interpretation (by the archive)
- not applicable