5-year mental health and eating pattern outcomes following bariatric surgery in adolescents: a prospective cohort study

Kajsa Järvholm, Gustaf Bruze, Markku Peltonen, Claude Marcus, Carl Erik Flodmark, Pia Henfridsson, Andrew J. Beamish, Eva Gronowitz, Jovanna Dahlgren, Jan Karlsson, Torsten Olbers

Research output: Contribution to journalArticlepeer-review


Background: Mental health problems are prevalent among adolescents with severe obesity, but long-term mental health outcomes after adolescent bariatric surgery are not well known. We aimed to assess mental health outcomes over 5 years of follow-up after Roux-en-Y gastric bypass surgery in adolescents who participated in the Adolescent Morbid Obesity Surgery (AMOS) study. Methods: This was a non-randomised matched-control study in adolescents aged 13–18 years who had a BMI of 40 kg/m2 or higher, or 35 kg/m2 or higher in addition to obesity-related comorbidity; who had previously undergone failed comprehensive conservative treatment; and were of pubertal Tanner stage III or higher, with height growth velocity beyond peak. A contemporary control group, matched for BMI, age, and sex, who underwent conventional obesity treatment, was obtained from the Swedish Childhood Obesity Treatment Register. Data on dispensed psychiatric drugs and specialist treatment for mental disorders were retrieved from national registers with complete coverage. In the surgical group only, questionnaires were used to assess self-esteem (Rosenberg Self-Esteem [RSE] score), mood (Mood Adjective Checklist [MACL]), and eating patterns (Binge Eating Scale [BES] and Three-Factor Eating Questionnaire-R21 [TFEQ]). This study is registered with ClinicalTrials.gov (NCT00289705). Findings: Between April 10, 2006, and May 20, 2009, 81 adolescents (53 [65%] female) underwent Roux-en-Y gastric bypass surgery, and 80 control participants received conventional treatment. The proportion of participants prescribed psychiatric drugs did not differ between groups in the years before study inclusion (pre-baseline; absolute risk difference 5% [95% CI −7 to 16], p=0·4263) or after intervention (10% [−6 to 24], p=0·2175). Treatment for mental and behavioural disorders did not differ between groups before baseline (2% [−10 to 14], p=0·7135); however, adolescents in the surgical group had more specialised psychiatric treatment in the 5 years after obesity treatment than did the control group (15% [1 to 28], p=0·0410). There were few patients who discontinued psychiatric treatment post-surgery (three [4%] receiving psychiatric drug treatment and six [7%] receiving specialised care for a mental disorder before surgery). In the surgical group, self-esteem (RSE score) was improved after 5 years (mixed model mean 21·6 [95% CI 19·9 to 23·4]) relative to baseline (18·9 [17·4 to 20·4], p=0·0059), but overall mood (MACL score) was not (2·8 [2·7 to 2·9] at 5 years vs 2·7 [2·6 to 2·8] at baseline, p=0·0737). Binge eating was improved at 5 years (9·3 [7·4 to 11·2]) relative to baseline (15·0 [13·5 to 16·5], p<0·0001). Relative changes in BMI were not associated with the presence or absence of binge eating at baseline. Interpretation: Mental health problems persist in adolescents 5 years after bariatric surgery despite substantial weight loss. Although bariatric surgery can improve many aspects of health, alleviation of mental health problems should not be expected, and a multidisciplinary bariatric team should offer long-term mental health support after surgery. Funding: Swedish Research Council, VINNOVA, Västra Götalandsregionen, ALF VG-region, Region Stockholm, Swedish Child Diabetes Foundation, Swedish Heart and Lung Foundation, Tore Nilsson's Foundation, SUS Foundations and Donations, Capio Research Foundation, and Mary von Sydow's Foundation.

Original languageEnglish
Pages (from-to)210-219
Number of pages10
JournalThe Lancet Child and Adolescent Health
Issue number3
Publication statusPublished - 2020

Subject classification (UKÄ)

  • Surgery
  • Psychiatry
  • Pediatrics


Dive into the research topics of '5-year mental health and eating pattern outcomes following bariatric surgery in adolescents: a prospective cohort study'. Together they form a unique fingerprint.

Cite this