All-cause and cause-specific mortality risk in individuals with eating disorders: systematic review and meta-analysis of relative risk and aggravating or attenuating factors.
Level 3 - non-randomized controlled study
Systematic review and meta-analysis of observational cohort and case-control studies
PubMed 41536100 · doi:10.1002/wps.70014
What was done
A PRISMA 2020-compliant systematic review and random-effects meta-analysis assessed all-cause and cause-specific mortality risk ratios (RRs) in cohort and case-control studies comparing individuals with eating disorders (EDs) to the general population or matched control groups. Study quality was evaluated using design-specific US National Institutes of Health tools. The meta-analysis included 83 studies with 307,710 patients with EDs and 15,719,076 general population controls (mean follow-up: 11.96 years; 94.35% female; mean age: 25.52 years; study quality: 65.0% good, 21.7% fair, 13.3% poor).
What was found
Any ED was associated with increased all-cause mortality versus the general population (RR=4.92, 95% CI: 4.03-6.00), ranging from RR=5.52 (95% CI: 4.47-6.82) in anorexia nervosa (AN) to a non-significant difference in binge eating disorder. Suicide-related mortality was elevated (RR=8.45, 95% CI: 5.73-12.47; AN RR=9.86, 95% CI: 5.63-17.27; bulimia nervosa RR=6.15, 95% CI: 2.52-15.04). Mortality from natural and non-natural causes was also increased (RR=3.47, 95% CI: 2.29-5.25 and RR=6.46, 95% CI: 4.62-9.04, respectively). All-cause mortality increased with lower body mass index, shorter follow-ups, male sex, any psychiatric comorbidity, and comorbid substance use, alcohol use, mood, or personality disorders.
Why it matters
This synthesis confirms substantial excess mortality across eating disorders, particularly from suicide and natural causes. It identifies high-risk clinical profiles—including AN diagnosis, low body mass index, male sex, and psychiatric comorbidities—that warrant close monitoring immediately after diagnosis.
Limits
The included literature consists of observational cohort and case-control studies subject to residual confounding. Over one-third (35.0%) of the analyzed studies were rated as fair or poor quality. The pooled cohort was predominantly female (94.35%), and findings for binge eating disorder showed no statistically significant mortality increase.