Mortality in British vegetarians: review and preliminary results from EPIC-Oxford.
Level 3 - non-randomized controlled study
Prospective non-randomized cohort studies with mortality follow-up
PubMed 12936946 · doi:10.1093/ajcn/78.3.533S
What was done
Authors described and reviewed three British prospective cohort studies: the Health Food Shoppers Study (~11,000 subjects, established 1970s), the Oxford Vegetarian Study (~11,000 subjects, established 1980s), and preliminary results from EPIC-Oxford (~56,000 subjects, established 1990s). Dietary exposure was assessed via short questionnaires in the first two studies and detailed food frequency questionnaires in EPIC-Oxford. Mortality was tracked via the National Health Service Central Register, reporting standardized mortality ratios (SMRs) against the UK population and death rate ratios (DRRs) comparing vegetarians to nonvegetarians adjusted for age, sex, and smoking.
What was found
Overall mortality in all three cohorts was markedly lower than the UK national average: SMRs were 59% (95% CI: 57%, 61%) for the Health Food Shoppers Study, 52% (95% CI: 49%, 56%) for the Oxford Vegetarian Study, and 39% (95% CI: 37%, 42%) for EPIC-Oxford. Comparing vegetarians with nonvegetarians within each cohort, all-cause DRRs were 1.03 (95% CI: 0.95, 1.13), 1.01 (95% CI: 0.89, 1.14), and 1.05 (95% CI: 0.86, 1.27), respectively. DRRs for ischemic heart disease mortality in vegetarians versus nonvegetarians were 0.85 (95% CI: 0.71, 1.01), 0.86 (95% CI: 0.67, 1.12), and 0.75 (95% CI: 0.41, 1.37).
Why it matters
Health-conscious British participants had substantially lower overall mortality than the general UK population regardless of diet. Within these cohorts, vegetarian diets were not associated with reduced all-cause mortality, though non-significant reductions in ischemic heart disease deaths were consistent with larger pooled analyses.
Limits
Observational prospective design subject to residual confounding beyond age, sex, and smoking. EPIC-Oxford data were preliminary with wide confidence intervals. Earlier cohorts relied on brief dietary questionnaires rather than detailed validated instruments. Causes of morbidity other than fatal endpoints were not evaluated.
Cited by
- supports The UK Shopper Study found that meat eaters and vegetarians had similar rates of all-cause mortality when healthy behaviors in vegetarians were accounted for.