Vegetarian eating: what the trials moved and what they did not

Two parts of this question have been through randomised trials, and they point in different directions. In 681 adults with type 2 diabetes across nine trials, vegetarian and vegan diets lowered HbA1c by 0.36 percentage points against omnivorous eating, with LDL down 0.16 mmol/L and BMI down 0.94 kg/m2, and no change in systolic blood pressure or fasting glucose. Everyone in those trials already had diagnosed type 2 diabetes, and that drop has not been shown in people without it. In ten trials covering 545 adults, the same style of eating showed no significant change in C-reactive protein (WMD -0.04, 95% CI -0.23 to 0.15), the inflammation marker it is most often credited with lowering. In most of those trials CRP was a secondary endpoint, and nobody can be blinded to what they are eating, so read that as an absence of a measured difference. An umbrella review of 30 reviews points the other way, linking vegetarian eating to lower CRP, and rates its own certainty low to very low. The cancer figure people quote, 13 percent lower total incidence, comes from seven prospective cohorts rather than trials, so it describes people who chose this way of eating along with everything else they chose. Iron stores read lower as well. In cross-sectional comparisons, vegetarian adults carried serum ferritin 29.71 ug/L below omnivores, and in men the gap was 61.88 ug/L. Ferritin tracks how much iron is stored, and a low reading is not the same thing as anaemia.

What your body absorbs

• Adult vegetarians carried serum ferritin 29.71 ug/L lower than omnivores, and the gap was widest in men at 61.88 ug/L. — Adults in 24 cross-sectional comparisons of vegetarians with non-vegetarian controls (n = 27 cross-sectional and 3 interventional studies, 24 pooled) Evidence C source
  Serum ferritin -29.71 ug/L (95% CI -39.69, -19.73); men -61.88 ug/L, premenopausal women -17.70 ug/L
• Habitual vegetarians took in 0.88 mg less zinc a day and had serum zinc 0.93 umol/L lower than non-vegetarians. — Males and females on habitual vegetarian diets in 26 studies with 48 comparisons (n = 34 studies reviewed, 26 pooled) Evidence C source
  Zinc intake -0.88 mg/day (SE 0.15), serum zinc -0.93 umol/L (SE 0.27); larger in women, vegans and developing countries

What cooking and storage change

• A review of literature from 2010 to 2023 ties phytates and oxalates in plant foods to poorer absorption of B12, DHA, EPA and iron, so planning, supplements and food preparation carry the risk. — Human studies of vegan and vegetarian diets and neurological outcomes published 2010 to 2023 (n = not stated in abstract) Evidence C source
  Antinutritional factors impair nutrient absorption; deficiencies of B12, DHA, EPA and iron linked to cognitive decline and mood disturbance

What it does to your health

• Across seven prospective cohorts, vegetarians had 13% lower total cancer incidence than non-vegetarians (RR 0.87, 95% CI 0.84 to 0.91). — Seven population-based prospective cohorts in the UK, the Netherlands and the USA, 10,210 to 472,377 participants each, followed 6 to 20.3 years (n = 17 publications from 7 cohorts) Evidence C source
  RR 0.87 (95% CI 0.84-0.91) for total cancer, I2 = 0%, WCRF grading probable
• In adults with type 2 diabetes, vegetarian or vegan diets lowered HbA1c by 0.36 percentage points (95% CI -0.54 to -0.19) against omnivorous diets. — 681 adults with type 2 diabetes in 9 randomised controlled trials (n = 9 RCTs, 681 participants) Evidence A source
  HbA1c WMD -0.36% (95% CI -0.54, -0.19), LDL -0.16 mmol/L, BMI -0.94 kg/m2; no change in systolic blood pressure or fasting glucose
• Pooling 11 randomised comparisons in 545 adults, vegetarian diets showed no significant change in C-reactive protein against omnivorous diets (WMD -0.04, 95% CI -0.23 to 0.15), with CRP mostly a secondary endpoint. — 545 adults in 10 randomised trials of at least 3 weeks, lacto, lacto-ovo and vegan arms (n = 10 RCTs, 11 effect sizes, 545 participants) Evidence A source
  CRP WMD -0.04 mg/L (95% CI -0.23, 0.15), p = 0.683, I2 = 0.0%

Safety, limits and interactions

• In 1,994 Seventh-day Adventists, who use fortified foods and supplements more often than vegetarians in general, plant-based eaters and omnivores showed no difference in serum B12 (MD -9.85 pmol/L, 95% CI -45.64 to 25.94), with intake heterogeneity at I2 90%. — 1,994 Seventh-day Adventists, vegan or vegetarian, with omnivore controls, in 4 observational studies (n = 4 studies, 1,994 participants) Evidence C source
  Serum B12 MD -9.85 pmol/L (95% CI -45.64, 25.94), intake MD 3.31 mcg/d (95% CI -4.70, 11.32); I2 50% and 90%
• Pooling 26 cross-sectional studies of 23,783 adults, vegetarians and vegans scored higher on orthorexia symptoms than omnivores (Cohen's d 0.46, 95% CI 0.33 to 0.60). — 23,783 adults, 72.0% female, mean age range 19.6 to 51.0 years, in 26 cross-sectional studies (n = 26 studies, 23,783 participants) Evidence C source
  Standardised mean difference 0.46 (95% CI 0.33, 0.60), odds ratio 1.99 (95% CI 1.21-3.25); I2 81.0% and 92.8%

Who this works differently for

• Among 48,626 children in 59 studies, vegetarian and vegan eating went with higher odds of iron deficiency, anaemia and B12 deficiency, while group averages stayed inside paediatric reference ranges. — Children under 18 without chronic disease: 7,280 lacto-ovo-vegetarians, 1,289 vegans, 40,059 omnivores (n = 59 studies, 48,626 participants) Evidence C source
  Higher odds of iron deficiency and anaemia in lacto-ovo-vegetarians and of B12 deficiency in vegans; lower height, weight, BMI z-scores, total and LDL cholesterol
• Across 45 studies, vegetarian diets raised intake of the plant omega-3 precursor ALA yet left EPA and DHA concentrations lower, most of all on vegan diets. — Humans on defined vegetarian diets, with attention to pregnancy and growth, in 45 studies (n = 45 studies) Evidence C source
  Higher LA and ALA intake, significantly lower EPA and DHA status; microalgae oil raised serum DHA

What is still unknown

• An umbrella review of 30 reviews found an inverse association between vegetarian diets and CRP, and rated the certainty of that evidence low to very low. — 30 systematic reviews covering 225 primary studies in adults with at least one chronic condition (n = 30 reviews, 225 primary studies) Evidence C source
  Inverse association of vegetarian diet with CRP, certainty of evidence low to very low
• Over about 7 years in 10,219 adults of five ethnic groups, sticking to a vegetarian diet showed no link with change in kidney function, while each extra gram of fibre a day went with a smaller eGFR decline. — 5,776 women and 4,443 men of Dutch, South-Asian Surinamese, African Surinamese, Turkish and Moroccan origin in Amsterdam, median follow-up 81 months (n = 10,219 participants) Evidence C source
  No association of vegetarian adherence with eGFR or ACR change; fibre +1 g/day linked to 0.10 (95% CI 0.02, 0.17) smaller eGFR decline in women

The bottom line

If you eat this way, iron and B12 are the two to watch, and they behave differently. The ferritin gap above is measured in blood and it is widest in men. B12 is the one that can be managed: in 1,994 Seventh-day Adventists, plant-based eaters and omnivores showed no difference in serum B12 at all, and the authors put that down to fortified foods and supplements. That zero belongs to that group. Adventists reach for fortified foods and supplements more often than vegetarians in general, so the result does not carry across on its own, and intake varied so much between those studies that heterogeneity came out at I2 90%. Zinc runs about 0.88 mg a day lower with serum zinc 0.93 umol/L lower, and that gap was larger in women, in vegans and in studies from developing countries. A review of work published between 2010 and 2023 points at phytates and oxalates in plant foods as part of why B12, iron, DHA and EPA land short, which puts food preparation alongside the shopping list. Children are a separate case. Across 59 studies and 48,626 children, vegetarian and vegan eating went with higher odds of iron deficiency and anaemia, and with B12 deficiency in vegan children, while group averages stayed inside paediatric reference ranges and height, weight and BMI z-scores sat lower. Take that to your paediatrician before you take it to a forum. Long-chain omega-3 is the other predictable shortfall. Across 45 studies, intake of the plant precursor ALA went up while EPA and DHA in the blood stayed lower anyway, most of all on vegan diets. Some of those studies tested microalgae oil and recorded higher serum DHA in the people taking it. What to do with that finding is a conversation with your doctor. One finding here is uncomfortable and we are leaving it in. Pooling 26 cross-sectional studies of 23,783 adults, vegetarians and vegans scored higher on symptoms of orthorexia, a standardised difference of 0.46 and odds of 1.99. Cross-sectional means nobody can say which came first, and the studies disagreed with each other a lot. Three things on this page have no evidence card behind them, and saying so is more useful than filling the space. There is no USDA table here because the database measures foods and this is a way of eating, so we cannot show you a composition row or a typical portion. We found nothing testing the claims that circulate about vegetarian eating, which is partly an artefact of where claims are registered: the American and European registers are indexed by substance rather than by diet. And the strength of the evidence needs a flag of its own. An umbrella review of 30 reviews covering 225 primary studies found vegetarian diets inversely associated with CRP and rated that certainty low to very low, which sits against the randomised pooling quoted higher up. Kidney function is the one place we have a clean null. Over about seven years in 10,219 adults of Dutch, Surinamese, Turkish and Moroccan background living in Amsterdam, sticking to a vegetarian diet showed no link with change in kidney function, while each extra gram of fibre a day went with a 0.10 smaller decline in eGFR in women.

Sources

1. vgtn-r1-05 · systematic review and meta-analysis of cross-sectional studies
   The meta-analysis which combined data of 24 cross-sectional studies showed that adult vegetarians have significantly lower serum ferritin levels than their non-vegetarian controls (-29.71 µg/L, 95% CI [-39.69, -19.73], p < 0.01)
   Crit Rev Food Sci Nutr, 2018 · checked 2026-09-23
2. vgtn-r1-06 · systematic review and meta-analysis of studies in humans
   Dietary zinc intakes and serum zinc concentrations were significantly lower (-0.88 ± 0.15 mg day(-1), P < 0.001 and -0.93 ± 0.27 µmol L(-1), P = 0.001 respectively; mean ± standard error) in populations that followed habitual vegetarian diets compared with non-vegetarians.
   J Sci Food Agric, 2013 · checked 2026-09-23
3. vgtn-r1-12 · critical systematic review of the literature
   However, deficiencies in critical nutrients such as vitamin B12, DHA, EPA, and iron have been consistently associated with an increased risk of cognitive decline, mood disturbances, and neurodegenerative disorders. Additionally, the presence of antinutritional factors like phytates and oxalates may further impair nutrient absorption, necessitating careful dietary planning and supplementation.
   Nutrients, 2025 · checked 2026-09-23
4. vgtn-r1-01 · systematic review and meta-analysis of prospective observational studies
   The summary RRs (95% CIs) for vegetarians vs. non-vegetarians was 0.87 (0.84-0.91, I2 = 0%, n = 4 studies) for total cancer incidence
   Eur J Epidemiol, 2026 · checked 2026-09-23
5. vgtn-r1-02 · systematic review and meta-analysis of randomized controlled trials
   Nine RCTs (681 participants) were included in this meta-analysis. The results indicated that vegetarian or vegan diets could reduce glycosylated hemoglobin, type A1C (HbA1c) (weighted mean difference [WMD] = -0.36%; 95% CI: -0.54, -0.19; P < .001)
   Nutr Rev, 2025 · checked 2026-09-23
6. vgtn-r1-03 · systematic review and meta-analysis of randomized controlled trials in adults
   The meta-analysis of 11 eligible studies involving 545 participants, illustrated that vegetarian diets did not significantly change CRP levels in comparison to the control group (Weighted Mean Difference (WMD): -0.04, 95% CI: -0.23, 0.15; p = 0.683)
   BMC Nutr, 2026 · checked 2026-09-23
7. vgtn-r1-09 · systematic review and meta-analysis of observational studies
   Meta-analysis showed no significant differences in serum vitamin B12 levels (MD: -9.85 pmol/L; 95% CI: -45.64 to 25.94 pmol/L; P = 0.54, I2 = 50%) or daily intake (MD: 3.31 mcg/d; 95% CI: -4.70 to 11.32 mcg/d; P = 0.42, I2 = 90%) between plant-based Adventists and omnivore controls, although there was high heterogeneity between the studies.
   Am J Health Promot, 2025 · checked 2026-09-23
8. vgtn-r1-10 · systematic review and meta-analysis of cross-sectional studies
   Adults who followed VVDs had moderately higher ON symptoms compared to omnivores (standardized mean differences using Cohen's d index = 0.46; 95% confidence interval [CI]: 0.33, 0.60; inconsistency index [I 2] = 81.0%)
   Int J Eat Disord, 2026 · checked 2026-09-23
9. vgtn-r1-07 · systematic review and meta-analysis of nutritional and health outcomes
   Searches in MEDLINE, Embase, Scopus, and Web of Science up to March 2025 identified 59 studies including 48,626 participants (7,280 lacto-ovo-vegetarians, 1289 vegans, and 40,059 omnivores)...increased odds of iron deficiency and anemia were observed in lacto-ovo-vegetarians, and vitamin B12 deficiency in vegans
   Crit Rev Food Sci Nutr, 2026 · checked 2026-09-23
10. vgtn-r1-08 · systematic review of PUFA status in humans
    A total of 45 studies were involved; the review found that vegetarian diets increased linoleic acid (C18:2 n-6, LA) and alpha-linolenic acid (C18:3 n-3, ALA) intake but resulted in significantly reduced concentrations of eicosapentaenoic acid (C20:5 n-3, EPA) and docosahexaenoic acid (C22:6 n-3, DHA), particularly in vegan diets.
    Lipids, 2026 · checked 2026-09-23
11. vgtn-r1-04 · umbrella review of systematic reviews and meta-analyses
    Additionally, a significant inverse association was identified between a vegetarian diet and CRP levels, with low to very low CoE.
    Nutr Rev, 2026 · checked 2026-09-23
12. vgtn-r1-11 · multiethnic prospective cohort study
    We did not find an association between vegetarian diet adherence and eGFR or ACR change for both sexes. However, a 1 g/day higher fiber intake at baseline was associated with a lower decrease in eGFR of 0.10 (95% confidence interval [CI] 0.02, 0.17) in women
    J Ren Nutr, 2026 · checked 2026-09-23