Cohort studyindependent cross-checkabstract checked★ key study I-VEG-107 in-depth review
Vegetable benefit for the heart may be overestimated
Raw and Cooked Vegetable Consumption and Risk of Cardiovascular Disease: A Study of 400,000 Adults in UK Biobank.
Overview
Feng Q et al.. Front Nutr 2022 · PMID 35265657 · DOI 10.3389/fnut.2022.831470
In 400,000 UK adults, raw vegetables were linked to 11% lower CVD risk, but adjustment for lifestyle and social factors explained over 80% of the association.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Separate raw and cooked vegetable effects on heart disease.
- How was the study done?
Prospective cohort with confounding analysis.
- How many people, animals or samples were studied?
399,586 adults.
- What was actually found?
Raw (not cooked) vegetables associated with lower CVD; mostly explained by confounding.
- How large was the observed effect?
HR 0.89 (0.83-0.95).
- What do the results mean in everyday life?
Eating vegetables may matter less for heart disease than many studies suggest.
- What are the study's limitations?
Crude dietary measure.
- What does it say about Elawady's specific claim?
Weakens the strength of one argument against the claim, but says nothing in favour of eliminating vegetables.
- Do the results apply to Germany and Europe?
UK data; directly applicable.
- What can explicitly not be concluded from this study?
Cannot show that a diet with almost no vegetables is healthy.
Scientific in-depth review
1. Full reference
- Authors
- Feng Q et al.
- Journal
- Front Nutr
- Year
- 2022
- Identifier
- PMID 35265657 DOI 10.3389/fnut.2022.831470
- Verification
- verified – PubMed efetch 2026-10-26; Crossref 2026-10-26
- Retracted
- no
- Corrections
- none
- Extent checked
- abstract only
- Conflicts of interest and funding
COI statement: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
2. Research question
Are raw and cooked vegetable intakes independently associated with CVD incidence and mortality, and how much of this is explained by confounding?
3. Study design and population
- Study design
- Cohort study – Prospective cohort (UK Biobank) with assessment of residual confounding via likelihood-ratio statistics
- Population
UK Biobank adults without prior CVD, mean age 56, 55% women
- Country
United Kingdom
- Period
Baseline 2006-2010; 12 years follow-up
- Sample size
399,586 (18,052 CVD events; 4,406 CVD deaths)
- Inclusion and exclusion criteria
not stated in abstract
4. Methods and endpoints
- Exposure / intervention
Raw and cooked vegetables (tablespoons/day; means 2.3 and 2.8)
- Comparator
Lowest intake
- Endpoints
Major CVD incidence and mortality
- Statistical methods
Multivariable Cox regression; % reduction in likelihood-ratio statistic after confounder adjustment
5. Actual results
- Results
Raw vegetables highest vs lowest: CVD HR 0.89 (0.83-0.95), CVD mortality 0.85 (0.74-0.97); cooked vegetables 1.00 (0.91-1.09) and 0.96 (0.80-1.13). Adjustment reduced the LR statistics for raw vegetables by 82% and 87%.
- Absolute risks
not reported
- Authors' conclusion
Higher raw, not cooked, vegetable intake was associated with lower CVD risk; residual confounding likely accounts for much, if not all, of the associations; the burden attributed to low vegetable intake should be reappraised.
6. Strengths
- Very large cohort
- Explicit quantification of confounding
7. Weaknesses
- Crude intake measure (tablespoons, single baseline)
- Healthy-volunteer cohort
- Wide categories; low intakes overall
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- simplified assessment (abstract only)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Measurement error may bias toward null; confounding analysis is informative.
9. Assessment of the statistical interpretation
CIs log-symmetric (stats/C-VEG-16_mortality.py). The 82-87% LR reduction indicates strong confounding of the crude association.
10. Relevance to the specific claim
The strongest observational evidence questioning a causal vegetable-CVD link; does not test vegetable elimination.
11. Independent contrary findings
Wang 2021 and Aune 2017 meta-analyses find consistent inverse associations; fibre RCTs show risk-factor benefits.
12. Transferability to Germany and Europe
Directly European (UK).
13. Overall methodological judgement
Important caveat: the size of the vegetable-CVD benefit is uncertain. It does not show that avoiding vegetables is harmless, because nutrient adequacy (fibre, folate, potassium, vitamin C) is a separate issue.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | direct |
| Reproducibility | not replicated |
| Consistency with the overall evidence | inconsistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Whether better intake measures change the conclusion.
15. Linked fact checks
- Avoiding (nearly) all vegetables, salads and legumes is compatible with good health. – Studies with mixed or unclear results