Very high refined-grain intake and death/heart disease in 21 countries

Overview

Swaminathan S et al.. BMJ 2021 · PMID 33536317 · DOI 10.1136/bmj.m4948

People eating about 7 portions/day of refined grains had 27% higher mortality and 33% more major heart events than those eating very little.

Does the study support the specific claim? Partly · Methodological quality: high

Original publication

Plain-language summary

Scientific in-depth review

1. Full reference

Authors

Swaminathan S, Dehghan M, Raj JM, Thomas T, Rangarajan S, Jenkins D, Mony P, Mohan V, Lear SA, Avezum A, Lopez-Jaramillo P, Rosengren A, Lanas F, AlHabib KF, Dans A, Keskinler MV, Puoane T, Soman B, Wei L, Zatonska K, Diaz R, Ismail N, Chifamba J, Kelishadi R, Yusufali A, Khatib R, Xiaoyun L, Bo H, Iqbal R, Yusuf R, Yeates K, Teo K, Yusuf S

Journal

BMJ

Year

2021

Identifier

PMID 33536317 DOI 10.1136/bmj.m4948 · PMC7856570

Verification

verified – PubMed E-utilities efetch, 2026-10-25

Retracted

no

Corrections

none

Extent checked

full text

Conflicts of interest and funding

ICMJE form: support as detailed in the paper; no financial relationships with interested organisations (PubMed record).

2. Research question

Are intakes of refined grains, whole grains and white rice associated with mortality, major CVD, lipids and blood pressure?

3. Study design and population

Study design

Cohort study – Multinational prospective cohort (PURE), Cox frailty models with random intercepts by centre

Population

Adults aged 35-70 y, urban and rural, 21 low-, middle- and high-income countries

Country

21 countries (North America/Europe, South America, Africa, Middle East, South Asia, South East Asia, China); Egypt not identified in the sections read

Period

Recruitment 2003 onward, follow-up to July 2019 (median 9.5 y)

Sample size

148,858 enrolled; 137,130 analysed

Inclusion and exclusion criteria

Excluded baseline CVD

4. Methods and endpoints

Exposure / intervention

Refined grains (refined wheat flour products: white bread, pasta/noodles, breakfast cereals etc.), whole grains, white rice (analysed separately) by country-specific FFQ

Comparator

Lowest intake category (<50 g/day refined grains)

Endpoints

Composite of death or major CVD; total mortality; major CVD; lipids; blood pressure

Statistical methods

Multivariable Cox frailty models

5. Actual results

Results

Composite events 9.2% (n=12,668). Refined grains >=350 g/day vs <50 g/day: total mortality HR 1.27 (95% CI 1.11-1.46; P trend 0.004); major CVD HR 1.33 (1.16-1.52; P trend <0.001); composite HR 1.28 (1.15-1.42). Higher refined grain intake associated with higher systolic BP. No significant associations for whole grains or white rice. Full text: median refined grain intake 116 g/day overall; per 200 kcal: white bread HR 1.05 (1.03-1.07), whole wheat bread 1.01 (0.98-1.03), cooked rice 0.98 (0.96-1.00).

Absolute risks

Event proportions by category not extracted; overall 9.2% composite events.

Authors' conclusion

High intake of refined grains was associated with higher risk of mortality and major CVD events; globally, lower consumption of refined grains should be considered.

6. Strengths

• Very large, multinational, diverse intake range
• Prospective design, standardised outcomes
• White rice analysed separately

7. Weaknesses

• Diet measured once at baseline by FFQ
• Residual confounding acknowledged by authors
• Effect seen in extreme intake category

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

Observational; confounding by socioeconomic status and overall diet pattern possible; single baseline FFQ.

9. Assessment of the statistical interpretation

Relative risks for the top versus bottom category; the top category (>=350 g/day) is about three times the median intake. Association, not causation.

10. Relevance to the specific claim

Main supporting study for C-WHT-03; its mechanism discussion contradicts C-WHT-01.

11. Independent contrary findings

A-WHT-02, I-WHT-02 (refined grains not clearly associated).

12. Transferability to Germany and Europe

Europe included as a region; direction plausibly transferable.

13. Overall methodological judgement

Robust large cohort showing an association at very high intakes; does not support the archive's mechanism.

14. Open questions

Region-specific estimates (Middle East) and Egypt-relevant intake levels.

15. Linked fact checks

• Heavy intake of refined grains (white bread etc.) is associated in the largest population studies with higher mortality and major cardiovascular disease (archive: PURE, +27% mortality, +33% major cardiovascular events at >=350 g/day). – Studies supporting the claim
• Refined white flour turns into a sticky glue-like paste in the stomach, sticks to the stomach wall, impairs stomach contractions and thereby causes reflux and indigestion, whereas whole-grain flour does not stick and is digested easily. – Studies contradicting or limiting the claim
• Rice in all forms (including white rice) is allowed without restriction; 'rice is sugar' and two spoons of rice a day sustain life. – Studies with mixed or unclear results

Use by the Tayebat archive

Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.

What the archive says about the study

Massive meta-analysis of 137,000 people across 21 countries: >=350 g/day refined grains raised mortality by 27% and major CVD by 33% (AR headline adds: and raises systolic blood pressure).

A. What was actually studied?

Association of refined grains, whole grains and white rice with mortality and CVD in one prospective cohort.

B. What did the authors find?

HRs as in results; no association for whole grains or white rice.

C. What conclusions do the authors themselves draw?

Lower refined grain consumption should be considered.

D. What does the archive derive from it?

Uses the study as confirmation of Elawady's 'sticky gel' mechanism and of white-flour avoidance.

E. Is this derivation scientifically justified?

Partly. The numbers are reported correctly, but the study is called a meta-analysis, associations are framed as causal ('raised'), the extreme-category nature is not explained, and the study's null finding for white rice (allowed freely by the system) is omitted. The authors' proposed mechanism (rapid digestion and absorption) contradicts the archive's glue/malabsorption mechanism.

Adequacy of the interpretation (by the archive)

partly correct but over-interpreted