INDEPENDENT THINKING. WHOLE-FOOD LIVING.NO ANIMALS / NO OIL
THENANODIET
← Blood pressure / Health Desk

Vegetarian diets and blood pressure: why trials and observational studies differ

The 2014 review found lower blood pressure in both evidence streams, but their estimates should not be merged into one promise.

Whole-food ingredients prepared without added oil

Illustrative food image, not a photograph of the study intervention.

Two evidence streams in one review

Yokoyama and colleagues reviewed controlled clinical trials and observational studies of vegetarian diets and blood pressure. Of 258 studies identified, seven clinical trials and 32 observational studies met their inclusion criteria. Participants were adults older than 20, and the outcome was a mean difference in blood pressure rather than a count of cardiovascular events. The investigators pooled results using a random-effects model. Keeping the two evidence streams separate is central to reading this review: a controlled dietary comparison and an observational comparison of people who already eat differently do not provide the same basis for interpreting a lower average pressure.

Yokoyama Y, Nishimura K, Barnard ND, et al.. JAMA Internal Medicine 174(4):577–587, 2014. PMID 24566947 · DOI

The controlled-trial estimate

The seven controlled trials included 311 participants with a mean age of 44.5 years. Vegetarian diets were associated with systolic pressure 4.8 mm Hg lower than omnivorous comparison diets, with a 95% confidence interval from 3.1 to 6.6 mm Hg lower. The corresponding diastolic difference was 2.2 mm Hg lower, with a confidence interval from 1.0 to 3.5 mm Hg lower. Both pooled comparisons had P values below 0.001. Reported I-squared was zero for each outcome, but that does not make 311 participants a large evidence base or establish identical effects for every vegetarian dietary pattern.

Yokoyama Y, Nishimura K, Barnard ND, et al.. JAMA Internal Medicine 174(4):577–587, 2014. PMID 24566947 · DOI

The observational estimate was larger and less consistent

The 32 observational studies included 21,604 participants, whose mean age was 46.6 years. Compared with omnivorous diets, vegetarian diets were associated with systolic pressure 6.9 mm Hg lower and diastolic pressure 4.7 mm Hg lower. The confidence intervals were 4.7–9.1 mm Hg lower for systolic pressure and 3.1–6.3 mm Hg lower for diastolic pressure. However, heterogeneity was substantial: I-squared was 91.4% and 92.6%, respectively. These observational averages should not replace the smaller trial estimates in a headline that implies an intervention effect. The review reported different results for different study designs, not one universal expected reduction.

Yokoyama Y, Nishimura K, Barnard ND, et al.. JAMA Internal Medicine 174(4):577–587, 2014. PMID 24566947 · DOI

A blood-pressure review, not a vegan outcome trial

The abstract concludes that vegetarian diets are associated with lower blood pressure and could provide a nonpharmacologic approach to reducing it. Its numbers concern systolic and diastolic pressure, not demonstrated reductions in heart attacks or mortality. The exposure category is vegetarian diets; the abstract does not establish an effect specific to a strictly vegan, zero-extracted-oil pattern. A careful summary therefore keeps the 4.8/2.2 mm Hg trial differences distinct from the 6.9/4.7 mm Hg observational differences. The finding supports further attention to dietary patterns without turning a pooled group average into a guaranteed personal response or a claim about unmeasured clinical endpoints.

Yokoyama Y, Nishimura K, Barnard ND, et al.. JAMA Internal Medicine 174(4):577–587, 2014. PMID 24566947 · DOI