Plant-based diets and CRP: a marker, not a verdict
A pooled difference in C-reactive protein did not translate into broad proof of reduced inflammation.

Illustrative food image, not a photograph of the study intervention.
What was actually pooled
Menzel and colleagues reviewed associations of vegan and vegetarian diets with circulating inflammatory biomarkers compared with omnivorous diets. Their searches extended through April 2020 and initially identified 1,073 publications. Twenty-one cross-sectional studies met the inclusion criteria. That design is central to the interpretation: the review compared dietary groups without randomly assigning people to change their diets. A pooled association can describe differences between those groups, but it cannot establish that diet alone produced them. The review also considered multiple biomarkers, so its findings should not be compressed into a single unqualified claim that every measure of inflammation was lower on plant-based diets.
Menzel J, Jabakhanji A, Biemann R, Mai K et al.. Scientific reports, 2020. PMID 33303765 · DOI
The vegan CRP estimate
For vegan diets compared with omnivorous diets, the pooled mean difference in C-reactive protein was −0.54 milligrams per liter. The 95% confidence interval ran from −0.79 to −0.28 milligrams per liter, with P below 0.0001. This is an absolute difference in the measured biomarker, not a percentage reduction in disease risk. It also compares groups in cross-sectional research, rather than measuring the effect of a prescribed dietary change in a randomized trial. The result supports a specific statement about lower observed CRP in vegan groups. It does not supply a personal treatment target or establish that a particular plant-based recipe will change someone's CRP.
Menzel J, Jabakhanji A, Biemann R, Mai K et al.. Scientific reports, 2020. PMID 33303765 · DOI
Vegetarian findings were less pronounced
The vegetarian comparison showed a smaller pooled CRP difference: −0.25 milligrams per liter, with a 95% confidence interval from −0.49 to 0.00 and P of 0.05. The authors described this association as less pronounced than the vegan finding. Keeping the two dietary categories separate avoids overstating the evidence. The abstract also reports a substantially larger association in patients with impaired kidney function, but that subgroup should not be used to promise the same magnitude of change in a general audience. These are different comparisons within a cross-sectional evidence base, not a controlled contest establishing a universally superior diet or a predictable individual response.
Menzel J, Jabakhanji A, Biemann R, Mai K et al.. Scientific reports, 2020. PMID 33303765 · DOI
Do not turn one signal into every signal
Beyond CRP, the review examined markers including interleukin-6, interleukin-18, tumor necrosis factor-alpha, and several adhesion or metabolic markers. The abstract reports no substantial effects for the other inflammatory biomarkers. It also notes that most biomarkers had been investigated only in single studies, leaving a limited basis for broader conclusions. The useful takeaway is therefore narrow: vegan and vegetarian patterns were associated with lower CRP, with a clearer pooled association in the vegan comparison. Neither prevention of a clinical disease nor a general reduction across the inflammatory system was demonstrated by this review. The authors called for further research rather than treating the biomarker evidence as settled.
Menzel J, Jabakhanji A, Biemann R, Mai K et al.. Scientific reports, 2020. PMID 33303765 · DOI