Plant-Forward Diets and Cognitive Aging in China

A promising cognition signal, with an environmental asterisk

A more plant-forward diet may track with better cognitive aging in older Chinese adults. It does not yet prove that changing your meals will prevent dementia, and it does not guarantee a smaller environmental footprint. Those distinctions matter because the newest study on the subject reached both conclusions at once.

In a 2025 observational analysis of older adults in the Chinese Longitudinal Healthy Longevity Survey, higher adherence to a China-adapted planetary health diet index (PHDI) was associated with lower incidence of cognitive impairment during follow-up: hazard ratio (HR) 0.539. A higher animal diet index (ADI) was associated with higher risk, with an HR of 1.379. Higher grain intake was the only individual component reported separately and was associated with lower risk, HR 0.729. The published information cited here does not report confidence intervals or the cohort size for these estimates, so it cannot establish their precision or statistical significance. Still, the direction of the findings fits several other prospective Chinese studies of plant-forward eating patterns and late-life cognition.

The catch is in the paper's subtitle: Li et al.'s 2025 study frames the question as a balance between health benefits and environmental sustainability. In its modeling, greater PHDI adherence also came with higher environmental burdens. A diet label is not a life-cycle assessment.

What the main study actually tested

The PHDI was a China-adapted dietary score, not a direct test of the original EAT-Lancet reference diet. It captured closer adherence to a locally defined planetary-health-style pattern; the ADI captured a more animal-food-oriented pattern. Researchers then examined subsequent cognitive impairment in older adults rather than assigning participants to diets.

That design can identify a useful association. People with higher PHDI scores had lower observed risk of incident cognitive impairment, while those with higher ADI scores had higher observed risk. It cannot tell us how much of that difference came from the food itself. Education, income, social conditions, physical health, frailty, appetite changes, activity, and access to care can all shape both diet and cognitive outcomes. In older populations, early illness can also alter eating before cognitive impairment becomes apparent.

The sensible reading is narrower than "the planetary health diet prevents dementia": a plant-forward dietary pattern was associated with better cognitive outcomes in this cohort.

The outcome also matters. These studies examine cognitive impairment, cognitive decline, or cognitive function—not one identical, adjudicated dementia endpoint across every paper. The results speak to cognitive aging markers, not confirmed dementia prevention in every setting.

Why the result is plausible, but still observational

The 2025 finding is not alone. A 2024 CLHLS cohort analysis by Lin et al. found that stronger adherence to a Chinese version of the MIND diet was associated with reduced cognitive decline in older Chinese individuals. MIND-style eating emphasizes plant foods and brain-health-oriented choices, but this was also cohort evidence rather than a randomized diet trial.

A 2022 study developing the cMIND diet similarly linked higher adherence with lower cognitive impairment in older Chinese people. Cultural adaptation is useful here: it tests whether the signal appears in a dietary framework built around Chinese eating patterns rather than assuming a Western score travels perfectly.

Two later analyses add context rather than certainty. A 2023 prospective Chinese cohort study reported that an animal-based dietary pattern interacted with green space in relation to cognitive function. And a 2025 10-year cohort analysis linked plant-based dietary patterns and living conditions to cognitive function among Chinese adults aged 65 and older who had normal cognition at baseline. Diet may matter, but residential and social conditions matter too. A bowl of beans cannot repair every disadvantage outside the kitchen.

Evidence beyond cognition points in a related direction without answering the same question. In a 10-year cohort of 2,004 older adults, a 2024 study examined plant-based diet indexes and activities-of-daily-living disability. Functional disability is not cognitive impairment, so it should not be folded into the cognition result. It does, however, support the broader possibility that diet quality tracks with healthier aging.

Why "better for the planet" does not automatically follow

Planetary-health diet scores combine nutrition with an environmental ambition, but their real-world footprints depend on local agriculture, food processing, transport, cooking, portion sizes, and the metric used. Greenhouse-gas emissions, land use, and water use can point in different directions. A lower-carbon diet may use more water; a healthier score may not improve every environmental measure.

China-specific evidence makes that problem concrete. A 2023 prospective study of Chinese adults linked adherence to a planetary health diet with mortality outcomes while reporting mixed environmental effects, including potentially higher water-use burdens. A 2024 Chinese cohort study likewise examined plant-based and planetary-health diets, mortality, and environmental burden, with environmental results that depended on the dietary pattern and context.

Food-level variation is large enough to make broad labels unreliable. A 2025 dataset of 1,000 Cantonese and Sichuan dishes found substantial variation in nutritional and environmental measures across dishes, shaped by ingredients and cooking techniques. "Plant-based" does not specify whether a meal is water-intensive, highly processed, or environmentally light. It specifies a direction of eating, not a footprint receipt.

This is not a China-only complication. A 2025 assessment of eight diet-quality indexes and seven sustainability metrics found that healthier diets were not uniformly more sustainable. A systematic review of Mediterranean-diet studies also found that environmental conclusions varied with the footprint measured and the comparator diet.

Conflicting health evidence deserves more than a footnote

The cognition literature does not license a universal health claim for every planetary-health diet score. A 2024 global systematic review and meta-analysis reported associations between the emerging EAT-Lancet planetary health diet and increased major cardiovascular disease and all-cause mortality risk. That finding concerns different populations, outcomes, diet definitions, and pooled observational studies; it does not directly refute the Chinese cognitive-impairment association. It does rule out a lazy conclusion that every version of this framework improves every health outcome.

Different studies can disagree because they score diets differently, compare different foods, follow different populations, and measure different outcomes. Those are not minor technicalities. Replacing red meat with legumes, refined starches, sugary drinks, or heavily processed plant foods produces different health and environmental questions.

What to do with this now

For someone trying to improve the odds of healthy cognitive aging, the evidence supports a practical, modest move: build more meals around grains, vegetables, fruit, legumes, and nuts, and reduce heavier reliance on animal foods. That approach matches the direction of several Chinese observational studies, including the 2025 PHDI analysis.

Keep the claim proportionate. No randomized trial in this evidence base shows that a China-adapted planetary-health diet prevents dementia. No result here predicts an individual's cognitive future. And no dietary score guarantees a lower environmental burden without examining the foods, sourcing, and footprint metric involved. The current evidence offers a credible observational signal for cognition and a conditional environmental story. The next decisive evidence would be randomized diet trials with cognitive endpoints, repeated dietary and cognitive measurements, and direct environmental comparisons of actual Chinese dietary patterns.