A China-adapted planetary diet tracked 14% lower deaths

The planetary health diet was designed to feed ten billion people without wrecking the planet. It was also designed globally, which means it corresponds to no actual national cuisine, and asking a household in Chengdu to eat it as written is asking them to eat like nobody they know.

A study in the European Journal of Nutrition took the obvious next step, which had not previously been done for China. It rebuilt the diet around Chinese food, scored nearly 20,000 adults against the result, and asked whether the people closest to it lived longer.

They did, by a margin that only just clears the bar.

What the planetary health diet asks for

The reference version comes from a 2019 commission known as EAT-Lancet, and it is a plate rather than a rule list. Half a plate of fruits, vegetables and nuts. The other half consists of primarily whole grains, plant proteins (beans, lentils, pulses), unsaturated plant oils, modest amounts of meat and dairy, and some added sugars and starchy vegetables.

It emphasizes a plant-forward diet where whole grains, fruits, vegetables, nuts and legumes comprise a greater proportion of foods consumed. Animal foods are not banned: meat and dairy constitute important parts of the diet but in significantly smaller proportions than whole grains, fruits, vegetables, nuts and legumes.

The projected health effect is large. Global uptake of the planetary health diet can reduce approximately 11 million premature adult deaths annually.

That figure is a model, not an observation, which is precisely why studies like this one matter.

Why the planetary health diet needed a Chinese version

A reference diet written for the world specifies grams of food groups. It does not specify what those grams look like on a table in Hunan, where the staple grain, the oil, the vegetables and the protein sources all differ from the assumptions baked into a global average.

So the researchers built one. They constructed a 14-component CPHDS (range 0-140) based on the 2019 EAT-Lancet Commission recommendations, the latest Chinese Dietary Guidelines, and existing dietary scoring frameworks, while incorporating adaptations specific to Chinese dietary practices and nutritional characteristics. CPHDS is the Chinese Planetary Health Diet Score.

Then they applied it. Using data from the China Health and Nutrition Survey (1991-2015), we included 19,660 adults without cardiovascular disease or cancer at baseline.

Diet was not captured once and forgotten. Dietary intake was repeatedly assessed using three consecutive 24-hour dietary recalls, which is a stronger record than the single food-frequency questionnaire most large cohorts rely on.

What the score tracked in deaths

Nobody scored well. The mean CPHDS was 50.3 (SD 11.3) points, out of 140.

Over a median follow-up of 9.1 years, 1654 deaths were recorded.

Set the top quarter of scorers against the bottom, and the top showed a 14% lower risk of all-cause mortality. Read across the whole range rather than the extremes, each 10% increment in CPHDS was associated with a 7% lower mortality risk.

Both point the same way, and both should be read with the confidence they earn. The top-versus-bottom estimate reaches almost to no difference at its upper edge, and the trend test sits a whisker inside the conventional threshold. This is a real signal at the boundary of what this cohort can detect.

What the diet score did to emissions, water and land

The environmental half is where the numbers are firm.

A higher CPHDS was associated with reduced greenhouse gas emissions, water use and land use, each estimated with ranges so narrow they leave no ambiguity about direction.

That contrast is worth noticing. The same score predicts the environmental outcome sharply and the mortality outcome loosely, which is what you would expect: emissions follow mechanically from what people ate, while death follows from everything about a life.

The authors’ conclusion holds both: greater adherence to the Chinese Planetary Health Diet was associated with reduced diet-related environmental burdens and a lower risk of all-cause mortality in Chinese adults.

What a diet score cannot show

This is observational. People who eat closer to a plant-forward pattern differ from people who do not in income, education, smoking and much else, and no adjustment removes all of it.

Diet was self-reported, even if repeatedly, and 24-hour recalls miss what people did not eat that day.

The cohort also spans 1991 to 2015 in a country whose food supply changed enormously across that period, which is a strength for variation and a complication for interpretation.

The paper is paywalled, so the authors’ own limitations section was not available for this piece.

What this planetary diet means for your kitchen

The score does not transfer. It was built for Chinese foods and would misjudge a Mediterranean or a Northern European plate.

What transfers is the shape underneath it, and it is the same shape almost every line of nutrition evidence points at: eating a variety of foods, including vegetables, fruits, and whole-grain products, with meat present in smaller proportions than the plant foods around it.

The novel part of this study is that the same shift tracked two things at once. Nutrition advice usually asks people to choose between what is good for them and what is good for the planet. On this evidence, in this population, the two moved together.