Vegetarian diet and likelihood of becoming centenarians in Chinese adults aged 80 y or older: a nested case-control study - 2026 [Paper][Observational]
Inverse associations of vegetarian diet with morbidity and mortality have been observed; however, the role of vegetarian diet on exceptional longevity remains unrevealed.
Objectives - This study aims to examine the association between a vegetarian diet and likelihood of becoming a centenarian in adults aged ≥80 y.
Methods - This prospective nested case-control study included 5203 participants aged 80+ y from the Chinese Longitudinal Healthy Longevity Survey, a nationally representative cohort initiated in 1998. Participants were classified as omnivores and vegetarians, and further into vegetarian subgroups (pesco-vegetarians, ovo-lacto-vegetarians, and vegans) based on consumption of animal-derived foods. The primary outcome was living to 100 y old by the end of follow-up (2018). Multivariable unconditional logistic regression models were used to evaluate the association analysis.
Results - The study identified 1459 centenarians and matched them with 3744 noncentenarians (who had deceased before reaching 100 y). Relative to omnivores, vegetarians had a lower likelihood of becoming centenarians [odds ratio (OR): 0.81, 95% confidence interval (CI): 0.69, 0.96], and similar patterns were observed for vegans (OR: 0.71, 95% CI: 0.54, 0.98), but not for pesco-vegetarians (OR: 0.84, 95% CI: 0.64, 1.09) and ovo-lacto-vegetarians (OR: 0.86, 95% CI: 0.67, 1.09). The significant association was seen in individuals with BMI <18.5 kg/m2 (OR: 0.72, 95% CI: 0.57, 0.91), but not for those with BMI ≥18.5 kg/m2 (OR: 0.92, 95% CI: 0.73, 1.17) (P-interaction = 0.08).
Conclusions - Targeting individuals of advanced age (80+ y) in China, we found that individuals following a vegetarian diet had a lower likelihood of becoming centenarians relative to omnivores, underscoring the importance of a balanced, high-quality diet with animal- and plant-derived food composition for exceptional longevity, especially in the underweight oldest-old.
https://doi.org/10.1016/j.ajcnut.2025.101136 - Not open access, but....
5 replies
I'm just going to say that this study uses the standard confidence interval of 95%. This means it is a 19 out of 20 chance that the true value is between those two values. That means that there is a 1 in 20 that it is outside those. With several of those values being very close to 1 it is not a strong prediction.
That said it seems that having fish, eggs, and milk in your diet is better than not, but the difference is small enough with large enough uncertainty and lots of overlapping ranges that I would say this study doesn't show much of anything. It is fairly weak and should not be used to hold any belief strongly.
I agree with the conclusion based on previous studies and prior probability, but this study is just not very well powered and doesn't really move the needle for me.
it's not nothing - the paper has directionally interesting data, the CIs didn't cross one... if nothing else these observational findings need to be accounted for in efforts to remove animal foods, especially in older populations
It did cross 1 for the ovolacto and pesceterian groups, as well as those with a higher BMI. That said, yes, something like this can suggest a likely direction and be used to justify further studies, but the statistical power is simply too low to make any actual conclusions based on this study.
That said, it would be less likely for this result to occur if animal based foods caused a negative outcome, so if you were to take any conclusions from this then concluding anything positive towards vegan or vegetarian diets would be unreasonable given this data.
Also those ranges are really large, 0.54 to 0.98, with a value of 0.71. That is anywhere between halving the likelihood or making almost no change to the likelihood of making it to 100 for vegans. 0.98? It is saying a 95% chance that the true value is between a 2% reduction of likelihood of reaching 100 or a 46% reduction in likelihood of reaching 100. But any further detail is quite uncertain. Is it a 29% reduction in the chance of reaching 100? Probably something close to that, but it really could be no change or basically a halting of the chance. We don't know from this study.
is applicable to just about every study in human nutrition. Solid read of the stats, and I agree with your assessment.
Omnivores have 40% higher odds of living to 100 years old vs vegans. (of chinese people who are 80 years old)
The less meat you eat, the lower your odds of reaching 100.
Normal limitations of observational studies apply - hypothesis generating, cannot establish causation, associations only, FFQs