An interview with Nicholas Poh-Jie Tan about his recent paper with a massive three-sample database showing that the link between depression and meat avoidance is weak and not likely to be causal, casting serious doubt on past claims of a relationship.
Nicholas, you recently published a Psych Science paper, “Associations between meat consumption and depression are small and unlikely to be causal”. What were the central aims of this paper, what motivated the research, and what did you find?
The central aim of our paper was to test whether eating less meat causes people to feel depressed. We were interested in this topic because existing research on this topic has largely been cross-sectional, rather than longitudinal – for example, see the systematic reviews by Jain et al. (2022) and Dobersek et al. (2021) – and has not provided evidence about the causal direction.
I also found this topic interesting because I was beginning to see some influential individuals like Jordan Peterson advocating that eating meat could be used to treat or protect against depression (Peterson is one of the leading advocates of the ‘all-meat’ diet). This is problematic given the lack of evidence for such claims. Although we were interested in the predictors and consequences of animal consumption in general, we focused on meat because of these kinds of claims, and because of the kind of data we used.
In our study, we examined the causal relationship between eating meat and depression across three very large longitudinal panel study datasets (total N = 77,678). We replicated past research by finding that there was a very small significant correlation of around .05 between eating less meat and feeling more depressed.
There was a very small significant correlation of around .05 between eating less meat and feeling more depressed.
This association was much weaker than the association between other correlates of depression such as time spend on social media, alcohol use, and exercise, which suggested that animal consumption is not a major factor in depression in general.
We also found that the between-person relationship was curvilinear such that there appeared to be a slight uptick in depression at very high ends of meat consumption, but a slight downtick at very low ends. This suggests that, although there is a positive correlation, people who eat excessive amounts of meat may be higher risk for depression than people who eat typical amounts, whereas people who abstain from meat might be at similar or even possibly lower risk than reducetarians.
We also found that during years where individuals ate less meat than usual, they were significantly more likely to feel depressed, however, this effect was very small, around .01. Perhaps most important, eating less meat than usual at one time point did not significantly predict feeling more depressed at the next time point. If there was a causal relationship, you would expect that dietary changes would precede mood changes in this way – but that’s not what we found.
Your research confronts the claim that there might be a relationship between avoiding meat and depression. What are some of the more common theories or arguments for why depression and meat consumption might relate?
Several arguments have been made for a causal relationship between meat avoidance and depression. One hypothesis is that meat provides nutrients like iron and vitamins B12 and D that protect against depression. Despite studies that have already ruled out this nutrients deficiency explanation, such claims still persist and circulate on social media.
There are also non-causal hypotheses. For instance, it could be that people who become depressed have a lower appetite and thus eat less meat, along with other kinds of food. Third variable explanations also exist. Though some previous studies have ruled out gender as a third variable, other third variable explanations have been made. For instance, climate change concerns might cause people to be both depressed and to eat less meat. Another third variable explanation is that people stop eating meat because they realise how horrific the animal agriculture industry is, and this causes them to be depressed.
It looks like some researchers have theorised that meat reduction, specifically, might be causally linked to depression. Why might being in a process of meat reduction cause depression? Did you find any evidence to support this idea?
Meat reducers are usually attitudinally ambivalent. On the one hand, they are still attached to eating animals, but on the other hand they are compelled by reasons not to. Because people make meal choices multiple times per day, they face this conflict regularly. It is fairly well-established that conscious ambivalence is distressing – that is why dissonance reducing strategies are so common and effective, because they help people avoid being aware of ambivalence. It is intuitive that a person who concentrates on this conflict and their internal ambivalence will tend to be more distressed than either a person who is fully committed to not consuming animals, or a person who consumes animals regularly and doesn’t give it much thought.
In your study, what was the reason to treat meat consumption as a continuous measure, as opposed to a categorical measure (i.e., meat consumers vs. avoiders)?
We chose to treat meat consumption as a continuous variable for three reasons. First, we used representative samples, and the proportion of meat abstainers in the general public is very low. Dividing people into those who do and do not eat meat would have resulted in very uneven samples.
Second, we were interested in variation among meat eaters, from people who eat meat once in a while to those who eat it a lot, because nutrition-based hypotheses generally propose that the more meat you eat the less prone to depression you should be. Third, it is well-established that dichotomizing continuous variables sacrifices reliable variance, and doing so in this study would have reduced power to detect effects.
In the study you compared three different samples from the Netherlands, Germany, and Australia, and each sample had methodological differences. How comparable were the measures from the three datasets? Could differences in measurement across the datasets affect the results?
Overall, the methods and results were highly comparable across the three samples. This can be seen from the consistency in the pattern of effects across all three samples and when the three samples were combined into a single “mega sample”.
Finding similar patterns in three samples from two regions of the world, using data at different timescales with slightly different measures, gives us confidence that our findings are relatively robust and generalisable.
Your team found small correlations between meat consumption and depression at the “between-person level”. You also found that “within-person associations” between meat consumption and depression were “exceedingly small”. Can you explain the difference between these different levels of analysis – “between-” vs. “within-person” correlations – and what they mean for your results?
Between-person associations compare average differences between individuals (e.g., people who eat more meat tend to be less depressed). Between-person associations are typically computed at a single time point (i.e., cross-sectionally), however, in our study we computed our between-person association by averaging across several time points, which led to more reliable estimates.
The between-person association from our study is analogous to the findings from past research conducted to date examining the relationship between meat consumption and depression. Importantly, the between-person association does not provide evidence of causality and could be confounded by other between-person variables (e.g., sociodemographic variables, personality traits).
Within-person associations examine how changes over time in one person, relative to their own average, are related. This metric asks: during years when a person eats more meat, are they are also more depressed, relative to their own typical level of depression? If there was a causal relationship between meat consumption and depression, we would expect that people would become more depressed during times when they eat less meat. In particularly, we would expect that a person who eats less meat today would become more depressed at some point in the future. However, we found no evidence to support this kind of lagged within-person effect.
Do you see your findings as a sizable step in silencing concerns about the relationship between meat avoidance and depression?
Our study suggests that changes in animal consumption is very unlikely to cause depression. To the extent that concerns about this are based on evidence, this study, along with other emerging research, should silence such concerns. But such concerns are often not based on evidence.
Our study suggests that changes in animal consumption is very unlikely to cause depression.
Your findings are based on self-reports of meat consumption, which we know can be unreliable – for example, with people reporting eating less meat than they actually consume. Have there been studies of meat consumption and mental health that have moved away from self-report measures of meat consumption?
Yes! Conner et al. (2025) conducted a study in which they delivered either meat or plant-based meat alternatives to participants in two experimental conditions. They measured wellbeing and nutrition indicators from blood samples. They found no differences between the groups. This finding aligns with our study in suggesting that reducing animal consumption does not increase risk for depression.
What are the next steps for research into meat consumption and mental health? Are there other controversial or outstanding questions that need empirical attention?
It might be interesting to examine other kinds of psychopathology. For instance, people with eating disorders typically restrict certain food groups, and animal products is one food group that is commonly restricted. Such individuals are often at risk of depression.
More experimental work, such as the study by Conner et al., would be useful as well. Culture may be an important moderator, and almost all research on this topic has been conducted with Western samples.
Finally, there is some anecdotal evidence from the advocacy community that people burn out because it is distressing to think about animal agriculture and animal suffering day to day. It would be interesting to turn this question around and ask about how a vegan worldview comes with emotional risks, and to figure out how those risks can be mitigated, so people can stay engaged with the movement.
(For resources on how to manage vegan advocacy burnout, you may wish to visit the Center for Effective Vegan Advocacy; see also ‘Tools and Practices’ at End Infighting.)
The featured article can be found here:
Poh-Jie Tan, N., Krämer, M. D., Haehner, P., Bleidorn, W., & Hopwood, C. J. (2026). Associations between meat consumption and depression are small and unlikely to be causal. Psychological Science, 37(3). https://doi.org/10.1177/09567976261426226
If you’d like to get in touch with the Nicholas about his research: Nicholas.tan@psychologie.uzh.ch
Cover photo by K. Mitch Hodge
Interview questions and blog by Jared Piazza at PHAIR Society.
PHAIR Society 