Human health

Observational studies on the risks of animal-sourced foods have yielded mixed and inconclusive results

Observational evidence is uncertain and inconsistent

The epidemiological evidence in favour of plant-based diets might seem compelling at first glance, but does it hold up under scrutiny? Not really, at least not beyond conjecture. Self-reported dietary data, which forms the backbone of nutritional epidemiology, are plagued by inaccuracies, memory-based errors, and biases, undermining the reliability of its conclusions. Moreover, the associations presented are based on weak relative risks, which fail to meet the thresholds necessary for causal claims, especially when confounding factors are likely to be present. Lastly, selective reporting of evidence tends to exaggerate risks while ignoring neutral or protective findings, perpetuating an overly simplistic narrative about the relationship between diet and health.

What is the case against animal-sourced foods?
Argument 1 - Beware of problematic input data
Argument 2 - Weak associations are only indicative
Argument 3 - Lifestyle confounding and healthy-user bias
Argument 4a - Cherry picking (neutral outcomes)
Argument 4b - Cherry picking (protective outcomes)

Additional commentary on the specific case of the Blue Zone argument can be found elsewhere on this website.

What is the case against animal-sourced foods?

Some observational studies suggest potential health benefits of vegan and vegetarian diets, such as lower risks of obesity, cardiovascular disease, diabetes, certain cancers, and mortality. Conversely, high consumption of red and processed meats has sometimes been associated with increased disease and mortality risks. However, the evidence is inconclusive, and direct causal relationships cannot be established due to underlying methodological limitations and biases, as argued below.

Argument 1

Beware of problematic input data

Epidemiological research commonly relies on self-reported dietary data, mostly through food frequency questionnaires, which are prone to memory-based errors, biases, and inaccuracies.

Pre-defined food categories in these studies may fail to align with regulatory definitions or lack standardization, resulting in inconsistent classifications, particularly for ‘red’, ‘white’, and ‘processed’ meats. Self-reported dietary patterns, such as vegetarianism, can also be misclassified due to vague definitions and the inclusion of semi-vegetarians to increase sample sizes.

Social desirability bias further compromises data reliability, as respondents tend to report diets that align with perceived health norms—for example, obese individuals often underreport calorie intake, while self-identified (semi-)vegetarians may underreport occasional meat consumption.

Finally, survivorship bias further distorts findings by overrepresenting individuals genetically predisposed to succeed on vegetarian or vegan diets, while the majority find it challenging to maintain such diets over the long term. Consequently, studies on plant-based diets may disproportionately reflect the outcomes of a minority who can sustain these diets successfully.

CLICK HERE for evidence

Food frequency questionnaires have been criticized

The input information for epidemiological assessments is usually obtained from surveys and food frequency questionnaires, providing an often seriously problematic way of data collection for a number of reasons [Goris & Westerterp 1999; Black et al. 2001; Herbert et al. 2002; Schatzkin et al. 2003; Bernstein & Loftus 2009; Lutomski et al. 2011; Archer et al. 2013, 2015, 2018a, 2018b; Speakman 2020; Händel et al. 2021; Young et al. 2022; Brown et al. 2023].

In summary, such methods are highly prone to ‘misreporting’ because people 1) cannot accurately estimate the amount of food they are eating [], 2) have fallible memories for their intake, 3) may deliberately falsify reports, 4) may react during the period of recording by changing their intake, and 5) may report atypical eating behaviour that is unrepresentative of what they routinely eat [Bajunaid et al. 2025]. Additional problems are due to imperfect manners in which investigators convert this information into energy and nutrients [Bajunaid et al. 2025].

Self-reported energy intake, for instance, tends to be very inaccurate and its misreporting correlates with the misreporting of (macro)nutrients and, likely, of specific foods, which then adds bias into the relationship between diet and health outcomes. [Bajunaid et al. 2025].

When combining such unreliable self-reports with the uncertainty of food composition data, needed to estimate the intake of specific nutrients, the bias is such that it can only be addressed by using biomarkers [Ottaviani et al. 2024].

Because of the severity and ubiquity of the misreporting problem, some researchers are of the opinion that there is no further benefit in publishing studies in which participants estimate their own dietary intake [Dhurandhar et al. 2015]. The result only being that this leads to a ‘fictional discourse' on diet-disease relations [Archer et al. 2018b], based on ‘pseudoscientific’, ‘fatally flawed’, and ‘inadmissible’ evidence [Archer et al. 2015, 2018a].

Problematic pre-defined food categories

The use of pre-defined food categories in questionnaires also leads to problems. In the case of meat, the categories and descriptors used for muscle foods display heterogeneity and often do not match those of public regulatory definitions [O'Connor et al. 2020]. The difference between what is considered ‘red’ and ‘white’ meat is particularly problematic [Keeton & Dikeman 2017]. Processed and (red) meats are often grouped together in 'all meat' even though they may lead to very different results [Männistö et al. 2010]. Even within the category of ‘processed meat’, there is considerably heterogeneity, which is expected to lead to different health outcomes [De Smet & Van Hecke 2024]. The same reservation is valid for plant-based foods as well, as ‘plant-based diets’ display a gradient from 'healthy' to 'unhealthy' options [Elliott et al. 2023].

Poor categorization of ‘vegetarianism’

An underreported intake of meat has been reported for self-defined vegetarians and other health-conscious groups [Haddad & Tanzman 2003]. In a UK survey, only 1/4 respondents who considered themselves to be reducing meat consumption had cut down on a variety of meats over the year [Richardson 1993]. Moreover, some of the studies claiming to look at 'vegetarian diets' are in reality combining true vegetarians with semi-vegetarians to avoid low sample sizes. Therefore, the term 'vegetarian' in self-reports requires caution [Juan et al. 2015]. In the Oxford Vegetarian Study, 23% of 'non-meat-eaters' ate meat occasionally but less than once a week, or ate fish, or both [Appleby et al. 1999].

Social desirability bias

While memory-based self-reporting issues undermine data robustness, social desirability bias may lead to unreliable data during self-reporting because the respondents tend to have their reporting affected by expected health norms.

For instance, whereas actual intake of sucrose based on biomarker assessment is positively related with being obese, self-reported intake tends to be negatively associated [Kuhnle et al. 2015]. An analysis of the NHANES cohort showed a that fewer calories were reported on average by individuals with obesity than people that were not obese [Ford & Dietz 2013]. A high BMI value is not only associated with an underreporting of energy intake, but also with overreporting protein intake and underreporting fat intake [Bajunaid et al. 2025]. Whereas the actual intake of fruits and vegetables (based on vitamin C as a biomarker) is associated with less obesity, this is not the case when intake is self-reported [Bingham et al. 2007].

Survivorship bias

Survivorship bias is also likely, since many vegans and vegetarians (up to 70-80%) rapidly revert to eating animal-sourced foods, a third even within three months of their change of diet [Faunalytics 2014, 2015]. Only 12- 24% of current vegans may be in the diet for >5 years [Kerschke-Risch 2015; FCN 2018;VOMAD 2019], 7% for >10 years, and 3% for >20 years [VOMAD 2019]. Studies looking into 'plant-based' diets may have a selection bias for the minority of subjects that do well on such diets. Several genes, especially those with functions in lipid metabolism and brain function, are associated with vegetarianism and may underlie the ability to subsist on a vegetarian diet [Yaseen et al. 2023].

Argument 2

Weak associations are only indicative

Robust methodological frameworks, such as the Bradford-Hill criteria and GRADE protocol, are required for evaluating the causal effects of dietary factors on health outcomes.

These studies typically report very low relative risks (RR), often below thresholds considered meaningful in evidence-based medicine. Generally, RR values less than 2 are poorly informative and should not be used to infer strong causal claims, especially when there is a suspicion of confounding. Only strong risk factors, like the effect of visceral fat on colorectal cancer (RR of 6), provide RRs of sufficient magnitude to assume a reliable causal relationship.

Comprehensive assessments have determined that the evidence supporting both the risks of red and processed meats and the benefits of plant-based diets is methodologically weak, inconsistent, and of (very) low certainty. Low RR values also parallel trivial absolute risks, which are rarely disclosed.

CLICK HERE for evidence

The problem with low levels of relative risk (RR)

Observational studies looking into the protective and harmful associations between plant-based eating or animal source food consumption, respectively, typically comes with low relative risk (RR) levels, if any. Moreover, the absolute risk (AR) is usually trivial and often not reported [cf., Johnston et al. 2019]. Taking into account that there can be an abundance of false-positive findings [Bofetta et al. 2008; Young & Karr, 2011], and because of the large bias and uncertainties [see below], such low risk levels would not be considered as strong evidence in most epidemiological research outside nutrition. Even if a threshold theoretically depends on the risk prevalence in the reference group [Sainani 2011], RR values <2 should be considered as poorly informative [Shapiro 2004]. They should not be used to infer strong causal claims, especially when there is a strong suspicion of confounding [McAfee et al. 2010; Alexander & Cushing, 2011; Alexander et al. 2015; Klurfeld 2015; Feinman 2018; Leroy et al. 2018; Händel et al. 2021]. Hazard ratios (HR), specifically, come with their own issues regarding causal inference [Hernán 2010; Uno et al. 2014].

RR levels are overshadowed by those of more important risk factors

To put things in perspective, the association of colorectal cancer with visceral fat is coupled to a RR of 5.9 for the highest compared to the lowest tertile [Yamamoto et al. 2010], overshadowing the value for meat eating (RR of 1.2 per 100g/d of red meat or 50g/d of processed meat). Likewise, the presence of the metabolic syndrome at baseline is to be seen as meaningful as it is associated with increased risk of colorectal cancer (HR 2.2) in postmenopausal women [Kabat et al. 2012], acting as a prognostic factor [Shen et al. 2010] and driven by hyperinsulinemia (e.g., via angiogenesis) [Liu et al. 2014]. It is within this gradient, created by differences in strength of association and the absence or presence of plausible mechanisms, that certainty of evidence should be established. To do so, methodological frameworks are needed, such as the Bradford-Hill criteria and GRADE protocol [BMJ Best Practice].

Causality cannot be inferred for red and processed meats

Application of Bradford-Hill criteria led to the conclusion that the causality of the associations between red and processed meat intake and cardiovascular disease can not be inferred due to weakness of associations and a lack of coherence with experimental evidence (although potential causality was mentioned for type-2 diabetes in the case of processed meats) [Hill et al. 2020, 2022]. The alleged link of processed meat with colorectal cancer has been criticized for demonstrating serious or critical risks of bias (risk of confounding, missing data, selective outcome reporting bias) [Händel et al. 2020] and in the case of red meat for displaying an unclear dose-response effect and weakening evidence over time [Alexander et al. 2015].

Low-certainty evidence is suggestive at best

Evidence for the protective effects of plant-based diets is typically of ‘low’ to ‘very-low’ certainty, for instance with respect to cardiovascular risk in Western cohorts [Kaiser et al. 2021; Selinger et al. 2023]. Using GRADE, very-low certainty evidence suggested there was no association between vegetarian diet and all-cause or cerebrovascular mortality, although a lower risk of ischemic heart disease mortality was reported, possibly affected by the healthier behaviours of vegetarians rather than their actual diet [Jabri et al. 2021]. For vegan diets specifically, combined evidence from two cohort studies showed a non-significant and low-certainty trend for lower all-cause mortality [Selinger et al. 2023].

Similarly, after the most comprehensive GRADE assessment of the evidence, the observational evidence used to incriminate red and processed meats for increased mortality or a variety of morbidities was deemed to be of 'low-' to 'very-low-' certainty [Han et al. 2019; Johnston et al. 2019; Vernooij et al. 2019; Zeraatkar et al. 2019; see elsewhere]. In support of this finding, another GRADE-based approach concluded that the recommendation to reduce the consumption of processed meat and meat products in the general population to lower the risk of cancer 'seems to be based on evidence that is not methodologically strong' [Händel et al. 2021]. Likewise, an umbrella review classified evidence for the link between red and/or processed meats and colorectal cancer as merely 'suggestive' (i.e., above 'weak' but below 'highly suggestive' and 'strong' [Papadimitriou et al. 2021]. Use of 'burden of proof risk function' methodology similarly indicated that the evidence is 'weak' for the link between unprocessed red meat and colorectal cancer, breast cancer, type-2 diabetes, and ischemic heart disease, and absent for stroke [Lescinsky et al. 2022].

Argument 3

Lifestyle confounding and healthy-user bias

Observational studies on the health effects of animal-source foods are influenced by the broader context of diet quality, lifestyle habits, and socio-demographics, which make it difficult to accurately determine the relationship between food intake and health outcomes.

Research indicates that red meat intake may increase disease risk in populations with low fruit and vegetable intake, but can have neutral to protective effects when combined with higher fruit and vegetable consumption. Healthier vegetarians and healthier meat eaters tend to have lower health risks compared to less healthy meat eaters, highlighting the importance of the dietary context.

‘Healthy-user’ bias complicates the analysis further. Individuals who moderate their meat intake are likely more health-conscious and socioeconomically advantaged, which skews results by indirectly linking low meat intake with better health outcomes. As a result, studies comparing vegetarians and meat eaters with similar socio-demographic characteristics tend to show no differences in mortality. Also, meta-analyses excluding US Seventh-Day Adventist cohorts, known for their health-focused lifestyles, show weaker or no benefits of vegetarian diets.

When statistically adjusting for confounders like smoking and physical activity, associations with health risks are often rendered non-significant. For example, accounting for body weight seems to dissipate the link between meat intake and inflammation. However, residual confounding remains practically unavoidable.

Finally, associations between red and processed meat consumption and health risks observed in North America may not apply globally. Cultural and regional differences seem to affect the findings, being weaker in Europe and Asia. When red meat and/or processed meats are part of wholesome traditional diets, health outcomes are benign, as shown for the Southern European Atlantic diet. The worldwide PURE cohort even demonstrated a protective association between red meat intake and cardiovascular health and mortality at the global level.

CLICK HERE for evidence

Role of lifestyles and the background diet

It is primordial yet very difficult to unambiguously disentangle specific dietary effects from the overall complexity and quality of diets and lifestyles within observational data sets. For example, less healthy vegetarians have no cardiovascular benefits compared with less healthy meat eaters, while both healthier vegetarians and healthier meat eaters have a lower risk than less healthy meat eaters [Petermann-Rocha et al. 2023]. Merely eating ‘plant-based' may not associate with protective shifts in markers of cardiometabolic health unless dietary quality is accounted for [Elliott et al. 2023].

Likewise, higher red meat intake increases cancer risk at low levels of fruit and vegetable intake, yet may show a neutral to protective association in people who also report an increased consumption of fruits and vegetables [Maximova et al., 2020]. This may or may not point to protective interactions, but can as well be indicative of more intricate confounding; in any case, such findings demonstrate that context is primordial. Similarly, for the Women's Health Initiative (WHI) [Zheng et al. 2022], UK Biobank [Petermann-Rocha et al. 2023], and ATTICA cohorts [Damigou et al. 2023], risk elevations seem to be due to dietary patterns rather than to the actual consumption of red meat per se. The Southern European Atlantic diet, as found in parts of Spain and Portugal, was identified as a protective eating pattern despite the fact that it deviates from the Mediterranean model due to the regular inclusion of red meat and pork products [Carballo-Casla et al. 2024].

In a population of adolescent girls, those girls consuming higher levels of both red meat and healthy plant foods (fruits and nonstarchy vegetables) had more favourable lipid markers at the end of adolescence than girls consuming lower levels of both food groups [Bradlee et al. 2013]. The authors concluded that: 'lean red meat may be included in a healthy adolescent diet without unfavourable effects on lipid values'.

The need to control for confounders

The overall health status of the investigated populations is also of primordial importance. The association between meat eating and inflammatory response or disease becomes non-significant or is at least substantially attenuated after adjustment for excess body weight [Montonen et al., 2013; Chai et al. 2017; Papier et al. 2021; Wood et al. 2023]. In prediabetic adults, association of processed meat consumption with the deterioration of glycemic markers and triacylglycerol levels disappeared when adjusted for weight increase [Zhu et al. 2022]. In the EPIC-Heidelberg cohort, the main ASFs were not meaningful as determinants for overall, cardiovascular, or cancer-related mortality risk, independently of smoking, alcohol consumption, and excess body weight [Bajracharya et al. 2023]. When medical history and dietary habits were taken into account in the Greek ATTICA cohort, positive association between the consumption of red meat (but not processed meat) and cardiovascular disease became non-significant [Damigou et al. 2023]. Yet, whereas such well-characterized factors as obesity can be statistically accounted for as such, at least to some degree, it is impossible to sufficiently correct for all lifestyle confounders. Even anxiety has recently been mentioned as a potential confounder of the associations between meat eating and colorectal cancer [Beslay et al. 2020]. Socio-demographic elements, in particular, are particularly difficult to account for.

With respect to mortality, vegetarianism has been associated with an advantage for studies of lower quality, while studies of good methodological quality do not tend to imply such an association [Dinu et al. 2017].

Healthy user bias

In observational studies, the problem of healthy user and related biases (e.g., healthy adherer) is a fundamental one [cf. Shrank et al. 2011]. It explains at least part of the deviations between the results of observational studies and randomized controlled trials [see also elsewhere]. Western upper-middle classes are inclined to eat less meat, as they are susceptible to what is perceived as virtuous eating, obediently following official advice [Leroy & Hite 2020]. Self-declared vegetarians tend to be more health conscious and socioeconomically advantaged [Bedford & Barr 2005], while plebeian eating patterns are less prone to dietary guidance. Meat intake thus parallels a higher intake of ultraprocessed foods, obesity, smoking, and lower physical activity [Alexander et al. 2015; Fogelholm et al. 2015; Grosso et al. 2017; Mihrshahi et al. 2017; Turner & Lloyd 2017; Hur et al. 2019]. Such associations captured by epidemiology then serve as positive feedback to reinforce the dietary recommendations that created them in the first place [Leroy & Hite 2020]. Essentially, the dietary guidelines behave as a self-fulfilling prophecy, reinforcing an animal/plant binary that started out as an ideological construct in the 19th century [see elsewhere].

When comparing people with similar socio-demographic characteristics, no differences in mortality are seen between vegetarians and meat eaters [Chang-Claude et al. 2005; Appleby & Key, 2016; Appleby et al. 2016; Mihrshahi et al. 2017]. Also, omission from meta-analyses of the atypical cohort of US Seventh-Day Adventist studies, describing a specific subset of communities following healthy lifestyles with a low degree of generalizability, results in the weakening or even disappearance of the beneficial associations between vegetarianism and cardiovascular health and all-cause mortality [Kwok et al. 2014; Dinu et al. 2017; FCN 2018].

To further illustrate this point, lifestyle confounding appears when comparing distinct cultural groups. For processed/red meats, associations found in North-America are not necessarily valid elsewhere, where they may be neutral or protective [see below; O'Sullivan et al. 2013; Wang et al. 2016; Grosso et al. 2017; ACC 2018; Iqbal et al. 2021]. Worldwide, bivariate correlation analyses revealed that meat intake is positively correlated with life expectancies and acted as one of the significant predictors of life expectancy, while carbohydrate crops showed weak and negative correlation [You et al. 2021]. Similarly, associations between egg consumption and type 2 diabetes typify US studies but vanish in European (neutral) and Asian (protective) studies [Djoussé et al. 2016; Tamez et al. 2016; Wallin et al. 2016; Drouin-Chartier et al. 2020].

Argument 4a

Cherry picking - dismissal of neutral outcomes

Developing a case against animal-sourced foods is often based on cherry-picked evidence from observational studies. In many studies there is no unambiguous association between their avoidance and improved health.

The evidence on unprocessed red meat, for instance, is conflicting, with some of the large studies showing no association. Poultry, eggs, and dairy display mostly neutral or even protective associations with health, while butter's associations with mortality, cardiovascular disease, and diabetes are relatively small or neutral.

CLICK HERE for evidence

Researchers with a priori health beliefs may display biased behaviour

Developing a case against animal source foods, in particular red and processed meats, often leads to a biased selection of evidence. This was, for instance, blatantly the case for the Nordic Nutrition Recommendations. Contradictory evidence was excluded from the evaluation, whereas even the sources cited did not support the recommendation to restrict red meat to ≤350 g/w based on colorectal cancer risk [Pouzou et al. 2024]. Such attempts are typically characterized by selective reporting, exaggerated cause-and-effect statements, and a refusal to refer to studies that find no associations, as has also been found in a critique of an influential meta-analysis favoring vegan and vegetarian diets [i.e., Dinu et al. 2017, as analyzed by Fenton & Gillis 2017].

Red and processed meats

The above-mentioned problem of a priori bias leading to questionable scientific behaviour seems particularly relevant for the case of red and processed meats. A critical analysis of studies looking into the link between processed meats and colorectal cancer has shown that '75% of the eligible studies had a moderate to serious risk of missing data, and about half of the studies had issues with bias in the selection of the reported results' [Händel et al. 2021]. One very large study within Harvard’s Pooling Project of Prospect Studies of Diet and Cancer did not find a link between red and processed meat intake and colorectal cancer, but was only abstracted and never published [Cho & Smith-Warner 2004], which has raised suspicion given the dietary beliefs prevailing at Harvard's School of Public Health [Butterworth 2007].

Many authors were indeed unable to affirm a clear link between veganism, vegetarianism, or red meat avoidance and lower mortality or morbidity due to various diseases of modernity, or only obtained harmful associations for some meat types (mostly processed ones) but not others, in particular of the unprocessed type [Heilbrun et al. 1989; Thun et al. 1992; Missmer et al. 2002; Flood et al. 2003; Huncharek et al. 2003; Key et al. 2003, 2009; Cho et al. 2004; Sato et al. 2006; Lee et al. 2008; Alexander & Cushing 2009; Alexander et al. 2009, 2010a,b,c; 2011; Wallin et al. 2011; Kappeler et al. 2013; Parr et al. 2013; Lee et al. 2013; Rohrmann et al. 2013; Kwok et al. 2014; Lippi et al. 2015; Mihrshahi et al. 2017; Hur et al., 2019; Mejborn et al. 2020; Wang et al. 2021; Iqbal et al. 2021; Zheng et al. 2022; Zhu et al. 2022; Blackie et al. 2023; Damigou et al. 2023; Silva de Medeiros et al. 2023; Myrmel et al. 2024; Nanri et al. 2024].

A significant association of colorectal cancer with unprocessed red meat only starts at ≥80 g/d, or 90 g/d if studies with combined consumption of red an processed meats are excluded [Pouzou et al. 2024], which is well below the average for high-income countries. Even above these levels, the evidence remains too uncertain and could as well be an artefact due to bias. The global PURE study did not find problematic associations between unprocessed red meat and cardiovascular disease or mortality [Mente et al. 2023; Mozaffarian 2023]. Similarly, the PREVIEW study indicated that only processed meat but not total or unprocessed red meat, poultry, dairy products, or eggs showed an association with weight regain and adverse risk factors [Zhu et al. 2022]. In the MESA study, neither processed not unprocessed forms of red meat were associated with any markers of inflammation [Wood et al. 2023]. In the EPIC cohort, processed meat intake above 20 g/d was associated with all-cause mortality, but this was only significant for men and for smokers and not for females or 'never smokers' [Rohrmann et al. 2013]. Mendelian randomization studies were unable to find causal effects of red and processed meats on cardiovascular disease outcomes or type-2 diabetes [Hu et al. 2024, Li et al. 2024].

As a result, the common narrative that red meat causes colon cancer and heart disease becomes messy. Even if vegetarians may display lower mortality from ischemic heart disease than meat eaters in some large prospective studies, no protective associations with mortality from cerebrovascular disease and various cancers were found in those same datasets (including colon cancer) [Key et al. 1999]. Also, no association was found between red meat and heart disease or diabetes (although processed meats paralleled higher risk) [Micha et al. 2010].

Poultry, eggs, and dairy

When arguing for restriction of animal source foods based on observational studies, poultry should be exonerated based on neutral [Lupoli et al. 2021; Zhu et al. 2022] or even protective findings [see below]. The same is valid for eggs, with the occasional association being limited mostly to a US context [Rong et al. 2013; Shin et al. 2013; Tran et al. 2014; Alexander et al. 2016; Djoussé et al. 2016; Tamez et al. 2016; Wallin et al. 2016; Dehghan et al. 2020; Xia et al. 2020; Krittanawong et al. 2021; Zhu et al. 2022; Myers & Stevenson Ruxton 2023]. The case against dairy leads to inconsistent yet mostly neutral results [Jayaraman et al. 2019, Zhu et al. 2022] and would overall lead to a net protective effect [Scrafford et al. 2020], while even the persistent narrative stating that it would cause prostate cancer is unlikely [Preble et al. 2019]. Cheese intake, for instance, is neutral for prostate cancer while offering protective associations for various other outcomes (see below) [Zhang et al. 2023]. For butter, a meta-analysis has revealed relatively small or neutral overall associations with mortality, CVD, and diabetes [Pimpin et al. 2016].

Total ‘animal protein’

With respect to total animal protein, no significant association with all-cause, cardiovascular, or cancer mortality was found, although intake of plant protein was associated with lower all-cause and cardiovascular but not cancer mortality [Nahghsi et al. 2020]. Also, no association with bladder cancer was found when substituting animal-based protein with plant-based protein, or the reverse [Beigrezaie et al. 2024].

The German Nutrition Society concluded that there is insufficient evidence for a type-2 diabetes risk increase with higher intakes of animal protein and a risk decrease with plant protein intake, mentioning inconsistency, lack of clear biological plausibility, poor methodological quality, contradictory information from intervention studies, and general uncertainty [Schulze et al. 2023]. A GRADE analysis of the relationship between vegan diets and cardiovascular disease indicated variable associations and weak to very weak quality of evidence [Kaiser et al., 2021].

Argument 4b

Cherry picking – dismissal of protective outcomes

The epidemiological case against animal-sourced foods is compounded by the selective reporting of data that supports a particular narrative while dismissing outcomes that contradict prior expectations and beliefs. For instance, there are many instances where vegetarian and vegan diets have been linked to poorer health outcomes and lower quality of life. Likewise, various studies associate the consumption of animal source foods with improved health.

To be consistent, those who rely on nutritional epidemiology to argue that we need more ‘plant-based’ eating would also need to acknowledge that some animal-sourced foods come with protective associations – implying that their underconsumption may be suboptimal for certain aspects of health.

The global PURE study, for instance, has associated health outcomes to the protective effects of fish, eggs, full-fat dairy, and poultry, while even red meat turned out to be benign. In several studies, red meat consumption has been specifically associated with a range of benefits, including lower morbidity, increased telomere length, and decreased risk of depression.

Conversely, some studies have linked vegetarian and vegan diets to poorer health outcomes and lower quality of life. However, it is essential to recognize that both pro-meat and pro-plant studies have limitations in establishing causal relationships.

CLICK HERE for evidence

Protective associations of omnivorous diets

Global analysis (based on the PURE study cohort) indicates that the healthiest dietary setups for cardiovascular protection include substantial amounts of animal-sourced foods besides plant foods, with whole-fat dairy and fish mentioned among the "protective" components, while meat is referred to as mostly neutral [Mente et al. 2023; Mozaffarian 2023].

Protective associations of animal-sourced foods (other than red meat)

Embracing observational data to incriminate omnivore diets would at the same time undermine the premise that vegan diets are the best choice to avoid chronic disease and death. Avoidance of animal-sourced foods that are associated with benefits would then have to be considered as a suboptimal dietary design for some populations, as could be argued for:

Fish and seafood [He et al. 2004; Bouzan et al. 2005; Larsson & Orsini 2011; Chowdhury et al. 2012; Zheng et al. 2012; Jiang et al. 2016; Zhao et al. 2016; Schwingschackl et al. 2017; Qin et al. 2018; Rimm et al. 2018; Black et al. 2019; Kim et al. 2019; Godos et al. 2024]

White meat [Shi et al. 2015; Xu et al. 2020; Lupoli et al. 2021; Alegria-Lertxundi et al. 2022; Damigou et al. 2023]

Eggs [Gopinath et al. 2020; Zhuang et al. 2020; Xu et al. 2020; Krittanawong et al. 2021; Pan et al. 2024]

Dairy [Shimizu et al. 2003; Aune et al. 2013; Gao et al. 2013; Astrup 2014; Tapsell 2015; Gijsbers et al. 2016; Lovegrove & Hobbs 2016; Wu & Sun 2017; Yoshida et al. 2019; Scrafford et al. 2020; Bhavadharini et al. 2020; Haugsgjerd et al. 2020; Jin et al. 2020; Poppitt 2020; Papadimitriou et al. 2021; Alegria-Lertxundi et al. 2022; Feng et al. 2022; Gaeini et al. 2022; Neisius et al. 2022; Yuan et al. 2022; Kim et al. 2023; Tanpowpong et al. 2023; Yuan et al. 2023; Villoz et al. 2023; Zhang et al. 2023; Anderson & Alpass 2024; Bjørklund Holven & Sonestedt 2024; Torres-Gonzalez et al. 2024; Xie et al. 2024; Zhuang et al. 2025].

Protective associations of red and processed meats

In contrast to the common narrative, the eating of red meat (or derived products) has in some instances been associated with protective outcomes, including a lower morbidity or mortality [Lee et al. 2013; ACC 2018; Yen et al., 2018; Black et al. 2019a, b], increased telomere length, potentially increasing life span [Kasielski et al. 2016, You et al. 2021], decreased risk of cognitive decline and neurogenerative disease [Xu et al. 2020; Zhang et al. 2021; Talebi et al. 2023], and lower incidence of depression [Kohl et al. 2022], or the effects of unprocessed red meat on waist circumference [Zandvakili et al. 2024].

Risk associations of vegan and vegetarian diets

Several studies have even linked vegetarian and vegan diets to poorer health outcomes and lower quality of life [Sebeková et al. 2001; Krajcovicová-Kudlácková et al. 2002; Ingenbleek & McCully 2012; Burkert et al. 2014; Orzylowska et al. 2016; Buscail et al. 2017; Ghoshal & Singh 2017; Iguacel et al. 2019, 2020; Tong et al. 2018; Vanacore et al. 2018; Acer et al. 2019; Borude 2019]. Outcomes include the reduced healing of scars and other wounds [Fusano et al. 2020; Posso et al. 2025], problems of dental health [Smits et al. 2020], decreased bone health [Tong et al. 2020; Zheng et al. 2023], and issues related to psychological well-being [Dobersek et al. 2020; Dobersek et al. 2021; Ocklenburg & Borawski 2021; Fazelian et al. 2021; Brytek-Matera 2021]. While meat-containing diets are more frequently linked to ischemic heart attacks, haemorrhagic strokes occur more frequently with vegetarian/vegan diets [Tong et al. 2019; Grüngreiff et al. 2020]. As such studies may be equally criticized for their lack of causal proof, they should at least serve as counter material in the overall debate.

Finally, potentially problematic associations have been reported for the inclusion of plant-based alternatives for animal-sourced foods [see elsewhere].

Share this page on: