Observational claims in nutritional sciences often fail to hold up when tested in randomized controlled trials. The latter are valuable for assessing cause-and-effect relationships, but are not infallible. Interventions trials with red meat, eggs, and dairy do not usually indicate harm. They are typically showing neutral or even beneficial effects on most of the risk markers for chronic diseases. Even if vegetarian diets may sometimes have positive effects on a limited set of these factors, such data should be approached with caution due to varying study designs and conflicting results. Experimental (animal) studies also fail to consistently support the alleged harmful effects of red and processed meats.
This subsection addresses the following five questions:
- Why are human randomized control trials (RCTs) with animal source foods useful but not infallible?
- What do RCTs tell about the effect of animal source foods on morbidity and mortality?
- What do surrogate endpoints in RCTs tell about the health effects of animal source food intake?
- Are RCTs with 'plant-based' diets indicative of health benefits?
- What do experimental (animal) studies say about the health effects of animal source foods?
_________________________________
Why are randomized controlled trials (RCTs) with animal source foods useful but not infallible?
Randomized controlled intervention trials are useful for establishing cause-and-effect relationships and are considered high up in the evidence hierarchy. However, such trials are not infallible. Problems may arise when, for instance, the normal dietary context is overlooked, or when biomarkers are not robust enough to evaluate the actual health impact.
Further reading (summary of the scientific literature):
To support assumptions of cause and effect, well-designed and meaningful randomized controlled trials (RCTs) are needed [Gerstein et al. 2019]. Past experience has shown that claims from observational epidemiology frequently fail to hold up when tested in RCTs [Young & Karr, 2011]. Although RCTs are considered to be higher up in the hierarchy of scientific evidence, it needs to be stressed that they are certainly not fail-safe and can at times be seriously flawed [Krauss 2018]. Problems arise when the normal dietary context is overlooked or when biomarkers are not robust, as is often the case in nutritional sciences. In such cases, it is not justified to claim a causal link between the consumption of ASFs and negative health outcomes [Turner & Lloyd 2017; Kruger & Zhou 2018].
What do RCTs tell about the effect of animal source foods on morbidity and mortality?
The evidence from randomized controlled trials with respect to the impact of a low or high intake of animal source foods on cardiometabolic outcomes or mortality is overall of low to very-low certainty. The causality of the associations between certain animal source foods and morbidity or mortality outcomes obtained from observational studies remains therefore highly uncertain.
Further reading (summary of the scientific literature):
It has been concluded that there is only low- to very-low-certainty evidence from RCTs to suggest that diets restricted in red meat affect major cardiometabolic outcomes and cancer mortality and incidence - so that causality of the claims coming from observational studies remains highly uncertain [Zeraatkar et al. 2019; Hill et al. 2020]. Also, lower red and processed meat intakes do not reduce the rate of Crohn's disease flares [Albenberg et al., 2019].
What do surrogate endpoints in RCTs tell about the health effects of animal source food intake?
Intervention trials with animal source foods do not show harmful impact on blood pressure or markers for glycaemic control, inflammation, or oxidative stress, and may even yield beneficial outcomes. When saturated fat content is high, there may be a slight increase in LDL-C, but effects on other lipid markers are mostly neutral to protective, especially when animal source foods are consumed as part of a healthy dietary setup. Comparing red meat to poultry and fish, there are no significant differences in lipid profiles. Eggs also do not have adverse effects on vascular function or major blood markers. Palaeolithic diets rich in meat, fish, and eggs show positive effects on cardiovascular risk markers and anthropometric measures. Regarding dairy consumption, it can have anti-inflammatory properties and may offer protective effects on cardiovascular risk factors, although there may be slight increases in LDL-C and triglycerides in some cases.
Further reading (summary of the scientific literature):
Interventions with red meat do not point to harmful effects on blood lipid markers and blood pressure [O’Connor et al. 2017]. Neutral or even net beneficial findings are common when intervention studies provide animal source foods within a healthy dietary context, as shown for instance in a Norwegian study using advanced lipidomics [Monfort-Pires et al. 2023]. When a Mediterranean-type diet containing 14-156 g/d of lean beef is contrasted to an average American diet, a decrease in total, LDL, and non-HDL cholesterol, as well as apoB, is obtained [Fleming et al. 2021]. Even if higher LDL-C levels may sometimes be obtained, particularly when saturated fat content is high, this is likely attenuated by a more benign particle profile (more buoyant LDL particles) [Bergeron et al. 2019]. Neither do such interventions worsen markers for inflammation, oxidative stress, thrombosis tendency, or glycemic control [Mann et al. 1997; Hodgson et al. 2007; Turner et al. 2017; O'Connor et al. 2021]. There seems to be no link between red meat intake and markers for type-2 diabetes, including insulin resistance, fasting glucose and insulin, glycated haemoglobin, pancreatic beta-cell function, or GLP-1, with red meat intake even modestly reducing postprandial glucose [Sanders et al. 2023]. In one study, increases in some inflammatory markers (IL-6 and TNF-α, but not CRP) were seen for Wagyu beef compared to lean kangaroo meat after 2 h [Arya et al. 2010], which is a normal transient response after a high-fat meal [Emerson et al. 2017]. The lack of association between red and processed meats with markers of inflammation has also been confirmed by observational data [Wood et al. 2023].Conversely, red meat avoidance or restriction fails to show a clear benefit. In hypocaloric treatments of patients at risk of type-2 diabetes, no favourable effects were obtained for weight loss, body-fat composition (including liver fat), or glucose metabolism outcomes [Willmann et al. 2019]. During ad libitum weight maintenance diets in the treatment of obesity, similar effects on body weight, energy metabolism and cardiovascular risk factors were found for 25g/d versus 150g/d beef intake [Magkos et al. 2022]. In healthy volunteers, halving the intake of red and processed meats decreased total cholesterol and LDL-C, but less favourable haematological parameters and a decrease of HDL-C were also found [Simpson et al. 2019].When comparing red meat with poultry, lipid profiles are not significantly affected [Maki et al. 2012; Bergeron et al. 2019; Guash-Ferré et al. 2019]. Relative to fish, lipid profiles are either similar [Maki et al. 2012], or slightly affected due to a decrease in total and LDL cholesterol as well as HDL-C [Guash-Ferré et al. 2019]. Oily fish is indeed known to increase HDL-C, besides a decrease in triglyceride levels [Alhassan et al. 2017].Intervention studies with eggs show no adverse impact on vascular function [Emamat et al. 2020] and no impact on total cholesterol, LDL-C, triglycerides, fasting glucose, insulin, or C-reactive protein [Richard et al. 2017]. Although some authors did identify an increase in total and LDL-C, an increase in HDL-C but not in triglycerides was also found [Rouhani et al. 2018; Wang et al. 2019], leading to an unaltered LDL-C/HDL-C ratio [Sikaroudi et al. 2020]. Moreover, eggs may help to achieve an optimal body composition through improved satiety, increased muscle protein synthesis, and lower fat mass [Myers & Stevenson Ruxton 2023].Similar effects are found in RCTs with Palaeolithic diets, rich in meat, fish, and eggs. Such diets do not have harmful effects on cardiovascular risk markers and improve anthropometric indexes, lipid profile, blood pressure, and inflammatory markers, despite the need for further evidence [Almeida de Menezes et al. 2019; Ghaedi et al. 2019].Interventions with dairy also lead to anti-inflammatory properties [Moosavian et al. 2020], at least in humans not suffering from milk allergy, and particularly so in subjects with metabolic disorders [Bordini et al. 2015;]. Depending on the product used, dairy can have protective effects on cardiovascular risk factors (increase in HDL-C and a decrease in total cholesterol and waist circumference), although it can also slightly increase LDL-C and triglycerides [Duarte et al. 2021]. A systematic review with network meta-analysis of RCTs indicated that there is little robust evidence that a higher dairy intake (irrespective of fat content) has detrimental effects on anthropometric outcomes, blood lipids, and blood pressure [Kiesswetter et al. 2023].
Do RCTs with 'plant-based' diets show health benefits?
There is insufficient information from intervention trials to draw conclusions on the causal effects of vegetarian diets on risk factors for chronic diseases. Such diets may lead to reductions in caloric intake and LDL-C but may also decrease HDL-C, increase triglycerides, and increase postprandial glucose and insulin responses (all depending on the type of plant-based diet). Moreover, comparing plant-based diets to animal-based diets in subjects with type-2 diabetes has not consistently shown clear advantages. Replacing
animal proteins with plant proteins might negatively affect bone health
markers in healthy adults, possibly due to lower protein quality and
reduced vitamin D and calcium intake. Overall, more research is needed
to draw definitive conclusions about the benefits of plant-based diets
on various health outcomes.
Further reading (summary of the scientific literature):
Interventions with plant-based diets are sometimes suggestive of benefit on cardiovascular risk factors [Guash-Ferré et al. 2019] or glucose metabolism in type-2 diabetic or obese subjects [Barnard, 2009]. A meta-analysis of interventions studies with vegetarian and vegan diets did indicate a reduction of total and LDL cholesterol and apoB, but not triglycerides [Koch et al. 2023]. Another meta-analysis indicated that vegetarian and vegan diets may reduce weight and fasting glucose, but not blood pressure or other metabolic or anthropometric outcomes, while stressing that the evidence is of low certainty [Melgar et al. 2023]. Even if vegetarian and vegan diets may decrease LDL-C, they also can decrease HDL-C and increase triglycerides in some cases [Yokoyama et al. 2017]. Taken together, decreases in total cholesterol and LDL-C have inferior predictive power for cardiometabolic health over high HDL-C and low triglycerides [Jeppesen et al. 2001; Varbo & Nordestgaard 2018; Hosadurg et al. 2018; Castañer et al. 2020].Generalizing findings on vegetarian intervention studies should thus be done with caution, because of (1) a focus on Western eating habits and (2) a low credibility of evidence of certain studies [Chew et al. 2023]. Comparison groups have typically not included optimal omnivorous diets [Barnard & Leroy 2020], while results are conflicting [Osland Johannesen et al. 2020]. A Cochrane review concluded that there is 'insufficient information to draw conclusions' about the effects of vegan dietary interventions on risk factors for cardiovascular disease [Rees et al. 2021].Intervention studies that compare plant-based and animal-based diets have failed to find clear advantages in subjects with (or at risk of) type-2 diabetes [Markova et al. 2019; Hassanzadeh-Rostami et al. 2020; Maki et al. 2020]. In a non-diabetic population, appetites were similar but the plant-based intervention reduced caloric intake and fat mass, while increasing postprandial glucose and insulin responses [Hall et al. 2020]. One plant-based intervention study with Beyond Meat products lowered serum TMAO compared to a meat intervention [Crimarco et al. 2020], but the clinical value of this marker is questionable [see elsewhere] and a secondary analysis of the same study failed to show a consistent improvement of inflammation biomarkers [Crimarco et al. 2022]. Another trial with a variety of plant-based imitation meats failed to show cardiometabolic health benefits and even indicated worse outcomes for glycaemic homeostasis [Toh et al. 2024]. Also, partial replacement of animal with plant proteins may negatively affect bone health markers among healthy adults, potentially due to lower protein quality and/or because of the concomitant decrease in vitamin D and calcium intake [Itkonen et al. 2020].
Intervention studies with experimental animals do not provide consistent evidence for the alleged harmful effects of red and processed meats. The International Agency for Research on Cancer (WHO) concluded that there is inadequate evidence in animals for the carcinogenicity of consuming red meat, processed meat, and haem iron.
Further reading (summary of the scientific literature):
Intervention studies with experimental animals fail to show consistent evidence for the alleged harmful effects of red and processed meats; one study even showing a protective effect of bacon on rat colon carcinogenesis [Parnaud et al 1998]. This was also confirmed by the IARC/WHO panel, stating that there is 'inadequate evidence' in experimental animals for the carcinogenicity of consumption' of red meat, processed meat, and haem iron [IARC 2018].
