Is meat bad for you?

The objection

“People have eaten meat for as long as there have been people, and it’s the easiest way to get protein, iron and B12. The scare stories come from food questionnaires, and the people who eat the most bacon also smoke more, drink more and move less, so of course they get ill more often. In 2019 a panel of researchers went through the whole lot and told adults to carry on eating meat as they do. Nobody has ever shown that a chicken breast did anyone harm.”

The answer

It depends on the meat and on how much of it you eat. In most of the large pooled studies processed meat has the strongest links with disease, unprocessed red meat smaller and less certain ones, and poultry the weakest, although the most cautious method of pooling rated the evidence on processed meat and heart disease lower than that on red meat. The objection is right that nearly all of this evidence is observational and that meat carries useful nutrients, and a serious group of researchers did argue in 2019 that the evidence was too uncertain to justify telling people to change. Where it goes wrong is in reading uncertainty about the size of the harm as proof that there is none, since that panel’s own reviews pointed the same way as everyone else’s.

The firmest finding is on cancer. In 2015 the International Agency for Research on Cancer (IARC), the World Health Organization’s cancer agency, classified processed meat as carcinogenic to humans (Group 1), on sufficient evidence that it causes bowel cancer, and red meat as probably carcinogenic (Group 2A), on limited evidence in people backed by strong evidence of how it could cause cancer [1]. For red meat it also saw links with cancers of the pancreas and prostate [1]. Its experts concluded that each 50 g of processed meat eaten daily raises the risk of bowel cancer by 18% [1], the same figure as a 2011 meta-analysis of prospective studies (Chan et al.), which found a relative risk of 1.18 (95% CI 1.10 to 1.28) per 50 g of processed meat a day and 1.17 (1.05 to 1.31) per 100 g of red meat [2]. Does red and processed meat cause cancer? covers the classification, the mechanisms and the arguments over how much is too much.

Poultry comes out better, mainly on heart disease. In a 2023 meta-analysis of 13 published studies with 1,427,989 people (Papier et al.), poultry showed no link with ischaemic heart disease across 10 studies [3]. In the largest pooled study of type 2 diabetes, published in 2024, poultry carried an extra risk per 100 g a day close to that of unprocessed red meat, though it weakened under other modelling assumptions [4]. In UK Biobank, people who ate more poultry had more reflux, gastritis and duodenitis, diverticular disease, gallbladder disease and diabetes even after allowing for body mass index, with 14% more diabetes for each extra 30 g a day (hazard ratio 1.14, 95% CI 1.07 to 1.21; Papier et al., 2021) [5]. Allowing for BMI had substantially weakened most of the links with every kind of meat, and the authors suggested that leftover confounding, or extra body fat lying on the path to disease, might explain some of what remained [5]. Is chicken healthier than red meat? sets the two side by side, cholesterol and food poisoning included.

Meat does supply nutrients. The NHS calls it a good source of protein, vitamins and minerals, says red meat provides iron, zinc and B vitamins, and lists meat as one of the main sources of vitamin B12 in the diet [6]. In the same UK Biobank analysis, people who ate more unprocessed red meat had less iron-deficiency anaemia, about 20% less for each 50 g a day (hazard ratio 0.80, 95% CI 0.72 to 0.90) [5]. The World Cancer Research Fund, which advises limiting red meat, calls it a valuable source of protein, iron, zinc and vitamin B12 but not an essential part of a healthy diet, and says people who eat no meat can get enough of these nutrients through careful food choices [7]. For a vegan, the NHS says, that means getting B12 from fortified foods such as some breakfast cereals and soya drinks, or from a supplement [20]. Do we need meat to be healthy? looks at how.

Does meat cause heart disease?

Pooled cohort studies link both processed and unprocessed red meat to more heart disease, while poultry showed no link across 10 studies, but the most cautious way of pooling such studies rates the evidence on red meat as weak and on processed meat as weaker still. Papier and colleagues’ 2023 meta-analysis of 13 published studies, which covered 32,630 cases of ischaemic heart disease, found an 18% higher risk for every extra 50 g of processed meat a day (relative risk 1.18, 95% CI 1.12 to 1.25, 10 studies) and a 9% higher risk for every extra 50 g of unprocessed red meat (1.09, 1.06 to 1.12, 12 studies) [3]. It combined the studies with fixed-effects models, which don’t widen the uncertainty when studies disagree [3].

Burden of Proof studies build that disagreement into their uncertainty. The 2022 one on unprocessed red meat estimated the same 1.09 for 50 g a day, but with a range of 0.99 to 1.18, and rated the link with heart disease two stars out of five, on the boundary between weak evidence and none, while finding no evidence of a link with either ischaemic or haemorrhagic stroke (Lescinsky et al., 2022) [8]. The 2025 one on processed meat (Haile et al.) rated its links with type 2 diabetes and bowel cancer two stars and gave no conservative estimate of harm for heart disease [10], and its authors wrote that processed meat and heart disease was the only pairing in their study that did not get two stars [19].

EPIC-Oxford, which followed 48,188 UK adults for 18 years, compared people who ate meat with people who ate none. After allowing for lifestyle differences, its vegetarians had a 22% lower rate of ischaemic heart disease than meat-eaters (hazard ratio 0.78, 0.70 to 0.87), equivalent to 10 fewer cases per 1,000 people over ten years (Tong et al., 2019) [9]. Allowing also for self-reported high cholesterol, high blood pressure, diabetes and body mass index brought the hazard ratio to 0.90 (0.81 to 1.00), which can’t rule out no difference [9]. The same study found a 20% higher rate of stroke in vegetarians (1.20, 1.02 to 1.40), about three more cases per 1,000 over ten years, mostly haemorrhagic stroke [9], which Is a vegan diet healthy? looks at alongside the other long-term studies.

Does eating meat cause type 2 diabetes?

Eating more meat goes with more type 2 diabetes in the largest pooled analysis so far, most clearly for processed meat. The InterConnect project (Li et al., 2024) analysed individual records from 31 cohorts in 20 countries, covering 1,966,444 adults and 107,271 new cases over a median of 10 years, using the same analysis in every cohort and allowing for body mass index among other factors [4]. Each extra 50 g of processed meat a day went with a 15% higher rate of type 2 diabetes (hazard ratio 1.15, 95% CI 1.11 to 1.20), each extra 100 g of unprocessed red meat with 10% (1.10, 1.06 to 1.15) and each extra 100 g of poultry with 8% (1.08, 1.02 to 1.14) [4]. The links showed up in North America, Europe and the Western Pacific, while in the other regions the estimates were too imprecise to tell [4]. The Burden of Proof method reads the evidence more cautiously: it rated processed meat’s link with diabetes two stars, with a risk at least 11% higher on average at 0.6 to 57 g a day than at none (Haile et al., 2025) [10], and unprocessed red meat’s link two stars as well, on the boundary between weak evidence and none [8].

How big is the risk for one person?

The risk to any one person is small, which is why both sides of the argument can quote the same studies. Cancer Research UK put the results of a study of half a million UK adults into numbers in 2024: of every 10,000 people eating under 11 g of red and processed meat a day, 45 were diagnosed with bowel cancer over almost seven years, and eating 79 g a day, similar to the UK average, meant 14 extra cases per 10,000, after allowing for smoking, alcohol, physical activity, body mass index and other differences between the groups [11]. In EPIC-Oxford, the gap in heart disease between meat-eaters and vegetarians was about 10 cases per 1,000 over ten years [9]. IARC’s head of monographs, Kurt Straif, said in 2015 that an individual’s risk of bowel cancer from processed meat remains small but rises with the amount eaten, and that the number of people who eat it makes its effect on cancer worldwide a matter of public health [1].

Long-term cohort studies link diets high in red and processed meat to more type 2 diabetes, cardiovascular disease and cancer, and to higher death rates overall, as Qian and colleagues summarised in 2020 [12]. The authors of the 2025 Burden of Proof analysis, who rated processed meat’s links with diabetes and bowel cancer only two stars out of five, still concluded that because these diseases are so common, there is merit in continuing to recommend limiting it [10].

Why did a 2019 review tell people to keep eating meat?

The NutriRECS panel judged the evidence too uncertain, and the effect on each person too small, to ask people to change, although its own reviews pointed towards harm. In 2019 the panel, 14 members from seven countries including three community members, suggested that adults carry on eating unprocessed red meat and processed meat at their current levels, a weak recommendation based on low-certainty evidence that also weighed people’s own values and preferences about eating meat (Johnston et al., 2019) [13]. Its review of 55 cohorts with more than 4 million participants found that eating three fewer servings of processed meat a week went with very small falls in deaths from all causes and from cardiovascular disease, and in stroke, heart attack and type 2 diabetes, with similar results for unprocessed red meat apart from deaths from all causes (Zeraatkar et al., 2019) [14]. It rated that evidence low certainty, listing poor adjustment for confounding and recall bias in how diets were measured among its limitations [14].

Other researchers disputed the method. Qian and colleagues argued in 2020 that the GRADE system the panel used was designed mainly for clinical treatments such as drugs and starts by rating all observational evidence as low quality, when trials long enough to test the effects of diet on heart disease or cancer are rarely feasible [12]. They pointed out that the panel’s own analysis of 14 cohorts with 669,530 people found a 22% lower risk of type 2 diabetes with less processed meat (relative risk 0.78, 0.72 to 0.84), which it then downgraded for differences between the studies, and that NutriGRADE, a version of the system adapted for nutrition, rated the evidence linking red and processed meat to diabetes high quality and to deaths moderate [12]. The 2022 Burden of Proof study took a middle position on unprocessed red meat, finding evidence of harm but calling it weak and insufficient for stronger or more conclusive recommendations [8].

Most dietary guidelines recommend eating patterns relatively low in red and processed meat, Qian’s paper noted in 2020 [12], and the UK’s still do. The NHS, which last reviewed its page in 2024, says anyone eating more than 90 g of red and processed meat a day should cut down to 70 g [6], and the World Cancer Research Fund advises no more than about three portions of red meat a week, about 350 to 500 g cooked, and very little, if any, processed meat [7]. The US Dietary Guidelines for 2025 to 2030 go the other way: they ask Americans to eat a variety of protein foods from animal sources, red meat included, and give no limit for it, though they keep saturated fat to no more than 10% of daily calories [15].

Does it matter what replaces the meat?

What replaces meat matters, and a 30-year study of US men that modelled swaps found lower heart disease risk when red meat was replaced with nuts, legumes and soya, whole grains or dairy. Among those 43,272 men, followed from 1986 to 2016, each daily serving of red meat went with a 12% higher risk of coronary heart disease (hazard ratio 1.12, 95% CI 1.06 to 1.18), and eating a daily serving of nuts, legumes and soya in its place went with a 14% lower risk (0.86, 0.80 to 0.93) (Al-Shaar et al., 2020) [16]. Whole grains and dairy in place of red meat were also linked to lower risk, as were eggs in place of processed red meat [16]. Swaps within meat count too: in InterConnect, replacing processed meat with unprocessed red meat or poultry went with less type 2 diabetes [4], and in EPIC-Oxford people who ate fish but no meat had 13% less ischaemic heart disease than meat-eaters (hazard ratio 0.87, 0.77 to 0.99) [9]. Qian and colleagues listed the NutriRECS panel’s lack of attention to what replaces red meat, whether plant protein or refined carbohydrate, among its main weaknesses [12].

If you take medicine for diabetes, blood pressure or cholesterol, speak to your GP before making big changes to how you eat, and don’t stop a statin or any other medicine without your doctor’s advice.