In this guide
Few topics in nutrition generate more confident, conflicting headlines than meat. One year red meat is a killer; the next, a team of scientists says the evidence is too weak to tell you to cut back. Add the vegetarian and the carnivore corners of the internet and it becomes almost impossible for an ordinary person to know what to put on the plate.
A new analysis from the American Cancer Society's Cancer Prevention Study-II, published in Current Developments in Nutrition in July and discussed widely since, adds a long-running dataset to the argument. It followed people for 38 years and asked a sharper question than most: what happens if you remove red and processed meat but keep white meat, or remove meat altogether? We read the paper's summary and the coverage, set it against the wider evidence, and focused on what it means for the thing most of our readers care about: getting enough protein from sources you can trust.
The short version: after full adjustment, low intake of red and processed meat was the only dietary category significantly associated with every mortality outcome studied, and the differences were modest (for cancer mortality, about 2–5% lower risk in several subgroups). Diets with white meat only or no meat weren't significantly linked to heart-disease mortality, and in a few subgroups results were counter-intuitive. It's observational, based on a single baseline questionnaire, with very few vegetarians (about 0.4%) and white-meat-only eaters (about 1%). The practical takeaway isn't to eat less protein. It's to choose your protein sources with care, starting with processed meat.
Why Meat Studies Keep Contradicting Each Other
There are three reasons. First, meat research is mostly observational: researchers ask large groups what they eat and follow them for health outcomes. That can reveal patterns, but can't prove cause, because people who eat a lot of meat also differ in smoking, alcohol, exercise, income, vegetables and healthcare use. Second, the categories are blurry. "Red meat" can mean a grass-fed steak or a gas-station hot dog, and a vegetarian can be a lentil-and-vegetable eater or someone living on crisps and white bread. Third, effects are small. Small effects are easy to blur with confounding, so different studies, with different populations and adjustments, land in different places.
Keep those three reasons in mind through the rest of this article, because they explain both the study's strengths and its surprises.
The 38-Year Study at a Glance
| Item | Details |
|---|---|
| Title | Associations between white meat-only and vegetarian diets with mortality from all causes, heart diseases, cancers, and stroke in the American Cancer Society's Cancer Prevention Study-II |
| Journal and date | Current Developments in Nutrition, 17 July 2026 (doi:10.1016/j.cdnut.2026.109438); summarised by News-Medical on 18 August 2026 |
| Cohort and follow-up | American Cancer Society's Cancer Prevention Study-II; about 38 years of follow-up |
| Diet measure | A brief baseline diet questionnaire that classified meat intake |
| Reference group | High red/processed meat intake (the comparison for all other groups) |
| Outcomes | All-cause, heart disease, cancer and stroke mortality |
| Design | Observational cohort analysis using statistical models adjusted for other factors |
How People Were Grouped
| Group | Definition | Share of participants |
|---|---|---|
| High red/processed meat | Six or more days a week | 37% |
| Moderate | Four to under six days a week | 31% |
| Low | Half a day to under four days a week | 31% |
| White meat only | Poultry or fish, no red or processed meat | about 1% |
| Vegetarian / no meat | No meat | about 0.4% |
Look at the last two rows. Over 98% of participants ate red or processed meat, because the cohort began in the early 1980s, when avoiding it was rare. That makes the white-meat-only and vegetarian groups tiny, which matters for how much weight their results can carry.
What They Found
- Low red/processed meat intake was the only category that stayed significantly associated with each mortality outcome after full adjustment in the overall analyses.
- White-meat-only and vegetarian diets weren't significantly associated with heart-disease mortality compared with high red/processed meat intake.
- For cancer mortality, low red/processed meat intake was linked with about 2–5% lower risk across several subgroups.
- Stroke: among former smokers, low intake was associated with about 5% lower stroke mortality (hazard ratio 0.95; 95% CI 0.91 to 0.99).
- White meat only was associated with the lowest all-cause mortality among men and among people with a healthy weight.
- A post hoc comparison using mixed diets (red and white meat) as the reference found that red-meat-only diets were associated with a 7–11% higher risk across mortality outcomes.
The authors' own wording is careful: excluding red and processed meat was associated with possible additional benefits in some groups, but the findings don't establish causation.
The Surprising Bits, and Why Not to Over-Read Them
Two results caught attention. Among former smokers, a vegetarian diet was associated with higher all-cause mortality than high red/processed meat intake, and in people under 50, with higher cancer mortality. It would be easy to turn those into headlines saying "vegetarian diets are dangerous". That would be a mistake, for several reasons.
- The group was tiny. About 0.4% of participants were vegetarian, so subgroup estimates (by smoking status and age) rest on very few people and can swing widely.
- "Vegetarian" in 1982 isn't "vegetarian" today. A baseline questionnaire can't tell whether someone ate lentils and vegetables or refined carbohydrates and little else. Plant-based food quality, and the availability of fortified products, have changed enormously.
- One measurement, 38 years. People change diets over decades. A single baseline answer can't capture that.
- Subgroup splits invite chance findings. When you slice a cohort into many smoking, age and sex groups, some odd results will appear by luck alone.
- Other cohorts disagree. Studies in populations with many vegetarians, such as the Adventist Health Study-2, generally point the other way, and found moderately higher risks of all-cause and cardiovascular mortality with red and processed meat even at low intakes.
The fair reading is "the data are inconclusive for the rare groups", not "vegetarian diets raise risk".
Small Effects and the Healthy-User Problem
A 2–5% relative difference in cancer mortality sounds small because it is. It's the kind of difference that can be partly explained by who chooses to eat less red and processed meat. In 1982, those people were more likely to be health-conscious in other ways, such as exercising, smoking less and eating more vegetables. Statistical adjustment helps, but it can't perfectly remove "healthy-user" bias.
That doesn't make the finding meaningless. A small effect that's consistent across many studies, as this one is with others, is more credible than a one-off. But it does mean you shouldn't treat 2–5% as a guarantee. A better way to use it is as one nudge among many: if you can shift a bit of your intake toward fish, poultry, legumes, nuts and dairy at little cost to your enjoyment, the evidence leans toward it being worth doing.
Zooming Out: The Wider Evidence
| Source | What it found |
|---|---|
| Umbrella review of processed meat (Frontiers in Public Health, 2026) | Highest versus lowest intake: all-cause mortality relative risk about 1.15 (95% CI 1.10–1.21); cancer mortality about 1.08 (1.06–1.11); cardiovascular mortality about 1.15 (1.07–1.24) |
| NIH-AARP Diet and Health Study (over half a million people) | Red and processed meat linked with modest increases in total, cancer and cardiovascular mortality; higher white meat intake linked with a small decrease in total and cancer mortality |
| Adventist Health Study-2 (low-meat population) | Moderately higher all-cause and cardiovascular mortality with red and processed meat even at low intakes; processed meat alone wasn't significantly associated in that analysis |
| IARC (2015) | Processed meat classified as carcinogenic to humans (Group 1); red meat as probably carcinogenic (Group 2A). These describe strength of evidence, not size of risk |
| CPS-II (2026, this study) | Low red/processed intake associated with lower mortality; vegetarian and white-meat-only groups too small for firm conclusions |
Taken together, the pattern is remarkably consistent for processed meat and more modest for unprocessed red meat. Most of the heat in the debate comes from how big the effect is and how much is explained by lifestyle, not from whether there's any signal at all.
Processed Versus Unprocessed: What Guidelines Say
The most defensible single swap is the one the evidence supports best: cut back on processed meat (bacon, sausages, ham, salami, hot dogs, many deli meats). For unprocessed red meat (beef, lamb, pork), the World Cancer Research Fund recommends eating no more than about 350–500 g (cooked weight) a week, which is roughly three portions, and eating little, if any, processed meat. UK health advice similarly suggests people who eat a lot of red and processed meat cut down. You don't need to eliminate red meat to follow these guidelines, but you do get more benefit from trimming the processed end.
What This Means for Protein Targets and Muscle
The meat debate sometimes slides into "so should I eat less protein?" The answer is no. This study is about sources, not amounts. Protein needs, which rise with age, training and dieting, as we explain in our guides, don't go away because you swap a sausage for a fillet of fish. If you're a lifter or someone dieting, the goal remains roughly 1.2–1.6 g per kg of body weight, higher for some athletes, and you can reach it without leaning on processed meat.
Use the protein intake calculator for your number and the macro calculator to fit it into your day. Our guides on the best protein sources for muscle gain and building muscle naturally show how to hit it with whole foods.
Plate Swaps That Keep Protein High
| Instead of… | Try… | Protein (g, approx.) |
|---|---|---|
| Bacon and sausage breakfast | Three-egg omelette with spinach, or Greek yoghurt with berries and nuts | 18–25 |
| Ham or salami sandwich | Tinned tuna or salmon, or chicken breast, with salad on wholegrain bread | 25–32 |
| Hot dog or burger as a quick dinner | Grilled chicken thighs, fish, or a lentil-and-vegetable curry | 25–35 |
| Pepperoni pizza | Pizza with vegetables and chicken, or a chickpea-and-egg dish | 20–30 |
| Processed-meat snack | Cottage cheese, edamame, a boiled egg or a handful of nuts | 7–20 |
A sample day for a 75 kg person aiming for about 120 g of protein, without processed meat: eggs and Greek yoghurt at breakfast (about 30 g), a tuna-and-bean salad at lunch (about 35 g), a chicken or tofu stir-fry at dinner (about 35 g), plus cottage cheese and nuts in between (about 20 g). To check how a recipe adds up, paste it into the recipe nutrition calculator.
Who May Genuinely Benefit From Some Red Meat
"Eat less red and processed meat" isn't "eat none". Unprocessed red meat is a rich source of well-absorbed iron, vitamin B12, zinc and leucine, and some people have a harder time getting those elsewhere. These groups often benefit from keeping modest amounts, or from careful planning if they don't eat it:
- Menstruating women, teenage girls and pregnant women, who have higher iron needs.
- Older adults with small appetites, for whom compact, nutrient-dense protein helps hit targets.
- People with iron-deficiency anaemia, who should follow their clinician's advice.
- Vegans and vegetarians, who should plan for B12, iron, zinc and iodine, supplementing where needed.
The Bottom Line
The 38-year American Cancer Society analysis fits an established pattern: processed meat is the clearest concern, unprocessed red meat is a more modest one, and small differences in mortality are linked with low red and processed meat intake. It adds a few surprising subgroup results, which are best understood as noise from tiny groups and a crude 1982 questionnaire rather than as evidence that vegetarian diets are harmful.
For your plate, the useful message is practical, not dramatic. Keep your protein target. Trim processed meat first. Keep unprocessed red meat to a few portions a week. Make fish, poultry, eggs, dairy, beans, lentils, tofu and nuts your everyday protein, and don't let anyone convince you that you have to choose a tribe.
Frequently Asked Questions
Is white meat healthier than red meat?
Observational studies generally associate higher white meat intake with equal or slightly lower mortality than red and processed meat, but the effects are modest and affected by confounding. In the 2026 CPS-II analysis, white-meat-only diets were associated with the lowest all-cause mortality among men and healthy-weight people, though the group was only about 1% of participants.
Does the study show vegetarian diets are risky?
No. The vegetarian group was about 0.4% of participants, the questionnaire was a single baseline measure from the early 1980s, and unusual results appeared only in small subgroups, such as former smokers and people under 50. Other cohorts with many vegetarians show different patterns.
How much red meat is okay per week?
The World Cancer Research Fund recommends eating no more than about 350–500 g of red meat a week (cooked weight), roughly three portions, and little if any processed meat. You don't have to eliminate it, but trimming processed meat has the strongest support.
Is processed meat worse than fresh red meat?
Evidence generally treats processed meat as the clearer concern. The International Agency for Research on Cancer classes processed meat as carcinogenic to humans (Group 1) and red meat as probably carcinogenic (Group 2A), reflecting the strength of evidence, not the size of risk.
Should I eat less protein to avoid these risks?
No. The research concerns protein sources, not total protein. Most active adults, older adults and dieters benefit from roughly 1.2–1.6 g per kg of body weight. You can reach that with fish, poultry, eggs, dairy, legumes, tofu and nuts.
What are good protein swaps for bacon and sausages?
Eggs, Greek yoghurt, cottage cheese, smoked salmon in moderation, chicken, tinned fish, tofu and beans. For lunches, tuna, chicken or hummus-based fillings replace deli meats well.
Can lifters build muscle without processed meat?
Yes. Whole-food proteins, dairy and plant proteins support muscle gain well. Aim for about 1.6–2.2 g per kg of body weight with 25–35 g per meal, using the protein intake calculator to set your target.
Why do studies on meat disagree?
Because they're mostly observational, categories are blurry (a steak and a hot dog both count as "red meat"), and effects are small enough to be influenced by lifestyle differences between eaters and non-eaters. Consistency across many studies matters more than any single result.
Who should be careful about cutting red meat?
People with higher iron needs, such as menstruating or pregnant women, older adults with small appetites and anyone with iron deficiency. They can still reduce processed meat and plan other sources of iron, B12 and zinc, with professional advice if needed.
References & Sources
- Associations between white meat-only and vegetarian diets with mortality from all causes, heart diseases, cancers, and stroke in the American Cancer Society's Cancer Prevention Study-II (2026). Current Developments in Nutrition. doi:10.1016/j.cdnut.2026.109438; News-Medical (18 August 2026).
- A comprehensive evaluation of epidemiological evidence on processed meat intake in relation to cancer, cardiovascular, metabolic diseases and all-cause mortality: an umbrella review (2026). Frontiers in Public Health. doi:10.3389/fpubh.2026.1763155.
- Sinha, R. et al. (2009). Meat intake and mortality: a prospective study of over half a million people. Archives of Internal Medicine, 169(6), 562–571.
- Alshahrani, S. M. et al. (2019). Red and processed meat and mortality in a low meat intake population. Nutrients, 11(3), 622.
- Bouvard, V. et al. (2015). Carcinogenicity of consumption of red and processed meat. The Lancet Oncology, 16(16), 1599–1600.
- World Cancer Research Fund / American Institute for Cancer Research (2018). Diet, nutrition, physical activity and cancer: a global perspective (Third Expert Report).
- MacrosMeasure (2026). Free Protein Intake Calculator.
- MacrosMeasure (2026). Free Recipe Nutrition Calculator.
Calculate your own numbers — free.
18 science-backed calculators covering everything in this article and more.