Bacon and sausage: a carcinogen label built on observational data

The hardest evidence on this page, a meta-analysis of 41 randomized studies with about 87000 subjects, is about behavior rather than health. Interventions aimed only at red and processed meat cut consumption with a standardized mean difference of 0.25, and the authors add that it remains unclear whether those interventions also reduce other meat and animal products. Everything about disease comes from cohorts. An umbrella review of 54 meta-analyses links each extra 50 g a day of processed meat to 17 percent higher colorectal cancer risk and 72 percent higher gastric cancer risk, all of it low to very low certainty by GRADE. WHO classifies processed meat as carcinogenic to humans and puts the colorectal figure at about 18 percent per 50 g eaten daily, drawing on an analysis of 10 studies. The Group 1 label describes how sure the agency is that the link exists, not how large the risk is. In USDA survey measures, 50 g is roughly eight cooked bacon slices, at the 6 g slice weight that table records for beef bacon, or two and a half breakfast sausage links, every day.

How much is actually in it

• By the NIH Office of Dietary Supplements, phosphate additives common in processed foods add on average 67 mg phosphorus per serving and are absorbed at about 70 percent. — NIH Office of Dietary Supplements Phosphorus fact sheet, Food section Evidence E source
  +67 mg phosphorus per serving versus similar foods without additives, absorption about 70% for additive phosphorus versus 40% to 70% for phosphorus naturally in food
• By the UK composition table, grilled back bacon has 287 calories, 21.6 g fat and 23.2 g protein per 3.5 oz (100 g), while raw pork sausages have 309 calories, 25.0 g fat and only 11.9 g protein. — CoFID 2021 entries Bacon rashers back grilled code 19-500 and Sausages pork raw code 19-510, per 3.5 oz (100 g) Evidence E source
  Bacon 287 calories, fat 21.6 g, saturated 8.10 g, protein 23.2 g, water 50.4 g; pork sausage 309 calories, fat 25.0 g, saturated 9.15 g, protein 11.9 g, carbohydrate 9.6 g, sodium 470 mg

What a real portion looks like

• In the USDA survey food table one cooked beef bacon slice weighs 6 g while one breakfast size sausage link weighs 20 g and one sausage patty 35 g. — USDA FNDDS survey foods Beef bacon cooked food code 21601000 and Sausage NFS food code 25221400 Evidence E source
  Beef bacon 1 slice cooked 6 g, 1 oz cooked 28 g, 1 cup pieces 84 g; sausage 1 breakfast size link 20 g, 1 patty 35 g, 1 bun-size link 75 g, quantity not specified 60 g

What your body absorbs

• In a randomized double-blind trial of 20 Japanese adults over 8 weeks, fish oil enriched sausages eaten at breakfast lowered serum triglycerides while the same sausages eaten in the evening did not. — 20 healthy Japanese adults aged 20 to 60, parallel groups, 4 and 8 weeks of intake (n = 20) Evidence B source
  Triglycerides and total saturated fatty acids fell significantly only in the breakfast group, omega-3 PUFA rose equally in both groups

What cooking and storage change

• By the USDA nutrient retention table, broiling or frying bacon is assigned 45 percent thiamin retention, the largest loss of any nutrient in that method, while zinc is assigned 100 percent. — USDA Table of Nutrient Retention Factors Release 6, group Pork Products code 10, method BACON,BROILED/FRIED, retention code 1456, table dated Sep-75 Evidence E source
  Thiamin 45%, calcium 75%, iron 80%, vitamin C 80%, vitamin B6 80%, folate 80%, choline 80%, potassium 85%, phosphorus 90%, sodium 90%, niacin 90%, vitamin B12 90%, magnesium 95%, copper 95%, riboflavin 95%, zinc 100%

What it does to your health

• Observational data link each extra 50 g per day of processed meat to 17 percent higher colorectal cancer risk and 72 percent higher gastric cancer risk. — Umbrella review of 34 articles and 54 meta-analyses of observational studies, searched to May 2025 (n = 34 articles, 54 meta-analyses) Evidence C source
  Per 50 g/day: gastric cancer +72%, colorectal cancer +17%, prostate cancer +4%, COPD +8%
• Observational data link each 50 g per day of processed meat to 15 percent higher incidence of type 2 diabetes over a median 10 years. — 1 966 444 adults from 31 cohorts in 20 countries, 107 271 incident cases, median follow-up 10 years (n = 1 966 444) Evidence C source
  HR 1.15 (95% CI 1.11 to 1.20) per 50 g/day of processed meat
• Observational data link processed meat to 1.65 times the risk of colorectal polyps, while the association with colorectal adenomas was not significant. — 31 observational studies of meat subtypes and precancerous colorectal lesions, published to 31 December 2024 (n = 31 studies) Evidence C source
  Colorectal polyps RR 1.65 (95% CI 1.05 to 2.61), colorectal adenomas RR 1.14 (95% CI 0.93 to 1.41)
• Observational data link one daily serving of processed meat to a 0.26 kg per year gain in measured body weight. — 381 291 adults in 3 cohort studies of measured body weight, systematic review to June 2024 (n = 381 291) Evidence C source
  +0.26 kg/year per 1 daily serving (95% CI 0.03 to 0.48), waist +0.14 cm/year (95% CI 0.02 to 0.26)
• Observational data link a diet of mostly processed meat products such as bacon and sausage to 2.89 times the 20 year risk of incident cardiovascular disease compared with rarely eating any meat. — 1988 adults free of cardiovascular disease at baseline in Athens, Greece, 718 incident cases over 20 years of follow-up (n = 1988) Evidence C source
  Mostly processed meat HR 2.89 (95% CI 1.05 to 7.89), mostly red meat HR 1.22 (95% CI 0.81 to 1.82), mostly white meat HR 0.35 (95% CI 0.17 to 0.71)

Safety, limits and interactions

• WHO states that processed meat such as bacon and sausage is carcinogenic to humans and that every 1.8 oz (50 g) portion eaten daily raises colorectal cancer risk by about 18 percent. — General population, WHO summary of an analysis of data from 10 studies (n = 10 studies) Evidence E source
  About +18% relative risk of colorectal cancer per 50 g of processed meat per day
• In a 12 week randomized trial of 113 healthy adults aged 18 to 50, a meat based diet raised fecal water genotoxicity by 15.97 percent DNA damage. — 113 healthy adults aged 18 to 50 randomized, 103 completed, 12 weeks free-living (n = 113) Evidence B source
  Fecal water genotoxicity +15.97% DNA damage (95% CI 4.61 to 27.32, p = 0.006), no significant change with alpha-tocopherol arm or pesco-vegetarian arm
• Observational data link eating raw or undercooked meat, including sausages and salami, to 1.2 to 1.3 times the risk of Toxoplasma gondii infection compared with thoroughly cooked meat. — Cohort, case-control and cross-sectional studies of meat eaters of all animal species, seroprevalence 1.3% to 88.6% Evidence C source
  Risk ratio 1.2 to 1.3, odds ratio 1.7 to 3.0 versus thoroughly cooked meat

What people believe that the data does not support

• By meta-analysis of 41 randomized studies with about 87000 subjects, interventions aimed only at red and processed meat cut consumption with a standardized mean difference of 0.25. — 35 papers, 41 randomized studies, 112 interventions, approximately 87 000 subjects, consumption measured at least one day after intervention (n = 87 000) Evidence A source
  Red and processed meat only interventions SMD 0.25 (95% CI 0.11 to 0.38), all 112 interventions pooled SMD 0.07 (95% CI 0.02 to 0.12)
• In a randomized trial of 3518 US adults, health and environmental warning labels and a 30 percent tax each cut the number of processed meat items selected in an online grocery store. — 3518 US adults in a naturalistic online grocery shopping task, October 2021, four arms (n = 3518) Evidence B source
  All three intervention arms selected fewer processed meat items than control, driven by reductions in processed pork

Who this works differently for

• Observational data link children's processed meat consumption to 1.72 times the odds of acute leukemia. — 15 studies, 1 cohort and 14 case-control, children under 18 and their mothers, searched to April 2022 (n = 15 studies) Evidence C source
  Children's processed meat OR 1.72 (95% CI 1.08 to 2.72), vegetables OR 0.61 (95% CI 0.39 to 0.92)
• In a case control study of 232 children with leukemia in Los Angeles County, breakfast meats such as bacon, sausage and ham showed no persistent association after adjustment, while 12 or more hot dogs a month gave odds of 9.5. — 232 cases and 232 controls, children from birth to age 10 in Los Angeles County, cases 1980 to 1987 (n = 232) Evidence C source
  Hot dogs OR 9.5 (95% CI 1.6 to 57.6) for 12 or more per month, breakfast meats not among persistent associations

What is still unknown

• No randomized trial has measured disease outcomes for processed meat, and every dose response estimate behind the cancer numbers rests on low to very low certainty evidence by GRADE. — Umbrella review of 34 articles and 54 meta-analyses on processed meat and health outcomes, searched to May 2025 (n = 54 meta-analyses) Evidence C source
  All dose-response associations rated low to very low certainty, authors call for prospective studies and randomized controlled trials

The bottom line

There is no USDA composition entry for this pair, so the closest numbers are from the UK table. Per 100 g, grilled back bacon has 287 calories, 21.6 g fat, 8.10 g of it saturated, and 23.2 g protein. Raw pork sausage has 309 calories, 25.0 g fat, 9.15 g saturated, 11.9 g protein, 9.6 g carbohydrate of which 6.8 g is starch, and 470 mg sodium. That is the real gap inside the pair. At almost the same energy the sausage carries about half the protein, although the two rows are not like for like: the bacon entry is grilled and the sausage entry raw. USDA survey measures put one cooked beef bacon slice at 6 g, one breakfast link at 20 g, one patty at 35 g and one bun size link at 75 g, which are measure weights rather than what anyone was observed to eat. The health numbers are cohort numbers and they point one way. Each 50 g a day of processed meat tracks with 15 percent higher type 2 diabetes incidence among 1.97 million adults in 31 cohorts followed a median of 10 years. Median intake across those cohorts ran from 0 to 49 g a day, so 50 g sits at the upper edge of what people actually eat. One daily serving tracks with 0.57 lb (0.26 kg) a year more measured body weight in 381291 adults, graded low certainty by the reviewers, on an interval of 0.07 to 1.1 lb (0.03 to 0.48 kg) and heterogeneity of 93 percent. Among 1988 adults in Athens followed for 20 years, eating mostly processed meat came with 2.89 times the risk of cardiovascular disease against rarely eating meat. That interval runs from 1.05 to 7.89, and only about one person in eight in the cohort ate that way. A meta-analysis of 31 observational studies put processed meat at 1.65 times the risk of colorectal polyps, while the figure for adenomas, 1.14, did not reach significance. Two randomized trials touch the mechanism instead of the disease. In 113 healthy adults aged 18 to 50 randomized for 12 weeks, a meat based diet raised fecal water genotoxicity by 15.97 percent DNA damage, a marker in stool water rather than a case of anything. That trial tested a whole diet, which is a wider thing than bacon and sausage. Twenty Japanese adults ate fish oil enriched sausages for 8 weeks, and triglycerides fell only in the group eating them at breakfast, a pilot result about fish oil and meal timing. Cook it through, and treat that as the real caution if you are pregnant or your immune system is suppressed. Eating raw or undercooked meat, sausages and salami included, tracks with 1.2 to 1.3 times the risk and 1.7 to 3.0 times the odds of Toxoplasma infection against thoroughly cooked meat. The infection usually passes unnoticed in everyone else, which is why those two groups are the ones who need to care. For children, a meta-analysis of 15 studies puts processed meat at 1.72 times the odds of acute leukemia, with 14 of the 15 asking about diet after the diagnosis. In the older Los Angeles study of 232 children, breakfast meats such as bacon, sausage and ham showed no persistent link after adjustment, while 12 or more hot dogs a month gave odds of 9.5 on an interval running from 1.6 to 57.6. We found no data on interactions with medicines. Frying or grilling bacon is assigned 45 percent thiamin retention in the USDA factor table, the largest loss in that method, against 100 percent for zinc, which records no loss rather than a measurement. Both values are group level figures for pork products from a table dated September 1975, and no factor exists there for sausage. Phosphate additives, common in processed foods, add about 67 mg phosphorus a serving and are absorbed at around 70 percent, by an NIH fact sheet that describes processed foods in general without naming bacon or sausage. No randomized trial has measured disease outcomes for processed meat, and the certainty behind every dose response number above is low to very low.

Sources

1. bcsg-daily-r2-15 · fact sheet
   Foods containing these additives have an average of 67 mg more phosphorus per serving than similar foods not containing the additives, and these additives contribute to overall phosphorus intakes in the United States [18,19]. ... The absorption rate for the phosphorus in phosphate additives is approximately 70% [24].
   NIH Office of Dietary Supplements, Phosphorus fact sheet for health professionals · checked 2026-09-25
2. bcsg-daily-r2-16 · reference dataset
   Bacon rashers, back, grilled | CoFID 19-500 | Energy 287 kcal | Fat 21.6 g | Satd FA 8.10 g | Protein 23.2 g | Water 50.4 g || Sausages, pork, raw | CoFID 19-510 | Energy 309 kcal | Fat 25.0 g | Satd FA 9.15 g | Protein 11.9 g | Carbohydrate 9.6 g | Starch 6.8 g | Sodium 470 mg
   CoFID 2021, Public Health England, codes 19-500 and 19-510 · checked 2026-09-25
3. bcsg-daily-r2-18 · reference dataset
   Beef, bacon, cooked | FNDDS 21601000 | fdc_id 2705856 | 1 slice, cooked (approx .8 oz, raw) 6201 = 6 g | 1 oz, cooked 40040 = 28 g || Sausage, NFS | FNDDS 25221400 | fdc_id 2706190 | 1 breakfast size link 64361 = 20 g | 1 patty 61633 = 35 g | 1 bun-size or griller link 64362 = 75 g
   USDA FNDDS, Beef bacon cooked food code 21601000 and Sausage NFS food code 25221400 · checked 2026-09-25
4. bcsg-daily-r2-13 · randomized trial
   Serum concentrations of TG and total saturated FA were significantly decreased in the BF-FO group, whereas those of ω-3 PUFA were significantly and identically increased in both groups.
   Nutrition, 2021 · checked 2026-09-25
5. bcsg-daily-r2-17 · reference dataset
   BACON,BROILED/FRIED | Pork Products (10) | Retn_Code 1456 | thiamin_mg 45% | iron_mg 80% | sodium_mg 90% | zinc_mg 100%
   USDA Table of Nutrient Retention Factors, Release 6, 2007, code USDA-RETN-R6:1456 · checked 2026-09-25
6. bcsg-daily-r2-02 · umbrella review
   Dose-response analysis revealed that an increase in PM consumption of 50 g/day was associated with a 72% higher risk of gastric cancer, 17% higher risk of colorectal cancer, 4% higher risk of prostate cancer, and an 8% higher risk of chronic obstructive pulmonary disease (COPD).
   Front Public Health, 2026 · checked 2026-09-25
7. bcsg-daily-r2-03 · cohort meta-analysis
   Greater consumption of each of the three types of meat was associated with increased incidence of type 2 diabetes, with HRs of 1·10 (95% CI 1·06-1·15) per 100 g/day of unprocessed red meat (I2=61%), 1·15 (1·11-1·20) per 50 g/day of processed meat (I2=59%), and 1·08 (1·02-1·14) per 100 g/day of poultry (I2=68%).
   Lancet Diabetes Endocrinol, 2024 · checked 2026-09-25
8. bcsg-daily-r2-04 · meta-analysis
   Processed meat consumption was correlated with increased colorectal polyp risk (1.65; 1.05, 2.61) but not colorectal adenoma risk (1.14; 0.93, 1.41).
   Nutr Rev, 2026 · checked 2026-09-25
9. bcsg-daily-r2-05 · cohort meta-analysis
   Three studies (n = 381 291) were included in the random-effects model of measured body weight, resulting in a summary increase of 0.26 kg/y per 1 daily serving of processed meat (95% confidence interval [CI], 0.03-0.48; τ2: 0.03; I2: 92.95%).
   Nutr Rev, 2025 · checked 2026-09-25
10. bcsg-daily-r2-06 · cohort
    In multivariate analysis, compared to never/rarely consuming any type of meat, consuming mostly processed meat [HR: 2.89, 95%CI: 1.05, 7.89], but not red meat [HR: 1.22, 95%CI: 0.81, 1.82], was positively associated with incident CVD during 20 years of follow-up, while consuming mostly white meat was inversely associated with incident CVD [HR: 0.35, 95%CI: 0.17, 0.71].
    Meat Sci, 2023 · checked 2026-09-25
11. bcsg-daily-r2-01 · agency position
    An analysis of data from 10 studies estimated that every 50 gram portion of processed meat eaten daily increases the risk of colorectal cancer by about 18%.
    WHO Q and A, Carcinogenicity of the consumption of red meat and processed meat · checked 2026-09-25
12. bcsg-daily-r2-07 · randomized trial
    FW genotoxicity increased significantly after MBD (+ 15.97%DNA damage; 95% CI 4.61 to 27.32; p = 0.006), with no significant within-group changes in MBD-T or PVD.
    Sci Rep, 2025 · checked 2026-09-25
13. bcsg-daily-r2-10 · meta-analysis
    Our results suggest that individuals who eat raw or undercooked meat have, respectively, 1.2-1.3 times the risk and 1.7-3.0 times the odds of T. gondii infection compared to those who thoroughly cook meat, regardless of the animal species they consume.
    Vector Borne Zoonotic Dis, 2021 · checked 2026-09-25
14. bcsg-daily-r2-08 · meta-analysis of RCTs
    Interventions aiming to reduce only consumption of red and processed meat were more effective (SMD = 0.25; 95 % CI: [0.11, 0.38]), but it remains unclear whether such interventions also decrease consumption of other forms of MAP.
    Appetite, 2026 · checked 2026-09-25
15. bcsg-daily-r2-09 · randomized trial
    Compared to control, all three interventions led participants to select fewer items with processed meat (driven by reductions in processed pork) and (for the tax and warning labels + tax interventions only) fewer items with unprocessed meat (driven by reductions in unprocessed beef).
    Int J Behav Nutr Phys Act, 2024 · checked 2026-09-25
16. bcsg-daily-r2-11 · meta-analysis
    Our results showed that children's consumption of processed meat was positively associated with AL (OR = 1.72; 95 % CI 1.08, 2.72), whereas the consumption of vegetables was found to be inversely associated with AL (OR = 0.61; 95 % CI % 0.39, 0.92).
    Clin Nutr ESPEN, 2025 · checked 2026-09-25
17. bcsg-daily-r2-12 · case-control
    When the risks were adjusted for each other and other risk factors, the only persistent significant associations were for children's intake of hot dogs (odds ratio [OR] = 9.5, 95 percent confidence interval [CI] = 1.6-57.6 for 12 or more hot dogs per month, trend P = 0.01), and fathers' intake of hot dogs (OR = 11.0, CI = 1.2-98.7 for highest intake category, trend P = 0.01).
    Cancer Causes Control, 1994 · checked 2026-09-25
18. bcsg-daily-r2-19 · umbrella review
    All the above dose-response associations are based on low to very low certainty evidence as assessed by the GRADE system. ... Future research should include more well-designed prospective studies and randomized controlled trials to further explore the associations between PM consumption and various health outcomes.
    Front Public Health, 2026 · checked 2026-09-25