Processed meat, in the following sections, tends to refer to pink and cured meat products that use the preservative sodium nitrate. This extends to hams, hot dogs, bacon products, and various pre-packaged deli meats
Although there are similarities between red meat products and processed meats, the minor differences include cooking possibly making carcinogens and processed meat having a higher nitrate content
In the following sections, hazard ratios (HRs) and confidence intervals (CIs) are used. A hazard ratio is a simple way of demonstrating the increased risk (usually by comparing one group against another defined group, such as high meat intake against no meat intake) and while 1.00 means no significant difference, 2.00 is a doubling and 4.00 a quadrupling while 0.50 and 0.25 are a half and quarter; negative values cannot occur, and a hazard ratio of 2-4 as a minimum to show a magnitudal difference is usually desirable (hazard ratios between 1-2 are quite small in magnitude for survey research)
Confidence intervals (almost always set at 95%) are a mathematical way of saying "we are 95% sure that the true value lies in this range". If the range at all crosses the 1.00 point, then the results are not statistically significant regardless of how large the range may be (due to being unable to rule out no change)
Processed Meat and Cancer Occurrence
Processed meat has been associated with an increased risk of cancer (generally referring to all) with hazard ratios of 1.11-1.12 (95% CI of 1.04-1.19 and 1.06-1.19 divided by gender) for the most recent evidence comparing the highest and lowest quintiles against each other. Smaller sample sizes may have hazard ratios in the 2 range (2.14-2.75). Other cancers that have been associated with increased cancer risk include breast (fixed relative risk for every additional 30g was 1.03 with a 95% CI of 1.00-1.06), pancreatic (HR of 1.19 and 95% CI of 1.04-1.36) and a failure to find a relationship with prostate cancer (1.05 and a 95% CI of 0.99-1.12).
In particular, processed meat appears to be associated with increased risk for colorectal cancers, with higher risk than red meat (poultry not associated and fish is protective) with one meta-analysis of epidemiological research concluding an increased risk of 1.18 (95% CI of 1.10−1.28) for every 50g increase in processed meat consumption; the increase was similar to 100g increases in red meat.
Processed meat intake does appear to be associated with an increased risk of cancer (colorectal cancer is most well known to be associated with processed meat, but seems to be just as much an increased risk as cancer overall). Despite the statistical significance, the effect size is very small
Processed Meat and All-cause Mortality
Consumption of processed meat has been linked to an increase rate of all cause mortality, with the most recent data (448,568 persons, of which 9,861 died from cancer during the study period) noted that processed meat remained associated with an increased risk of all cause mortality after controls were in place, with a calibrated hazard ratio (HR) of 1.18 (95% CI of 1.11-1.25) when comparing an intake of above 160g daily against the reference intake (10-19.9g).
Both dietary deprivation of meat and excessive levels of meat intake appear to be associated with increased risk for death, but the HR was quite low.
Comparisons to other Meats
Although the most recent (and largest) data has failed to find any negative link between red meat and mortality, a result which has been noted elsewhere, some studies do indeed find an association; all studies seem to use similar controls for the most predictable confounds (caloric intake, obesity, social status, smoking status, etc.) with the only major difference being locale (European data having no association and American data showing a link).
For poultry products, there does not appear to be an increased risk of death (a slight protective effect that becomes statistically insignificant after weighting).
Processed meat and red meat both appear to be associated with increased mortality and death, with processed meat slightly more reliably associated. Both meat categories still have fairly small magnitudes (never surpassing a doubling of risk), and both poultry and fish are not associated with these adverse effects
Sodium nitrate is a preservative commonly associated with pink or cured meats which contains dietary nitrate, a molecule that has cardioprotective properties but may also convert into a class of molecules known as nitrosamines. In vitro, nitrosamines are known to be carcinogenic. The nitrate found in processed meat products is the same as that from vegetable products, although their biological effects seem to differ.
The nitrates themselves do not differ between sources, and sodium nitrate from processed meat is the same 'nitrate' that is found in vegetables and drinking water
Nitrates can 'nitrosylate' other compounds by donating a nitroso group (donating nitric oxide to a molecule to change its structure), and nitrosylating amines (amino acids) will create nitrosamines. 'Nitrosamine' is a categorical term rather than a single molecule, and while there are select nitrosamines that are carcinogenic (cancer causing) some may be inert.
Other products of nitrosylation include S-nitrosothiols and some minerals such as iron from heme (Nitrosyl iron, or FeNO), heme being the ring shaped structure in hemoglobin (red blood cells carrier of oxygen) that holds iron in place.
Nitrate can nitrosylate other compounds by binding to them, and when this occurs to amino acids then nitrosamines are formed
Nitrosyl compounds have been detected in the feces of otherwise healthy volunteers following consumption of red or processed meat products, with their production occurring in a fairly dose dependent manner; furthermore, a nitroso compound specific DNA adduct (thought to be a reliable biomarker of causing cancer) has been detected in the colon of humans consuming red meat.
Following consumption of processed and red meats, there does appear to be production of nitroso compounds in humans. At least one study has confirmed DNA damage in the colon from this
Heme is an issue that applies to both processed meats and red meats, but not poultry nor fish products (at least to a concernable degree).
Role of Heme
An observation that has been noted is that white meats (poultry) do not form nitroso compounds following ingestion to the degree that red and processed meats do and there are studies that despite finding associations with processed meat and cancer, fail to find any association with poultry.
Currently, it is thought that the higher presence of heme in red meat products relative to white is an explanation. It appears that nitric oxide (from sodium nitrate) has affinity for heme and production of nitrosylheme (some sources refer to it as nitrosyl iron) can act as a catalyst for further nitroso compound formation; this is a function not attributed to isolated iron, but to heme containing iron. This product, when detected in the feces, is a predictor of total nitroso formation alongside transit time (longer time to defecation being associated with higher production) and reduced fecal weight (more fecal weight being associated with less production).
Overall, longer exposure times to higher nitroso concentrations appear to dose-dependently increase the risk of forming DNA adducts (a biomarker of DNA damage and thus cancer risk). Without heme, the production rate appears to be quite slow and thus risk is minimized; even at high levels of intake, the epidemiological research does again suggest that while the risk is present it is not dire.
A current theory is that iron from heme (but not free iron from other sources) appears to itself be nitrosylated and then augment production of other nitroso compounds. Thus when nitrates and heme are coingested, production of nitroso compounds is accelerated. It appears to be dose-dependent
It is thought that the presence of heme in a meat product would act as an almost required catalyst for formation of nitroso compounds (implicated both red meat and processed meat), and in this situation only a higher intake of nitrates would lead to dose-dependent increases in nitroso compound formation.
This current theory is in line with the epidemiological evidence that notes that white meat is not associated with any cancer or cardiovascular mortality, that red and processed meat are both implicated (but processed to a slightly higher degree), and that nitrate intake from vegetables (without heme) are very unreliably correlated with harm.
This theory is able to explain the inability of poultry and vegetables to be associated with an increased cancer risk, despite the latter having nitrate content and the former being a meat product
Processed meats, through a combination of nitrates and heme (neither one alone appearing to be too much of a problem) appear to dose-dependently form carcinogenic nitrosamines. This is a phenomena very well researched in vitro (outside of living systems) and in animal models, with at least one human study noting that DNA damage has occurred in the colon following consumption of processed meat products. Due to the current state of research (survey research showing a connection, but no interventions existing due to ethical complications of possibly giving a human cancer intentionally) and the biological plausibility, it is reasonable to assume that processed meats do cause cancer.
However, in the survey research the hazard ratios and overall risk is fairly small. While this increased risk does exist (in a dose-dependent manner), due to its size it is wholly possible that dietary or supplemental interventions to reduce cancer risk could eliminate the risk. The possibility of one unforeseen or additional variable negating the risk is why low hazard ratios tend not to be used, despite technically being an increased risk