Ultra Processed Food and Cancer Risk: What Research Says
An evidence-based review of epidemiological research examining associations between ultra-processed food consumption and cancer incidence. This guide presents findings from major cohort studies alongside their limitations and practical context.
Important Medical Disclaimer
This guide summarizes epidemiological research for educational purposes only. It is not medical advice. Cancer risk is influenced by genetics, environment, lifestyle, and many factors beyond diet. The studies cited are primarily observational and demonstrate statistical associations, not proven causation. Do not interpret this information as diagnostic or prescriptive. Consult an oncologist or qualified healthcare provider for cancer-related health concerns.
Understanding the Research Landscape
Cancer is among the most complex diseases studied by modern medicine. It arises from the interplay of genetics, environmental exposures, lifestyle factors, and -- increasingly studied -- dietary patterns. Over the past decade, a growing body of epidemiological research has examined whether consumption of ultra-processed foods is statistically associated with cancer incidence. This guide presents the major findings from that research, alongside the substantial limitations that must inform any interpretation.
It is essential to understand what this research does and does not show. Large population studies (cohort studies) follow thousands or hundreds of thousands of people over years, tracking their dietary habits and health outcomes. When researchers find that people who consume more ultra-processed food also have higher rates of certain cancers, this is a statistical association. It does not, by itself, prove that ultra-processed foods caused those cancers. Many other factors -- smoking, alcohol, obesity, physical inactivity, socioeconomic conditions, genetic predisposition -- also influence cancer risk, and disentangling these variables is one of the central challenges in nutritional epidemiology.
This guide draws on peer-reviewed research published in journals including The BMJ, eClinicalMedicine (The Lancet), the American Journal of Clinical Nutrition, and others. For a broader overview of UPF research across all health outcomes, see our health effects guide.
A note on language: Throughout this guide, we use phrases such as “research suggests,” “studies have found associations,” and “linked to increased risk.” These phrases reflect the observational nature of the evidence. We deliberately avoid language that implies proven causation, because the current evidence does not support such claims.
The NutriNet-Sante Cohort: A Landmark Study
The first large-scale study to specifically examine ultra-processed food consumption and cancer risk was published in 2018 by Fiolet and colleagues, drawing on data from the French NutriNet-Sante prospective cohort. This study generated significant media coverage and catalyzed further research into the topic.
Study: “Consumption of ultra-processed foods and cancer risk”
Authors: Fiolet T, Srour B, Sellem L, et al.
Journal: The BMJ, 2018;360:k322
Cohort: 104,980 French adults, followed for a median of 5 years
Key finding: A 10% increase in the proportion of ultra-processed food in the diet was associated with a 12% increase in overall cancer risk (HR 1.12, 95% CI 1.06-1.18) and an 11% increase in breast cancer risk (HR 1.11, 95% CI 1.02-1.22).
What This Study Found
The researchers tracked dietary intake using repeated 24-hour dietary records and classified foods according to the NOVA system. After adjusting for known cancer risk factors including age, sex, smoking status, physical activity, BMI, family history of cancer, alcohol intake, and overall dietary quality, they found that participants who consumed higher proportions of ultra-processed food had statistically higher rates of overall cancer and breast cancer during the follow-up period.
To contextualize the absolute risk: among the 104,980 participants, 2,228 cancer cases were diagnosed during follow-up. The reported hazard ratio of 1.12 for a 10% increase in UPF proportion means a relatively small absolute difference when applied to an individual. If the baseline cancer rate in this population was approximately 2.1% over the follow-up period, a 12% relative increase corresponds to approximately 2.4% -- a difference of roughly 0.3 percentage points.
Key Limitations
Observational design: As a cohort study, it identifies associations but cannot prove causation. People who eat more UPFs may differ from those who eat fewer in ways that independently affect cancer risk.
Self-reported diet: Dietary intake was assessed through participant-reported records, which are subject to recall bias and underreporting.
Residual confounding: Despite statistical adjustments for known confounders, variables such as income, education, stress, and environmental exposures may not be fully accounted for.
Relatively short follow-up: A median of 5 years is relatively short for cancer research, given that many cancers develop over decades.
Follow-up analyses from the same cohort have examined specific UPF subcategories and additional cancer sites, generally finding consistent but modest associations. The NutriNet-Sante study remains an active cohort and continues to produce research on this topic.
Subsequent Large-Scale Studies
Following the NutriNet-Sante publication, several other large cohort studies examined the relationship between ultra-processed food consumption and cancer. These studies span different countries, populations, and methodologies, providing additional data points but also reflecting the same fundamental limitations inherent to observational research.
UK Biobank Analysis
Authors: Chang K, Gunter MJ, Rauber F, et al.
Journal: eClinicalMedicine (The Lancet), 2023
Cohort: ~197,000 UK adults from the UK Biobank
Key finding: Higher UPF consumption was associated with increased risk of cancer across 34 anatomical sites. The strongest associations were reported for ovarian cancer (HR 1.19 per 10% increase) and brain cancer. Each 10% increase in UPF intake was associated with a 2% increase in overall cancer incidence and a 6% increase in cancer mortality.
Italian Moli-sani Cohort
Authors: Bonaccio M, Di Castelnuovo A, Costanzo S, et al.
Journal: American Journal of Clinical Nutrition, 2022
Cohort: ~22,000 Italian adults, median follow-up of 12 years
Key finding: Found that higher consumption of ultra-processed food was associated with increased all-cause and cardiovascular mortality. Cancer-specific mortality associations were examined but showed less consistent results than cardiovascular findings.
EPIC Cohort (European Prospective Investigation into Cancer and Nutrition)
Scope: Multi-country European cohort spanning 10 countries
Participants: ~500,000 adults
Key finding: Analyses from the EPIC cohort have examined UPF consumption in relation to colorectal cancer and other cancer sites. Results have generally found modest positive associations, though effect sizes vary by cancer type and analysis method.
Harvard Cohorts (Nurses' Health Study / Health Professionals Follow-Up Study)
Authors: Wang L, Du M, et al.
Journal: The BMJ, 2022
Cohort: ~200,000 US health professionals, up to 28 years follow-up
Key finding: Higher UPF consumption was associated with increased colorectal cancer risk, particularly in men. The association was strongest for ready-to-eat or ready-to-heat mixed dishes and sugar-sweetened beverages.
Brazilian Cohort Studies
Context: Brazil, where the NOVA classification system originated, has produced several analyses examining UPF consumption and health outcomes
Key finding: Brazilian population studies have contributed data on UPF consumption patterns and mortality, though cancer-specific analyses from these cohorts are more limited. These studies are notable for examining populations with different dietary patterns than European and North American cohorts.
Research Summary: Key Studies at a Glance
The following table summarizes the major epidemiological studies examining ultra-processed food consumption and cancer outcomes. Each entry includes both the primary finding and the key limitation -- both are essential for interpreting the research accurately.
| Study | Year | Cohort Size | Cancer Types | Finding | Key Limitation |
|---|---|---|---|---|---|
| Fiolet et al. (NutriNet-Sante) | 2018 | 104,980 | Overall, breast | 10% UPF increase associated with 12% higher overall cancer risk | Short follow-up (5 yr median); self-reported diet |
| Chang et al. (UK Biobank) | 2023 | ~197,000 | 34 sites, ovarian, brain | 10% UPF increase associated with 2% higher incidence, 6% higher mortality | Single dietary assessment at baseline; UK-specific population |
| Wang et al. (Harvard cohorts) | 2022 | ~200,000 | Colorectal | Higher UPF linked to colorectal cancer risk, especially in men | Health professional cohort may not represent general population |
| Bonaccio et al. (Moli-sani) | 2022 | ~22,000 | Cancer mortality (general) | UPF associated with all-cause mortality; cancer findings less consistent | Smaller cohort; Italian dietary context differs from US/UK |
| EPIC multi-country analyses | Various | ~500,000 | Colorectal, various | Modest positive associations across cancer types | Dietary assessment methods vary across 10 countries |
| Lane et al. (BMJ Umbrella Review) | 2024 | ~10 million (pooled) | Multiple | Cancer evidence classified as “suggestive” (not “convincing”) | Review of existing studies; inherits limitations of source data |
Interpreting this table: The consistency of associations across independent studies in different countries is noteworthy. However, consistency of observational findings does not, by itself, establish causation. All of these studies share fundamental methodological limitations including reliance on self-reported dietary data and the impossibility of fully controlling for confounding variables. The BMJ umbrella review -- which synthesized all available evidence -- classified the cancer associations as “suggestive,” placing them below the “convincing” threshold met by cardiovascular disease and type 2 diabetes.
Potential Mechanisms Under Investigation
If ultra-processed food consumption is genuinely associated with increased cancer risk beyond confounding, what biological mechanisms might explain this? Researchers have proposed several hypotheses, each supported by varying levels of evidence. These are areas of active investigation, not established causal pathways.
Food Additives Under Regulatory Review
Some food additives commonly found in ultra-processed foods are under active review by regulatory and research bodies. Titanium dioxide (E171), used as a whitening agent in candies, chewing gum, and some processed foods, was banned as a food additive in the European Union in 2022 after the European Food Safety Authority concluded it could not rule out genotoxicity concerns. It remains permitted in the United States. Certain synthetic food dyes are under ongoing evaluation. However, the mere presence of an additive in a food does not mean it causes cancer at typical consumption levels -- dose, duration of exposure, and individual variation all matter.
Evidence level: Substance-specific; some additives have regulatory action, most have not been linked to cancer. See our ingredients to avoid guide for more detail on specific additives.
Contaminants from Processing and Packaging
Ultra-processed foods often undergo extensive industrial processing and are packaged in plastics and other materials from which chemicals can migrate into food. Bisphenol A (BPA) and phthalates are endocrine-disrupting compounds that have been detected in packaged foods and are under investigation for potential links to hormone-sensitive cancers. The extent to which typical dietary exposure to these compounds contributes to cancer risk remains an active research question. Regulatory agencies have set exposure limits, but debate continues about whether current limits are sufficiently protective.
Evidence level: Active research area; endocrine disruption is established, but the cancer risk from food-related exposure at typical levels is not yet quantified.
Acrylamide Formation During Processing
Acrylamide is a chemical compound that forms when starchy foods are cooked at high temperatures (above 120 degrees Celsius / 248 degrees Fahrenheit) -- a process common in the manufacture of chips, crackers, cereals, and many baked goods. The International Agency for Research on Cancer (IARC) classifies acrylamide as “probably carcinogenic to humans” (Group 2A), based primarily on animal studies showing that high doses cause cancer in rodents. Human epidemiological evidence is less consistent, and typical dietary exposure is far lower than doses used in animal studies.
Evidence level: IARC Group 2A (probable carcinogen); animal evidence is strong, human epidemiological evidence is limited and inconsistent at typical dietary exposure levels.
Nutrient Displacement
Diets high in ultra-processed foods tend to be lower in fiber, vitamins, minerals, and phytonutrients -- compounds found in fruits, vegetables, legumes, and whole grains that have established or proposed protective roles against certain cancers. High fiber intake, for example, is associated with reduced colorectal cancer risk in multiple studies. When ultra-processed foods displace whole foods in the diet, the resulting nutrient shortfall may reduce the body's protective capacity. This mechanism does not require any direct carcinogenic effect from UPFs themselves -- the issue is what is not being eaten.
Evidence level: Well-supported; the protective role of dietary fiber against colorectal cancer is classified as “probable” by the World Cancer Research Fund.
The Obesity Pathway
Obesity is an established risk factor for at least 13 types of cancer, according to the Centers for Disease Control and Prevention. These include cancers of the breast (post-menopausal), colon, rectum, endometrium, esophagus, kidney, liver, pancreas, and others. Research -- including the NIH feeding trial by Hall et al. (Cell Metabolism, 2019) -- has found that ultra-processed food consumption is associated with higher caloric intake and weight gain. If UPFs contribute to obesity, they may indirectly increase cancer risk through this well-established pathway. This may explain some or all of the observed associations between UPF consumption and cancer in epidemiological studies.
Evidence level: The obesity-cancer link is well-established. The UPF-obesity link is supported by the NIH RCT and consistent observational data. The indirect UPF-obesity-cancer pathway is plausible but not yet quantified.
Important context: These are hypothesized mechanisms, not established causal pathways. Multiple mechanisms may operate simultaneously, and their relative contributions are unknown. It is also possible that the observed associations between UPF consumption and cancer are largely or entirely explained by confounding factors rather than by any direct or indirect mechanism. More research, particularly randomized controlled trials, is needed to clarify these questions.
What the Research Does Not Show
Given the sensitivity of the topic, it is critically important to be explicit about what the current research does not demonstrate. Overinterpreting observational associations can lead to unnecessary anxiety and poorly informed decisions.
UPFs do not “cause” cancer
No study has demonstrated a causal relationship between ultra-processed food consumption and cancer in humans. The existing research shows statistical associations in observational data. This is an important scientific distinction. Many factors that are correlated are not causally related.
Occasional UPF consumption does not meaningfully change individual cancer risk
The studies examine habitual dietary patterns over years, not individual meals or occasional consumption. The absolute risk differences reported are small. Eating an ultra-processed snack or meal occasionally is not something the research suggests should cause concern.
The absolute risk increases are small
Media coverage often emphasizes relative risk increases (such as “12% higher cancer risk”) without providing absolute context. A 12% relative increase applied to a low baseline risk yields a very small absolute change. For comparison, smoking increases lung cancer risk by approximately 1,500-3,000%, and heavy alcohol consumption increases certain cancer risks by 200-500%. The UPF-associated risk increases, even if real, are far smaller than those from established carcinogenic behaviors.
Confounding variables cannot be fully eliminated
People who consume more ultra-processed food may differ from those who consume less in many ways beyond their diet. They may have lower incomes, less access to healthcare, higher stress levels, less physical activity, and different environmental exposures. While researchers statistically adjust for measured confounders, unmeasured or incompletely measured confounders can still influence results.
“Ultra-processed” groups vastly different foods together
Under the NOVA classification, a fortified whole-grain bread, a diet soda, a protein bar, and a frozen pizza all fall into the same “ultra-processed” category. Treating these as equivalent in terms of cancer risk is a significant oversimplification. Some studies have begun to examine UPF subcategories, finding that associations vary substantially -- but most of the headline-level research treats UPFs as a single group.
To be direct: If you are anxious about cancer risk after reading media coverage of UPF research, it is worth understanding that the absolute risk changes described in these studies are small, that the evidence is classified as “suggestive” rather than “convincing,” and that many other factors have far greater influence on cancer risk. This is not a reason for complacency about diet, but it is a reason to maintain perspective.
Practical Context: Diet Within the Bigger Picture
To put the UPF-cancer research into proper perspective, it helps to understand where diet fits within the broader landscape of known cancer risk factors. The following factors have strong, well-established evidence supporting their role in cancer risk.
Established Cancer Risk Factors With Strong Evidence
Tobacco Use
The single largest preventable cause of cancer. Linked to at least 15 cancer types. Responsible for approximately 30% of all cancer deaths in the United States.
Alcohol Consumption
Classified as a Group 1 carcinogen by IARC. Associated with cancers of the mouth, throat, esophagus, liver, breast, and colon. Risk increases with quantity consumed.
Obesity
Established risk factor for at least 13 cancer types. Mechanisms include chronic inflammation, elevated insulin levels, and hormonal changes.
UV Exposure
Ultraviolet radiation from the sun and tanning beds is the primary cause of skin cancers, including melanoma. A well-established and dose-dependent relationship.
Certain Infections
HPV (cervical cancer), Hepatitis B and C (liver cancer), and H. pylori (stomach cancer) are infectious agents with established causal roles in specific cancers.
Physical Inactivity
Regular physical activity is associated with reduced risk of colon, breast, and endometrial cancers. The evidence is rated as “convincing” by the World Cancer Research Fund.
What Major Health Organizations Recommend
The American Cancer Society (ACS) and the World Cancer Research Fund (WCRF) both issue dietary recommendations for cancer prevention. Their guidance emphasizes eating patterns rather than individual foods or ingredients.
Eat a diet rich in whole grains, vegetables, fruits, and beans. The WCRF recommends making these the foundation of daily meals. Dietary fiber from whole plant foods is associated with reduced colorectal cancer risk.
Limit consumption of red and processed meats. The IARC classifies processed meat as a Group 1 carcinogen (sufficient evidence of carcinogenicity in humans) and red meat as Group 2A (probably carcinogenic).
Limit sugar-sweetened beverages and highly processed foods. Both the ACS and WCRF recommend limiting these, consistent with but not exclusively based on UPF research.
Maintain a healthy weight. Given the established link between obesity and multiple cancers, maintaining a healthy body weight is one of the most evidence-based cancer prevention strategies.
Reducing ultra-processed food consumption may align with these recommendations -- since doing so often means eating more whole foods, less added sugar, and less processed meat -- but it is not a cancer prevention strategy on its own. The research on UPF and cancer is one component of a much larger body of nutritional evidence.
For practical guidance on shifting toward less processed dietary patterns, see our guide to reducing ultra-processed food and our food swaps guide.
The bottom line: Diet is one of many factors that influence cancer risk. The current evidence suggests that habitual consumption of high amounts of ultra-processed food is associated with modestly increased cancer risk in population studies, but this association is not proven to be causal, and the absolute risk changes are small. Known risk factors with far stronger evidence -- tobacco, alcohol, obesity, physical inactivity -- should be the primary focus for anyone concerned about cancer prevention. Consult an oncologist or your healthcare provider for guidance tailored to your individual risk profile.
Frequently Asked Questions
Do ultra-processed foods cause cancer?
The current research does not establish that ultra-processed foods cause cancer. Several large observational studies have found statistical associations between higher UPF consumption and increased incidence of certain cancers, but observational studies cannot prove causation. Confounding factors such as smoking, obesity, socioeconomic status, and overall lifestyle make it difficult to isolate the role of food processing itself. The BMJ umbrella review (2024) classified cancer associations as "suggestive" rather than "convincing." If you have concerns about cancer risk, consult an oncologist or your primary care physician.
Which cancers have been most associated with UPF consumption in studies?
The strongest statistical associations reported in published research involve colorectal cancer, breast cancer, and ovarian cancer. The NutriNet-Sante cohort (Fiolet et al., BMJ, 2018) found associations with overall cancer risk and breast cancer. The UK Biobank analysis (Chang et al., eClinicalMedicine, 2023) reported associations across 34 cancer sites, with the strongest for ovarian and brain cancers. However, these are observational associations with meaningful limitations, and findings vary across studies. No single cancer type has a "convincing" level of evidence linking it to UPF consumption specifically.
How large is the increased cancer risk associated with UPFs?
In the most widely cited study (Fiolet et al., BMJ, 2018), a 10% increase in the proportion of ultra-processed food in the diet was associated with approximately a 12% increase in overall cancer risk. This is a relative risk increase, not an absolute one. To put this in context: if the baseline absolute risk of developing a particular cancer over a given period is 5%, a 12% relative increase would raise it to approximately 5.6%. These figures come from observational data with known limitations, and the absolute risk change for any individual is small. Other established cancer risk factors such as tobacco use carry far larger relative and absolute risk increases.
Should I change my diet based on this research?
Major health organizations including the American Cancer Society and the World Cancer Research Fund recommend diets rich in whole grains, vegetables, fruits, and beans, and limited in processed and red meats, sugar-sweetened beverages, and highly processed foods. These recommendations are based on the totality of nutritional evidence, not just UPF research. Reducing consumption of ultra-processed foods may align with these broader dietary recommendations, but it is not a cancer prevention strategy on its own. Any dietary changes should be discussed with a healthcare provider who understands your individual health history and risk factors.
Are food additives in UPFs carcinogenic?
Most food additives approved for use in the United States and European Union have been evaluated for safety by regulatory agencies. However, some substances found in or associated with ultra-processed foods are under ongoing investigation. Titanium dioxide (E171) was banned as a food additive in the EU in 2022 based on genotoxicity concerns. Acrylamide, which forms during high-temperature processing of starchy foods, is classified as "probably carcinogenic to humans" (Group 2A) by the International Agency for Research on Cancer. Sodium nitrite in processed meats can form nitrosamines, which are established carcinogens. These are substance-specific findings and do not mean all food additives are carcinogenic. Regulatory evaluations are ongoing.
Continue Learning
UPF and Health Overview →
Comprehensive review of 200+ studies on ultra-processed food and health outcomes across all categories
Ingredients to Avoid →
Research-based guide to food additives, preservatives, and processing agents under scientific scrutiny
Reducing UPF →
Practical strategies for reducing ultra-processed food in your diet with category-by-category swaps
Disclaimer: All tools and data visualizations are provided for educational and informational purposes only. They are not intended as health, medical, or dietary advice. Product formulations change frequently — always check the actual label for current ingredients and nutrition facts before making purchasing decisions. Consult healthcare professionals for personalized dietary guidance.