Level 1 - Absolute Beginner
Ultra-processed food is food made in a factory. Think of chips, soft drinks, and sweet cereal. Many people think this food can cause cancer.
A new study looked at 461,682 older adults in the United States. They filled in a food form in the 1990s. Scientists followed them until 2018.
The result was a surprise. Overall, people who ate more ultra-processed food did not get more cancer of the colon. The colon is part of the body that helps with food.
But one kind of food was different. Processed meat, like some sausages, was linked to more cancer. Foods with whole grains, fiber, or milk were linked to less cancer.
- factory
- A big building where things are made.
- cancer
- A serious illness where cells grow in a bad way.
- colon
- A long part of the body that helps with food.
- study
- Careful work to find out something new.
- whole grain
- Bread or rice with the whole seed, not only the white part.
- fiber
- A part of plant food that is good for your body.
- sausage
- A food made from chopped meat.
- surprise
- Something you do not expect.
Level 2 - Elementary
A big American study has found no overall link between ultra-processed food and colorectal cancer. The work was published on October 15 in the journal PLOS Medicine. It was led by Erikka Loftfield at the National Cancer Institute.
The team used data from 461,682 adults aged 50 to 71. They joined the NIH-AARP Diet and Health Study in 1995 and 1996, and were followed until 2018. People who ate the most ultra-processed food were about as likely to get colorectal cancer as people who ate the least.
The details matter, though. Ultra-processed meat was linked to a higher risk. Ultra-processed foods that came from whole grains, fiber, dairy, and calcium were linked to a lower risk. The authors say the results support existing advice: eat whole grains and dairy, and avoid processed meat.
The study has limits. It watched people but did not run an experiment, so it cannot prove cause. People described their food once, at the start, and most of them were older and white. A 2025 study of health workers found a higher risk for men, so the question is still open.
- colorectal
- Of the colon and the rectum, parts of the large intestine.
- journal
- A magazine where scientists publish their work.
- ultra-processed
- Made with many industrial steps and added ingredients.
- risk
- The chance that something bad will happen.
- dairy
- Milk and food made from milk.
- authors
- The people who wrote a study.
- limit
- A weakness or a boundary.
- cause
- The reason something happens.
Level 3 - Intermediate
A large American study has challenged a common assumption about diet and cancer. Published on October 15 in PLOS Medicine, the work found no overall association between eating ultra-processed food and developing colorectal cancer. When researchers compared the highest fifth of intake with the lowest, the hazard ratio was 0.97, which means no meaningful difference.
Led by Erikka Loftfield of the National Cancer Institute, the team analysed 461,682 adults aged 50 to 71 who joined the NIH-AARP Diet and Health Study in 1995 and 1996 and were tracked through 2018. The sheer size and long follow up give the result weight, even though it is only one study.
The headline figure hides important differences between food types. Ultra-processed meat was tied to higher risk, whereas ultra-processed products built on whole grains, fiber, dairy, and calcium were tied to lower risk. In other words, the category may be too broad to be useful, and what is in the product may matter more than how industrial it is.
Caveats apply. This was an observational study, so it cannot show cause and effect. Diet was recorded once, with a questionnaire, at the start, and the participants were mostly older and white, which limits how far the findings travel. The study also sits uneasily with a 2025 analysis of about 200,000 health professionals in which men in the top group had a 29 percent higher risk, so the evidence remains mixed.
- assumption
- Something accepted as true without proof.
- association
- A link between two things.
- hazard ratio
- A number comparing how likely an outcome is in two groups.
- follow up
- The period during which participants are tracked.
- observational
- Based on watching, not on an experiment.
- caveat
- A warning about the limits of something.
- participants
- People who take part in a study.
- mixed
- Pointing in different directions.
Level 4 - Advanced
Few nutritional ideas have spread as fast as the suspicion that ultra-processed food drives cancer, which makes a new American study a useful brake. Published on October 15 in PLOS Medicine, it found no overall association between ultra-processed food and colorectal cancer, with a hazard ratio of 0.97 comparing the highest and lowest fifths of intake.
The analysis, led by Erikka Loftfield of the National Cancer Institute, drew on 461,682 adults aged 50 to 71 enrolled in the NIH-AARP Diet and Health Study in 1995 and 1996 and followed through 2018. At that scale and length, a null finding is informative rather than merely inconclusive.
The more interesting result lies beneath the average. Ultra-processed meat was tied to a higher risk, while ultra-processed products rich in whole grains, fiber, dairy, and calcium were tied to a lower one. That pattern suggests that the label, which groups industrial breakfast cereal with reconstituted sausage, is a crude proxy for the nutrients and additives that actually matter, and that guidance to avoid processed meat while favoring fiber and dairy remains sound.
Humility is warranted on both sides. An observational design cannot establish causation; a single baseline food questionnaire cannot capture decades of changing habits; and a cohort that is largely older and white cannot speak for younger, more diverse populations. A 2025 study of around 200,000 health professionals found a 29 percent higher risk among men in the top group, so the two papers are best read as a disagreement still to be resolved.
- brake
- Something that slows a trend down.
- null finding
- A result showing no effect or no difference.
- inconclusive
- Not leading to a clear answer.
- reconstituted
- Made by rebuilding from parts.
- proxy
- A measure used in place of the thing you really want to measure.
- additive
- A substance added to food for taste, color, or shelf life.
- causation
- The relationship in which one thing makes another happen.
- cohort
- A group of people followed together over time.
Talk about this article with KatePractice this story's vocabulary in a real spoken conversation with your AI tutor