Swap one-in-ten ultra-processed foods to lower bowel cancer risk — QIMR data

A study has come out of QIMR Berghofer — one of Australia’s leading medical research institutes — and it is doing something useful: it is forcing a more specific question than “is ultra-processed food bad for you?” Researcher Erikka Loftfield’s analysis, drawing on data from 460,000 US adults followed for more than 20 years, found that replacing just one-in-ten ultra-processed food servings with minimally processed alternatives was linked to a 2% lower bowel cancer incidence. But the headline figure is not the most important thing the study says.

The more important finding is what it did not find. Total ultra-processed food intake — the blunt “how much UPF are you eating” metric — was not independently associated with bowel cancer risk. The signal was in substitution, and more pointedly, in what you are substituting.

The NOVA trap

NOVA, the food classification system that divides foods into four groups by degree of industrial processing, is useful. It is also blunt. When a study uses NOVA and finds that “ultra-processed foods” are harmful, it is grouping together: hot dogs and fruit yoghurt. Sliced deli meats and wholegrain bread. Flavoured crackers and fortified breakfast cereal.

This study found that some categories in the NOVA “ultra-processed” bucket were actually associated with lower bowel cancer incidence: breakfast cereals, wholegrain breads, dairy products including fruit yoghurts. The category with consistently higher risk was ultra-processed meats — hot dogs, processed deli items — which is not a new finding. Red and processed meat have been a Tier 1 carcinogen (bowel cancer specifically) since 2015 according to the World Health Organization’s International Agency for Research on Cancer.

The NOVA system captures something real. Food made in factories, with emulsifiers, preservatives, and industrial ingredients, is often nutritionally poor in ways that matter. But treating all NOVA Group 4 foods as equivalent misses the actual biology. This study’s contribution is to say: when it comes to bowel cancer, the food type matters more than the processing label.

Who should care about this

Bowel cancer is Australia’s second-most-common cancer killer. It is one of the few cancers where modifiable diet risk is both well-established and actionable. For a woman in her mid-forties — the age range where bowel cancer surveillance is beginning to be relevant, and where dietary patterns set over decades are either paying off or not — this study lands on practical ground.

It is also a study with real limitations. The data is from a US cohort (NIH-AARP Diet and Health Study) enrolled in 1995–96 and followed for 20+ years. US food processing patterns differ from Australian ones. The dietary assessment was questionnaire-based, which introduces error. This is an observational association; it cannot establish that the swap caused the reduction. A 2% reduction in incidence is modest.

2 cents

What the study does well is translate into a very manageable recommendation. Not “stop eating processed food” — a goal that is unrealistic for most people and poorly defined at the food level. Instead: make one-in-ten servings a swap.

Focus the swap on meat, specifically. Ultra-processed meats — the hot dogs, the salami, the processed ham — are where the risk evidence is clearest, and they are easy to reduce without upending a diet. This is also where RACGP bowel cancer prevention messaging has long sat: reduce red and processed meat, increase fibre.

The wholegrain angle is underused. Patients often frame “cutting processed food” as eliminating anything in a packet. This study is a prompt to clarify: a wholegrain breakfast cereal or a high-fibre bread is not the dietary enemy. In fact, these foods appear in the lower risk category. The nuance matters for patients who are already food-anxious and applying the UPF label too broadly.

Screening is the other conversation. Diet modification has a ceiling on bowel cancer prevention. The bowel cancer screening programme (NBCSP) sends free faecal immunochemical test (FIT) kits to Australians aged 45 to 74. If your patient is 45 or older and hasn’t returned one, that conversation is more impactful than any single dietary swap.

Verdict: maybe — the 2% figure is modest and the data is observational, but the substitution message is cleaner and more clinically useful than most nutrition headlines.

---

Sources cited

1. Replacing just 10% of ultra-processed foods with whole foods linked to drop in bowel cancer rate — The Conversation