The headline draws conclusions only very tenuously supported in the body:

>The study, published in the British Medical Journal, does not prove that ultra-processed foods cause disease. Nor does the effect appear particularly large, even in the most enthusiastic junk food consumers.

>Touvier said it was unclear how ultra-processed foods may harm health. Even when its poor nutritional value is taken into account, consumption is still linked to more disease and death, she said. One suspicion is that it displaces healthier, more nutritious foods, but additives and perhaps contaminants from processing and packaging may play a role too.

The researchers don't substantiate the cause of "heavily processed foods" being linked to early death at all; if they're not even sure if it's the content of processed foods or just bad diets, it's total fearmongering.

Muffins and ice cream don’t harm your health because they’re “made in a factory.”

The advice to cut out these foods is probably good, but I wish they didn’t tie it to this romantic, appeal-to-nature silliness.

The label "processed food" indicates not only that the food was probably made in a factory, but that its also full of preservatives and lacking in nutrients, generally. This article doesn't seem, to me, to be tying their argument to an appeal-to-nature.

Or they're just full of sugar. Ice cream isn't unhealthy because it 'lacks nutrients' it is unhealthy because it is basically just shoveling sugar into your mouth.

Typical Guardian clickbait.

From the article:

The study, published in the British Medical Journal, does not prove that ultra-processed foods cause disease. Nor does the effect appear particularly large, even in the most enthusiastic junk food consumers.

2 different studies are discussed in the article:

For the second study, also in the BMJ, a team at the University of Navarra in Pamplona monitored the eating habits and health of nearly 20,000 Spanish graduates from 1999 to 2014. Over the course of the study, 335 participants died. Once factors such as age, sex, body mass index and whether or not people smoked were taken into account, the trend was clear. The top quarter consumers of ultra-processed foods – who had more than four servings a day – were 62% more likely to have died than those in the bottom quarter, who ate less than two portions a day. For each additional serving, the risk of death rose 18%.

> The results suggest that 277 cases of cardiovascular disease would arise each year in 100,000 heavy consumers of ultra-processed foods, versus 242 cases in the same number of low consumers.

So, one is 14% more likely. But at 0.242% and 0.277% chance the correlation seems unusual to even mention.

I’m not a statistician. Is this typical of such studies?

Oh, and as for the model used, Cox multifactorial analysis is fine, except they forgot to add an unknown member to the parameter set. Without it, this cannot measure unknown unknowns and you cannot say if the analysis makes sense. If unknown covariate has higher hazard ratio than known covariate, something is off. Another way to validate is to calculate goodness of fit for each covariate.

The strong assumption in Cox model is that hazards are independent constant multiplicative which might not be true at all. (Such as limited detoxication capacities or hazards varying over time, or hazard clusters, or synergy between hazards. These are all examples of nonlinearities.)

In case anyone wants a primer: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2394368/

FAMD and MFA would show whether the simple nonparametric linear constant hazard model holds between paired variables.

This is correlation on a self selected group. A different statistical correction is needed for those. (These people are not being fed this diet despite their wishes, so there are lifestyle, metal health and other correlates.)

Thus causality can go either way with multiple factors interfering. The experiment is not powered enough to actually derive a conclusion. They'd need at least a 2 million sample sizes for each group to show anything other than noise, and that's with the ultra fancy multilinear math used in genetics to untangle multiple factors. (Advanced variants of FAMD.)

Welcome to land of observational studies.

A study found that processed meat consumption is inversely correlated with telomere length. That is likely due to the various chemicals and preservatives used in processing the meat, because unprocessed red meat did not have the same effect. https://www.ncbi.nlm.nih.gov/m/pubmed/27558579/

If you’re curious about diet’s effect on longevity other than causing obesity and diabetes, I suggest reading “The Telomere Effect.” It’s a great book which also covers things like the epigenetic effects of stress and how stressing a pregnant woman could even cause epigenetic changes in the baby. Note that some of the research is new and not rigorously validated yet.

Unless it grew, it probably gives you cancer, disease. What’s new? How many studies do we need to know that processed = not good for you?

imho - this is better summary:

BMJ Editorials: https://www.bmj.com/content/365/bmj.l2289 "Ultra-processed food and adverse health outcomes" (Published 29 May 2019)

the ingredients list in many ice creams is very small, 2-3 things

Which?

This is what a Vanilla ice cream tub looks like in Australia:

Water, Sugar, Milk Solids, Cream and/or Butter, Glucose Syrup (from Wheat), Maltodextrin, Vegetable Emulsifier (471), Stabilisers (410, 412, 415), Natural Vanilla Flavour, Natural Colour (160a), Reduced Fat Ice Cream contains minimum 4% Milk Fat

https://www.woolworths.com.au/shop/productdetails/74273/esse...

No idea about Australia, but where I live I get ice creams with an ingredient list that looks like this:

Milk, Figs & Sugar

Of course I can also get ice creams with all of the stuff you've listed, too, but the option exists.