Part 1 – Watted?

Guest post by Kip Hansen — January 4, 2025 — 1000 words/4 minutes

prelude: This is a new entry in a series discussing ongoing scientific controversies, a specific type of which is often referred to in the scientific media and elsewhere as “wars”—such as the Wars of the Monarchs, the Wars on Meat, the Obesity Epidemic, the War on Salt, the War on Sugar, and Great Barrier Reef War.

All of these controversies, these scientific wars, have common characteristics that I lay out in the book Modern Scientific Controversies Part 5: Common Elements”.

food war serves as a model for all controversies – in fact, it overlaps with three previously reported wars: War on sugar, war on meat and obesity epidemic.

notes: This article is just a brief introduction to a multi-part series Ultra-processed foods, UPF. There will be several similar short articles in this series.

######

Today’s nutrition and food news, including radio, television, blogs, is full of stories condemning consumption ultra-processed food – UPF. Normally I'd provide a series of links to the latest headlines, but I doubt anyone will miss them – they're everywhere. Here is a typical article asking the main questions: Ultra-processed foods: Are they bad for you?

this Department of Medicine The article cites a number of recent review papers published in major journals and uncritically quotes or explains their conclusions in statements such as: “A 2024 review published in the British Medical Journal (BMJ) examined 45 studies involving nearly 10 million participants. The review's authors concluded that eating more ultra-processed foods increases the risk of death from any cause and is linked to 32 health conditions, including heart disease, mental health disorders, type 2 diabetes and other problems.

The “review” mentioned is Lane et al. 2024 [ pdf here ]with a link to the BMJ editorial “Reasons to avoid ultra-processed foods”. The editorial included this statement:

“The evidence for all-cause mortality, obesity and type 2 diabetes was of high quality (this evidence was rated as moderate quality using the GRADE system, which initially considers all observational studies to be of low quality). Overall, the authors found that a diet rich in ultra-processed foods may be harmful to most, perhaps all, body systems”.

The study authors report their findings as follows:

As an introduction to the entire topic, let’s examine the actual evidence found in Lane et al. 2024 and compare it with the BMJ description and the authors' conclusions. I have provided the data from the paper (with annotations), as well as two explanatory graphs on the strength of the evidence.

Let me try to give a more pragmatic and unbiased explanation of these two tables of findings:

Each item highlighted in yellow shows reasons to doubt the validity of the study results for the following reasons:

1. An odds ratio (OR) between 1.02 and 1.07 is very small and of any significant importance in the real world, and certainly not the smallest clinically important difference.

2. In all graphs, any highlighted result with a confidence interval whisker containing “1” does not make a significant difference, or in plain English, “did nothing” and the effect cannot be considered positive or negative. These individual findings should not be included in any overall results of the study, nor should they be combined with other findings in the “overall score.”

3. Studies with GRADE ratings [“a method used to assess the quality of evidence in research studies”] of low or very low Rather than returning “strong evidence”, they provide low or very low quality evidence.

4. Research results with credibility scores of III, IV and V are not actually evidence. They are restatements of abstracts [and as we shall see are often inaccurate summaries] Regarding the evidence found in other systematic reviews, it is likely that it is what John PA Ioannidis describes as “a simple and accurate measure of general prejudice.”

5. All-cause mortality is neither caused nor prevented by diet. Given the numerous possible confounders, an odds ratio as low as 1.02 (1.01 to 1.03) is unlikely to be a significant result.

Lane 2024 has an interesting finding that deserves a closer look: the mental health finding. The OR is high enough, but the study used had a GRADE rating of low. I wanted to dig a little deeper and see which way the arrow of causation points – whether mental/emotional issues lead to increased UPF intake, or if consuming more UPF leads to mental/emotional health issues. Observational studies such as these cannot answer this question; they only find an association.

We will see the same pattern in all major systematic reviews of UPF questions: tiny OR/HR, OR/HR with CI containing “1”, studies included in the review even if they are low and very low Low quality, low credibility score Research Typethe study measured ambiguous, non-agreed, varying and unclear food ranges (labeled UPF) and claimed that the results apply to a wide heterogeneous food range.

The next article in this series will cover: Changes and uncertainties in UPF definitions.

######

Author comments:

I'm trying a new thing, making the articles in this series short enough for the average reader to read in the limited time they have. The result will be several shorter papers, around 1000 words, with a reading time of 4 or 5 minutes. Please tell me in the comments (starting with “Kip –”) if this is better. The benefit is that more readers can actually read all If they are interested. The disadvantage is that a shorter paper will cover a limited part of the overall topic – leaving many questions that must (hopefully) be answered in subsequent papers.

Lane et al. 2024 There are indeed some findings that may be of interest to researchers. But it used the language of “the strongest available evidence” in its conclusions, drawn from recognized research. low quality and very low quality. Such evidence is not strong (or strongest) – it may be the “strongest” among many weak pieces of evidence, but it should be stated as such. Many weak pieces of evidence do not add up to strong evidence.

These large observational studies can raise interesting questions but cannot and do not provide any clear answers.

The rest of this series will attempt to answer this question: What exactly are ultra-processed foods?

Thank you for reading.

######

Relevant