Here’s the truth the headlines keep dodging: “ultra-processed” has become the new scare word, and like every scare word before it, it is being used far more lazily than usefully.
A fresh study made headlines by suggesting that higher ultra-processed food intake around conception may be linked to reduced fertility in men and slightly smaller embryonic growth and yolk sac volume in women. Predictably, the media did what the media does. It turned a nuanced piece of research into a broad cultural warning about “processed food”, as if the category itself explains everything and as if the public hasn’t already been trained to hear “processed” and think “poison”.
But that is not what the evidence says. The study itself was more limited than the coverage made it sound. It found that higher maternal ultra-processed food intake was associated with smaller measurements at 7 weeks, but the associations weakened later in the first trimester. It also found that higher paternal UPF intake was linked with reduced fertility. That is worth discussing. It is not nothing. But nor is it a licence to flatten every food made in a factory into the same moral and nutritional category.
That flattening is one of the most irritating habits in modern nutrition discourse. It gives people the illusion of clarity while making them less informed. A hot dog, a fortified soy milk, a sugary energy drink, a high-fibre plant-based burger, margarine, baby formula, supermarket bread, and a can of beans can all end up caught in the same rhetorical dragnet. Then people wonder why the public is confused.
Of course they are confused. They are being told to fear a category instead of understand foods.
And that confusion matters, because when people hear “avoid ultra-processed food”, they do not all picture the same things. Some imagine fizzy drinks and processed meat. Others start side-eyeing tofu, soya milk, breakfast cereal, plant-based meat, wholegrain bread, or anything with more than three ingredients. The result is not better public understanding. The result is nutritional superstition.
This is where the conversation goes badly wrong.
Not all processing is harmful. Humans have been processing food for an astonishingly long time. Archaeological research has shown that our ancestors were grinding seeds, pounding tubers, cooking starches, and detoxifying bitter plant foods thousands of years before agriculture. Processing food is not some modern corruption of a pure ancestral diet. It is one of the reasons our species spread, adapted, and survived. We are not a species that fell from nutritional grace the moment somebody invented a machine. We are a species that has always transformed food to make it digestible, safe, nourishing, portable, and useful.
So the problem is not processing in itself. The problem is what is being processed, into what, for whom, and with what health consequences.
That distinction should be obvious, but the current UPF discourse often treats it as a nuisance. The category becomes the story. The ingredients, nutrient profile, and replacement effect get pushed aside.
Replacement effect matters enormously. A food does not exist in isolation. People do not eat categories. They swap one thing for another. And when researchers actually examine those swaps, the tidy morality tale around “ultra-processed” starts to wobble.
A review published in Current Nutrition Reports made this point clearly. Ultra-processed plant foods are not the same as ultra-processed animal foods, and in many cases they compare favourably not just with processed meat, but with supposedly “unprocessed” animal products too. Plant-based milks, plant-based meat analogues, and modern margarine are often lower in saturated fat, contain no cholesterol, provide fibre absent from animal foods, and lack heme iron, which has been repeatedly linked with higher chronic disease risk. Replacing cow’s milk with soya milk has been associated with lower total and LDL cholesterol, lower C-reactive protein, and lower breast cancer risk in substitution analyses. Replacing meat with plant-based analogues has been associated with reductions in cholesterol, body weight, TMAO, and ammonia. Replacing butter with margarine lowers cholesterol and is associated with lower cardiovascular risk.
That is the part many people do not want to hear.
The food advertised as “natural” is not automatically the healthier one. The food marketed as simple, traditional, and recognisable is not morally or biologically absolved by its familiarity. Red meat does not become a health food because it can be described as “unprocessed”. Dairy does not become benign because it came from a cow rather than a factory vat. Butter does not become cardioprotective because your grandmother used it.
This is one of the deepest flaws in the way UPF is discussed. It quietly smuggles in the idea that “natural” is virtuous and industrial is suspect. That may be emotionally satisfying, but it is not serious nutrition analysis.
Even some researchers and public health bodies that are otherwise concerned about UPFs have had to admit this. The UK government’s rapid update on processed foods reported that vegetarian alternatives were not associated with adverse health outcomes, while ultra-processed meat, animal products, and sweetened drinks tended to be associated with increased risk. That should have been a much bigger story than it was. Instead, the broader public conversation continues to be dominated by hand-wringing over whether an oat milk or veggie burger is too “processed”, while processed meat keeps slipping through under the comforting glow of familiarity.
This would be laughable if it were not so backwards.
Processed meat has been consistently linked with disease risk. Yet a lot of people who would never touch a vegan sausage because it has methylcellulose will happily eat carcinogenic pig flesh wrapped in plastic because at least it feels traditional. That is not a rational nutritional framework.
And yes, there are serious concerns around many ultra-processed foods. No sensible person needs to pretend otherwise. Diets dominated by industrially manufactured products high in salt, sugar, saturated fat, and low in fibre are not a great idea. A global review in The Lancet argued that UPFs pose a major health threat and called for strong public health action. Fine. But even that wider debate has critics pointing out that the category is too broad to tell us which foods are driving which risks. Correlation is not causation. Lumping unlike with unlike does not become rigorous because it is fashionable.
Some of the strongest signals in UPF research appear to be driven by products like sugary drinks and processed meat. Remove those, and the neat story starts to look much messier. That does not make the issue trivial. It makes it more important to get right. Because when public health messaging gets sloppy, people do not become wiser. They become easier to manipulate.
That is exactly what we are seeing now with plant-based foods. Plant-based meats and milks are being folded into the same threatening category as products with completely different nutritional profiles and effects. Research from Finland has even shown that current classification systems can miss meaningful differences in the biochemical composition of plant-based foods. Some forms of processing reduce useful compounds. Others preserve them. Some, as in fermentation, can improve bioavailability. Yet the public is encouraged to think in crude binaries: processed equals bad, unprocessed equals good. No wonder survey data found that so many people either believe all processed foods are unhealthy or do not know what to think at all.
This is the cost of dumbing down nutrition until it fits a headline.
It also has consequences beyond personal confusion. The more the public is taught to fear “ultra-processed plant foods” in the abstract, the easier it becomes to steer them back toward meat and dairy under the guise of common sense. Suddenly, the cholesterol-free burger with fibre is cast as suspicious, while the corpse of an animal gets framed as wholesome because it is less technologically novel. Suddenly, soy milk is interrogated for ingredients while cow’s milk escapes scrutiny despite its saturated fat, hormones, and links to adverse outcomes. Suddenly, “processed” becomes less a scientific descriptor than a cultural weapon used selectively against plant-based change.
That is why this latest fertility story needs to be handled carefully.
If people take from it that diets high in many ultra-processed products may be associated with poorer reproductive outcomes, fair enough. That is a reasonable point for further research and cautious reflection. If they take from it that men and women trying to conceive may benefit from a diet centred more on whole and minimally processed plant foods, also fair enough. That is hardly controversial.
But if they take from it that the answer is to recoil from plant-based milks, meat alternatives, or other processed plant foods while continuing to treat meat, dairy, and butter as sensible staples, then the message has been mangled beyond recognition.
Whole plant foods should absolutely be the foundation. Beans, lentils, grains, vegetables, fruit, nuts, seeds, potatoes, herbs, spices. That is where the strongest evidence keeps pointing. But transitional foods matter too. Real life matters. Accessibility matters. Convenience matters. The person swapping cow’s milk for soy milk or a beef burger for a plant-based one is not making some catastrophic health trade-off because a factory was involved. In many cases, they are making an improvement.
Public health messaging should be mature enough to say that plainly.
Not all UPFs are equally useful. Not all are equally risky. Not all are equally nutritious. Not all should be defended. Not all should be condemned. And the idea that “processing” itself is the main event is increasingly looking like a blunt instrument masquerading as insight.
The real questions are simpler and harder.
What is this food made of?
What does it replace?
What happens when people eat more of it?
Who profits from the confusion?
And why are animal products so often treated as the default safe option even when the evidence says otherwise?
The media loves a villain, and “ultra-processed food” is a convenient one. It sounds modern, sinister, and broad enough to carry any anxiety people already have about industrial life. But science is not supposed to exist to flatter our intuitions. It is supposed to sharpen them.
So let’s say it clearly: a category that places processed meat and fortified soy milk under the same ominous umbrella is not precise enough to carry public health messaging on its own. A label that encourages people to fear plant-based alternatives more than animal products is not helping. And a culture that hears “processed” and stops thinking is not becoming healthier. It is becoming easier to mislead.
The fertility study is worth attention. It is not worth turning into another anti-plant panic.
Because the real danger here is not just bad food. It is bad framing. And bad framing has a habit of protecting the very foods doing the most damage while demonising the ones that could help replace them.