A recent study called beer and fries part of a plant-based diet. That’s problematic for more reasons than you’d think

Earlier this month, a study was published in the Lancet, linking “plant-based ultraprocessed foods” (UPFs) to heart disease and early death. After years of doctors and other health experts telling us that plant-based eating is nutritionally superior, that eating more vegan foods will extend our lives and fight our risk of serious diseases, it’s surprising to hear the opposite. And media outlets like the New York Post and the Daily Mail took this supposed finding even further, singling out plant-based meat in particular with headlines like “Vegan fake meats linked to heart disease, early death.”

The problem is, that’s not really what the study found. At first glance, it might seem like the takeaway is that plant-based meat is actually worse for you than the animal meat it’s meant to replace. But the reality is that nutritional science is much, much less definitive and reliable than we’d like to think—and there are risks of making decisions based on shaky data, like those that come from this study.

For starters, plant-based meat actually only made up 0.2% of the food consumed in the study. There were 118,387 participants (ages 40-69) from the United Kingdom; each shared what they ate over at least two 24-hour periods (which researchers later paired with hospital and mortality records).

Most of the foods reportedly consumed were technically plant-based, but wouldn’t be the first things that come to mind when people hear the word “vegetarian” or “vegan.” The items that made up much greater proportions of the subjects’ food intake were things like French fries, soft drinks, cookies, even alcoholic beverages. All the study really suggests is that eating a diet of sugary soda, wine, and packaged chips is associated with poor health outcomes—not exactly groundbreaking stuff.

Then there’s the fact that due to the nature of this research, it wasn’t possible for the researchers to randomly assign similar individuals to different diets. Consequently, they can’t be certain that they’ve accounted for all the factors that may influence the health outcomes they tracked. Dr. Tom Sanders, Professor Emeritus of Nutrition and Dietetics at King’s College London, wrote that “[t]hose who consumed the highest amount of ultra processed food were more likely to be smokers/ex-smokers, to have higher levels of physical activity, [and] more likely to live in London.” He noted that researchers were able to adjust for those potentially confounding factors, but that things like home and car ownership, employment, and housing weren’t considered.

Unfortunately, this isn’t about just one study—there’s a broader issue when it comes to the way the media presents scientific findings to the public, especially those regarding nutrition. There’s an overarching issue where people extrapolate from scant data without even realizing they’re doing it; for instance, when a study finds that X food is associated with Y outcome, and readers jump to thinking that X necessarily causes Y.

Communications around nutritional science can be confusing for the public and even for journalists and other professionals, because the study of nutritional epidemiology—the relationship between nutrition and health—is a minefield of flawed methods and archaic assumptions. Most findings in this field are much, much more preliminary than the public realizes. For one thing, many of these studies—including the aforementioned one—use self-reported data. Even when people are trying to be honest and accurate, they’re generally not good at estimating how much they actually ate, how much salt they used while cooking, or how much dressing they used on their salads. And as journalist Tamar Haspel notes in the Washington Post, “You can use a 24-hour recall, which gives respondents a fighting chance of remembering what they actually ate but doesn’t give you a representative sample of overall diet.” The data itself is ultimately a collection of guesses—and that’s just the starting point.

Another problem is that nutritional science tends to use categorizations and measurements that are arbitrary and outdated. Many studies refer to BMI (body mass index), but there’s actually little backing up the BMI’s categorizations of normal, overweight, and obese. On top of that, you can find hordes of articles where scientists criticize BMI as an inaccurate measurement of health that doesn’t account for important confounding factors.

Similarly, the classification of foods as “ultraprocessed” or “minimally processed” comes from what’s called the NOVA system, which isn’t particularly well-suited to tell us about nutritional outcomes, despite being used in this recent study. As Max Elder, Managing Director of Food System Innovations, points out, NOVA “likely has no causal explanatory power,” meaning it doesn’t explain the health outcomes observed in the study. “The authors themselves confirm that food composition is a likely explanation.” In other words, what seems to matter is the contents of the foods, not the level of processing.

But of course, the average person skimming an article about this particular study is going to get the impression that the more processed something is, the worse it is for you, which isn’t necessarily the case (a win for factory farmed meat since it is, in a sense, the most processed food on Earth). Sure, kale is both less processed and more nutritious than a potato chip, but sometimes, it’s the reverse. Consider that in the study, tofu—a dietary staple in many cultures and a revered protein among health experts—was classified as ultraprocessed, whereas beer and wine were classified as unprocessed. As Dr. Duane Mellor, Dietitian and Spokesperson for British Dietetic Association notes, “no public health expert would suggest drinking beer and wine as … a good way to improve health!”

Because of issues with data collection and measurement, not to mention the influence of personal and professional biases, nutritional findings are all over the place. The fact is that the human body is complicated and it’s difficult to pick a metric that represents overall health. We can measure specific markers but none really give the full picture of a person’s well-being. Throw in the fact that everyone’s body is different and reacts differently to different foods, and it becomes difficult to confidently extract any sort of general, widely-applicable truth, especially on the margins.

Ultimately, we need to stop looking for big answers in small studies, and avoid painting in broad strokes where it’s inappropriate to do so. When it comes to plant-based meat, there are some ways in which it beats out animal meat, like if you’re looking to cut back on cholesterol and add in more fiber, and some ways in which it doesn’t — for instance, those trying to limit sodium. Of course, there are dozens of other nutrients, vitamins, and minerals that need to be considered, and in truth, the impact of food on our bodies probably can’t be reduced to them alone. And obviously, neither plant-based meat nor animal-based meat comes close to the associated healthfulness of fruits, vegetables, legumes, and grains. (This is largely irrelevant in practice, as plant-based meat is designed to compete with animal-meat and not salads).

Swapping out some animal meat for plant-based patties, nuggets, or filets might extend your life and protect you from disease—or it might not. Honestly, we just don’t have data to form a reliable conclusion on what the optimal human diet is for everyone (if there even is one single “optimal” diet for anyone, which is unlikely). But there are other metrics outside of nutrition, which we can use to evaluate foods. How much land and water was used to make your meal? How much carbon dioxide was released in the process of making or growing or raising it? Is it associated with air pollution and a risk of zoonotic disease and antibiotic resistance? Did any living being have to be tortured for it to end up on your plate?

In those respects, the answers are clear. Plant-based meat has obvious and dramatic advantages over factory farmed meat when it comes to environmental impact, animal welfare, and public health. Until nutritional science can offer us more reliable information on this topic, let’s make choices based on what we do know—swapping in plant-based meat for animal meat will benefit our planet, animals, and people at large. I’ll be grilling plenty of plant-based meat (the indulgent kind) this summer, and you should too.