This week in Milan, at the annual meeting of the European Association for the Study of Diabetes (EASD), two studies landed with a familiar message: what you eat and how you move may shape your risk of type 2 diabetes.
That message isn’t new. What’s interesting is the detail underneath, especially the parts that rarely make it into the headlines. So let’s go through it slowly, and honestly.
- Study 1 (France, 89,953 adults): People with the highest intake of Neu5Gc, a sugar in red meat, had a 63% higher type 2 diabetes risk than those with the lowest. The results are preliminary.
- Study 2 (Hong Kong team, UK data, 56,964 adults): Less ultra-processed food plus regular exercise was linked to lower risk, even in people with high genetic risk (49% lower).
- The fine print: Both studies are observational. Neither proves cause and effect, and the headline percentages shrink once you look closely.
- The takeaway: Move more, watch your drinks, and don’t let beef and lamb dominate the plate. Not one of these is a magic bullet.
Study 1: The sugar your body never learned to accept
The ELI5 version: Most mammals make a sugar molecule called Neu5Gc. Humans lost the ability to make it thanks to an ancient gene mutation. But we still eat it, mostly in beef and lamb, and a little in dairy. Once absorbed, it becomes part of our tissues, and our immune system treats it like a foreign object. It makes antibodies against it, which keeps the immune system mildly switched on.
Scientists call this low-grade, long-running irritation xenosialitis. It has already been linked to heart disease and cancer. The new question was whether it also nudges people toward diabetes.
Who was studied? French researchers used data on 89,953 adults from NutriNet-Santé, a long-running cohort. The average age was 42, and 79% were women. None had diabetes or cancer at the start. Each person filled in three 24-hour food diaries, and the team matched those against a database of Neu5Gc levels in foods.
What happened? Over a median of 7.2 years, 966 people developed type 2 diabetes. Compared with the lowest-intake group:
- Highest fifth of Neu5Gc intake: 63% higher risk
- Fourth fifth: 41% higher risk
- Second and third fifths: no meaningful difference
Look at that pattern again. Risk didn’t creep up gradually. It stayed flat, then jumped. The researchers suggest a possible threshold: the immune system copes until tissue levels pass a certain point, then gets overwhelmed. That is a hypothesis, not a finding.
The hidden realities inside Study 1
1. The scariest number has an asterisk. When the researchers also accounted for total red meat eaten, the link pointed the same way but stopped being statistically significant. The authors’ reading is that Neu5Gc may explain part of red meat’s link to diabetes, not all of it. Heme iron, another red meat component, is on their list to investigate.
2. It’s not just about weight. Body weight explained only about 13% of the association. If that holds up, most of the effect isn’t simply “meat makes you heavier.” The authors point to direct inflammatory effects on insulin-producing cells and insulin-sensitive tissues, though they say this still needs confirming.
3. Dairy got a pass, at least here. Dairy-derived Neu5Gc showed no link to diabetes. Milk contains less of the molecule and is usually eaten in smaller servings. Meanwhile, Neu5Gc from all sources combined showed a smaller 37% higher risk, mostly driven by red meat. The source seems to matter.
4. Nobody found a “safe amount.” The top group ate a median of 97 g of red meat a day (men) and 73 g (women). The bottom group ate close to none. The authors are blunt: these describe what people ate, not a limit to aim for.
5. The absolute picture is calmer than “63%.” About 966 of nearly 90,000 people developed diabetes over roughly seven years, around 1 in 90. A relative risk jump applies to that small base. It is worth taking seriously, but it isn’t a 63% chance of getting diabetes.
And the authors themselves say the results are preliminary, not yet validated, and may change once the analysis is finalized.
Study 2: “Our genes are not our destiny”
The question: Can eating fewer ultra-processed foods and exercising regularly help even people who carry a high genetic risk?
Who was studied? A team from City University of Hong Kong analyzed 56,964 UK Biobank participants, with an average age of 69.5 and 42.3% men. Ultra-processed food (UPF) was measured from 24-hour dietary recalls. Exercise was measured by wrist trackers worn for a week, which is far more reliable than asking people how active they are. Over about 7.8 years, 926 people developed type 2 diabetes.
Key findings:
- Every 10% of the diet (by weight) from ultra-processed foods came with 7% higher risk.
- Meeting the WHO guideline of 150+ minutes of moderate-to-vigorous activity a week was linked to 39% lower risk.
- Among people with high genetic risk, low UPF intake (about 11% of food by weight or less) was linked to a 35% lower risk, and meeting activity guidelines to 33% lower.
- Combining both habits, compared with doing neither: 49% lower risk (high genetic risk), 70% (intermediate), 59% (low).
The hidden realities inside Study 2
1. The biggest payoff wasn’t in the highest-risk group. The combined benefit was largest in the intermediate genetic group (70%), not the high-risk group (49%). Genes may load the dice, but they don’t set a ceiling on what habits can do.
2. Exercise may be doing more of the heavy lifting. Independent experts reviewing the study noted that activity was measured objectively and showed the larger associations. Diet was self-reported, which is messier. Even among people who ate a lot of ultra-processed food (about 33% or more of their diet by weight), meeting activity guidelines was linked to lower risk at every genetic level. That included a 31% lower risk in the high-genetic-risk group.
3. It may be about your drinks more than “processed” as a label. When ultra-processed foods were split into subgroups, only ultra-processed drinks stayed significantly linked to diabetes after full adjustment. Reviewers suggested overall diet quality, including sugary drinks, may matter more than the level of processing itself. So a “processed” sticker doesn’t automatically make a food villainous, and a “natural” one doesn’t clear your beverage habits.
4. The absolute effect is small. One statistician estimated that the 7% relative rise per 10% of diet works out to roughly one extra case per 1,000 people over about eight years. Real, but modest.
5. Nobody actually changed anything. Participants were measured once. The study can’t say whether changing your habits changes your risk, and its genetic analysis covered only people of European ancestry, so it’s unclear how far the findings travel.
The contradiction
The two studies pull in slightly different directions. Study 1 suggests something specific in meat may matter beyond calories and weight. Study 2’s experts suggest that overall diet quality may matter more than any single food category. Both can be partly true. Neither solves it, and that’s the honest outcome.
What this means for real life
Based on what the research supports (and where it’s still just a hypothesis), here are practical, low-drama moves. This is general information, not personal medical advice.
- Rotate your protein rather than banning meat. The French team’s suggested next step is testing whether favoring poultry or plant proteins over beef and lamb lowers risk. That is unproven, but swapping a few meals a week is a low-risk experiment.
- Audit your drinks first. If Study 2’s subgroup finding holds, your glass may matter more than the ingredients label on your snack.
- Treat 150 minutes a week as the floor, not the goal. That’s about 30 minutes, five days a week. It is the habit with the strongest evidence here, and it worked even for people eating plenty of processed food.
- Don’t wait for a genetic test. Family history isn’t a verdict. The intermediate-risk result suggests habits can matter a lot.
- Stack habits. The authors emphasized addressing lifestyle factors together rather than one at a time.
- Ignore “safe amount” claims. The researchers said outright that their meat figures aren’t a threshold.
The bottom line
These studies don’t prove that red meat or processed food causes diabetes, and one of them isn’t even final. What they do offer is a consistent and useful direction: less beef and lamb-heavy eating, fewer ultra-processed drinks and foods, and more movement. For most people, that is neither shocking nor complicated, and it has the advantage of being something you can start doing today.