Nine million people in the UK are living with at least two long-term health conditions right now. That’s roughly one in seven adults juggling combinations like type 2 diabetes and osteoarthritis, or chronic kidney disease and heart disease. For years, doctors have suspected these pairings weren’t random, that something fundamental was causing conditions to cluster together in the same people. Now researchers have quantified exactly what that something is, at least for certain combinations. Obesity.
The GEMINI study represents the largest investigation of its kind, examining 71 conditions that frequently occur together. Using genetics as a tool to unpick cause from mere correlation, the team analyzed data from hundreds of thousands of people across multiple international datasets, looking for shared genetic underpinnings between disease pairs. What emerged was a pattern: for 61 of those 71 conditions, obesity genetics played a role. But more striking still, for 10 specific pairs of conditions, obesity genetics explained the entire genetic overlap between them.
“We’ve long known that certain diseases often occur together, and also that obesity increases the risk of many diseases,” says Professor Jack Bowden at the University of Exeter Medical School, who led the research. “This largescale study is the first to use genetics to quantify the role of obesity in causing diseases to occur in the same individuals.”
Type 2 diabetes and osteoarthritis, for instance. Or gout and osteoarthritis. The genetic connection between these pairs vanishes when you account for body mass index – suggesting obesity is the main driver of why they so frequently appear together in the same patients. The researchers went further, calculating how much weight reduction would actually matter at a population level. For every thousand people who currently have both chronic kidney disease and osteoarthritis, dropping BMI by 4.5 units would have prevented 17 of them from developing both conditions. For type 2 diabetes and osteoarthritis, it’s nine people per thousand.
BMI of 4.5 units is the difference between, say, moderately overweight and normal weight. Not dramatic, but meaningful.
The study used a clever approach, really. Rather than relying on observational data – which is riddled with confounding factors and the chicken-and-egg problem of whether obesity causes disease or disease causes obesity – the team turned to genetics. Genetic variants associated with higher BMI are determined at conception and remain constant throughout life, making them a kind of natural experiment. If higher BMI genetics correlate with both conditions in a pair, and that correlation disappears when you adjust for obesity, you’ve found your culprit.
Professor Jane Masoli, also at the University of Exeter Medical School and a consultant geriatrician, emphasizes the healthcare implications: “Currently nine million people in the UK live with two or more long-term conditions. Understanding how to prevent diseases accumulating is a key national research and healthcare priority.” The research strengthens arguments for public health programs targeting obesity, she notes, and reinforces why lifelong obesity management features so prominently in NHS prevention strategy. “Our work shows that this could reduce the risk of accumulating multiple health conditions, supporting people to live longer, healthier lives.”
Importantly, the team also identified disease pairs where obesity isn’t the main culprit – pairs with strong genetic correlations that persisted even after accounting for BMI. These now become candidates for investigating other shared mechanisms. Perhaps inflammation, or metabolic pathways, or environmental factors we haven’t yet pinpointed.
The findings matter because multimorbidity is becoming increasingly common, particularly as populations age. More than half of people over 65 have several long-term conditions simultaneously. Each additional condition doesn’t just add linearly to healthcare costs and quality of life impacts, it multiplies them through interactions and treatment complications.
What’s striking about this research is how it quantifies the preventable burden. Seventeen people per thousand with chronic kidney disease and osteoarthritis. Nine per thousand with diabetes and osteoarthritis. These aren’t hypothetical numbers – they’re based on what would happen if the study population reduced their BMI by one standard deviation. Scale that up to a population of millions and you’re talking about tens of thousands of people who could potentially avoid developing multiple chronic conditions through weight management.
The GEMINI collaborative – which brought together people living with multimorbidity, healthcare professionals, statisticians and geneticists – is continuing to dig into why certain conditions cluster together. They’re quantifying other modifiable risk factors beyond obesity and searching for novel genes and pathways that might point to new interventions. All their data, results and code are freely available online for other researchers to use, an increasingly common practice in large-scale genetic studies that helps accelerate follow-up work.
For clinicians, this kind of research provides more precise tools for patient conversations. Instead of vague warnings about obesity increasing disease risk, doctors can now point to specific condition combinations where the link is particularly strong – and quantify roughly how much risk reduction comes from how much weight loss. It’s the difference between “lose weight, it’s good for you” and “reducing your BMI by this much could prevent these specific diseases from accumulating.” One feels like generic advice. The other feels like a personalized prevention strategy.
Study link: https://www.nature.com/articles/s43856-025-01347-y
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.