Obesity is widely recognized as a leading contributor to diabetes, heart disease, and premature death, although the level of risk varies significantly among individuals. An innovative clinical risk prediction algorithm has been created, categorizing obesity into five distinct diagnostic profiles, each associated with different health risks and treatment needs.
Obesity is a chronic, complex, and recurring condition characterized by abnormal and/or excessive fat accumulation that poses health risks. The World Health Organization (WHO) estimates that complications from obesity claim the lives of at least 2.8 million individuals each year. Since 1990, the prevalence of obesity in adults has more than doubled, with the WHO estimating that in 2022, 890 million adults were classified as obese. “That’s about 1 in 8 individuals globally,” comments Dr. Abd Tahrani, a member of the IMI SOPHIA patient advisory board.
“As healthcare professionals, we encounter a significant challenge in identifying which individuals with obesity are at the highest risk of developing complications and should be prioritized for intervention. Precision medicine, which is making strides in enhancing the prediction, prevention, diagnosis, and treatment of various health issues, can help tackle these challenges,” states Dr. Carel le Roux, a Professor of Metabolic Medicine at University College Dublin.
A study conducted by the IMI SOPHIA consortium published in the journal Nature Medicine introduces a new precision prediction algorithm that identifies previously unrecognized subtypes of obesity associated with an increased likelihood of developing type 2 diabetes and heart disease.
“On a broad scale, increased weight is typically detrimental to health. However, a closer examination reveals more complicated patterns at the individual level, which can be utilized to enhance disease prediction,” adds Dr. Ewan Pearson, a Professor of Diabetes Medicine at Dundee University.
“For instance, an individual’s fat or sugar levels in the blood can significantly differ from what one might anticipate based solely on their body weight, affecting their risk for complications from obesity,” explains Dr. Daniel Coral, the study’s lead author from Lund University Diabetes Centre in Sweden. “Standard clinical prediction tools often overlook this nuance, leading to about 20% of individuals who may require early intervention being missed. The algorithm we have developed could assist healthcare providers and patients in the future,” Coral elaborates.
The researchers analyzed data from 170,000 adults across the UK, the Netherlands, and Germany, compiling detailed clinical information. By employing an artificial intelligence technique known as “machine learning,” the team formulated robust algorithms that categorize obesity into five different diagnostic profiles, each showing varied risks for complications related to obesity.
“Obesity is both widespread and diverse, meaning that the health risks faced by one individual with obesity can considerably differ from those of another. Identifying who has the greatest health risks is crucial, as this can facilitate more targeted, precise, and timely preventative measures and treatments,” says Dr. Paul Franks, a Professor of Genetic Epidemiology at Lund University Diabetes Centre and the paper’s senior author.
This research was spearheaded by scientists from Lund University Diabetes Centre in Sweden, along with collaborators from the Maastricht Centre for Systems Biology and Erasmus MC University Medical Centre in the Netherlands, as part of the IMI SOPHIA consortium.
Key Facts
- Approximately 80% of individuals had health indicators that align with what is expected based on their body weight.
- About 8% of women exhibited higher-than-anticipated blood pressure for their weight, along with increased “good” cholesterol (HDL) and a lower waist-to-hip ratio (WHR), indicating more fat around their hips and less around their waist than expected. This pattern was not observed in men.
- Around 5% of women and 7% of men presented a profile marked by high “bad” cholesterol (LDL), elevated triglycerides (fat in the blood), higher WHR (indicating more waist fat), and increased blood pressure compared to what was expected for their weight.
- About 5% had elevated liver enzymes (ALT) and higher WHR corresponding to their weight.
- About 4% showed an increased level of inflammation (measured by CRP) than anticipated for their weight.
- Approximately 2.5% had high blood sugar levels coupled with lower LDL for their weight.
