UK researchers have developed a new tool that can identify individuals most at risk of obesity-related diseases, potentially helping the National Health Service (NHS) prioritise access to limited weight-loss medications. The tool, named Obscore, uses interpretable machine learning to provide a personalised risk assessment based on 20 health, lifestyle, and demographic features.
Addressing the Obesity Crisis
Recent data indicates that approximately two-thirds of adults in England are overweight or obese, a situation that has raised significant concern among health experts. The new tool aims to move beyond the current reliance on body mass index (BMI) and specific obesity-related health problems to determine who should receive interventions such as weight-loss injections.
Professor Nick Wareham from the University of Cambridge, a co-author of the study, emphasised that the tool is not about expanding the use of particular therapies. 'It is about developing and validating a score that can help with more rational resource allocation. So, can we prescribe therapy to those people who are most likely to need it and most likely to benefit from it – which is what we should do within the NHS,' he explained.
How the Tool Works
The research, published in the journal Nature Medicine, involved applying interpretable machine learning to data from nearly 200,000 participants of the UK Biobank project, all of whom had a BMI of 27 or higher, categorising them as overweight or obese. The analysis identified 20 key features—including age, sex, total cholesterol, and creatinine levels—that could predict the 10-year risk of 18 different obesity-related complications, ranging from gout to stroke.
For each condition, participants were placed into one of five equal-sized risk categories, from low to high. The team then calculated the proportion of individuals in each category who developed the condition over a decade. The validity of Obscore was tested using UK Biobank data and two independent health studies.
Key Findings
The researchers found that individuals with the same age, sex, and BMI could have vastly different risks for various obesity-related conditions. This supports the idea that the tool could help inform strategies for prioritising weight-loss interventions. Notably, for some conditions like type 2 diabetes, those in the highest risk category included a considerable proportion of people who are overweight rather than obese.
'These constitute a population of individuals who may be overlooked if we only look at BMI and not other risk factors,' said Kamil Demircan, a co-author from Queen Mary University of London.
Real-World Application
The team also applied a version of the tool to data from a randomised control trial for the weight-loss drug tirzepatide, confirming that individuals predicted to be at highest risk for obesity-related conditions experienced similar weight loss to others.
However, Professor Naveed Sattar from the University of Glasgow, who was not involved in the study, noted that many obesity-related conditions are closely interrelated and that robust risk scores already exist for some. He also pointed out that several metrics used in the study are not routinely available within the NHS.
'Overall, this work represents a thoughtful attempt to move towards more holistic risk prediction across multiple obesity-related conditions,' Sattar said. 'But substantial further development and validation will be required before such an approach can be translated into routine clinical practice.'
