UK researchers develop tool to identify people at highest risk of obesity-related diseases

UK researchers have developed Obscore, a machine learning tool that predicts 10-year risk for 18 obesity-related complications using 20 health and lifestyle features from nearly 200,000 UK Biobank participants with BMI of 27 or greater. The tool, published in Nature Medicine by Prof Nick Wareham and colleagues, could help the NHS prioritize limited weight-loss medications more effectively than current criteria based solely on high BMI and existing health problems. The research reveals that individuals with identical age, sex, and BMI can have vastly different disease risks, with some overweight people facing higher type 2 diabetes risk than obese individuals. However, Prof Naveed Sattar notes that several Obscore metrics are not routinely available in the NHS and robust risk scores already exist for some conditions, raising questions about practical implementation amid England's obesity crisis affecting two-thirds of adults.

Development and functionality of the Obscore risk prediction tool

• ▪Obscore was developed by applying interpretable machine learning to data from almost 200,000 UK Biobank participants with a BMI of 27 or greater
• ▪Obscore places participants in one of five equal-sized risk categories from low to high risk for each obesity-related condition
• ▪Obscore uses 20 health, lifestyle and demographic features including age, sex, total cholesterol, and creatinine levels to predict obesity-related complications
• ▪Prof Nick Wareham of the University of Cambridge was a co-author of the Obscore study
• ▪UK researchers developed a tool called Obscore to identify people at highest risk of obesity-related diseases
• ▪Obscore predicts the 10-year risk of 18 different obesity-related complications, from gout to stroke

Potential applications for prioritizing weight-loss interventions

• ▪Access to weight-loss medications on the NHS is currently limited and based on having a high BMI and particular obesity-related health problems
• ▪For type 2 diabetes, the highest risk category identified by Obscore included a considerable proportion of people who are overweight rather than obese
• ▪Participants with the same age, sex and BMI can have very different risks for various obesity-related conditions according to the Obscore study
• ▪About two-thirds of adults in England are overweight or obese according to recent data
• ▪Obscore could help prioritize who should receive weight-loss interventions such as weight-loss jabs on the NHS
• ▪Kamil Demircan from Queen Mary University of London was a co-author of the Obscore study

Validation results and limitations of the approach

• ▪Obscore was applied to data from participants in a randomised control trial for the weight-loss drug tirzepatide
• ▪Obscore was validated using UK Biobank data and datasets from two independent health studies
• ▪Naveed Sattar is a professor of cardiometabolic medicine at the University of Glasgow
• ▪Several of the metrics used in the Obscore study are not routinely available within the NHS
• ▪The Obscore study was published in the journal Nature Medicine
• ▪Robust and more easily implemented risk scores already exist for some obesity-related conditions according to Naveed Sattar

Perspective of Obscore developers (Prof Nick Wareham and Kamil Demircan)

• ▪Obscore reveals that people with identical age, sex, and BMI can have vastly different risks for obesity-related conditions

Perspective of NHS and healthcare policymakers

• ▪Two-thirds of adults in England are overweight or obese, creating significant demand for weight-loss interventions