OBSCORE: Identifying people at high risk of obesity-related diseases

Obesity is a major global health challenge. According to the World Health Organization, in 2022, one in eight people in the world were living with obesity.

A chronic, complex disease that is influenced by genets, biological characteristics, the environment, comorbidities, behaviours and more, obesity is associated with over 250 other health conditions.

At Queen Mary’s Precision Healthcare Research University Research Institute (PHURI), researchers are exploring how we can use data and computational methods to identify those with obesity or overweight who are most at risk of developing other health conditions.

Twenty measures for personalised care

Body Mass Index (BMI) is widely used as a tool to identify those with overweight or obesity who are most at risk of developing obesity-related complications. However, this score has many limitations if used alone, and so there is a need for a clear, universal framework that incorporates or works alongside BMI.

A team at Queen Mary’s PHURI, led by Professor Claudia Langenberg, carried out a project to change this. They sought to create a tool that could accurately predict an individual’s risk developing obesity-related diseases that is simple to use in clinical settings

Working in collaboration with the Berlin Institute of Health at Charité, the team analysed health data from 200,000 participants with overweight or obesity, for whom data is held in the UK Biobank. They then used machine-learning to evaluate more than 2,000 measures of health data – measures such as blood test results, molecular data, lifestyle factors – to determine which were most effective at predicting disease or complications.

From this analysis, the team identified 20 health measures that most effectively predict an individual’s future risk of developing one or more of 18 obesity-related diseases such as cancer or heart problems. Interestingly, they found substantial differences in risk profiles among individuals in the same BMI category, meaning two people of similar weight can have very different risks of developing other conditions. They also found that those identified as being at the highest risk of developing obesity-related conditions were not always those with the highest BMI, further demonstrating that BMI alone isn’t an effective way to determine risk. 

Their findings, recently published in Nature Medicine, led to the development of the OBSCORE model – a tool which is simple to use in clinical settings, and which was validated in the independent Genes & Health and the European Investigation into Cancer (EPIC) – Norfolk studies.

An open tool for all  

The team used their findings to create an open access risk prediction tool that can be accessed by healthcare professionals to assess for risk of developing obesity-related complications. Professionals can input the 20 health data measures for their patient(s) to generate a personalised risk profile for the 18 obesity-related conditions listed in the Nature Medicine paper.  

Using the OBSCORE model through this tool, in parallel with BMI scores, can help healthcare professionals better choose the most effective treatment for their patient(s) (eg surgery, pharmacological intervention, dietary changes) and to prioritise treatments to those who need them most.

In addition, following further validation and evaluation of cost-effectiveness in appropriate clinical trials, OBSCORE could help identify which people living with overweight or obesity may benefit most from early intervention, closer monitoring, or intensified treatment, which could not only help the NHS but also save lives.

Access the OBSCORE tool here.