The increased prevalence of type 2 diabetes mellitus (T2DM) is a major public health issue, affecting 400 million individuals worldwide. Currently, management of T2DM comprises anti-diabetic drugs and lifestyle modifications, such as diet and exercise. Being able to predict the risk of this disease would help to delay onset of T2DM. 

Satoh and colleagues developed and validated a predictive model of the onset of T2DM using health check-up data from the general population. They analysed data from 463,248 adults who had regular health checks, developing a prediction model from 308,832 of these and validating the model with the remainder. The model assigns scores based on a person’s health and lifestyle characteristics to estimate their diabetes risk. Over a follow-up period of around five years, roughly 17% of people in both groups developed T2DM. When tested, the model showed a moderate ability to predict who would develop the disease. 

The model assigns scores based on age, sex, body mass index, blood pressure, lipid profiles, liver enzymes, kidney function and lifestyle habits to estimate risk of diabetes. Because it uses information that is already collected in regular health exams, this tool could help identify people at higher risk and support earlier prevention efforts.

Read the full article in Scientific Reports https://doi.org/10.1038/s41598-025-21831-8