A heart-risk score for people with type 2 diabetes, built in China and checked in Scotland
A five-year cardiovascular risk model built from routine Nanjing hospital data held up in a Scottish population (C-statistic 0.716 externally).
The question
People with type 2 diabetes face a higher risk of cardiovascular disease, but most risk scores were built on Western populations. Can routine hospital data from China give a useful score, and does it work elsewhere?
What we did
We used records of 84,630 people newly diagnosed with type 2 diabetes at a Nanjing hospital to build basic and extended five-year risk models. We then tested them on a Scottish national diabetes register of 119,891 people.
What we found
The extended model had a C-statistic (how well the model ranks people by risk, where 0.5 is chance and 1 is perfect) of 0.723 in internal validation and 0.716 in the Scottish data. That is a moderate ability to separate higher- from lower-risk people, and it transferred across populations of very different ancestry.
Why it matters
A model built from routinely collected data can still travel across countries, which is the evidence needed before anyone considers using it to guide prevention.