Most recently, Australian researchers has found out that the two calculators which estimated CHD risk provided inappropriate results. Those are Framingham equation and United Kingdom Prospective Diabetes Study risk equation.
The Framingham equation underestimated actual CHD risk by a third while the United Kingdom Prospective Diabetes Study (UKPDS) risk equations overestimated the risk of CHD by up to seven fold.
Why these results suffered such large deviations?
The reason most supposedly is the demographic differences between the UK and US patient cohorts that provided the source data for the calculators and the West Australian patient population.
The Framingham CHD risk equation showed both poor discrimination calibration whereas UKPDS CHD risk equations showed modest discrimination but poor calibration.
Hence, the conclusion is these two estimators must not be used for risk analysis in people with type 2 diabetes.