Updating and Improving Cardiovascular Risk Prediction in Type 2 Diabetes

heart, cardiovascular
heart, cardiovascular
Investigators hypothesized that cardiovascular risk prediction equations derived before widespread screening would now significantly overestimate risk in screen-detected patients.

Cardiovascular risk prediction equations for people with type 2 diabetes derived from populations without widespread screening should not be applied to populations where screening is increasingly common because they will significantly overestimate risk in many screen-detected patients, according to findings of a study published in The Lancet.

The ongoing PREDICT open cohort study in New Zealand automatically recruited 63,362 participants via health care practitioners completing assessments using the PREDICT decision support software from October 2004 to December 2016.

The PREDICT-1° Diabetes subcohort had 46,652 participants aged 30 to 74 years with type 2 diabetes and no known cardiovascular disease, heart failure, or significant renal impairment.

The study authors prespecified the primary outcome using the total cardiovascular disease outcome in the Framingham equations. They found that “each additional year of age was associated with an increased estimated 5-year cardiovascular disease risk of approximately 5%, in relative terms, and every year since diabetes diagnosis was independently associated with a further 2% increased risk.”

By comparing the PREDICT-1° Diabetes equations with the original New Zealand Diabetes Cohort Study equations, the study authors observed that the PREDICT-1° Diabetes equations demonstrated excellent calibration across all risk groups in both sexes, and the original equations overpredicted observed 5-year risk in all groups in both sexes. The medians and interquartile ranges (IQR) of the original equations were higher than the new PREDICT-1° Diabetes equations in the following regard: 3 times higher in women: original (median, 14.2%; IQR, 9.7-20.0) vs new (median, 4.0%; IQR, 2.3-6.8), and 2 times higher in men: original (median, 17.1%; IQR, 4.5-20.0) vs new (median, 7.1%; IQR, 4.5-11.2).

There were 4114 (8.8%) first cardiovascular disease events during follow-up (median, 5.2 years; IQR, 3.3-7.4): 1213 (29.5%) nonfatal myocardial infarctions; 963 (23.4%) strokes or transient ischemic attacks; 666 (16.2%) cases of congestive heart failure; and 362 (8.8%) cases of peripheral vascular disease. It was noted that 369 (9.0%) of these events were fatal.

“Our findings strongly suggest that most cardiovascular risk assessment equations derived in people with diabetes internationally will need to be validated and updated in contemporary diabetes populations to better inform both non-pharmacological and pharmacological management decisions,” the study authors concluded.


Pylypchuk R, Wells S, Kerr A, et al. Cardiovascular risk prediction in type 2 diabetes before and after widespread screening: a derivation and validation study. Lancet. Published online June 2, 2021. doi:10.1016/S0140-6736(21)00572-9