New Method Improves Diabetes Complication Risk Estimation
Risk Equations for Complications Of Type 2 Diabetes (RECODe) predicts the risk for complications in patients with type 2 diabetes more effectively than existing risk prediction equations, according to a recent study.
Existing risk equations for diabetes-related complications have substantially misestimated the risks for these complications.
_______________________________________________________________________________________
RELATED CONTENT
Stomach Pain Is Common After 15 Years of Diabetes Duration
Diabetes Treatment Improves Risks of All Cause, CV Mortality
_______________________________________________________________________________________
To develop new equations, the researchers assessed data on several previous studies. Data on 9635 patients enrolled in the 2001-2009 Action to Control Cardiovascular Risk in Diabetes study (ACCORD) was evaluated to develop and validate existing risk equations. Data on 1018 patients in the 1996-2001 Diabetes Prevention Program Outcomes Study (DPPOS) was used to validate the equations for microvascular events, and data on 4760 patients enrolled in the 2001-2012 Action for Health in Diabetes (Look AHEAD) study was used to validate the equations for cardiovascular events.
The researchers defined microvascular outcomes as nephropathy, retinopathy, and neuropathy, and cardiovascular outcomes as myocardial infarction, stroke, congestive heart failure, and cardiovascular mortality. All-cause mortality was also included as an outcome.
A cross-validating machine learning method was implemented to identify and select predictor variables from demographic characteristics, clinical variables, comorbidities, medications, and biomarkers into Cox proportional hazards models for each outcome. Additionally, model discrimination, calibration, and the net reclassification index were examined in order to compare the new risk equations with older ones.
Results indicated that all equations evaluated in the study had moderate internal and external discrimination, as well as high internal and external calibration. However, the new equations had demonstrated better discrimination and calibration vs the UK Prospective Diabetes Study Outcomes Model 2 in terms of microvascular and cardiovascular outcomes, and the American College of Cardiology/American Heart Association Pooled Cohort Equations in terms of the risks for fatal or non-fatal myocardial infarction or stroke.
“RECODe might improve estimation of risk of complications for patients with type 2 diabetes,” the researchers concluded.
—Christina Vogt
Reference:
Basu S, Sussman JB, Berkowitz SA, Hayward RA, Yudkin JS. Development and validation of Risk Equations for Complications Of type 2 Diabetes (RECODe) using individual participant data from randomised trials. The Lancet Diabetes Endocrinol. 2017;5(10):788-798. http://dx.doi.org/10.1016/S2213-8587(17)30221-8.