Diabetes Care Improved With New Risk Score

A simple risk model that uses body mass index, triglycerides, and glycated hemoglobin effectively predicted 5-year glycemic responses of patients with newly diagnosed type 2 diabetes. This model “can be used in practice as a quick, easy and accurate tool to determine patients' care needs and provide tailored diabetes treatment,” according to the researchers.

Using data from 2 large Dutch diabetes registries, the researchers identified all adult patients with newly diagnosed type 2 diabetes between January 2006 and December 2014. They stratified patients into development (n=10,528) and validation (n=3777) cohorts and used latent growth mixture modelling to determine distinct 5-year glycemic trajectories. In addition, the researchers used machine learning models to predict these trajectories using patient characteristics that are easily obtained during routine visits.
________________________________________________________________ 
RELATED CONTENT
Diabetes Risk Prediction Models: Which One Is Most Effective?
Could an Oral Glucose Tolerance Test Predict Prediabetes Risk?
________________________________________________________________

Within the development cohort, the model determined 3 trajectories of glycemic control and identified 76.5% of patients with stable or adequate glycemic control, 21.3% with improved glycemic control, and 2.2% with deteriorated glycemic control. These trajectories were discernible among patients in the validation cohort. Additionally, this model had a receiver-operating characteristics area under the curve of 0.96 in the validation cohort after it was trained on the development cohort, which indicated excellent accuracy.

Furthermore, the researchers found that the most important predictors of glycemic trajectories were triglyceride levels, glycated hemoglobin, and body mass index.

“The developed model can effectively explain heterogeneity in future glycemic response of patients with type 2 diabetes,” the researchers concluded. “It can therefore be used in clinical practice as a quick and easy tool to provide tailored diabetes care.”

—Melissa Weiss

Reference:

Hertroijs DFL, Elissen AMJ, Brouwers Martijn CGJ, et al. A risk score including body mass index, glycated haemoglobin and triglycerides predicts future glycaemic control in people with type 2 diabetes [published online November 24, 2017]. Diabetes Obes Metab. https://doi.org/10.1111/dom.1314.