PracticeUpdate Diabetes June 2019

CONFERENCE COVERAGE 20

Stage of Cardiometabolic Disease Proves Predictive of Incident Diabetes The risk prediction tool is specific to race and gender. T he weighted score on the cardio- metabolic disease scale confers high model discrimination using availa- ble clinical information, and can be used to quantify race- and sex-specific risk of type 2 diabetes.

Altered levels of these laboratory values may indicate metabolic syndrome. The val- ues are altered typically in both black and white men and women. In the REGARDS cohort, 1614 incident cases of diabetes were observed. C-statistics ranged from 0.72 (95% confidence inter- val 0.72–0.75) for black men to 0.79 (95% CI 0.79–0.81) for white women. Cardiomet- abolic risks differed more by race than sex (P = .0001). The screening method is the first risk pre- diction tool for black individuals derived from a large scientific US sample. Cardiometabolic disease scoring is unique in that it incorporates the observation that incident type 2 diabetes is not a linear function of age. High area under the curve values highlight metabolic syndrome and insulin resistance as its central pathological mechanism in genesis of type 2 diabetes. Dr. Wilkinson explained that the prevention of type 2 diabetes has become impera- tive to stem the rising rates of this disease, particularly among black Americans. The at-risk pool is large, however, and a clini- cally meaningful metric for risk stratification to guide intervention remains a challenge. Dr. Wilkinson concluded that weighted cardiometabolic disease stage scoring confers high model discrimination using available clinical information, and can be used to quantify race- and sex-specific type 2 diabetes risk. “The results highlight possible differences in contribution to cardiometabolic disease risk between black and white men and women. They also highlight how crucial simple, easy-to-use risk prediction tools for chronic disease are. We use objective data that are easy to collect, using black and white American adults,” she noted. “Through this analysis,” she added, “we hope to refine this tool further so clini- cians will be able to target those at highest risk for developing diabetes simply and accurately. “These tools can be expanded,” she asserted, “beyond black and white Amer- ican adults, and should be expanded. Additionally, the binary risk prediction model used is insufficient. We are develop- ing a prediction calculator based on the full range of metabolic markers, as opposed to above and below cut-off thresholds as we used in this study.” www.practiceupdate.com/c/82854

Results of this analysis of the popula- tion-based REasons for Geographic and Racial Differences in Stroke (REGARDS) study were reported at AACE 2019. Lua Wilkinson, PhD, RD, of the University of Alabama, Birmingham, and colleagues set out to predict diabetes risk from nationally sampled data of white and black American adults age ≥45 years. “Our current models,” Dr. Wilkinson told Elsevier’s PracticeUpdate , “of predicting whether someone will develop diabetes are not ideal. Most of the data we use leans heavily on data collected on Caucasians. This data includes measures such as family history of disease or physical activity levels, which may not be known or accurate.” The 12,123 black and white popula- tion-based cohort taken from REGARDS (2003–2007), was observed through 2016. A sex- and race-stratified cardiometabolic score using data assessed regularly taken during primary care visits was created to predict who is likely to develop diabetes up to 10 years after the visit. The tool uses blood glucose, blood pres- sure, high-density lipoprotein cholesterol, waist circumference, triglycerides, and age.

" The results highlight possible differences in contribution to cardiometabolic disease risk between black and white men and women. "

PRACTICEUPDATE DIABETES