Objectives: In type 2 diabetes (T2D), the most common causes of death are cardiovascular (CV) related, accounting for >50% of deaths in some reports. As novel diabetes therapies reduce CV death risk, identifying patients with T2D at highest CV death risk allows for cost-effective prioritization of these therapies. Accordingly, the primary goal of this study was to quantify the risk continuum for CV death in a real-world T2D population as a means to identify patients with the greatest expected benefit from cardioprotective antidiabetes therapies.
Methods: This retrospective study included patients with T2D receiving services through an integrated health-care system and used data generated through electronic medical records (EMRs). Quantifying the risk continuum entailed developing a prediction model for CV death, creating an integer risk score based on the final prediction model and estimating future CV death risk according to risk score ranking.
Results: Among 59,180 patients with T2D followed for an average of 7.5 years, 15,691 deaths occurred, 6,033 (38%) of which were CV related. The EMR-based prediction model included age, established CV disease and risk factors and glycemic indices (c statistic = 0.819). The 10% highest-risk patients according to prediction model elements had an annual CV death risk of ∼5%; the 25% highest-risk patients had an annual risk of ∼2%.
Conclusions: This study incorporated a prediction modelling approach to quantify the risk continuum for CV death in T2D. Prospective application allows us to rank individuals with T2D according to their CV death risk, and may guide prioritization of novel diabetes therapies with cardioprotective properties.
Keywords: cardiovascular death; diabetes; diabète; décès d’origine cardiovasculaire; modèle prédictif; prediction model.
Copyright © 2021 Canadian Diabetes Association. Published by Elsevier Inc. All rights reserved.