Premium
Building clinical risk score systems for predicting the all‐cause and expanded cardiovascular‐specific mortality of patients with type 2 diabetes
Author(s) -
Liu ChiuShong,
Li ChiaIng,
Wang MuCyun,
Yang SingYu,
Li TsaiChung,
Lin ChengChieh
Publication year - 2021
Publication title -
diabetes, obesity and metabolism
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.445
H-Index - 128
eISSN - 1463-1326
pISSN - 1462-8902
DOI - 10.1111/dom.14240
Subject(s) - medicine , hazard ratio , proportional hazards model , receiver operating characteristic , diabetes mellitus , type 2 diabetes , blood pressure , cohort , retrospective cohort study , framingham risk score , confidence interval , disease , endocrinology
Aim To develop and validate risk score systems by examining the effects of glycaemic and blood pressure variabilities on the all‐cause and expanded cardiovascular‐specific mortality of people with type 2 diabetes. Materials and methods This retrospective cohort study consisted of 9692 patients aged 30‐85 years, diagnosed with type 2 diabetes and enrolled in a managed care programme of a medical centre from 2002 to 2016. All the patients were randomly allocated into two groups, namely, training and validation sets (2:1 ratio), and followed up until death or August 2019. Cox's proportional hazard regression was performed to develop all‐cause and expanded cardiovascular‐specific mortality prediction models. The performance of the prediction model was assessed by using the area under the receiver operating characteristic curve (AUROC). Results Overall, 2036 deaths were identified after a mean of 8.6 years of follow‐up. The AUROC‐measured prediction accuracies of 3‐, 5‐, 10‐ and 15‐year all‐cause mortalities based on a model containing the identified traditional risk factors, biomarkers and variabilities in fasting plasma glucose, HbA1c and blood pressure in the validation set were 0.79 (0.76‐0.83), 0.78 (0.76‐0.81), 0.80 (0.78‐0.82) and 0.80 (0.78‐0.82), respectively. The corresponding values of the expanded cardiovascular‐specific mortalities were 0.85 (0.80‐0.90), 0.83 (0.79‐0.86), 0.80 (0.77‐0.83) and 0.79 (0.77‐0.82), respectively. Conclusions Our prediction models considering glycaemic and blood pressure variabilities had good prediction accuracy for the expanded cardiovascular‐specific and all‐cause mortalities of patients with type 2 diabetes.