|
KMID : 0667720070440000498
|
|
Report Natlonal Institute of Health
2007 Volume.44 No. 0 p.498 ~ p.499
|
|
|
A prediction model for type 2 diabetes high-risk group using both the epidemiological and genetic factors
|
|
Cho Nam-Han
Go Min-Jin
|
|
|
Abstract
|
|
|
Purpose : In this study we developed a risk model for type 2 diabetes using both epidemiological and genetic risk factors in an effort to prevent diabetes and reduce diabetes-related morbidity and mortality.
Methods : A total of 472 cases and 456 age-matching controls collected from the prospective epidemiological cohorts in Ansung and Ansan were examined. Area, hypertension, waist circumference, total cholesterol level, fasting insulin and ALT were selected by chi-square test as the epidemiological risk factors. The 381 single nucleotide polymorphisms (SNPs) from 84 candidate genes were analyzed by chi-square validating SNPs of TCF1 (rs1169288), ACE (rs4362), CD36 (rs3211908), IL4 (rs2243250), LPL (rs343), VAMP3 (novel, +9326A>G), CAP1 (novel, +12922C>A),
ICAM1 (rs5498), LDLR (rs3786730), NOS2A (rs2297518), SELE (rs4786), UCP3 (novel, +834T>C) as the genetic risk factors for diabetes. Multiple logistic regression model was constructed using selected risk factors to predict the onset of diabetes.
Results : Risk factors used to derive a multiple logistic regression model involve area, hypertension, waist circumference, total cholesterol level, fasting insulin, ALT, IL4 (rs rs2243250), LPL (rs343), ICAM1 (rs5498) and SELE (rs4786). The analyses based on ROC curve of the model demonstrated that Nagelkerke R-square and the area of ROC curve were improved from 0.388 to 0.420 and from 0.820 to 0.834, respectively in the model using both the genetic and epidemiological risk factors compared to that using only the epidemiological factors.
Conclusions : Our result strongly suggests that the addition of genetic risk factors for the type 2 diabetes improves the accuracy of the predictive model, which facilitate the possible accomplishment of the personalized prevention.
|
|
|
KEYWORD
|
|
|
Type 2 diabetes, Epidemiological risk factor, Genetic risk factor, Single nucleotide polymorphism, Multiple logistic regression model, ROC curve
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|
|