Identification of Type 2 Diabetes Risk Factors Using Phenotypes Consisting of Anthropometry and Triglycerides based on Machine Learning

The hypertriglyceridemic waist (HW) phenotype is strongly associated with type 2 diabetes; however, to date, no study has assessed the predictive power of phenotypes based on individual anthropometric measurements and triglyceride (TG) levels. The aims of the present study were to assess the association between the HW phenotype and type 2 diabetes in Korean adults and to evaluate the predictive power of various phenotypes consisting of combinations of individual anthropometric measurements and TG levels. Between November 2006 and August 2013, 11 937 subjects participated in this retrospective cross-sectional study. We measured fasting plasma glucose and TG levels and performed anthropometric measurements. We employed binary logistic regression (LR) to examine statistically significant differences between normal subjects and those with type 2 diabetes using HW and individual anthropometric measurements. For more reliable prediction results, two machine learning algorithms, naive Bayes (NB) and LR, were used to evaluate the predictive power of various phenotypes. All prediction experiments were performed using a tenfold cross validation method. Among all of the variables, the presence of HW was most strongly associated with type 2 diabetes (p <; 0.001, adjusted odds ratio (OR) = 2.07 [95% CI, 1.72-2.49] in men; p <; 0.001, adjusted OR = 2.09 [1.79-2.45] in women). When comparing waist circumference (WC) and TG levels as components of the HW phenotype, the association between WC and type 2 diabetes was greater than the association between TG and type 2 diabetes. The phenotypes tended to have higher predictive power in women than in men. Among the phenotypes, the best predictors of type 2 diabetes were waist-to-hip ratio + TG in men (AUC by NB = 0.653, AUC by LR = 0.661) and rib-to-hip ratio + TG in women (AUC by NB = 0.73, AUC by LR = 0.735). Although the presence of HW demonstrated the strongest association with type 2 diabetes, the predictive power of the combined measurements of the actual WC and TG values may not be the best manner of predicting type 2 diabetes. Our findings may provide clinical information concerning the development of clinical decision support systems for the initial screening of type 2 diabetes.

[1]  S. Lear,et al.  The relationship between waist circumference and metabolic risk factors: cohorts of European and Chinese descent. , 2002, Metabolism: clinical and experimental.

[2]  Isabelle Lemieux,et al.  Hypertriglyceridemia and Waist Circumference Predict Cardiovascular Risk among HIV Patients: A Cross-Sectional Study , 2011, PloS one.

[3]  C. Craig,et al.  Adiposity, physical fitness and incident diabetes: the physical activity longitudinal study , 2007, Diabetologia.

[4]  P. Zimmet,et al.  Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional report of a WHO Consultation , 1998, Diabetic medicine : a journal of the British Diabetic Association.

[5]  R. Ness-Abramof,et al.  Hypertriglyceridemic waist: a simple clinical phenotype associated with coronary artery disease in women , 2012 .

[6]  S. Yusuf,et al.  Anthropometric measures and glucose levels in a large multi-ethnic cohort of individuals at risk of developing type 2 diabetes , 2010, Diabetologia.

[7]  Luigi Bouchard,et al.  The “hypertriglyceridemic waist” phenotype and glucose intolerance in pregnancy , 2010, Canadian Medical Association Journal.

[8]  Gary M. Weiss.  Mining with rarity: a unifying framework , 2004, SKDD.

[9]  Claus Christiansen,et al.  Enlarged Waist Combined With Elevated Triglycerides Is a Strong Predictor of Accelerated Atherogenesis and Related Cardiovascular Mortality in Postmenopausal Women , 2005, Circulation.

[10]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[11]  Michael I. Jordan,et al.  On Discriminative vs. Generative Classifiers: A comparison of logistic regression and naive Bayes , 2001, NIPS.

[12]  Alfonso T. Perez,et al.  Hypertriglyceridemic Waist Phenotype Predicts Increased Visceral Fat in Subjects With Type 2 Diabetes , 2009, Diabetes Care.

[13]  Kang Li,et al.  BMI is Strongly Associated With Hypertension, and Waist Circumference is Strongly Associated With Type 2 Diabetes and Dyslipidemia, in Northern Chinese Adults , 2012, Journal of epidemiology.

[14]  Ahmad Esmaillzadeh,et al.  The association of hypertriglyceridemic waist phenotype with type 2 diabetes mellitus among individuals with first relative history of diabetes , 2011, Journal of research in medical sciences : the official journal of Isfahan University of Medical Sciences.

[15]  Ahmad Esmaillzadeh,et al.  Prevalence of the hypertriglyceridemic waist phenotype in Iranian adolescents. , 2005, American journal of preventive medicine.

[16]  D. Gaudet,et al.  Hypertriglyceridemic waist: A marker of the atherogenic metabolic triad (hyperinsulinemia; hyperapolipoprotein B; small, dense LDL) in men? , 2000, Circulation.

[17]  Harry Zhang,et al.  The Optimality of Naive Bayes , 2004, FLAIRS.

[18]  John Halloran.  Classification: Naive Bayes vs Logistic Regression , 2009 .

[19]  A Berber,et al.  Anthropometric indexes in the prediction of type 2 diabetes mellitus, hypertension and dyslipidaemia in a Mexican population , 2001, International Journal of Obesity.

[20]  Jie Cao,et al.  Is an appropriate cutoff of hypertriglyceridemic waist designated for type 2 diabetes among Chinese adults? , 2010, Clinical nutrition.

[21]  J. Woo,et al.  Prediction of hypertension, diabetes, dyslipidaemia or albuminuria using simple anthropometric indexes in Hong Kong Chinese , 1999, International Journal of Obesity.

[22]  A Lev-Ran,et al.  Human obesity: an evolutionary approach to understanding our bulging waistline , 2001, Diabetes/metabolism research and reviews.

[23]  D. Mikhailidis,et al.  The hypertriglyceridemic waist phenotype is a predictor of elevated levels of small, dense LDL cholesterol , 2006, Lipids.

[24]  Xu Lin,et al.  Hypertriglyceridemic waist, cytokines and hyperglycaemia in Chinese , 2012, European journal of clinical investigation.

[25]  J. Després,et al.  The hypertriglyceridemic waist phenotype versus the National Cholesterol Education Program-Adult Treatment Panel III and International Diabetes Federation clinical criteria to identify high-risk men with an altered cardiometabolic risk profile. , 2009, Metabolism: clinical and experimental.

[26]  J. Christman,et al.  Hypertension and type 2 diabetes comorbidity in adults in the United States: risk of overall and regional adiposity. , 2001, Obesity research.

[27]  Joohon Sung,et al.  Which Obesity Indicators Are Better Predictors of Metabolic Risk?: Healthy Twin Study , 2008, Obesity.

[28]  T. K. Young,et al.  Hypertriglyceridemic-waist phenotype and glucose intolerance among Canadian Inuit: the International Polar Year Inuit Health Survey for Adults 2007–2008 , 2011, Canadian Medical Association Journal.

[29]  B. Balkau,et al.  Prescreening tools for diabetes and obesity-associated dyslipidaemia: comparing BMI, waist and waist hip ratio. The D.E.S.I.R. Study , 2006, European Journal of Clinical Nutrition.

[30]  S. Lear,et al.  The Use of BMI and Waist Circumference as Surrogates of Body Fat Differs by Ethnicity , 2007, Obesity.

[31]  P Zimmet,et al.  Ethnic comparisons of the cross‐sectional relationships between measures of body size with diabetes and hypertension , 2008, Obesity reviews : an official journal of the International Association for the Study of Obesity.

[32]  Ambady Ramachandran,et al.  Cutoff values for normal anthropometric variables in asian Indian adults. , 2003, Diabetes care.

[33]  S. Allahverdian,et al.  Cardiovascular risk factors in males with hypertriglycemic waist (Tehran Lipid and Glucose Study) , 2004, International Journal of Obesity.

[34]  N. J. Wareham,et al.  The hypertriglyceridemic-waist phenotype and the risk of coronary artery disease: results from the EPIC-Norfolk Prospective Population Study , 2010, Canadian Medical Association Journal.

[35]  James J. Chen,et al.  Class-imbalanced classifiers for high-dimensional data , 2013, Briefings Bioinform..

[36]  The American Association of Clinical Endocrinologists Medical Guidelines for the Management of Diabetes Mellitus: the AACE system of intensive diabetes self-management--2000 update. , 2000, Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists.

[37]  Félix M. Puchulu,et al.  Definition, Diagnosis and Classification of Diabetes Mellitus , 2018 .

[38]  J. Shaw,et al.  Independent and opposite associations of waist and hip circumferences with diabetes, hypertension and dyslipidemia: the AusDiab Study , 2004, International Journal of Obesity.

[39]  Xuefeng Yu,et al.  Visceral adiposity index, hypertriglyceridemic waist and risk of diabetes: the China Health and Nutrition Survey 2009 , 2014, International Journal of Obesity.

[40]  Ki-Hong Chun,et al.  Sex Difference of Type 2 Diabetes Affected by Abdominal Obesity versus Overall Obesity , 2010, Yonsei medical journal.

[41]  Denis Prud'homme,et al.  Postprandial hyperlipidemia: another correlate of the "hypertriglyceridemic waist" phenotype in men. , 2003, Atherosclerosis.

[42]  Paul Poirier,et al.  Hypertriglyceridemic waist: a useful screening phenotype in preventive cardiology? , 2007, The Canadian journal of cardiology.

[43]  J. Y. Kim,et al.  A Comparison of the Predictive Power of Anthropometric Indices for Hypertension and Hypotension Risk , 2014, PloS one.

[44]  R. Pandey,et al.  Waist circumference cutoff points and action levels for Asian Indians for identification of abdominal obesity , 2006, International Journal of Obesity.

[45]  S. Feld,et al.  The American Association of Clinical Endocrinologists Medical Guidelines for the Management of Diabetes Mellitus: The AACE System of Intensive Diabetes Self-Management - 2002 Update , 2002 .

[46]  Yanqing Zhang,et al.  SVMs Modeling for Highly Imbalanced Classification , 2009, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[47]  X. Qi,et al.  Waist‐to‐height ratio is the best indicator for undiagnosed Type 2 diabetes , 2013, Diabetic medicine : a journal of the British Diabetic Association.

[48]  Daisuke Kunii,et al.  Anthropometric Characteristics, Dietary Patterns and Risk of Type 2 Diabetes Mellitus in Vietnam , 2005, Journal of the American College of Nutrition.

[49]  Lincoln A Sargeant,et al.  Predicting incident diabetes in Jamaica: the role of anthropometry. , 2002, Obesity research.

[50]  J. Després,et al.  Contribution of abdominal obesity and hypertriglyceridemia to impaired fasting glucose and coronary artery disease. , 2002, The American journal of cardiology.

[51]  D. Dimić,et al.  The hypertriglyceridemic waist phenotype and metabolic syndrome by differing criteria in type 2 diabetic patients and their relation to lipids and blood glucose control. , 2011, Endokrynologia Polska.

[52]  J. Després,et al.  Relation of the "hypertriglyceridemic waist" phenotype to earlier manifestations of coronary artery disease in patients with glucose intolerance and type 2 diabetes mellitus. , 2007, The American journal of cardiology.

[53]  I. Okosun,et al.  Abdominal obesity, hypertriglyceridemia, hypertriglyceridemic waist phenotype and risk of type 2 diabetes in American adults , 2008 .

[54]  Barbara E Ainsworth,et al.  The hypertriglyceridemic waist phenotype among women. , 2003, Atherosclerosis.

[55]  Boncho Ku,et al.  Prediction of Fasting Plasma Glucose Status Using Anthropometric Measures for Diagnosing Type 2 Diabetes , 2014, IEEE Journal of Biomedical and Health Informatics.

[56]  Haibo He,et al.  Learning from Imbalanced Data , 2009, IEEE Transactions on Knowledge and Data Engineering.