Factors associated with incidence of type II diabetes in pre-diabetic women using Bayesian Model Averaging

authors:

avatar maryam mahdavi , avatar Mehrabi Yadollah ORCID , * , avatar davood khalili , avatar Ahmad Reza Baghestani , avatar farideh bagherzadeh khiabani , avatar samaneh mansoori


how to cite: mahdavi M, Yadollah M, khalili D, Baghestani A R, bagherzadeh khiabani F, et al. Factors associated with incidence of type II diabetes in pre-diabetic women using Bayesian Model Averaging. koomesh. 2017;19(3):e152905. 

Abstract

Introduction: Diabetes is a chronic disease which usually begins with impaired glucose tolerance. This step is known as pre-diabetes. People with pre-diabetes are at greater risk for diabetes. Typically for the variable selection, stepwise approach is used which does not take into account model uncertainties. In this study, Bayesian Model Averaging (BMA) method was used to sort out the above shortcoming. Materials and Methods: The study population was 734 pre-diabetic women with 20 years and older participated in Tehran Lipid and Glucose Study (TLGS). In this study, the stepwise and BMA variable selection methods were employed in logistic regression. Then area under curve (AUC) for both methods was computed and compared with Delong test. All analyses was done using R version 3.1.3. Results: BMA selected the fasting plasma glucose, 2 hours’ blood glucose, and family history of diabetes, body mass index and aspirin use at baseline as risk factors for diabetic. In addition to these factors, stepwise method selected diastolic blood pressure, history of past 3 months’ hospitalization, thyroid drug use and education. Although the number of variables selected by BMA (5 variables) was less than that of stepwise (9 variables), AUC for the two methods was not significant. Conclusion: It seems that the BMA provide better model for screening of diabetes because with selecting fewer variables, prediction ability of the model is preserved

References

  • 1.

    Rathmann W, Giani G. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004; 27: 2568-2569.

  • 2.

    Azimi-Nezhad M, Ghayour-Mobarhan M, Parizadeh M, Safarian M, Esmaeili H, Parizadeh S, et al. Prevalence of type 2 diabetes mellitus in Iran and its relationship with gender, urbanisation, education, marital status and occupation. Singapore Med J 2008; 49: 571.

  • 3.

    Hajihasani A, Bahrpeyma F, Bakhtiari A. Effects of eccentric and concentric exercises on postural sway in type 2 diabetic patients. Koomesh 2016; 17: 493-500 (Persian).

  • 4.

    Masaeli N, Attari A, Molavi H, Najafi M, Siavash M. Normative data and psychometric properties of the quality of life questionnaire for patients with diabetes mellitus. Koomesh 2010; 11: 263-269 (Persian).

  • 5.

    Roche MM, Wang PP. Factors associated with a diabetes diagnosis and late diabetes diagnosis for males and females. J Clin Translat Endocrinol 2014; 1: 77-84.

  • 6.

    Esmaeil-Nasab N, Abdolrahimzadeh A, Ebrahimi A. The cross-sectional study of effective factors on type 2 diabetes control in a diabetes care center in Sanandaj. Iran Epidemiol J 2010; 6: 39-45 (Persian).

  • 7.

    Sajadi F. The prevalence of type 2 diabetes and its relation with CVD risk factors in the population of Isfahan. Med J Mashhad Univ 2003; 46: 68-73 (Persian).

  • 8.

    Navaei L. Study of the prevalence of diabetes and impaired glucose tolerance in rural areas of Tehran province. 2001; 4: 92-99. (Persian).

  • 9.

    Mansoori F, Namdaritabar H, Shahrezaee A, Rezaei R, Alikhani A, Montazer MJ, Gazerpoor F. Diabetes mellitus in over-thirty-year-old individuals in Kermanshah province (2002). J Kermanshah Univ Med Sci 2004; 8. (Persian).

  • 10.

    Barnett KN, Ogston SA, McMurdo ME, Morris AD, Evans J. A 12year followup study of allcause and cardiovascular mortality among 10 532 people newly diagnosed with Type 2 diabetes in Tayside, Scotland. Diabeti Med 2010; 27: 1124-1129.

  • 11.

    Fahimfar N, Khalili D, Mohebi R, Azizi F, Hadaegh F. Risk factors for ischemic stroke; results from 9 years of follow-up in a population based cohort of Iran. BMC Neurol 2012; 12: 117.

  • 12.

    Annest A, Bumgarner RE, Raftery AE, Yeung KY. Iterative bayesian model averaging: A method for the application of survival analysis to high-dimensional microarray data. BMC Bioinformatics 2009; 10: 72.

  • 13.

    Noble Jr RB. Multivariate applications of Bayesian model averaging [dissertation]. Citeseer; 2000.

  • 14.

    Wiegand RE. Performance of using multiple stepwise algorithms for variable selection. Stat Med 2010; 29: 1647-1659.

  • 15.

    Hoeting JA, Madigan D, Raftery AE, Volinsky CT. Bayesian model averaging: a tutorial. Stat Sci 1999; 14: 382-401.

  • 16.

    Hoeting J, Raftery AE, Madigan D. A method for simultaneous variable selection and outlier identification in linear regression. Computational Statistics & Data Analysis. 1996; 22(3): 251-270.

  • 17.

    Genell A, Nemes S, Steineck G, Dickman PW. Model selection in medical research: a simulation study comparing Bayesian model averaging and stepwise regression. BMC Med Res Methodol 2010; 10: 108.

  • 18.

    Wang D, Zhang W, Bakhai A. Comparison of Bayesian model averaging and stepwise methods for model selection in logistic regression. Stat Med 2004; 23: 3451-3467.

  • 19.

    Azizi F, Madjid M, Rahmani M, Emami H, Mirmiran P, Hadjipour R. Tehran lipid and glucose study (TLGS): rationale and design. Iran J Endoc Metab 2000; 2: 77-86 (Persian).

  • 20.

    Stekhoven DJ. MissForest: nonparametric missing value imputation using random forest. Astrophysics Source Code Library 2015; 1: 05011.

  • 21.

    Filzmoser P, Gschwandtner M, Filzmoser MP, LazyData T. Package mvoutlier; 2015.

  • 22.

    Leamer EE. Regression selection strategies and revealed priors. J Am Stat Assoc 1978; 73: 580-587.

  • 23.

    Madigan D, Raftery AE. Model selection and accounting for model uncertainty in graphical models using Occam's window. J Am Stat Assoc 1994; 89: 1535-1546.

  • 24.

    Raftery AE. Bayesian model selection in social research. Soc method 1995; 25: 111-163.

  • 25.

    Furnival GM, Wilson RW. Regressions by leaps and bounds. Technometrics 1974; 16: 499-511.

  • 26.

    Raftery A, Hoeting J, Madigan D. Model selection and accounting for model uncertainty in linear regression models: Citeseer; 1993.

  • 27.

    Kass RE, Raftery AE. Bayes factors. J Am Stat Assoc 1995; 90: 773-795.

  • 28.

    Kass RE, Wasserman L. A reference Bayesian test for nested hypotheses and its relationship to the Schwarz criterion. J Am Stat Assoc 1995; 90: 928-934.

  • 29.

    Jeffreys H. The theory of probability: OUP Oxford; 1998.

  • 30.

    Thompson WR. Variable selection of correlated predictors in logistic regression: investigating the diet-heart hypothesis [dissertation]. Florida State Univ; 2009.

  • 31.

    Clyde M. Bayesian model averaging and model search strategies. Bayesian Statistics 6 JM Bernardo, JO Berger, AP Dawid and AFM Smith. Oxford: University Press; 1999; 157-185.

  • 32.

    TAY PL. Iterative Bayesian Model Averaging For Patients Survival Analysis [dissertation]. Universiti Teknologi Malaysia; 2010.

  • 33.

    Lipkovich IA. Bayesian model averaging and variable selection in multivariate ecological models [dissertation]. Virginia Tach Univ; 2002.

  • 34.

    Montgomery JM, Nyhan B. Bayesian model averaging: Theoretical developments and practical applications. Politic Anal 2010; 18: 245-270.

  • 35.

    Bagherzadeh-Khiabani F, Ramezankhani A, Azizi F, Hadaegh F, Steyerberg EW, Khalili D. A tutorial on variable selection for clinical prediction models: feature selection methods in data mining could improve the results. J Clin Epidemiol 2016; 71: 76-85.

  • 36.

    Flack VF, Chang PC. Frequency of selecting noise variables in subset regression analysis: a simulation study. Am Stat 1987; 41: 84-86.

  • 37.

    Mohamadi SM, Rashidi M, Afkhami Ardakani M. Risk factors for type 2 diabetes. J Shahid Sadoughi Unive Med Sci 2011; 19: 266-280 (Persian).

  • 38.

    Mehrabi Y, Khadem-Maboudi A, Hadaegh F, Sarbakhsh P. Prediction of diabetes using logic regression. Iran J Endocrinol Metab 2010; 12: 16-24 (Persian).

  • 39.

    Kaykha M, Ghorbani MJ, Amini M. The prevalence of type 2 diabetes, pre-diabetes and metabolic syndrome risk factors in first-degree relatives of patients with type 2 diabetes. J Kerman Univ Med Sci 2013; 20: 115-128 (Persian).

  • 40.

    Soori R, Rashidi M, Choobineh S, Ravasi AA, Baesi K, Rashidy-Pour A. Effects of 12 weeks resistant training on MTNR1B gene expression in the pancreas and glucose and insulin levels in type 2 diabetic rats. Koomesh 2017; 19: 46-55 (Persian).##.