artificial neural networks versus bivariate logistic regression in prediction diagnosis of patients with hypertension and diabetes

background: diabetes and hypertension are important non-communicable diseases and their prevalence is important for health authorities. the aim of this study was to determine the predictive precision of the bivariate logistic regression (lr) and artificial neutral network (ann) in concurrent diagnosis of diabetes and hypertension. methods: this cross-sectional study was performed with 12000 iranian people in 2013 using stratified-cluster sampling. the research questionnaire included information on hypertension and diabetes and their risk factors. a perceptron ann with two hidden layers was applied to data. to build a joint lr model and ann, sas 9.2 and matlab software were used. the auc was used to find the higher accurate model for predicting diabetes and hypertension. results: the variables of gender, type of cooking oil, physical activity, family history, age, passive smokers and obesity entered to the lr model and ann. the odds ratios of affliction to both diabetes and hypertension is high in females, users of solid oil, with no physical activity, with positive family history, age of equal or higher than 55, passive smokers and those with obesity. the auc for lr model and ann were 0.78 (p=0.039) and 0.86 (p=0.046), respectively. conclusion: the best model for concurrent affliction to hypertension and diabetes is ann which has higher accuracy than the bivariate lr model.