Introduction: The serum 25(OH)D response to vitamin D supplementation, differs between individuals. The goal of this study was the evaluation of the relationship between socioeconomic and demographic factors with the hugeness of response to vitamin supplementation, defined by statistical analysis artificial neural network (ANNs). Method: The prospective interventional study was conducted on 529 participants aged 12-19 years old. All participants were administrated to receive nine vitamin D capsules (50000IU vitamin D) over nine weeks. The response variables were the following: the differences between the concentrations of vitamin D before and after intervention. ANNs algorithm, combination of transfer functions hyperbolic tangent or sigmoid transfer in hidden layers and linear, softmax, hyperbolic tangent or sigmoid transfer function in output layer were assessed. The best ANNs algorithm to predict response to vitamin D levels had acceptable sensitivity, specificity and accuracy area under the ROC curve. Result: Among various sociodemographic factors which affect the increase in serum vitamin D amounts in response to supplementations, baseline serum vitamin D (28.1%), BMI (13.8%), physical activity (12.1%), age (7.6%), mother\\\\\\\\\\\\\\\'s education (6.4%), and father\\\\\\\\\\\\\\\'s occupation (5.8%) be important variables.Conclusion: This interventional study provides specific sociodemographic-recommendations to achieve 25(OH)D targets in cases with severe vitamin D deficiency, perhaps indicating that a higher dose is require to obtain optimal Vit D levels in some individuals.

کلمات کلیدی