Diagnosing Thyroid Disorders: Clinical Symptoms vs. Laboratory Tests

Introduction: The main goal of this study was to diagnose the two most common thyroid disorders, hyperthyroidism and hypothyroidism; and also to evaluate the predictive ability of laboratory tests against the individual clinical symptoms.Materials & Methods: In this study, the data from patients with thyroid dysfunction who referred to Imam Khomeini Clinic and Shahid Beheshti Hospital in Hamadan were collected. The data contained 310 subjects in one of3 classes: normal, hyperthyroidism and hypothyroidism. Collected variables included demographics , symptoms of hypothyroidism and hyperthyroidism, as well as laboratory tests. In order to compare the predictive ability of the clinical signs and laboratory tests, different multinomial logistic regression models were fitted to the data. These models were compared in terms of the mean of the accuracy and AUC.Results: The results related to logistic model which included all variables (demographic, symptomatic and laboratory tests), indicated that the difference between variables of history, TSH and Score.Hyper (defined as sum of symptoms) in the hyperthyroid people compared to healthy people are statistically significant. Additionally, the difference in TSH and Score.Hypo in the hypothyroid group is significant in comparison with healthy subjects. Also, the results showed that multinomial logistic model based on symptomatic variables works well in thyroid disorder prediction with about 90% accuracy.Conclusion: The results of this study indicate that predictive performance of multinomial logistic model based on symptomatic variables was better than laboratory variables.