Utility of a statistical model of cognitive styles in attention deficit hyperactivity disorder

The purpose of this study was to determine the best statistical model of cognitive styles, based on the MFFT-20, CEFT and Stroop tests to predict attention deficit hyperactivity disorder (ADHD), analyzing the validity of the model for the diagnosis of the disease. We studied 100 ADHD cases (DSM-IV criteria) and 100 controls, age ranging between 7 and 11 years. Controls were randomly recruited and matched in age, gender and sociodemographic area with ADHD cases. On average, ADHD cases showed more impulsiveness (d: 1.28), less cognitive flexibility (d: 0.91) and more field dependence (d: 1.62) than controls. The logistic regression model that predicts ADHD best is made up of age, CEFT, MFFT-20 and Stroop tests and the formula derived from the model shows 85% sensitivity and 85% specificity for ADHD, regarding the DSM-IV criteria as the standard. The statistical model of cognitive styles presents valid indicators to diagnose ADHD, contributing to an increase in the objectivity of its analysis.