Evaluation of psychometric properties of the Persian version of Conner’s early childhood questionnaire-parent’s form in pediatrics 3-6 years

Abstract

According to the statistic and diagnostic manual of mental disorders, attention deficit - hyperactivity disorder is one of the most common psychiatric diagnoses among children and adolescents (1). Different studies have estimated the prevalence of ADHD in children before the age of admission to school about 1.9-18.2% (2, 3, 4, 5, 6). Conners has provided a dedicated tool for screening of the disorderfor 2 to 7 years old which is in 2 forms of parent and teacher. The main aim of this research is to determine the validity, reliability, and diagnostic values of the Persian version of the Conners early childhood - Parent form questionnaire. Method: Conners Behavioral Questionnaire, Early Childhood- Parent Form, has 110 items with 3 items of impairment and 2 additional items which are totally 115 items. This questionnaire consists of 11 subscales. In order to prepare the Persian version first a research group translated the questionnaire into Farsi and after approving the formal and conceptual adaptation by the panel of experts, the questionnaire was translated back into English which corresponded to the original version. In order to determine the concurrent validity, the ADHD sub-scale of the SDQ questionnaire, along with the Persian version of the Conners tool ,were used for 1013 children. In this study the internal consistency for the ADHD subscale was examined and the reliability of the test was verified by completing 35 questionnaires in 2 turns within 2 weeks. To determine the diagnostic validity 334 children were interviewed.

Results: For all items in the Conners questionnaire, content validity was above 0.76, using modified Kappa. Concurrent validity of the Conners tool was estimated as 0.703 using the SDQ questionnaire, with Spearman correlation coefficient. in order to determine the diagnostic value, the sub-curve ROC for the sub-standard ADHD has been calculated using normal parent data and the normal mean of the parent-instructor which were 0.69 and 0.73, respectively and the area below the curvewere upgraded to 0.77 and 0.79, respectively, by removing 10% of people who had the greatest difference among parent and coach scores. With this analysis, the best cut-off suggested for normal parental scores is 26, with sensitivity and specificity of 79 and 70 percent. The internal reliability was calculated as 0.86 by Cronbach's alpha for the ADHD sub-scale. The retest reliability was estimated as 0.83 by calculating ICC.