Factor analysis on ADHD and autism spectrum disorder DSM-IV-derived items shows lack of overlap

Abstract

I studied with interest the review article about shared heritability of attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD). The authors mentioned that there is a considerable rate of co-morbidity of ADHD and ASD. It was concluded that there are similar familial/genetic factors for ADHD and ASD origins. It was recommended that in children both ASD and ADHD should be examined instead of only one of both conditions. The authors discussed that the co-occurrence of two disorders may be due to overlapping diagnostic criteria of ADHD and ASD resulting in inflated the rate of co-morbidity. Then they mentioned about their ‘‘vital concern’’ that ‘‘as far as the authors know, no studies exist that have performed a factor analysis on ADHD and ASD DSM-IVderived items to examine the extent of overlapping diagnostic criteria’’ [1]. There is an argument about co-morbidity of ADHD and ASD. Diagnostic and Statistical Manual of Mental Disorders 4th edition (DSM-IV) precludes making ADHD diagnosis in ASD [2]. However, there are many reports about co-morbidity of these two disorders [3, 4]. Herein, the results of a re-analysis of a study about ADHD and autism symptoms in a community sample of 1,600 school aged children are presented. More information about sampling method and the participants was presented before. Valid and reliable parent-completed, DSM-IV-referenced rating scales of ADHD symptoms and ASD symptoms were used [5, 6]. Exploratory factor analysis with varimax rotation was conducted on the 30 items (including 12 items of ASD and 18 items of ADHD). Kaiser–Meyer–Olkin measure of sampling adequacy was 0.92, and Bartlett’s test of sphericity was \0.001. The analysis of the items indicated that 52.5% of the variance was explained by the factor solution. Factor number was determined with eigenvalue more than 1. The loading of items on the six factors are shown in Table 1. A confirmatory factor analysis with two fixed number factors was also conducted. According to the results in both of the analyses, there were no overlapping diagnostic criteria of ADHD and ASD. Therefore, these results support the idea that ‘‘the high rate of co-morbidity of ADHD and ASD is most likely not due to the overlapping diagnostic criteria of ADHD and ASD’’ [1].