Psychometric Analysis of the RAADS Screen Scale for Adult ASD

Thursday, May 17, 2012
Sheraton Hall (Sheraton Centre Toronto)
3:00 PM
J. Eriksson and S. Bejerot, Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden
Background:  Autism spectrum disorders (ASD) manifest with a wide range of symptoms, some of which are also present in other psychiatric disorders. Thus, psychiatric patients can enter adulthood without receiving an ASD diagnosis, because it is cloaked by, or even mistaken for, another psychiatric disorder. There are few diagnostic instruments available for adults with ASD and no wide-spread screening instrument for quick assessment of the need for further evaluation.

Based on a Swedish translation of the Ritvo Autism and Asperger Diagnostic Scale-Revised (RAADS-R), an 18 item self-evaluation scale, RAADS Screen, was designed for screening for ASD in adults. The 18 questions were selected on two main premises; being the statements that best differentiated between the ASD group and the control group and to represent the four domains from RAADS-R in a proportion reflecting both the importance of symptoms in the DSM-IV criteria and the proportions of the original scale. The selection of statements was based on the data from the Swedish validation of RAAD-R. 

Objectives: To validate the psychometric properties of RAADS Screen for screening for ASD in adults in the psychiatric population.

Methods:  The RAADS Screen questionnaire was administered to 60 adults diagnosed with ASD and 744 adults with no ASD diagnosis; 594 volunteers without any DSM IV diagnosis and 150 psychiatric outpatients with a current DSM IV diagnosis other than ASD, distributed on the five groups: primary affective disorder, primary psychotic disorder, primary anxiety disorders, ADHD and emotionally unstable personality disorder.

The total score was compared between the ASD group and the psychiatric groups respectively and together.

Results:  Preliminary results show that RAADS Screen discriminate ASD in the psychiatric group as well as in the non-psychiatric group.  All psychiatric subgroups had significantly lower score than the ASD group. A cut-off value of 21/54 gives a sensitivity of .8 and a specificity of .65. In the non ASD sample, three factors could be identified with satisfactory internal consistency in the full sample (Cronbach’s α between .73 and .89).

Conclusions:  RAADS Screen is a valid and reliable instrument, useful for screening for ASD in the adult psychiatric population.

| More