A Large Scale Study of the Psychometric Characteristics of the Ibr Modified Overt Aggression Scale: Findings and Evidence for Increased Self-Destructive Behaviors in Adult Females with Autism Spectrum Disorder

Background:

Aggressive behaviors toward self and others are reported to be relatively common problems in persons with intellectual disabilities (ID), have major impacts on the family, and seriously affect the ability of such individuals to reside in more “normalizing” environments.  In order to begin to study the biological bases of some of these behaviors, we carried out a survey using a modification of the Overt Aggression Scale (OAS)  (Yudofsky, et al 1986). We added new items to assess self-deprecating statements and included information on antecedents and consequences of aggressive behaviors, clinical diagnoses, and known etiologies of ID to better define our samples.  We named this version the IBR (Institute for Basic Research) Modified Overt Aggression Scale or IBR-MOAS.  Due to this modification it was important to ensure it had adequate reliability and validity.

Objectives:

To document both the reliability and validity of the aggression scale of the IBR-MOAS using a sample of over 2,000 people.  Validity studies included developmental validity (showing variations in scores with age and ID level), diagnostic validity, and construct validity.  Diagnostically, we tested whether females show more self-injury, as reported by Crocker et al.  (2006) , and whether this effect would be stronger in those females who also have an autism spectrum disorder, controlling for level of intellectual disability.

Methods:

Two sets of samples were ascertained for this study.  The first sample consisted of 3,547 people community-based people who received services from OMRDD between 2006 and 2007. The second set consisted of a sample of 25 people who were seen in 2008 at the George A. Jervis Clinic because of problems with aggressiveness toward self and/or others.  Forms completed on this set were used for determining inter-rater and two-week test-retest reliability of the Aggression Scale of the IBR-MOAS.  A Likert measure of the frequency of occurrence of each of the aggression items during the past year was developed as follows: 0 = Never (never happens); 1 = Rarely (averages about once a year to once a month); 2 = Sometimes (averages about several times a month to several times a week); 3 = Often (averages about daily to several times a day); and U (Used to happen but not this past year).  For the present analyses, the U rating was coded as equivalent to Never.  Unlike the OAS and its modified versions, a weighted scoring system was not used.

Results:

Reliability ranged from good to excellent.  Aggression toward others and objects was highest in the youngest adults, in those in the moderate to severe range of ID, and in those with an autism spectrum diagnosis.  Self-injury was highest in those in the severe to profound range of ID and in those with autism, particularly the females.  Females with autism were also more likely to make the most self-deprecating statements.  

Conclusions:

Our data suggest that adult females with autism are a unique group and support the notion that mood and anxiety disorders play a role in self-destructive behaviors in this population.