The DD-CGAS As a Tool for the Assessment of Global Functioning in Treatment Outcome Research

Background:  Successful treatment should improve the client’s global functioning as well as alleviate targeted symptoms. The Developmental Disabilities Modification of the Children’s Global Assessment Scale (DD-CGAS; Wagner et al., 2007), developed as a measure of global functioning for children with ASD, is administered by a rater trained to assign ratings reliably using all available information . This is the first report on the DD-CGAS as a measure of change in a randomized-controlled treatment (RCT) study.

Objectives:  The purpose of this study was to further examine the utility of the DD-CGAS as a treatment outcome measure. Specifically, we report on the measure’s convergent and discriminant validity, its sensitivity to change with intervention, and procedures for establishing reliability among raters.

Methods:  Data for the present study were drawn from an RCT evaluating a psychosocial therapy program with higher functioning adolescents (n = 30) with ASD and co-occurring anxiety disorders.  The DD-CGAS was administered by independent raters trained to a pre-established reliability threshold (Wagner et al., 2007) who were blind to treatment assignment (treatment or wait-list).  Higher DD-CGAS scores indicate better functioning and are based on the domains of self care, communication, social behavior, and academic functioning.

Results:  Over a one-year period, eight raters were trained to reliability on the DD-CGAS. Baseline DD-CGAS scores were, as expected, negatively correlated with ASD-related social disability (r = -.388, p < .05). They were not, however, significantly correlated with adaptive behavior (r = -.23, ns) or parent-reported symptoms of anxiety (r = .02, ns) at baseline.  Bivariate correlations were calculated to analyze the relationship between the change in DD-CGAS scores from baseline to post-treatment with the change on parent-reported social disability and anxiety.  Improvement on the DD-CGAS was significantly correlated with improved social competence (r = .40, p < .05), but not with improvement in anxiety (r = .30, ns).  An independent samples t-test revealed that the mean change in DD-CGAS scores was greater for those participants in the active treatment condition than those in the wait-list control condition, t(30)=2.54, p< .05.  Finally, an independent samples t-test was used to test whether the change in DD-CGAS scores were higher for those participants who were considered treatment responders, based on the CGI-I (Guy, 1976), than for non-responders, t(30)= -2.00, p = .06.  Although not a statistically significant result, this finding demonstrates a trend toward treatment responders having higher DD-CGAS scores (mean difference = 4.81) than non-responders.

Conclusions:  Raters can be trained to reliability on the DD-CGAS in a fairly brief period of time.  The DD-CGAS demonstrated convergent validity with ASD-related social disability.  Discriminant validity of the DD-CGAS was evidenced by the non-significant correlation with adaptive behavior.  Although it showed sensitivity to treatment gains, it may be primarily tethered to core ASD problem areas and, as such, less relevant for treatment studies addressing comorbid problems in ASD.  Further examination is needed to evaluate its utility in treatment studies targeting a non-ASD problem domain.