Level 2 Pervasive Developmental Disorder Rating Scales

Background: Rating scales are an attractive means by which to screen for Pervasive Developmental Disorders (PDDs) due to their ease of administration and low cost. They are also used to determine eligibility for services or research studies and to measure change. Many rating scales have been developed for screening purposes, making it difficult for professionals to select the most appropriate and effective instrument for their target population. Although reliable diagnosis of PDDs can be made as young as 20-24 months (e.g., Cox et al., 1999; Stone et al., 1999), several studies have reported diagnoses to be delayed several years and to occur in older children (e.g., Wiggins et al., 2006; Sivberg, 2003).

Objectives: To review the peer-reviewed literature on the diagnostic validity of level 2, caregiver-completed rating scales for the screening of PDDs in school-aged children (ages 5-12 years).

Methods: PsycInfo database was searched with combinations of the terms “PDD,” “autism,” “Asperger Disorder,” “PDD-NOS,” “rating scale,” “screen,” and “screening instrument.” Inclusion criteria required that scales: (a) be developed post-ICD-10; (b) be PDD-specific; (c) have published diagnostic validity evidence in peer-reviewed journals; (d) be level 2 instruments; and (e) focus on school-aged children. Diagnostic validity was evaluated based on sensitivity and specificity values (values >.70 and .80, respectively, were considered good), positive and negative predictive power, and AUC (higher values were considered desirable). Special attention was paid to the choice of comparison groups for diagnostic validity.

Results: Five scales met inclusion criteria: the Social Communication Questionnaire (SCQ), Gilliam Autism Rating Scale- Second Edition (GARS-2)/Gilliam Autism Rating Scale (GARS), Social Responsiveness Scale (SRS), Autism Spectrum Screening Questionnaire (ASSQ), and Asperger Syndrome Diagnostic Scale (ASDS). Review of published studies indicated that the SCQ has received the most examination, with more studies of diagnostic validity than the other four scales combined (n=11). Available evidence for the remaining four scales was limited. Contrast groups varied significantly across studies and impacted results significantly. Overall, studies on the diagnostic validity of the SCQ indicated that it performs well at different cut-offs for different groups, although performance was poorest in younger individuals. No independent studies of the GARS-2 are available. Three of the four studies examining diagnostic validity of its predecessor, the GARS, indicated poor sensitivity. The SRS has limited evidence of diagnostic validity but appears promising. Limited support is available for the ASSQ, though it has demonstrated adequate sensitivity and high specificity. Only one independent study has examined the ASDS and it suggested poor diagnostic validity.

Conclusions: Results of this review indicate level 2 PDD caregiver-completed rating scales are in need of much more scientific scrutiny. The SCQ is a strong choice for screening, while the SRS and ASSQ show promise. The GARS and ASDS should be avoided in diagnostic contexts.