(PS2-34) Development of a Brief Revised Child Anxiety and Depression Scale

Treatment progress monitoring is an effective method for improving the impact of effective therapies, such as CBT for anxiety and depression (e.g., Chorpita, Bernstein, & Daleiden, 2014), especially among those otherwise at risk of poor outcomes (Lambert et al., 2018). Such methods depend upon low cost, accessible, efficient, valid, and reliable measures of treatment targets; however, with few exceptions the best evidence-based assessments for measuring youth anxiety and depression are lengthy and are often relevant only to a single clinical focus (e.g., depression or anxiety, but not both).  

To that end, Ebesutani et al. (2012) developed a 25-item version of the RCADS-47 (Chorpita et al., 2000), which has shown strong psychometric properties and has been adopted by NIMH as one of the Common Data Elements (National Institute of Mental Health Data Archive). Based on feedback from the RCADS and the interest in routine outcome monitoring in health care systems, we sought to develop an even briefer version with comparable psychometric quality to the RCADS 25- and 47 item versions. 

Using data from Trent et al. (2012) we examined RCADS scores from 12,802 youth in grades 2 to 12 in public schools across the state of Mississippi (N = 12,802). The sample was evenly distributed between boys and girls and ethnically representative of the state of Mississippi (see Trent et al.). We used hierarchical cluster analysis (HCA) and an item response theory (IRT) nominal response model to guide selection of the best candidate items. Results so far indicate that a scale of approximately 12-14 items would be capable of producing accurate estimates of anxiety in depression in youth. This process will be replicated in multiple other independent clinical and community samples from archival datasets available to our research team, including ~18,000 youth reports and ~8,000 caregiver reports.  We will ascertain discriminant validity utilizing samples that include diagnostic information. We will report reliability and validity data from multiple independent samples and expect to make a brief version (e.g., 14 items or fewer) to the general public by the time of ABCT conference in November. We plan to highlight characteristics of the newly created versions that are especially suitable to treatment research designs that include longitudinal modeling of change as well as to routine outcome monitoring in service settings to improve the quality of evidence-based care for youth with anxiety and depression.