Els van Meijel

29 Evaluation of the STEPP | Chapter 2 & Rapee, 2007; Silverman, Saavedra, & Pina, 2001). For a random sample of children in our study (12%), the audiotaped ADIS child and parent interviews were rated independently for inter-rater reliability. The result showed almost perfect agreement (κ = 0.88). The ADIS-C/P showed good reliability for the current sample. Cronbach’s alphas for ADIS-C/P were 0.84 for the child score and 0.80 for the parent score. Depending on the answer and the clinical interpretation of the interviewer, symptoms can be rated as present or absent. If the number of symptoms endorsed as ‘present’ is enough to meet DSM-IV criteria, impairment in daily functioning is rated on a 9-point Likert scale (0-8). A diagnosis of PTSD requires an impairment level of 4 or more and depends also on the clinician’s judgment of clinical severity. The diagnosis can be based upon either the child report (C) or the parent report (P). The interview also provides for a combined diagnosis, based on both the child and parent report. In cases of disagreement between the two interviews, the child receives a diagnosis if one of the two interviews yields a diagnosis. Partial PTSD is diagnosed when at least one symptom is present in each of three subscales – re-experiencing, avoidance and hyperarousal – resulting in substantial distress or impairment in one or more areas of functioning (Winston et al., 2003). The interviewers were extensively trained on administering and scoring the ADIS-C/P and were supervised by an experienced child and adolescent psychiatrist (RJL). The interviewers were blind to the outcome of the STEPP screening. Self-reported children’s posttraumatic stress symptoms The children completed the Dutch version of the Children’s Revised Impact of Event Scale (CRIES; Olff, 2005; Perrin, Meiser-Stedman, & Smith, 2005; Verlinden et al., 2014). This self-report measure gives a good indication of the presence of PTSD. It consists of 13 questions in the subscales re-experiencing, avoidance and hyperarousal, with answers on a 4-point Likert scale. Items are rated according to the frequency of their occurrence during the past week (Not at all=0, Rarely=1, Sometimes=3 and Often=5; range 0-65). We asked the children to focus on their accident when answering the questions. The validation and reliability of the Dutch version of the CRIES was evaluated by Verlinden et al. (2014). Children with PTSD had significantly higher scores than children without PTSD on the total scale of the CRIES (mean score 42.48 versus 19.4; p < .001). At a cut-off score of 30, the Dutch CRIES was significantly better than