Els van Meijel

30 Chapter 2 | Evaluation of the STEPP chance at identifying PTSD as measured by the ADIS (area under the ROC curve = 0.91; 95% CI, 0.88-0.94). The CRIES showed excellent test-retest reliability (κ = 0.85) and good reliability: Cronbach’s alpha for the total score was 0.89 (Verlinden et al., 2014). For the current sample Cronbach’s alpha was 0.87. The CRIES showed good agreement with the ADIS-C/P for the current sample. On the CRIES, 16% of the children scored positively; on the ADIS-C/P, 12% of the children were diagnosed with PTSD. These percentages were not significantly different from each other based on the results of the McNemar test of dependent proportions ( p = .18). Self-reported parental posttraumatic stress symptoms The parents completed the Dutch version of the Impact of Event Scale-Revised (IES- R; Horowitz, Wilner, & Alvarez, 1979; Weiss, 2007). The IES-R consists of 22 questions and contains the subscales re-experiencing, avoidance and hyperarousal. Scoring is on a 5-point Likert scale. Items are rated according to the frequency of their occurrence during the past week (Not at all=0, A little bit=1, Moderately=2, Quite a bit=3, Extremely=4; range 0-88). The focus is on the child’s accident. A total score of 23 or above indicates the likely presence of PTSD (Mouthaan, Sijbrandij, Reitsma, Gersons, & Olff, 2014). The Dutch IES-R showed adequate similarity with the total score of the Clinician-administered PTSD scale (CAPS; r = .75, p < .001) (Hovens et al., 1994; Mouthaan et al., 2014; Weathers, Keane, & Davidson, 2001) and good reliability for the current sample; Cronbach’s alpha was 0.96. Statistical analyses We used descriptive statistics to summarize the demographic, trauma-related and clinical characteristics of the sample. Differences between participants and non- participants were analyzed with Mann-Whitney tests for age and injury severity and a Pearson Chi-Square test for sex. Differences between those who completed the second assessment and those who dropped out after the first assessment were analyzed with Mann-Whitney tests for age and injury severity, a Pearson Chi-Square test for sex and a t-test for the STEPP scores. To evaluate the performance of the STEPP at predicting child and parent PTSD, Receiver Operating Characteristic (ROC) curve analyses and cross-tabulations were conducted. An ROC curve analysis represents the changes in accuracy (sensitivity and specificity)