Page 89 - Age of onset of disruptive behavior of residentially treated adolescents -Sjoukje de Boer

Also, in a comparable residential sample of psychiatric inpatients with severe disruptive behaviour of Van der Reijen et al. (2013), a dropout percentage of 59.1%
was found (14.2% withdrawal and 44.9% pushout). In this light, the dropout percentage of 34.4% (17.9% withdrawal and 16.5% pushout) in our study was on the
low end. The low percentage of pushout was probably partly due to what the facility describes as “retentive treatment” (Boon & Haijer, 2008) which, among other things,
reflects the way the facility regards the youngsters and their disruptive behaviour. Employees of the institution are well aware not to walk into the common pitfall to 5 send the youngsters away when they exhibit the behaviour for which they were
initially admitted.
Among the sample, there was a high prevalence of risk factors for dropout
known from outpatient settings, that is, the majority was male, two third had an externalising disorder, and about a quarter was not attending school or attended special education. With this in mind, one would actually expect a high percentage of dropout. In order to find predictors of dropout within this high risk group, it was of clinical relevance to investigate specific predictors for dropout, which would apply to psychiatric inpatients with disruptive behaviour. It was found that the dropouts differed significantly from the completers on only one of the known pre-treatment patient characteristics (i.e., sex) and on three of the behavioural characteristics, namely cannabis usage prior to admission, disruptive behaviour reported at school, and EO disruptive behaviour. Of all characteristics, after multivariate regression analysis only EO disruptive behaviour and cannabis usage prior to admission remained as predictors of dropout. The explained variance of 10% indicates that many other factors play a role. This, of course, raises the question to what the extent the findings can be used to predict behaviour of individual youths.
We found no differences between pushouts and withdrawals, or at least, not in the pre-treatment client characteristics or predictors that were examined. However, both groups may differ in factors that are characteristic but were not examined, such as parent or family factors (e.g., SES, parenting), therapist factors (e.g., the quality of the therapeutic alliance), and also motivation for treatment, having a purpose in life, or distrust of adults. We therefore think it is useful to distinguish between the two
87