Diagnosing childhood depression can pose a challenge, even for mental health specialists. Screening tools can aid clinicians within the initial step of the diagnostic process. For the first time, the Children׳s Depression Screener (ChilD-S) is validated in a mental health setting as a novel field of application beyond the previously examined pediatric setting. Based on a structured interview, DSM-IV-TR diagnoses of depression were made for 79 psychiatric patients aged 9–12, serving as the gold standard for validation. For assessing criterion validity, receiver operating characteristic (ROC) curves were calculated. Point prevalence of major depression and dysthymia was 28%. Diagnostic accuracy in terms of the area under the ROC curve was high (0.97). At the optimal cut-off point ≥12 according to the Youden׳s index, sensitivity was 0.91 and specificity was 0.81. The findings suggest that the ChilD-S is not only a valid screening instrument for childhood depression in pediatric care but also in mental health settings. As a brief tool it can easily be implemented into daily clinical practice of mental health professionals facilitating the diagnostic process, especially in case of comorbid depression.
Childhood depression is highly prevalent in mental health care, with rates of 8–10% (Ling et al., 1970 and Stegmann et al., 2013). Early recognition and timely treatment are crucial as childhood depression has negative effects on psychosocial functioning (Lima et al., 2013 and Maughan et al., 2013) and increases the risk of developing other psychiatric disorders in later life (Copeland et al., 2013).
Yet, diagnosis of depression in children is challenging, even for mental health specialists. Compared to other psychiatric disorders, interrater-reliability between professionals of child and adolescents psychiatry is lowest for the diagnosis of depression (Blanz and Schmidt, 1990 and Schmidt and Sinzig, 2006). The difficulty in making the correct diagnosis can be explained by the heterogeneous presentation of childhood depression. Unlike in the later course of development, the clinical impression of the symptomatology mostly is not a typical one. Often, anxiety and unspecific somatic symptoms that are not always easily traced back to depression are predominant (Garber et al., 1991). Only in adolescence, the clinical presentation approaches the characteristics of adulthood depression. In general, symptoms of depression are hard to be observed and especially young children struggle with giving insight in their inner life (Mehler-Wex, 2008). On top of this, childhood depression often occurs along with other mental disorders such as conduct disorders, hyperkinetic disorders and separation anxiety disorder (Breton et al., 2012 and Ryan et al., 1987). Comorbid depression is often shadowed by these more prominent disorders (Nijdam, 1986). The best method to assess comorbid disorders is the conduction of structured interviews that are rarely applied in clinical practice as they are comprehensive and time-consuming (Bruchmüller et al., 2011). Instead, mental health professionals rather rely on their clinical judgement, running the risk that the further exploration is led by their first impression (Crumlish and Kelly, 2009 and Nath and Marcus, 2006). As a result, comorbid disorders are missed (Zimmerman and Mattia, 1999 and Zimmerman and Mattia, 2001) since disorder-specific questionnaires are not applied (Zimmerman and Mattia, 2001).
Given this, short and time-economic diagnostic instruments such as screening tools can facilitate the diagnostic process. In a second step, children who scored positively on the screener can be followed-up by a further clinical exploration to verify or falsify a diagnosis of depression.
For the use in children, screening tools ideally have a simple wording and answering format.
There are only two depression-specific screening instruments that have solitarily been validated in children. The Children׳s Depression Inventory Short Version (CDI:S, Kovacs, 2003) and the Short Mood and Feelings Questionnaire (MFQ-SF, Angold et al., 1995) yielded good diagnostic accuracy in medical settings (Allgaier et al., 2012 and Katon et al., 2008). To the best of our knowledge, both screening tools have not been investigated in a mental health setting so far. The only study in a psychiatric setting using the full-version of the CDI found a low recognition rate of depression of 0.63 (Sorensen et al., 2005). Apart from disorder-specific screening tools there are other instruments covering a broader spectrum of mental disorders including subscales for depression, such as the Revised Child Anxiety and Depression Scale (RCADS, Chorpita et al., 2000). Again, only 74% of the subjects in a clinical sample were correctly classified as depressed by the RCADS (Chorpita et al., 2005). Yet, this clinical sample did not only include children but also adolescents.
In sum, none of these instruments yielded a satisfying performance in detecting depression in mental health care. Our newly developed screening tool, the Children׳s Depression Screener (ChilD-S, Frühe et al., 2012), reached a high recognition rate of 0.91 in pediatric hospital patients aged 9–12. Based on these previous findings showing that the ChilD-S discriminates well between depression and somatic disorders, the current study aimed at investigating if it can also differentiate between childhood depression and other mental disorders. For this purpose, the ChilD-S was validated in a mental health care sample.
