According to new Canadian research, there is insufficient evidence to show that the various short questionnaires physicians use to ask about symptoms of depression in children and adolescents accurately screen 6- to 18-year-olds for the disease. Researchers believe this calls into question the use of these assessment tools for this group and raises worries about possible misdiagnosis of the disease in this age range.

To assess the quality of the screening tools that are currently being used to identify depression in children or adolescents, researchers carried out a search of the medical evidence looking for studies that put the screening tools to the test. They identified only 17 studies where the test results from the screening tools were compared with results from a diagnostic interview to determine if the children or adolescents in the study actually had depression.

Lead author Dr Michelle Roseman, who is affiliated with the Jewish General Hospital’s Lady Davis Institute for Medical Research in Montreal, and colleagues then assessed the methodology and results of these 17 studies. They found that most of the studies were too small to make a valid determination about the accuracy of the screening tools and that the methods of most studies fell short of expected standards. They also found that there was inadequate evidence to recommend any single cutoff score for any of the questionnaires. (Patients scoring above a pre-defined cutoff score are considered likely to be depressed, whereas patients below the cutoff are not.)

Researchers suggest that, given the inaccuracy of the tools currently being used, some children could end up mislabeled as depressed, and that large, well-designed studies that present results across a range of cutoff scores are needed to properly assess the accuracy of depression screening tools in children.

The study, “Accuracy of depression screening tools to detect major depression in children and adolescents: A systematic review,” is published in the May 2016 issue of the Canadian Journal of Psychiatry.

The research was supported by the Canadian Institutes of Health Research, the Arthritis Society, the Mach-Gaensslen Foundation of Canada, and a Murray R. Stalker Primary Care Research Bursary.