Prediction model may help identify adolescents at risk for depression

Researchers have developed a tool with predictive potential for identifying adolescents who are at high or low risk for depression in young adulthood, according to findings from a Brazilian cohort that were published in Journal of the American Academy of Child & Adolescent Psychiatry.

“This research marks an important first step in developing an accessible tool that could help screen adolescents for depression and improve mental health worldwide,” Valeria Mondelli, MD, PhD, of the Institute of Psychiatry, Psychology and Neuroscience King’s College London, said in a press release. “Depression can have debilitating lifelong impacts and adolescents are especially vulnerable to its onset. Identification of those at high risk for developing depression in later adolescence could be valuable in devising effective early-intervention strategies to help prevent this illness.”

According to Mondelli and colleagues, although prediction models are now frequent in medical literature, most have been conducted in a single setting. Further, previous models are limited by heterogeneity between development and validation samples, which is a major obstacle for the generalization of models, they noted. In the present study, the researchers used data on 11 sociodemographic variables from 2,192 Brazilian adolescents aged 15 years, as well as mental health assessment of these adolescents at age 18 years, to develop a risk-assessment tool for major depressive disorder. They tested the tool’s generalizability in two representative cohort studies that included 1,144 British adolescents aged 12 years and 739 New Zealand adolescents aged 15 years.

The researchers found major discrepancies between the samples, which impacted the external prognostic performance of the model. However, the generalized model “kept a valid and beyond chance prognostic capacity in parsing future risk for depression among the adolescents in the independent cohorts, especially when heterogeneity was accounted for,” they wrote. Further, they noted that applying adaptations to the original model would enhance its external clinical utility and allow it to be applied in multiple other contexts despite significant differences in socioeconomic characteristics, cultural influences and assessment strategies.

“When developing any new risk assessment tool, it is important to consider how it will work in the real world,” Mondelli said. “As part of the [Identifying Depression Early in Adolescence] project, we are interviewing adolescents, parents and other stakeholders to understand their perceptions of this tool and the ethical implications of using it to stratify young people in terms of risk for developing depression. By doing this, we hope that we can ensure that it can help improve the mental health of adolescents worldwide.”

The researchers concluded that “future efforts should focus on developing better methodological approaches for incorporating heterogeneity in prognostic research.” – by Joe Gramigna

Disclosures: Mondelli reports research funding from Johnson & Johnson, which has interest in the development of anti-inflammatory strategies for depression. Please see the study for all other authors’ relevant financial disclosures.