Child and adolescent mental illness during COVID-19: A rapid review

Prior to the novel coronavirus (COVID-19) pandemic, the international prevalence of child and adolescent mental illness, across all mental disorders, was 13.4% (Polanczyk et al., 2015). As a result of COVID-19, children and adolescents have experienced unprecedented interruptions to their daily lives and it is anticipated that these disruptions may be precipitants of mental illness, including anxiety, depression, and/or stress related symptoms (Lee, 2020). In order for governments and policy makers to plan, and allocate resources for child and adolescents mental illness, a rapid review of the research is recommended (Tricco et al., 2017). The aim of the current study was to conduct a rapid review of child and adolescent anxiety, depression, and traumatic stress experienced during the COVID-19 pandemic. Risk and protective factors for child and adolescent mental illness were also examined. The review was registered with PROSPERO [CRD42020184903] and followed PRISMA guidelines. Searches were conducted in PsycINFO, Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, and MEDLINE up to May 26th, 2020 by a health sciences librarian. PsycArXiv, a repository of unpublished pre-prints, was also searched using the key terms “COVID-19″ and “mental health”. The initial search yielded 3405 non-duplicate abstracts, which were reviewed for study inclusion. Study inclusion criteria were as follows: (1) empirical study; (2) written in English; (3) data collected during COVID-19; (4) sample < 18 years of age; and (5) data on the prevalence of mental illness symptoms and/or factors associated with mental illness symptoms were available. Reliability between coders on 20% of titles/abstracts ranged from .78 to .96 %. In total, 175 full text articles were reviewed and six met full inclusion criteria (See Supplementary Material 1). A description of study characteristics can be found in Table 1 . Consistent with recommendations for rapid reviews (Garritty et al., 2020; Tricco et al., 2017), narrative results were extracted by a primary coder and 20% checked by a secondary coder. Reliability between coders was 100%. Table 1 Study characteristics.