The categorical, prototypical approach to organizing mental disorders used in traditional classifications systems, such as the Diagnostic and Statistical Manual of Mental Disorders (DSM; now in its 5th edition) has a number of limitations, such as a lack of specificity as demonstrated by the prevalence of comorbidity [7]. In contrast, empirical models of psychopathology use a broad range of quantitative approaches to generate coherent arrangements of signs and symptoms of psychopathology and capture the high rates of psychiatric comorbidity [11]. What results is a quantitively organized framework that facilitates investigation of the specificity and generality of risk and protective factors for psychopathology that is not achievable with traditional classification systems [9,12,13]. Two empirical models have emerged in recent years and received increasing attention in the literature.

Two previous reviews have examined risk and protective factors in relation to internalizing and externalizing dimensions in children and adolescents; however, to our knowledge, no previous reviews have investigated other broadband dimensions [17,18]. A mega-analytic synthesis of child, family, school, community, and cultural risk and protective factors correlated with internalizing behaviors, externalizing behaviors, or both found 4 risk factors and 3 protective factors common to both internalizing and externalizing disorders [17]. Although this suggests that additional factors examined were specific to either internalizing or externalizing, it is unclear from the review whether the studies included examined both internalizing and externalizing disorders simultaneously, only one of these, or specific behaviors or disorders within those disorder groupings. Thus, it is not possible to draw any conclusions about whether any of the identified risk and protective factors are transdiagnostic or disorder-specific.

McMahon and colleagues [18] conducted a systematic review of studies examining the relationship between internalizing and externalizing symptoms and a range of stressors, such as exposure to violence, abuse, poverty, and parental divorce, with the aim of evaluating the specificity of stressors. However, the review found little evidence that individual stressors were associated with specific internalizing or externalizing outcomes, with the exception of an association between sexual abuse and internalizing or post-traumatic stress disorder symptoms. This suggests that most stressors examined were transdiagnostic across internalizing and externalizing disorders. However, while stressors may be transdiagnostic risk factors and useful for identifying young people at risk of developing mental health problems, further investigation is needed to identify factors that can be addressed and modified through intervention. Further, it is unknown whether these transdiagnostic associations hold across other domains of psychopathology, such as psychotic-related disorders.

Research that takes into account a broad range of disorders and comorbidity is necessary to identify transdiagnostic risk and protective factors. Identifying the risk and protective factors for psychopathology in young people that occur across traditional diagnostic categories is of great clinical significance. Such factors may be useful for more efficient prediction and early identification of psychopathology, as some may provide useful targets for reducing overall risk for psychopathology, thus preventing a variety of mental disorders from subsequently emerging [19].

The aim of this systematic review is to identify transdiagnostic risk and protective factors for psychopathology in young people. This will be done by synthesizing and critically evaluating studies examining empirically based models of psychopathology.

The Population Exposure Comparator Outcome (PECO) framework was used to develop the research question and eligibility criteria for this review [22].

Searches will be conducted in Medline, EMBASE, and PsycINFO databases. An example search string developed for Ovid PsycINFO is shown in Table 1, which will be replicated for EMBASE and Medline databases. Reference lists of selected articles will also be hand searched to identify additional relevant articles not captured by the initial search strategy.

All titles and abstracts will be screened by one reviewer (SJL); the other reviewers (CC, NN, MS) will screen 25% of the titles and abstracts, which will be randomly selected. For all studies identified in the initial screen, the full-text articles will be reviewed and assessed against the eligibility criteria by two reviewers (SJL and MS). Disagreements at each stage of screening will be resolved through discussion or by a third reviewer (CC). A PRISMA flow chart will be created to show the results of each stage of the screening process.

Citations will be imported into the Covidence systematic review software [24], which will be used to remove duplicates and screen tittles, abstracts, and full texts. The following information will be extracted by the primary reviewer (SJL): publication details (authors, year of publication, country), study design (eg, cross-sectional, longitudinal), sample characteristics (sample size, mean age, ethnicity, sex), psychopathology measures (measures used), informant type (parent, self, other), risk or protective factor measures, data analysis strategy (techniques used, model specification, indicator type), outcome statistics (eg, test statistics, P values, effect size, model fit statistics, network centrality statistics). For longitudinal studies, additional information will be extracted regarding follow-up intervals and frequency. A summary of main findings will also be recorded.

Following data extraction, study quality will be assessed independently by two reviewers. Cross-sectional studies will be evaluated using the Joanna Briggs Institute Critical Appraisal Checklist for Analytical Cross-Sectional Studies, and longitudinal studies will be evaluated using the Joanna Briggs Institute Critical Appraisal Checklist for Cohort Studies [25].