Transdiagnostic risk factors of emotional disorders in adults: A systematic review

: Comorbidity is more the rule than the exception in mental health, specifically in the case of anxiety and depression. Transdiagnostic models studied the underlying processes to improve mental health treat-ment and understating. Objective: This systematic review searchs for evidence on transdiagnostic risk factors for anxiety and depression in the clinical population diagnosed with these psychopathological conditions, by analysing the different types or categories of factors identified. Methods: A systematic review was registered in PROSPERO (registration number CRD42022370327) and was designed according to PRISMA-P guidelines. Two independent reviewers with field knowledge assessed the study quality to reduce bias. Results: Fifty-three articles were examined, and the transdi-agnostic variables were grouped into three categories: psychological, biological, and sociocultural. Conclusions: The most studied category was that of psychological variables, especially cognitive processes, negative affect, and neuroticism, intolerance of uncertainty, anxiety sensitivity. Biological and sociocultural factors require more study to support their transdiagnostic approach


Introduction
The existence of common factors through mental pathology, specifically in the case of the more prevalent mental health problems in the world, anxiety, and depression (Santomauro et al., 2021), has become an increasingly important research topic in mental health (Littlefield et al., 2021).However, the research on the common vulnerability between them is not new.In 1991, Clark and Watson developed a tripartite model of anxiety and depression, indicating that these disorders are both strongly associated with negative affective ties.They also identified two other specific factors: (1) positive affective and (2) physiological hyperarousal.Only depression was characterized by low positive affective rates (anhedonia), while hyperarousal was exclusively a characteristic of anxiety.Therefore, a part of the variance of the different disorders is shared (general distress).Even though this model was discovered three decades ago, this line of investigation continues nowadays with transdiagnostic models.
Transdiagnostic approaches seek to identify core psychological and biological processes that underlie multiple forms of psychopathology (Nolen-Hoeksema & Watkins, 2011).
tive conception that goes beyond individual factors and that also analyzes environmental factors that can affect the appearance of anxiety or depression, would be necessary.This change to integrative approximation is both conceptualexplaining why certain psychopharmacological agents and psychological therapies are effective in different conditionsand applied -providing a common target of intervention if the mechanisms underlying several disorders were found - (Krueger & Eaton, 2015).
Therefore, it can be said that another group of factors to take into account in mental health are the sociocultural ones.Clinicians and researchers agree on the importance of sociocultural factors in health.Bi-directional effects exist between health outcomes and these types of factors (Zvolensky & Leventhal, 2016).Although in other types of disorders, such as eating disorders, the impact of sociocultural factors has been more studied, the field of emotional disorders has yet to be investigated as much.
To our knowledge, there are previous reviews that examined risk factors focused on young people (e.g., Lynch et al., 2021), on a specific transdiagnostic variable (e.g., Rosser, 2019), or specific diagnoses (e.g., Zimmermann et al., 2020).However, there are still no systematic review studies that have jointly analyzed the environmental, biological, and psychological transdiagnostic risk factors in adults with anxiety or depressive disorders.A systematic review of the transdiagnostic risk factors related to anxiety or depressive disorders allows to: 1. Underscore the risk factors for developing anxiety or depressive disorders, 2. Provide recommendations for designing transdiagnostic measures that evaluate the risk, as well as the nature of anxiety or depressive disorders and 3. Highlight future directions for transdiagnostic therapy of both anxiety and depression.
Therefore, this systematic review aims to search for evidence on transdiagnostic risk factors for anxiety and depression in the clinical population diagnosed with these psychopathological conditions by analyzing the different types or categories of factors identified.

Method
A systematic review was registered in PROSPERO (International Prospective Register of Systematic Reviews) under number CRD42022370327.It was designed by the PRISMA-P guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols; Page et al., 2021) to identify studies that examined the risk factors for depressive and anxious emotional disorders.The eligibility criteria were developed using the Population Exposure Comparator Outcome (PECO) framework.Empirical studies were included if they met the following criteria: 1. Participants were adults (beyond 18 years old) diagnosed with anxiety and depression in a clinical sample.
2. They examined any transdiagnostic risk factors variable, such as psychological, biological, and sociocultural characteristics, and their association with the diagnosis.3. Studies were not required to have a comparison group.4. To ensure the quality of the empirical studies detected, they were required to be peer-reviewed.
Articles were excluded if the sample did not fit the intended group of participants or the studies did not report peer-reviewed, original empirical findings, such as reviews, opinion pieces, and conference abstracts.For inclusion, no constraints were placed on the publication date in this review.For a complete list of inclusion and exclusion criteria for identifying relevant literature in this review, see Table 1.The following terms were entered into the database: "emotional disorder" OR "mood disorder" OR (anxiety AND depression) AND transdiagnostic AND "risk factor" OR predict* OR caus* OR vulnerab* OR predispos* OR susceptib* OR perfectionist OR rumination OR "negative affective" OR "anxiety sensitivity" OR "intolerance of uncertainty" OR neuroticism (see Appendix 1).Search strategies were adapted to each database (see Appendix 2).Hand-searching was used since research is supported by a hand-searching effort to complement database searches (Vassar et al., 2016).However, all the articles found with the manual search already appeared in the electronic search, so that no new ones could be included.
The last search was in December 2022, which yielded 2,307 studies, and 1,138 remained after removing duplicates.Search results were imported into Rayyan for screening  (Ouzzani et al., 2016).After the removal of duplicates, all titles and abstracts were screened independently by two reviewers.Disagreements at each screening stage were resolved through discussion between the two screening authors.

Data selection
Two independent reviewers with knowledge of the field assessed the study quality to reduce possible bias.The selection was performed in Rayyan based on the checklists from the Joanna Briggs Institute, explicitly using the Checklist for Analytical Cross-Sectional Studies for cross-sectional studies and Checklist for Cohort Studies for longitudinal studies.These checklists are tools designed to help researchers and practitioners critically appraise different research studies.Once studies were selected, researchers used the JBI Checklist (Moola et al., 2020) to appraise each critically included study.The checklist consists of specific questions relevant to different study designs (e.g., cross-sectional studies).Researchers evaluated each checklist item individually, and at the end of the checklist is the final decision on whether to include it, not to include it, or whether additional information is needed.After developing each checklist individually, the two reviewers met and decided whether or not to include each item.When there was uncertainty about the interpretation, it was resolved by discussion between the authors based on the checklist items.Cohen's kappa (Cohen, 1968) was calculated to compare the degree of agreement among the reviewers.The minimum overall agreement score (Cohen's kappa) between the two raters was .88,and the conflicting results were discussed until 100% agreement was reached in the evaluated cases.Thus, the selection of the articles was carried out in two phases.In the first, the title and abstract were examined, obtaining a score of .91 on Cohen's kappa, and in the second, the full text of those studies that had passed the first screening was reviewed, obtaining a score of .88 on Cohen's kappa.On the other hand, to study the quality of the included studies, the method used by Hoppen and Chalder (2018) was followed, according to which the number of items rated 'yes' was summed and then divided by the number of the maximum possible number of 'yes' ratings and, to find the percentage, this amount is multiplied by 100.

Figure 1
Summary of the study selection process.
Note.The search output used: "emotional disorder" OR "mood disorder" OR (anxiety AND depression) AND transdiagnostic AND "risk factor" OR predict* OR caus* OR vulnerab* OR predispos* OR susceptib* OR perfectionist OR rumination OR "negative affective" OR "anxiety sensitivity" OR "intolerance of uncertainty" OR neuroticism

Data extraction
At this stage, two reviewers independently obtained the information necessary to answer the research question posed in the study's objective from the selected studies.The information collected in the data collection form is detailed in Tables 3-5.In addition, the risk factors of anxiety or depressive disorders were analyzed according to the type of risk factors: psychological, biological, or sociocultural risk factors.

Results
The literature search in this review comprised 53 articles from 2011 to 2022.For an overview of the study selection process, see Figure 1.The studies included are presented in Tables 3 to 5.
The overall quality of the included studies was high, with a mean rating of 91.91% across all studies.Longitudinal studies demonstrated slightly higher quality with an average score of 93.39%, compared to 91.52% for cross-sectional ones (see Tables 3 to 5).Separating the quality of the three main types of risk factors, the average quality of the psychological risk factors was 91.23% of the biological risk factors was 94.44%, and of the cultural risk factors was 93.86%.The increased of BMI was associated with depression, suicide risk, and lower wellbeing.
Increased of waist circumference was associated with worse interpersonal functioning.
An increase in waist circumference was associated with a decrease in anxiety.

Psychological risk factors
Most studies (n = 39, 73.58%) examined psychological risk factors.To present the results in a structured way, they were grouped into six categories: cognitive processes, negative affect and neuroticism, intolerance of uncertainty, anxiety sensitivity, coping strategies, and other variables.

Biological risk factors
Nine (16.98%) articles examined the cross-sectional biological risk factors, with high average quality.The amplitude of brain waves was confirmed in three (5.66%)articles (Bedwell et al., 2015;Macedo et al., 2021;MacNamara et al., 2016) and the activation of brain regions were examined in four (7.54%) (Feurer et al., 2021;Park et al., 2021;Radoman et al., 2019;Sheena et al., 2021) and demonstrated that was a transdiagnostic factor.Only one (1.89%)publication analyzed and found Respiratory Sinus Arrhythmia (RSA) responsivity as a transdiagnostic measure (Kircanski et al., 2016), and the last one (Yusufov et al., 2021) associated the BMI with depression, suicide risk, and lower well-being.However, despite being the type of factors with the highest average quality, further study is needed to determine their transdiagnostic status.

Sociocultural risk factors
The less studied category was sociocultural risk factors (n = 5, 9.43%) publications.The low frequency of these studies also prevents us from drawing firm conclusions regarding their transdiagnostic status.This category is the most heterogeneous, and it groups childhood adversities (Dragan &Kowalski, 2020), gender, age, andmarital status (Muñoz-Navarro et al., 2021); the socialization process (Swerdlow et al., 2021;Taylor et al., 2020) and physical activity (Tull et al., 2018).When people suffered childhood adversities, the risk of being diagnosed with anxiety or depressive disorders was higher than people who did not suffer from them (Dragan & Kowalski, 2020).In addition, being a woman, single (instead of being married), having lower studies (Muñoz-Navarro et al., 2021), and being sedentary (Tull et al., 2018) are risk factors for anxiety or depressive disorders.The same study found that young adults (age 18-25) and adults (age 26-39) presented more anxiety or depressive disorders than older people.Moving on to the socialization process, feeling less socially connected was associated with diminished life satisfaction beyond clinical symptom severity (Taylor et al., 2020).Whereas depressive symptoms were related to perceived social quality, anxious ones were more tied to reported social behavior (Swerdlow et al., 2021).

Discussion
A systematic process outlined by the PRISMA guidelines (Page et al., 2021) was employed to identify all transdiagnostic risk factors for anxiety and depressive disorders in the clinical adult population.A variety of psychological, biological, and sociocultural factors emerged from the reviewed publications that appear to be transdiagnostically relevant in adults diagnosed with anxiety and depressive disorders.The results of this systematic review highlight various transdiagnostic factors that are markers of preventive or treatment interventions and promising future lines of investigation.
This review indicated that the most investigated transdiagnostic risk factors were psychological variables.Neuroticism or negative affectivity has shown its predictive power in the appearance of anxiety or depressive disorders, being a general factor in the transdiagnostic model (Paulus et al., 2015).It could represent the general distress shared in internalizing disorders.However, it works differently depending on whether it is a depressive disorder, an anxiety disorder, or a mixed (anxiety and depressive) disorder (Merino et al., 2016).These differences could be due to the links with cognitive processes such as rumination, repetitive negative thinking, attentional control, and worry that appear relevant in the development of anxiety and depressive disorders.Although they are cognitive processes, many studies have shown that ruminations are associated with depressive disorders (Hong & Cheung, 2015;Spinhoven et al., 2018;Strujis et al., 2018), whereas worry is a better predictor of general anxiety disorder (Spinhoven et al., 2015;Strujis et al., 2018).However, Kircanski et al. (2015) have shown that they both are transdiagnostic constructs between anxiety and depressive disorders.Thus, due to the correlation of these variables (Silveira et al., 2020), it could be integrated into a transdiagnostic dimension, such as repetitive negative thinking (Spinhoven et al., 2015;2018).Moving on to other cognitive variables, focusing (Hunt et al., 2022) and reflection (McEvoy & Brand, 2013) could be better predictors of depressive disorders.
The second and third most studied transdiagnostic variables were intolerance of uncertainty and anxiety sensitivity.
Although the meta-analytic structural equation model carried out by Hong and Cheung (2015) found anxiety sensitivity and intolerance of uncertainty as anxiety-specific vulnerability, in this review it has been found that such variables are transdiagnostic and, therefore, would correspond to a shared vulnerability between the diagnoses of anxiety and depression (Paulus et al., 2015;Toro-Tobar et al., 2019;2020) even after controlling for neuroticism (McEvoy & Mahoney, 2011;2012Naragon-Gainey & Watson, 2016).This result aligns with previous investigations where this construct was essential to the genesis of internalizing disorders (Griffith et al., 2010), among which are the disorders studied in this systematic review (Krueger & Eaton, 2015).It would be essential to investigate other transdiagnostic variables with less evidence, such as emotional regulation or experiential avoidance, which could explain the efficacy of the same treatment for anxious and depressive disorders and the development of transdiagnostic protocols as an effective and more efficient treatment alternative (Aguilera-Martín et al., 2022).
The second type of variable most investigated according to this review was biological ones, where the studies' average quality was excellent (94.44%).However, more research would be needed to have more empirical evidence.All biological studies are cross-sectional, so longitudinal studies could also help gain empirical weight.The typical way to examine the biological correlate is to evaluate the neurobiological waves or activation regions.The amplitude of waves or the amygdala activation was a transdiagnostic biological finding.However, other measures, such as RSA, BMI and waist circumference, require more investigation because there was only one article for each measure.Although these factors have been investigated more than sociocultural factors, epigenetics has demonstrated how the environment can produce changes in our body, which can affect mental health (Cecil et al., 2022).Thus, it has been shown how situations of severe stress produce changes in brain waves (Jin et al., 2021) or how child abuse produces structural, biochemical, and functional changes in the brain (Coley et al., 2021;Glaser, 2000); therefore, studies should also evaluate contextual dimensions that can promote these brain changes.
Among the factors investigated by studies in the present review, some underexplored factors warrant further investigation.Sociocultural were the less investigated.There are only four publications, and all the categories except the process of socialization, which was investigated twice, have only been investigated in one publication.In addition, this systematic review does not show positive results for emotional regulation, cognitive biases, or metacognition.It could be due to the studies that investigate these variables not using the inclusion variables that this systematic review has analyzed.
This systematic review also has limitations to be taken into account.It is important to note that the results of this review should be interpreted with caution due to several methodological complexities and uncertainties.First, the studies varied considerably in terms of transdiagnostic measures, and many of them were in self-report format, so despite having found considerable consistency in some factors, the diversity in the measures of psychopathology included in the studies makes it difficult to draw unified conclusions.
Secondly, it is essential to consider that approximately 80% of the studies analyzed are cross-sectional and that almost half of them were conducted in the United States, which makes it difficult to determine causality and generalization of the results obtained.On the other hand, the inclusion criteria applied in this review meant that only peerreviewed publications were used.Although this measure has been taken to guarantee the quality of the studies, bias may occur due to the file drawer problem.
Furthermore, future lines of investigation could focus on the study of those less investigated variables, for example, sociocultural factors, to obtain more excellent scientific evidence regarding the common factors of the different diagnoses.On the other hand, future studies could make an empirical model that can explain the standard part of anxiety and depressive disorders that could be attributed to transdiagnostic variables and specific variance that could differentiate between the specific disorders.These findings can also be extrapolated to improve assessments of anxiety or depressive disorders in adults through the need for dimensional assessments that take into account multiple constructs, such as the Multidimensional Emotional Disorders Inventory (MEDI) (Osma et al., 2021;Rosellini & Brown, 2019) or the need of use transdiagnostic interventions, such as the Unified Protocol (Barlow et al., 2018) to intervene on the vulnerability and maintenance mechanisms shared by all emotional disorders.In addition, the intervention in transdiagnostic variables can be more efficient since it allows its application in group treatment format and can be used in several diagnoses (Aguilera-Martín et al., 2022;Peris-Baquero et al., 2022).Finally, integrated and multidisciplinary interventions will also have to be investigated and carried out to reduce the symptomatology of people suffering from anxiety or depressive disorders and improve their quality of life.

Conclusion
To our knowledge, this is the first systematic review of transdiagnostic risk in patients diagnosed with anxiety or depressive disorders in adult samples.After reviewing several risk factors in high-quality studies, it cannot be stated with certainty that all of them are transdiagnostic risk factors for anxiety or depressive disorders.While some psychological factors such as cognitive processes, neuroticism, negative affective rates, intolerance of uncertainty, and anxiety sensitivity have demonstrated their transdiagnostic nature, the remaining variables need further investigation to reach a solid conclusion.Another significant result was the most relevant predictors in each disorder.Thus, it was intolerance to uncertainty for generalized anxiety disorder, social anxiety disorder, obsessive-compulsive disorder, and major depressive disorder, as well as anxiety sensitivity for panic disorder.In the case of posttraumatic stress disorder, both anxiety sensitivity and intolerance to uncertainty were the best variables.On the other hand, despite the transdiagnostic model advocated in this paper, an attempt has been made to synthesize which variables have been the most investigated for each disorder to guide future research.
These findings have important implications for prevention and intervention.Improving emotion regulation and self-regulation and reducing environmental conditions that foster stressful life events may be particularly salient targets for the prevention and intervention of general and specific dimensions of psychopathology.In addition, this publication allows a stronger foundation of knowledge for how to build a model for anxiety or depressive disorders using transdiagnostic and specific risk factors that could help to understand the development of psychopathology and how to prevent it.In addition, this review identified several methodological concerns that should be addressed in future research.In particular, there is a fundamental need for more comprehensive, longitudinal, and multidisciplinary studies to establish further causal relationships.Conducting such studies could result in a more substantial knowledge base that will drive the identification of robust relationships between transdiagnostic risk and mental disorders, which can facilitate the development of empirically supported approaches to prevention and intervention.

Table 1
Inclusion and exclusion criteria

Search strategy and selection criteria Literature
searches were performed in Scopus, Dialnet, and Web of Science [WoS: All databases, specifically WoS Core Collection, Current Contents Connect, Derwent Innovations Index, KCI Korean Journal Database, MEDLINE, Russian Science Citation Index, and SciELO Citation Index].

Table 2
Classification of the articles

Table 3
Studies that evaluated the psychological risk factors(n = 39; 73.58 % of the studies)