Abstract
Social media use is ubiquitous to the lives of children and adolescents. The body of research investigating its potential impact on mental health has risen exponentially. We systematically reviewed the present literature exploring potential linkages between social media use and symptoms of depression and anxiety in this vulnerable group. Using the Preferred Reporting Items for Systematic Reviews and Meta-analyses framework, articles were searched across Medline, EMBASE, CINAHL, and PsycINFO databases from inception to February 2024. Quantitative studies with social media as exposure and anxiety/depressive symptoms as outcomes in children and adolescents 5–18 years of age were included. Of the 4850 studies retrieved, 67 fulfilled the inclusion criteria. The most frequent measures of social media were “time spent on social media” and “frequency of use.” Depressive symptoms were the outcome of 61 studies, whereas anxiety was measured in 27 studies. Most studies were of fair quality (n = 53). A meta-analysis was not possible due to study heterogeneity. Our review shows that (1) problematic social media use is associated with depressive and anxiety symptoms among children and adolescents, (2) duration of social media use was more consistently linked with anxiety and depression in girls compared with boys, and (3) mediating and moderating mechanisms were sleep deprivation, social comparison, and feedback-seeking behaviors, exercise, social support, and type of social media use. Qualitative work and robust large-scale longitudinal observations using a person-specific approach are needed to further our understanding of the impact of social media use on depression and anxiety in children and adolescents.
Introduction
Social media platforms such as TikTok, Instagram, and Snapchat are now integral to daily life, particularly for young people. 1 Adolescents spend hours online daily, with almost 45% in the United States “constantly online”.2,3 Average teen online time has doubled since 2015.4,5 Membership of social networks offers evolving virtual profiles and novel opportunities for experimentation.6,7
Childhood and adolescence are vulnerable periods of rapid biological, psychological, and cognitive development. 8 Identity is being negotiated in relational, familial, sociocultural, and now digital contexts amid educational challenges and formation of career goals. 9 There is a high risk of developing low self-esteem 10 and mood disorders. 11 Depressive symptoms seem to increase from early to mid-adolescence sharply.12–14 Depressive disorders are the third leading cause of disability in young people from Western Europe, closely followed by anxiety disorders. 15
Social media’s impact on the mental health of children and adolescents is a growing concern.16,17 Studies have found a significant association between time spent on social media and depression and anxiety.18–21 Indeed, a meta-analysis showed that adolescents with problematic social media use are at a higher risk of developing depression.23,24
A grounded, practically useful, and clinically relevant understanding demands a deeper look into not only a myriad of potential mediating and moderating variables (such as sleep deprivation, gender, support mechanisms, and personality tendencies) but also overcoming research challenges in capturing the quality and content of digital social interactions. 18 Social media platforms are complex tools for social connectedness and support on the one hand 25 but also constant exposure to addictive tendencies, 26 cyberbullying, and identity conflicts on the other. It is not surprising therefore that users might experience psychological distress leading to mental illness. 27
This review presents a systematic narrative synthesis of studies which have reported on the relationship between social media and depression and anxiety among children and adolescents. Previous systematic reviews and meta-analyses have focused on adolescents alone, except one by Piteo and Ward, 22 which reviewed publications up to March 2019. The social media landscape is evolving rapidly. By bringing our understanding of the literature pertaining to social media’s potential impact on depression and anxiety among children and adolescents up to date, this review fills a current gap and provides an important contribution.
Materials and Methods
Protocol and registration
This systematic review follows the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines, and the protocol is registered on Prospero (Registration number: CRD42020199664).
Eligibility (inclusion and exclusion) criteria
All studies with participants aged 18 years or below were included where the “exposure” was social media use, and the “primary outcome” measured was depression and/or anxiety. Quantitative studies (randomized controlled trials, open-label trials, cohort studies, case–control studies, and cross-sectional studies) published in peer-reviewed journals in English were included if full texts were available. Studies with participants over 18 years were included only if separate analyses on participants under 18 years were presented.
Studies investigating outcomes, such as cyberbullying, cybervictimization, online harassment, suicidality and self-harm, body image, conduct disorder, and life satisfaction, were excluded to prevent their confounding effects on the relationship between social networking site (SNS) use and depression/anxiety. These outcomes, while important, could make it difficult to isolate the specific effects of SNS use on depression and anxiety. In addition, these factors encompass complex psychosocial elements that fall outside the scope of our study’s primary focus. By excluding them, we aimed to maintain a more focused and consistent approach to analyzing the relationship between SNS use and depression/anxiety. Similarly, studies measuring overall media consumption (e.g., online gaming and television) or smartphone use for academic purposes were excluded unless they explicitly measured SNS use as a distinct variable. Qualitative studies and articles not in English were also excluded.
Search strategy
Four electronic databases (Medline, PsycINFO, EMBASE, and CINAHL) were searched using a comprehensive strategy developed by the authors. This included subject headings and keywords relevant to each database, with various combinations of terms related to the participant population, outcome, and exposure [e.g., (Child*) AND (facebook) AND (depress*)]. Table 1 lists all combination terms searched, and search histories are available as a Supplementary Data.
Search Terms and Linkage
Search terms in all the above categories were selected carefully by authors in meetings with the Institution’s librarian familiar with the search databases and software for compilation of the searches. Search history can be found in Supplementary Data provided. The corresponding author can be contacted for more details on the search methodology.
Eligible articles were those published in English from database inception up to February 2024. Reference lists of prior systematic reviews on similar topics were also examined to identify additional relevant papers.
Data extraction
An initial list of studies was identified using a keyword search strategy. Articles retrieved from databases were collated using EndNote X9. After deleting duplicates, the authors (NB, PY, and SY) independently screened titles and abstracts to identify studies meeting the inclusion criteria. If necessary, full texts were reviewed. Disagreements were resolved through discussion. Key information from each study was extracted and tabulated separately for cross-sectional and longitudinal studies. Examples of tabulated information include the author (publication year), country/study setting, study population, social media variables (exposure), outcome measures, key findings, etc. (Tables 2 and 3).
Summary of Cross-Sectional Studies (n = 41)
Summary of Longitudinal Studies (n = 26)
Quality assessment
Study quality and risk of bias were assessed using an adapted Newcastle-Ottawa scale for cross-sectional studies 27 and the Newcastle-Ottawa scale for cohort studies. 28 Studies are rated on a scale of 0–10 based on (1) participant selection (representativeness of the sample, sample size, how exposure variable was measured and nonrespondent details), (2) comparability of subjects in different outcome groups (whether confounding variables were accounted for), and (3) assessment measures used for outcome variables and adequacy of follow-up (in case of longitudinal studies).
Quality assessment was done in two phases, with each phase performed by the authors in pairs. Disagreements were resolved by a third author. The resulting studies were tabulated in Excel.
Data analyses
The primary outcome of interest was depression and/or anxiety symptoms or disorder or diagnosis. The potential relationship between these outcomes and different measures of social media use among children and adolescents was explored using narrative synthesis. Because studies varied greatly in sample characteristics, social media measures, and outcome measures (depressive and/or anxiety symptoms) used, a meta-analysis could not be performed.
Results
The initial search yielded 4850 studies. After removal of duplicates (n = 1535), 3315 studies remained. Following the screening of titles and abstracts, 211 studies were found of potential interest. Upon review of full texts, 68 studies were eligible for the review. The PRISMA diagram included in this report (Figure 1) provides further details on reasons for exclusion at each stage.
PRISMA diagram of the study selection process. This Figure illustrates the PRISMA checklist process through which we refined our systematic search and excluded studies not fulfilling our inclusion criteria. All articles were identified through the research databases mentioned in the article. Sixty-seven studies were finalized for the narrative review. Reasons for exclusion of the remaining studies are detailed in the diagram. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-analyses.
Study characteristics
Sample sizes varied from 30 hospital-based clinic attendees 29 to 388,275 participants of a large community-based sample enrolled for the “Monitoring The Future” surveys. 30 Most studies (n = 40) focused on adolescents (12 years and above), whereas four studies exclusively targeting the pre-adolescent group. Two papers included participants above 18 years of age but were deemed suitable for this review because in one study most participants were aged 18 years or under, 32 and in another, outcomes were reported separately for an under-18 group. 33
The regional distribution of countries represented within the literature covered Europe (n = 21), North America (n = 20), Asia (n = 23), and Australia (n = 3). Most (n = 39) studies were undertaken in school settings. Secondary databases, such as the UK Millenium Cohort Study, were analyzed in many papers reviewed (n = 18). Others sampled the general population (n = 7) or used hospital-based settings (n = 6).
Table 2 and Table 3 provide details of the included cross-sectional studies (n = 41) and longitudinal studies (n = 26). Table 4 gives a summary of the findings. One study used secondary data from a randomized controlled trial. 34
Summary of Findings
Key to Table Columns: (1) Social Media Exposure Measured: The variable measured for social media exposure; (2) Direction of Effect: Whether the effect is positive, negative, or neutral based on the overall findings; (3) Key Findings: A brief summary of the key results for the outcome; (4) Strength of Evidence/Consistency: Overall quality of evidence for the outcome (e.g., high, moderate, and low) and whether the findings are consistent across studies or there is variation.
Study quality
Overall, the quality of the studies was fair (see Tables 2 and 3). Thirteen studies were deemed to be good quality and 53 were of fair quality and one study of poor quality. A cross-sectional design was used in 59.7% studies (n = 40). Longitudinal studies followed participants for variable time spans (ranging from 2 months to 5 years). Several did not account for lost-to-follow-up cases (n = 8). Self-report measures were used in most studies that are prone to recall and response bias. Response rates were reported in some studies only. Three studies did not report gender distribution of the studied sample.33,35,36 Sample size was not estimated and/or inadequate for some studies.29,37,38 Others did not consider important co-variates for analyses such as socioeconomic status or presence of mental illnesses. 39 One study adopted within-subjects analyses by using an individual fixed-effects model for assessing the association between social media and emotional disorders. 40
Outcome measures
The outcome measured by most studies was depression (n = 61). Anxiety symptoms were measured within 27 studies. Among those measuring depression, all except three studies41,42,43 adopted validated scales. The Center for Epidemiological Studies—Depression Scale was most used, followed by the Short Mood and Feelings Questionnaire. Scales used for assessing anxiety symptoms and social anxiety were Screen for Child Anxiety Related Emotional Disorders (SCARED), Kiddie Schedule for Affective Disorders and Schizophrenia, and Multidimensional Anxiety Scale for Children.
Assessment of social media use
Exposure to social media was assessed using various parameters (Figure 2). Most studies measured the duration of social media use (n = 33), calculated as self-reported time spent on social media platforms in a day or a week, with some reporting weekend/weekday patterns. Frequency of use in a typical week was also recorded (n = 18). Two studies measured “Intensity of social network usage,” whereas nine studies focused on or compared the type of social media platform(s). Four studies measured active and/or passive use of social media platforms. Importance given by participants to digital interactions was observed in two studies. “Problematic Social/Digital Media Use” was assessed in 17 studies, whereas addictive use was measured through different scales in eight studies. Four studies recorded social comparisons carried out by participants on social media.
Social media use measures used across included studies. This Figure shows a horizontal bar chart of the different social media-related parameters considered as “exposure” in the reviewed studies. The Y-axis represents the number of studies measuring each of these parameters. The most commonly used parameters were duration and frequency of social media use.
All studies relied on Likert-based self-report data except four studies; three of these used “time use diaries” to measure time spent on social media,44,45,101 and one study directly measured the immediate biological and psychological effect of social media exposure on a group of adolescents. 29
Main findings of the review
Is social media use associated with depressive symptoms?
Large nationally representative cross-sectional population-based good-quality data30,42,44–47,49 conclude upon a significant association between time spent on SNSs and the risk of depressive symptoms among adolescents. Daily social media use increased the likelihood of depressive symptoms in adolescents by 13%. 30 This association was significant after adjusting for age, ethnicity, socioeconomic status, and other family-related variables. School-based surveys36,41,43,50,51–56 also reported a positive association. Clinically depressed adolescents 57 spent more time on social media sites compared with their healthy counterparts. Boers et al. 34 followed adolescents in an alcohol/drug prevention program over four years, finding significant between-persons and within-persons positive correlation with regard to time spent on social media and depressive symptoms. Kandola 102 also found that participants using social media for most days of the week at age 11 had 13% higher depression scores three years later compared with less frequent users.
Gender appeared to moderate this link, with stronger effects for girls than boys. Twenge and Farley 46 observed that as little as 2 to 4 hours of daily social media use increased the risk of depression in girls, while 5 or more hours of use was linked to depression scores in boys. Kelly et al. 47 also reported similar results, with 3–5 hours and >5 hours of use increasing depression scores by 26% (vs. boys 21%) and 50% (vs. boys 35%), respectively. Barthorpe et al. 44 and McAllister et al. 45 using the same data (Millenium Birth Cohort, year 2015) found an association for girls but not boys.
Some studies found no correlation35,58–61 between social media use and depression. Other variables, such as sleep disturbance,47,48 exercise, 41 and perceived social support, 39 may mediate the relationship. For instance, sleep disturbance has been found to mediate the link between social media duration and depression. In addition, psychological security online mediated the connection between mobile social media use and depressive symptoms. 62
Problematic social media use (PSMU), addictive use and depression
Several studies58,63,64 focusing on problematic social media use found a modest effect and observed bidirectionality in association with depressive symptoms. Li et al.58,64 found a significant association in Chinese primary school students between online social networking addiction (OSNA) and depression. Nondepressed adolescents at baseline classified as “persistent OSNA” or “emerging OSNA” had higher risk of developing depression compared with those who had no OSNA; also participants depressed at baseline had increased risk of developing OSNA at follow-up. Similarly, a bidirectional association was seen in Estonian students between increases in problematic social media use (PSMU) and depressive symptoms, irrespective of gender.31,65 Other studies also revealed strong positive associations.66,67,68 Children diagnosed with ADHD and having problematic digital media use had significantly higher depression scores 69 compared with those who did not have PSMU.
Certain new social media platforms are more interactive and video based, such as Tiktok. A recent study 70 established that “addictive users” exhibited higher levels of depression and anxiety compared with “moderate users” and “nonusers.” “Moderate use” and “no use” were not linked to mental health and family environments were no different.
Quality of social media engagement and depressive symptoms
Studies that focused on quality of social media engagement found a direct link with increased depressive symptoms. First, passive social media use in Vietnamese high-school students was linked to depression scores. 71 Active social media use led to decreased depressive symptoms in girls only. 72 Second, social comparison and feedback seeking through social media use were positively associated with depression symptoms.73,74 Negative social comparison fully mediated the relationship between Qzone usage and depression in Chinese adolescents. 51 Third, emotional investment and perceived importance of social media were linked to depression. Depression scores were negatively associated with bridging subscale scores and positively associated with identity subscale scores for both boys and girls.75,76 Greater positive emotional responses were linked with higher depression scores a year later, more likely in girls than boys. 77
Is social media use linked to anxiety symptoms?
Population-based samples reported a significant correlation of SNS use with anxiety symptoms. For example, Rutter et al. 78 discovered that the frequency of checking and posting on Facebook was significantly correlated with anxiety. Oliva et al. 79 found regression estimates for anxiety scores to be significant when participants spent 2 or more hours daily on social media sites, and a small sample 80 also reported direct association with hours spent on social media in youth of color irrespective of racial discrimination and intergroup social contact. Studies in psychiatric outpatient clinics found adolescents diagnosed with generalized anxiety 32 or internalizing disorders 38 spent more time on social media sites compared with their healthy counterparts.
Across the literature, pattern of social media use seemed to differ among adolescents who developed anxiety. Night-time use in secondary school adolescents from Scotland 36 was associated with higher levels of anxiety. Thorisdottir et al. 81 also found that passive social media use (consuming information online with minimal effort) was associated with increased symptoms of anxiety after controlling for time spent on social media in a school-based group of Icelandic adolescents. A similar finding was reported by Fredrick, Nickerson, and Livingston 72 but for girls alone. Charmaraman et al. 82 measured social anxiety in a sample of American school students and could not find any association with positive (online support seeking and education) and negative online behaviors. Other confounding variables not measured could be the reason.
The type of social media platform used also affects the presence of anxiety symptoms. Iranian children using WhatsApp and Viber reported lower anxiety levels than users of other social messaging apps such as Instagram and Facebook. 33 Muzaffar and colleagues 32 focused on one platform, Facebook, finding generalized anxiety disorder was associated with increased and repetitive Facebook behaviors. However, no link was established with social anxiety, depression, or distress. Banjanin et al. 58 also studied Facebook-related behaviors in a subsample and found no association of number of friends or frequency of posting/sharing with anxiety scores.
PSMU and anxiety
PSMU usage was associated with anxiety symptoms in children and adolescents. Yıldız Durak 83 and Kılınçel and Muratdağı 84 studied social anxiety symptoms and general anxiety symptoms, respectively, through online school-based surveys in Turkey and found significant correlation with PSMU and social media addiction, respectively. Kim, Oh, and Huh 85 used indigenously developed and locally validated tools called “SNS Addiction Tendency” and “SNS Fatigue” and their relationship to anxiety symptoms in South Korean school-going adolescents. Addiction tendency and fatigue were both positively correlated with anxiety. Similarly, Chen et al. 67 and Chen et al. 68 assessed PSMU in Chinese pre-adolescents cross-sectionally and longitudinally to find a strong association with DASS scale scores. Among a sample of children diagnosed with ADHD, Shuai et al. 69 found those with problematic digital media use to have significantly higher scores on screening child anxiety-related emotional disorders.
Quality of social media engagement and anxiety symptoms
Emotional investment in social media platforms was linked to anxiety symptoms. Woods and Scott 36 found that Scottish adolescents showing emotional investment in SNS use had poorer sleep quality and higher levels of anxiety. Winstone 86 categorized types of social media users into high communicators, moderate communicators, broadcasters, and minimal. Broadcasters were at higher risk of depression, anxiety, and self-harm compared with other user types after adjusting for co-variates within sociodemographic profiles. Gingras 87 observed that active use of social media is also linked to anxiety, particularly for individuals with low or moderate extraversion traits.
Discussion
The past two decades have seen a staggering rise in studies looking into the effect of social media use on mental health. In this review, we have focused on understanding the relationship between different social media parameters and symptoms of depression and anxiety that are becoming increasingly prevalent in young people (children and adolescents). 93
Previous systematic reviews and meta-analyses have concluded upon a small to moderate effect size20,22,23 with cautious interpretation because most studies were cross-sectional, and heterogeneity of studies was high. A meta-analysis 23 also demonstrated a dose–response relationship between duration of use and emotional symptoms. Our review shows that (1) problematic social media use is associated with depressive and anxiety symptoms among children and adolescents, (2) duration of social media use was consistently linked with this outcome in girls in particular, and (3) mediating and moderating effects of sleep deprivation, social comparison and feedback-seeking behaviors, exercise, social support, and type of social media use were observed.
Problematic social media use and linkages to depression and anxiety
Problematic social media use signifies the experience of addiction-like symptoms with social media use. Studies in our review consistently reported a positive association between “problematic social media use,” “social media addiction,” or “OSNA” with symptoms of depression and anxiety. Longitudinal observations have revealed a bidirectional association. 90 Shannon and colleagues 94 conducted a meta-analysis and concluded there is a moderate effect size for anxiety, depression, and stress in adolescents and young adults. This addiction-like pattern is therefore important for clinicians to assess in child and adolescent psychiatric clinics. Further research on problematic social media use needs to explore it as a concept distinct from overuse. 95 Therefore, scales that measure motivation, behaviors, and impact on different areas of life should be used. Moreover, specific unifying terminologies such as “social media misuse” when describing problematic use may enable more reliable comparison.
Consideration of mediators and moderators
Several mediating factors have been considered when examining the connection between social media use and mental well-being.96,97 In our review, we found the mediating role of sleep,47,48 social comparison, 51 physical activity, 41 and family/friends support. 39 Future studies must consider these effects when investigating social media with emotional symptoms.
Being female was a significant moderator in the relationship between time spent on social media and emotional distress symptoms.30,44 One possible explanation for this could be that girls engage more in social comparison behaviors, 98 are more prone to cyberbullying,99,100 and are more likely to develop internalizing problems. However, more research is needed to understand sex differences.
The challenges of defining exposure to social media
Evaluating social media exposure empirically can be challenging. Time spent on social media and frequency of use were common measures but are crude markers at best. Active use of social networking sites revealed mixed results for depression,72,82 whereas passive use increased depressive symptoms after controlling for duration of use. 81 Night-time specific use was linked to depression through sleep disruption, as it is not uncommon for teenagers to spend the night scrolling SNSs with an inability to stop—a phenomenon called “vamping.” 91 Furthermore, the level of depressive symptoms depends on the type of social media platform. 33 Facebook-related behaviors particularly (number of postings, number of friends, etc.) and their association with emotional distress78,92 and with generalized anxiety disorder 32 were noted.
Comparing the same type of activities across platforms to understand any platform-specific effects is essential as well. Li et al. 88 developed the concept of “Social Networking Use Intensity” and found some evidence of an association between depression and Social Networking Use Intensity in Chinese student cohorts.39,89 Similar studies in other populations are needed. Neira and Barber 50 and Woods and Scott 36 measured the degree to which participants were emotionally invested in social media sites. But this construct needs to be explored qualitatively before operationalizing it for further research, 19 and the differences between social media intensity, overuse, and misuse need to be explored.
Shifting focus to social media content engagement and quality
Recent studies have started to move beyond the numeric and capture the quality of social media engagements.86,87 An hour spent passively scrolling through one’s Instagram account may have a different psychological impact to an hour of intimate conversation with a friend or communicating political views on a public forum. Valkenburg and colleagues 18 suggest that mock SM sites be used to measure engagement or software that allows for screen monitoring through multiple screenshot images, etc. Rodriguez, Aalbers, and McNally 103 went beyond group-level comparison of active and passive social media use to understand individual-level variations within the relationship between social media and depressive symptoms. They did this using a “person-specific approach” that treats individuals as complex integrative systems and uses multiple observations. There was significant variability between individuals in the strength and kind of relationship between social media use and depressive symptoms. Such approaches are necessary not only for the identification of high-risk individuals but also for tailoring therapeutic interventions accordingly.
Limitations and recommendations for future research
Several limitations need to be considered: (1) Half of the studies were cross sectional, hence limiting our ability to draw causal linkages; (2) most studies did not use randomized sampling, limiting generalizability; and (3) overall quality of studies was fair, with longitudinal studies lacking proper control groups. Future studies must use a longitudinal design with sufficient follow-up times, representative large samples, and comparable nonexposed control groups. Another limitation was reliance on self-report social media measures prone to bias. Future studies need to incorporate multiple objective assessment methods to enhance the quality of findings.
Our decision to include only participants below 18 years excluded some important articles. This limit was observed because adolescents’ developmental challenges, coping strategies, preoccupations, and aspirations differ from those of emerging adults. It is hoped that researchers consider this in future work. Moreover, future studies would benefit from consistent terminology for social media misuse.
It is recommended that studies adopt a mixed-methods approach, combining qualitative insights into user experiences with quantitative analysis of mental health symptoms. Mediating and moderating effects of factors such as sleep, personality, physical health, and family dynamics must be incorporated. A person-specific approach with multiple observations in large samples could be the key to understanding individual variations in social media use and mental health consequences. Although resource intensive, these steps are necessary to advance our understanding.
Implications for clinicians, educators, and policymakers
Social media use/misuse should be part of routine assessments in child and adolescent psychiatric assessment, focusing on the frequency/duration of use and the presence of addiction-like patterns. Certain groups, such as those with ADHD, may be at higher risk of problematic social media use so treatment care plans need to consider this. Educators should raise awareness of social media misuse, warning signs, and resulting depressive/anxiety symptoms. Policymakers should strengthen funding for mental health research and services.
Conclusion
Notwithstanding the heterogeneous nature of the literature under review and the limitations of associative rather than causal findings, this study adds to the weight of evidence to suggest that social media use is linked to depression and anxiety symptoms among children and adolescents. This is a complex field of research, and these detrimental mental health effects may result indirectly via sleep deprivation or other variables. Nevertheless, it is critical that policymakers, educators, and clinicians take the risks of social media use among young people seriously, especially as technology is advancing at an incredible pace, and social media use is ubiquitous. Future research must explore problematic social media use using a person-specific approach to overcome existing methodological challenges.
Footnotes
Acknowledgments
Dr Dennis Ougrin, MBBS, MRCPsych, PGDip(Oxon), PGCAPHE, PhD. Professor of Child and Adolescent Psychiatry, Queen Marry University of London, Visiting Professor of Child and Adolescent Psychiatry and Global Mental Health, Institute of Psychiatry, Psychology and Neuroscience, King’s College London; Mr Daniele Porricelli, Research Assistant—Youth Resilience Unit, Queen Mary University of London. Ms Milena Nikolajeva, Research Assistant—Youth Resilience Unit, Queen Mary University of London, Paul Lee, Librarian, South London, and Maudsley Trust Library.
Author Disclosure Statement
No competing financial interests exist.
Authors’ Contributions
N.S.: Conceptualization, Methodology, Resources, Writing—Original draft preparation, Formal analysis P.Y.: Methodology, Formal analysis, Resources, Writing—Review and Editing S.Y.: Formal analysis, Writing—Review and Editing.
Funding Information
No funding was received for this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
