Depressive disorder is a major global health concern, affecting over 332 million people in 2021. 1 The burden of depressive disorder spans the lifespan but has risen particularly sharply among adolescents and young adults. Individuals aged 10–24 years, who are undergoing rapid cognitive, emotional and social development, have experienced a marked increase in depressive disorder burden over the past decade, Reference Thapar, Eyre, Patel and Brent2,Reference Collishaw3 with years lived with disability (YLD) rates rising by 25.4% from 2011 to 2021. 1 These trends underscore an urgent need to identify key drivers of depressive disorder in this age group.
Although family history and genetics remain important, growing evidence shows that childhood trauma, especially childhood sexual abuse (CSA) and bullying victimisation, is a major driver of rising adolescent depressive disorder. Reference Sweeting, Garfin, Holman and Silver4 CSA has a global prevalence of 12.7%, with a higher prevalence among girls (18%) compared with boys (7.6%). Reference Stoltenborgh, van Ijzendoorn, Euser and Bakermans-Kranenburg5 CSA has consistently been identified as a significant risk factor for depression, with a recent meta-analysis reporting a pooled odds ratio of 2.66 for the association with adult depression. Reference Li, D’Arcy and Meng6 Bullying victimisation affects approximately 30% of middle school students globally, with higher prevalence among boys (34.8%) than girls (30.4%). Reference He, Chen, Xie, Chen, Zhang and Yuan7 It shows strong and persistent associations with depressive disorder extending into adulthood, with a meta-analysis of 165 studies reporting an odds ratio of 2.21. Reference Moore, Norman, Suetani, Thomas, Sly and Scott8 Epidemiological studies highlight the public health relevance of these exposures, suggesting that nearly a third of mental health disorders may be attributable to adverse childhood experiences. Reference Kessler, McLaughlin, Green, Gruber, Sampson and Zaslavsky9,Reference Keyes, Eaton, Krueger, McLaughlin, Wall and Grant10
Addressing CSA- and bullying victimisation-related depressive disorder in adolescents and young adults requires comprehensive national and global data. Understanding trends, sociodemographic patterns and projected depressive disorder burden is critical for evaluation of public health interventions. However, research remains limited due to the scarcity of large-scale studies capturing sensitive exposures and the lack of standardised methods to quantify CSA- and bullying victimisation-attributable depressive disorder. To fill this gap, we used Global Burden of Disease (GBD) 2021 data to examine the global epidemic situation of depressive disorder linked to CSA and bullying victimisation in individuals aged 10–24 years, aiming to inform policies and guide targeted prevention strategies.
Method
Data source
This study focuses on the burden of depressive disorder attributable to CSA and bullying victimisation among adolescents and young adults (aged 10–24 years) worldwide from 1990 to 2021, and projects YLD until 2050 based on the GBD 2021 study. The GBD 2021 project estimated the incidence, prevalence and disability-adjusted life years (DALYs) associated with 371 diseases and injuries across 204 countries and 21 regions from 1990 to 2021. 1 The data were obtained using the Global Health Data Exchange query tool (https://ghdx.healthdata.org/gbd-results-tool), including the incidence, prevalence and YLD of depressive disorder attributed to CSA and bullying victimisation in individuals aged 10–24 years worldwide from 1990 to 2021.
CSA was included as the only form of child maltreatment in GBD 2021, with available exposure prevalence and relative risk estimates linking exposure to depressive disorder. Other forms of child maltreatment (physical abuse, emotional abuse, neglect and exposure to domestic violence) lack standardised GBD risk factor modelling and population-attributable fraction data. Bullying victimisation was included given its well-documented association with depressive disorder and the availability of standardised GBD estimates.
Definitions
Depressive disorder
Depressive disorder involves the presence of at least one major depressive episode, characterised by either depressed mood or a loss of interest or pleasure in most activities for most of each day, lasting for at least 2 weeks. In the present study we followed the case definitions for depressive disorder used by GBD, classified according to the criteria set by DSM-IV-TR or ICD-10 code F32.
Risk factors
The risk factors associated with both CSA and bullying victimisation are defined within the context of GBD studies. CSA is defined as the experience of sexual intercourse or other contact abuse (e.g. fondling and other sexual touching) when the contact is unwanted or when the abuser is at least 5 years older than the victim, provided the victim is 15 years old or younger. Bullying victimisation is defined as the experience of being bullied by peers while attending school. This includes various subtypes of bullying victimisation, such as physical, verbal, relational and cyberbullying. It is important to note that the perpetrators in this context are restricted to peers and do not include siblings, intimate partners or adults (e.g. teachers). Definitions of CSA and bullying victimisation followed World Health Organization (WHO) and International Society for the Prevention of Child Abuse & Neglect standardised frameworks. Reference Krug, Mercy, Dahlberg and Zwi11,Reference Zolotor, Runyan, Dunne, Jain, Péturs and Ramirez12
Sociodemographic index
Sociodemographic index (SDI) is the index that combines the measures of income, average years of schooling and fertility rates for each location and year within the GBD framework. Initially introduced in GBD 2015, SDI employs the methodology of the human development index to gauge sociodemographic advancement. In this study, SDI is classified into five distinct categories: low SDI (0–0.45), low-middle SDI (0.46–0.61), middle SDI (0.62–0.69), high-middle SDI (0.70–0.81) and high SDI (0.82–1.00).
Statistical analysis
Age-standardised years lived with disability rates (ASYR) (number per 100 000 population) were used to describe the disease burden. Detailed GBD methodology has been reported previously, 1,13 with comprehensive statistical specifications provided in the Supplementary Materials available at https://doi.org/10.1192/bjp.2026.10862.
In this study, YLD were used as the primary metric, representing the number of healthy life-years lost due to living with a less-than-optimal state of health. Data on YLD for depressive disorders were sourced from the GBD 2021 study. 1 Within GBD, these estimates are produced by the Bayesian meta-regression tool DisMod-MR 2.1 and calculated as prevalence multiplied by disability weights (0, full health to 1, most severe), with full details given in the Supplementary Materials. Reference Burstein, Fleming, Haagsma, Salomon, Vos and Murray14,15
Microsimulation was subsequently used to obtain a final estimate of YLD after correcting for comorbidities. This adjusts disability weights when conditions coexist to avoid double-counting overlapping symptoms; full details are provided in the Supplementary Materials and published literature. 16
To minimise differences among various populations, we calculated age-standardised rates (ASRs) using the WHO 2000–2025 standards and the world standard population developed by the GBD 2021 study. 17 The ASR was calculated using the following formula: Reference Ahmad, Boschi-Pinto, Lopez, Murray, Lozano and Inoue18
${\rm{Age}} - {\rm{standardised}}\;{\rm{rate}} = {{\sum _{i = 1}^A{a_i}{w_i}} \over {\sum _{i = 1}^A{w_i}}}$
The reference standard population’s age-specific rates and number of persons (or weight) in each age group are represented by a i and w i , respectively (where i represents the ith age group).
Moreover, to evaluate the YLD attributable to the burden of CSA and bullying victimisation among adolescents and young adults aged 10–24 years with depressive disorder and their trends, we used joinpoint regression analysis to calculate the average annual percentage change (AAPC) and the corresponding 95% confidence interval. Reference Kim, Fay, Feuer and Midthune19 The annualised rate of change for 204 countries and 21 regions was used to present spatial trends. The trend data were obtained using the Joinpoint Regression Program software (version 5.0.2 for Windows) from the National Cancer Institute (Bethesda, Maryland, USA; https://surveillance.cancer.gov/joinpoint/), and the simplest joinpoint model that the data allowed was fitted. The significance of additional joinpoints was assessed by Monte Carlo permutation tests. Up to five joinpoints were permitted, with the final model selected by the Bayesian information criterion; 95% confidence intervals used the grid-search method. The average annual percentage change was calculated by weighting the annual percentage change of each refinement in the best-fit model. We compared age groups and SDI. Least squares regression and generalised additivity models were used to test for linear or non-linear relationships between SDI and ASRs or the corresponding mean percentage change per year. The proportional contribution of each risk factor was estimated by dividing the attributable YLD for that factor by the total YLD for depressive disorder. Plausibility was determined by a P value of 0.05.
Using data from the GBD study (1990–2021), we applied a Bayesian age-period-cohort (BAPC) model to project YLDs attributable to depressive disorder among adolescents and young adults aged 10–24 years for 2022–2050. Reference Riebler and Held20 The model assumed that YLD counts would follow a Poisson distribution, and then decomposed temporal trends into independent age, period and cohort effects. Uncertainty intervals (95%) were estimated through 1000 Monte Carlo simulations. Projections assume stable historical patterns; future policy changes or epidemiological shifts may affect accuracy. BAPC models based on integrated nested Laplace approximation methods are commonly used for data prediction. Reference Rue, Martino and Chopin21 The BAPC model has been shown to have relatively low error rates and achieve more sensitive predictions compared with other models such as Joinpoint, the APC model and the Nordpred model. In this study we used the BAPC (version 0.0.34) and INLA (version 21.02.23) software packages to develop the BAPC model. All statistical analyses were performed using R software (version 4.3.2 for Windows; R Foundation for Statistical Computing, Vienna, Austria; https://www.r-project.org/).
Results
Global distribution of depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation in 2021
In 2021, the ASYR per 100 000 populations for depressive disorder in adolescents and young adults (aged 10–24 years) attributable to CSA was 14.15 (95% uncertainty interval: 14.12–14.19). For depressive disorder attributable to bullying victimisation, the ASYR was 87.83 (95% uncertainty interval: 87.75–87.90) per 100 000 population worldwide (Supplementary Table 1). Among all risk factors contributing to depressive disorder, CSA accounted for 11.81% with bullying victimisation contributing to 73.31% (Fig. 1(a, c)).
(a) Temporal trend of age-standardised YLD rate for the burden of depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation, globally and by SDI (five categories: high, high-middle, middle, low-middle and low SDI) from 1990 to 2021, for all and men and women separately. (b) Average annual percentage change in age-standardised YLD rate for the burden of depressive disorder in adolescents and young adults, globally and by SDI category, including by gender, from 1990 to 2021. (c) Risk factors for depressive disorder among individuals aged 10–24 years in 2021 (both genders combined). YLD, years lived with disability; CSA, childhood sexual abuse; SDI, sociodemographic index; AAPC, average annual percentage change.
Fig. 1 Long description
Panel A: Three line graphs show the temporal trend of age-standardized years lived with disability (YLD) rate for the burden of depressive disorder in adolescents and young adults attributable to childhood sexual abuse (CSA) and bullying victimization (BV), globally and by sociodemographic index (SDI) categories from 1990 to 2021. The graphs are divided into men and women separately. The x-axis represents the year, and the y-axis represents the age-standardized YLD rate per 100,000. Different colored lines represent different SDI categories. Panel B: Two bar graphs display the average annual percentage change in age-standardized YLD rate for the burden of depressive disorder in adolescents and young adults, globally and by SDI category, including by gender, from 1990 to 2021. The x-axis represents the location, and the y-axis represents the average annual percentage change (AAPC) in percent. Different colored bars represent different genders. Panel C: A pie chart shows the risk factors for depressive disorder among individuals aged 10-24 years in 2021, with both genders combined. The chart is divided into sections representing different risk factors: bullying victimization, childhood sexual abuse, and intimate partner violence.
The highest ASYR for depressive disorder attributable to CSA and bullying victimisation was observed in high-SDI regions (CSA: 30.57, 95% uncertainty interval: 30.43–30.72; bullying victimisation: 143.67, 95% uncertainty interval: 143.35–143.99) in 2021. By contrast, the lowest YLD rate for depressive disorder attributable to CSA was found in those regions with high-middle SDI (9.09, 95% uncertainty interval: 9.02–9.16), whereas the lowest YLD rate for depressive disorder attributable to bullying victimisation was observed in the middle-SDI regions (65.23, 95% uncertainty interval: 65.11–65.35) in 2021 (Supplementary Table 1). The association between ASYR and SDI followed a U-shaped curve, with a higher ASYR in the high-SDI regions and lower ASYR in the high-middle-SDI regions in 2021 (Fig. 1(a)).
At the regional level, high-income North America exhibited the highest burden of depressive disorder attributable to CSA and bullying victimisation (Supplementary Table 2). Meanwhile, at the national level, Greenland had the highest ASYR attributable to both risks (Supplementary Table 3).
Temporal trend of depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation from 1990 to 2021
Figure 1(a) and Supplementary Table 1 show that the ASYR for depressive disorder in adolescents and young adults attributable to CSA increased from 13.69 (95% uncertainty interval: 13.66–13.72) in 1990 to 14.15 (95% uncertainty interval: 14.12–14.19) in 2021, with the ASYR for depressive disorder attributable to bullying victimisation rising significantly from 70.12 (95% uncertainty interval: 70.05–70.20) in 1990 to 87.83 (95% uncertainty interval: 87.75–87.90) in 2021 globally.
The ASYR for depressive disorder attributable to CSA decreased in regions with high-middle and middle SDI but increased in regions with other SDI levels. Meanwhile, the ASYR of depressive disorder attributable to bullying victimisation increased most rapidly in high-SDI regions, with an AAPC of 1.36% (95% confidence interval: 1.28–1.45%) (Supplementary Table 1 and Fig. 1(b)).
At the regional level, the ASYR for depressive disorder attributable to CSA and bullying victimisation both grew the fastest in high-income North America between 1990 and 2021 (AAPC: 1.47%, 95% confidence interval: 1.17–1.76% in CSA; 1.95%, 95% confidence interval: 1.88–2.02% in bullying victimisation) (Supplementary Table 2). At the country level, the USA exhibited the fastest-growing ASYR for depressive disorder attributable to CSA (AAPC: 1.50%, 95% confidence interval: 1.23–1.78%), with Niger showing the fastest-growing ASYR for depressive disorder attributable to bullying victimisation (AAPC: 3.03%, 95% confidence interval: 2.92–3.13%) from 1990 to 2021 (Supplementary Table 3).
Figure 2 shows that SDI was inversely associated with the faster increase in the AAPC of depressive disorder attributable to bullying victimisation. However, SDI was associated with a stable AAPC of depressive disorder attributable to CSA (bullying victimisation, P < 0.001; CSA, P < 0.05).
Association between age-standardised YLD rates of depressive disorder attributable to CSA and bullying victimisation among adolescents and young adults in 2021, and corresponding AAPC trends by SDI from 1990 to 2021. YLD, years lived with disability; CSA, childhood sexual abuse; AAPC, average annual percentage change; SDI, sociodemographic index.
The gender pattern of global burden of depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation
Globally, in 2021, the ASYR of depressive disorder in adolescents and young adults attributable to CSA was higher in women 18.27 (95% uncertainty interval: 18.22–18.32) than in men 10.24 (95% uncertainty interval: 10.20–10.28), whereas the ASYR attributable to bullying victimisation was higher in men 88.76 (95% uncertainty interval: 88.66–88.87) than in women 86.84 (95% uncertainty interval: 86.73–86.95). From 1990 to 2021, the ASYR of depressive disorder attributable to CSA declined in men but increased in women. By contrast, depressive disorder attributable to bullying victimisation showed an upward trend for both genders, with a faster increase in women (AAPC: 0.81%, 95% CI: 0.75–0.87%) than in men (AAPC: 0.64%, 95% CI: 0.60–0.68%) (Supplementary Table 1 and Fig. 3(a, b)).
(a) Trends in age-standardised YLD rates for depressive disorder attributable to CSA in adolescents and young adults, by gender, SDI and age group, from 1990 to 2021. (b) Trends in age-standardised YLD rates for depressive disorder attributable to bullying victimisation in adolescents and young adults by gender, SDI and age group, from 1990 to 2021. YLD, years lived with disability; CSA, childhood sexual abuse; SDI, sociodemographic index.
Fig. 3 Long description
Panel A: Six line graphs show trends in age-standardized YLD rates for depressive disorder attributable to CSA in adolescents and young adults, by gender, SDI, and age group, from 1990 to 2021. The x-axis represents the year, and the y-axis represents the age-standardized YLD rate per 100,000. Each graph is labeled by SDI category: Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI. Different colored lines represent different age groups: 10-14 years, 15-19 years, 20-24 years, and gender: Women and Men. Panel B: Six line graphs show trends in age-standardized YLD rates for depressive disorder attributable to bullying victimization in adolescents and young adults, by gender, SDI, and age group, from 1990 to 2021. The x-axis represents the year, and the y-axis represents the age-standardized YLD rate per 100,000. Each graph is labeled by SDI category: Global, High SDI, High-middle SDI, Middle SDI, Low-middle SDI, and Low SDI. Different colored lines represent different age groups: 10-14 years, 15-19 years, 20-24 years, and gender: Women and Men.
In regard to SDI, the ASYR of depressive disorder attributable to CSA was higher in women than in men across all SDI regions. For bullying victimisation, the ASYR was higher in women in those regions with high SDI and high-middle SDI, but higher in men in other regions. For CSA, the ASYR in men increased from 1990 to 2021 in the high-SDI region whereas it declined in other SDI regions. In women, the ASYR declined in those regions with high-middle and middle SDI but increased in other SDI regions (Supplementary Table 1 and Fig. 1(a)). In regard to bullying victimisation the ASYR increased for both genders, with the highest increase in the high-SDI region, showing an AAPC of 1.09% (95% CI: 1.05–1.13%) for men and 1.61% (95% CI: 1.53–1.68%) for women (Supplementary Table 1 and Fig. 3(a, b)).
The age pattern of the global burden of depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation
In 2021, the global ASYR for depressive disorder attributable to CSA in adolescents and young adults was relatively lower. The values were 5.24 (95% uncertainty interval: 5.19–5.30) in the 10–14 years age group, 16.97 (95% uncertainty interval: 16.87–17.07) in the 15–19 years age group and 21.16 (95% uncertainty interval: 21.05–21.28) in the 20–24 years age group. Additionally, the ASYR attributable to bullying victimisation was 41.30 (95% uncertainty interval: 41.15–41.46) in the 10–14 years age group, 104.57 (95% uncertainty interval: 104.32–104.83) in the 15–19 years age group and 122.26 (95% uncertainty interval: 121.98–122.54) in the 20–24 years age group (Supplementary Table 1 and Fig. 3(a, b)).
From 1990 to 2021, the global ASYR for depressive disorder attributable to CSA significantly increased in the 10–14 and 15–19 years age groups, particularly in the former (AAPC: 1.82%, 95% CI: 1.44–2.21%). By contrast, the ASYR for CSA in the 20–24 years age group declined (AAPC: −0.39%, 95% CI: −0.45 to −0.34%). The ASYR attributable to bullying victimisation increased in all age groups, with the largest increase also in the 10–14 years age group (AAPC: 0.93%, 95% CI: 0.87–0.99%) (Supplementary Table 1).
In regard to SDI, all three age groups had the highest ASYR for depressive disorder attributable to CSA in the high-SDI region. Moreover, the middle-SDI region had the lowest ASYR for the 10–14 years age group, with the high-middle-SDI region having the lowest ASYR in the 15–19 and 20–24 years age groups. For bullying victimisation, ASYR was also the highest in all three age groups in the high-SDI region, with the middle-SDI region having the lowest ASYR attributable to bullying victimisation for all three age groups (Supplementary Table 1 and Fig. 3(a, b)).
From 1990 to 2021, the ASYR for depressive disorder attributable to CSA increased more rapidly for women than for men in the 10–14 and 15–19 years age groups, whereas men had a greater increase than women in the 20–24 years age group. For bullying victimisation, women had a more rapid increase than men in all age groups (Supplementary Table 1 and Fig. 3(a, b)).
Projection of global ASYR of depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation to 2050
Based on our projections from 2022 to 2050, the global ASYR for depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation is expected to follow distinct trends. Specifically, the ASYR for depressive disorder attributable to CSA is projected to increase globally for both men (from 10.52 in 2022 to 91.53 in 2050) and women (from 18.71 in 2022 to 176.57 in 2050), whereas that for depressive disorder attributable to bullying victimisation is expected to decrease for both genders (men: from 88.40 in 2022 to 82.09 in 2050; women: from 86.01 in 2022 to 70.64 in 2050). These trends are based on demographic projections from 2022 to 2050 (Supplementary Table 4 and Fig. 4).
Projection of global age-standardised YLD rates for depressive disorder in adolescents and young adults attributable to CSA and bullying victimisation up to 2050, based on the BAPC model. YLD, years lived with disability; CSA, childhood sexual abuse; BAPC, Bayesian age-period-cohort.
Discussion
This study provides a comprehensive analysis of ASYR trends in depressive disorder burden due to CSA and bullying victimisation among adolescents and young adults between 1990 and 2021, and is projected to continue until 2050. Over the past three decades, the ASYR for depressive disorder related to both CSA and bullying victimisation has increased (CSA: from 13.69 in 1990 to 14.15 in 2021; bullying victimisation: from 70.12 to 87.83). Projections based on the BAPC model suggest that, by 2050, the ASYR for depressive disorder linked to CSA may continue to rise, whereas that related to bullying victimisation could decline.
Gender differences in the burden of depressive disorder attributable to CSA and bullying victimisation
The ASYR for depressive disorder due to CSA in women (18.27) is significantly higher than in men (10.24), consistent with previous research. Reference Weiss, Longhurst and Mazure22 This gender gap reflects biological, psychological and structural factors. Women’s heightened trauma sensitivity, including hypothalamic–pituitary–adrenal (HPA) axis reactivity, Reference Slavich and Irwin23 interacts with social contexts such as polyvictimisation and gender-based violence. Reference Finkelhor, Ormrod and Turner24 Poverty, limited access to care and gender-based barriers further restrict recovery, whereas stigma and scarce gender-sensitive services hinder help-seeking. Reference Lund, De Silva, Plagerson, Cooper, Chisholm and Das25 These combined biological and social disadvantages probably increase female vulnerability to CSA-related depressive disorder.
By contrast, men exhibit a slightly higher ASYR for depressive disorder due to bullying victimisation (88.76) compared with women (86.84). Bullying occurs in peer and school environments shaped by gender norms, and is often compounded by early life adversity. Reference Cosma, Bjereld, Elgar, Richardson, Bilz and Craig26 Adverse school climates and limited psychosocial support increase boys’ vulnerability, because socialisation may discourage emotional expression, impede help-seeking and lead to maladaptive coping that persists into adulthood and elevates depressive disorder risk, Reference Chaplin and Aldao27 potentially reinforced by stress-related neurobiological alterations following chronic peer victimisation.
These findings underscore the need for gender-responsive interventions. For women, trauma-informed, gender- and poverty-sensitive mental health programmes are crucial. For men, improving school environments, psychosocial support and reducing stigma can mitigate bullying victimisation-related depressive disorder. Broader policies addressing socioeconomic inequities and strengthening community support may reduce the depressive disorder burden for both genders.
Age-specific trends and long-term impact of CSA and bullying victimisation
The burden of depressive disorder associated with CSA demonstrated a bimodal distribution, peaking in the 15–19 and 20–24 years age groups. Late adolescence represents a critical period for identity formation and relational development; during this stage, CSA-related shame may disrupt self-concept, particularly in the absence of trauma-informed school support and familial validation. Reference Feiring, Simon and Cleland28 In emerging adulthood, stressors associated with independent living and exposure to workplace power dynamics may reactivate unresolved trauma. This pattern probably reflects both the cumulative effects of untreated trauma and the developmental challenges characteristic of these life stages.
The burden of depressive disorder related to bullying victimisation exhibited a distinct temporal pattern, reaching its zenith in the 20–24 years age group, while also showing a notable elevation among early adolescents aged 10–14 years. Early adolescence is marked by heightened peer dependency, rendering social exclusion particularly detrimental – especially in school environments that normalise bullying or lack adequate counselling resources. Reference Han, Ye and Zhong29 By emerging adulthood, maladaptive coping strategies developed during adolescence, such as emotional suppression, may compromise psychosocial functioning.
Biological mechanisms also contribute to these patterns, although they operate within broader social contexts. Both CSA and bullying victimisation activate the HPA axis and trigger neuroinflammatory pathways that affect hippocampal and prefrontal neurogenesis. Reference McLaughlin, Sheridan, Gold, Duys, Lambert and Peverill30,Reference Kuhlman, Chiang, Horn and Bower31 During early adolescence, the developing HPA axis increases vulnerability to bullying victimisation-related chronic stress, particularly in unsupportive school environments. Reference Kuhlman, Chiang, Horn and Bower31 During late adolescence, ongoing prefrontal cortex maturation may heighten susceptibility among CSA survivors, especially when familial support is limited. Reference Lupien, McEwen, Gunnar and Heim32 Additionally, chronic stress has been associated with reduced dendritic spine density and altered glutamatergic signalling, which may underlie the persistent depressive symptoms observed in affected individuals. Reference Price and Duman33
Regional variations in the burden of depressive disorder attributable to CSA and bullying victimisation
Despite advanced healthcare systems and better detection, high-SDI regions show higher ASYRs for depressive disorders, probably due to improved reporting, greater public awareness and accessible mental health services. Reference Walsh, Eggins, Hine, Mathews, Kenny and Howard34 However, these regional differences may reflect visibility rather than true prevalence. Even in high-SDI regions, most CSA cases remain undisclosed – less than 5% are officially recorded; Reference Martin and Silverstone35 a Finnish study reported only 12% disclosure. Reference Lahtinen, Laitila, Korkman and Ellonen36 Low- and middle-SDI regions face persistent or rising CSA-related depressive disorder, largely due to limited mental health resources, low public awareness and cultural norms that frame abuse as a family secret. 37 These factors underscore the need for strengthened prevention efforts, improved reporting mechanisms and expanded mental health support. Together, disparities in healthcare capacity, cultural norms surrounding disclosure and socioeconomic constraints jointly shape regional differences in the observed CSA-related depressive disorder burden.
The burden of depressive disorder attributable to bullying victimisation is notably higher in high-SDI regions, reflecting the convergence of traditional and cyberbullying facilitated by widespread digital access and heightened societal awareness. Reference Bannink, Broeren, van de Looij-Jansen, de Waart and Raat38 In these settings, persistent online exposure may intensify psychological harm beyond school environments. In contrast, middle- and low-SDI regions show rising but less pronounced bullying victimisation-related depressive disorder, probably reflecting underreporting, limited mental health infrastructure and differences in educational and social contexts. 37
These regional variations reflect the interplay of economic development, healthcare capacity, cultural norms and digital environments in shaping trauma exposure, recognition and detection. Resource constraints and cultural barriers obscure the true CSA burden in low- and middle-SDI settings, whereas the convergence of traditional and cyberbullying victimisation in digitally connected regions introduces new risks that intensify harm. 37 Moreover, polyvictimisation – common among CSA survivors – compounds trauma severity and depressive outcomes and further complicates detection and intervention, particularly where resources are limited. Reference Finkelhor, Ormrod and Turner24 Accordingly, the observed disparities call for region-specific interventions that account for local socioeconomic conditions, healthcare infrastructure and digital environments. Universal prevention and mental health support remain essential, but their implementation must be adapted to regional contexts, with attention to resource constraints, cultural barriers and emerging digital forms of violence.
Implications of projected trends in CSA- and bullying victimisation-related depressive disorder
CSA-related depressive disorder is projected to rise, probably reflecting persistent stigma, the private and often intrafamilial nature of CSA, weak protection systems, limited trauma-informed service and polyvictimisation. In the absence of early identification and sustained support, these factors may interact and accumulate over time, transforming early trauma into enduring vulnerability to depressive disorders. Reference Li, D’Arcy and Meng6 These projections highlight the need for strengthened prevention, improved detection and referral pathways and gender-responsive care. Future changes in public awareness, stigma reduction efforts and mental health policies could also alter these trends. Reference Assini-Meytin, Kaufman, Mathews, Palmer, Ingram and Letourneau39
By contrast, bullying victimisation-related depressive disorder is projected to decline, consistent with evidence that school-based anti-bullying programmes effectively reduce bullying victimisation. Reference Gaffney, Ttofi and Farrington40 Maintaining this decline will require continued scaling of effective interventions and cross-sector collaboration.
These forecasts assume stable reporting systems, policy enforcement and digital environments. Therefore, shifts in global health dynamics, including cyberbullying trends, mental health service capacity and policy reforms, may substantially influence projections for 2050.
Limitations
This study has several limitations. First, despite improvements in GBD 2021, estimates remain susceptible to biases from missing or low-quality data, and years-of-life-lost data for depressive disorders were unavailable; therefore, our analysis focused on YLD. Second, CSA- and bullying victimisation-related depressive disorder estimates depend on detection methods, screening quality and registry systems, which vary by socioeconomic level, probably leading to underestimated burden in low-SDI countries. Third, the GBD age threshold of 10 years excludes younger children who may experience CSA or bullying victimisation, whereas clinical depressive disorder is rare under the age of 10 years, and subclinical symptoms and developmental disruptions are not captured, potentially underestimating early life trauma impacts. Fourth, YLD capture functional disability but may not fully reflect social isolation, quality-of-life loss or educational consequences, particularly relevant for youth experiencing CSA-/bullying victimisation-related depressive disorder. Fifth, BAPC projections based on a normal approximation and only three age groups carry some uncertainty, with lower bounds falling below zero in certain years, and further data are needed to validate these projections.
Future directions
The burden of depressive disorder among adolescents and young adults attributable to CSA and bullying victimisation has increased over the past three decades. Projections show continued growth in CSA-related depressive disorder, whereas bullying victimisation-related depressive disorder may decline by 2050. Women are more affected by CSA, men by bullying victimisation, and adolescents aged 10–14 years show the fastest rise, reflecting heightened developmental vulnerability. These trends call for equity-focused strategies, particularly in low- and middle-SDI settings, including school-based prevention and trauma-informed, gender-sensitive care. Future research should examine developmental timing, polyvictimisation and cyberbullying to better guide interventions and reduce global mental health inequities.
Supplementary material
The supplementary material is available online at https://doi.org/10.1192/bjp.2026.10862
Data availability
The data that support the findings of this study are openly available in the GBD database at https://vizhub.healthdata.org/gbd-results/. The analytic code supporting the findings of this manuscript is not available for public access.
Acknowledgements
Thanks to Xiao Ming for his work in the GBD database. His excellent sharing of GBD database analysis procedure and other public databases made it easier for us to explore the GBD database. No artificial intelligence tools have been used in the research or writing process for this manuscript.
Author contributions
X.L. – study design, data interpretation, study supervision, drafting of the manuscript and acquisition of funding. Y.Z. – study design, data interpretation, manuscript revision. X.D., T.T., T.L. and Y.W. – data processing, statistical analysis and drafting of the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding
This work was supported by the Harbin Medical University Backbone Teacher Training Programme of China (grant no. 31021230096 to X.L.) and the Science Foundation of Heilongjiang Province (grant no. LH2023H018 to T.T.).
Declaration of interest
None.
Transparency declaration
The lead author and manuscript guarantor affirm that this manuscript provides an honest, accurate and transparent account of the study being reported. All important aspects of the study have been included, and no significant information has been omitted. We confirm that the methods and results presented align with the original study plan, and any discrepancies from the preregistered protocol have been clearly explained. We are committed to upholding the principles of academic integrity to ensure the validity and reproducibility of the research findings.
eLetters
No eLetters have been published for this article.