Adverse childhood experiences and mental health: the mediating role of perceived social support and coping strategies

Abstract

Background

Adverse childhood experiences (ACEs) are significant risk factors for poor mental health in adolescence. However, the mechanisms linking ACEs to depression, anxiety, and stress remain underexplored in low-resource settings like Bangladesh. To examine the associations between ACEs and mental health and investigate the mediating effects of perceived social support and coping strategies among adolescents.

Methods

A cross-sectional survey was conducted among 1139 high-school graduates in Bangladesh. Self-administered questionnaires assessed ACEs, perceived social support, coping strategies, and symptoms of depression, anxiety, and stress. Structural equation modeling evaluated direct and indirect pathways and logistic regression identified risk factors associated with depression, anxiety, and stress.

Results

ACEs were positively associated with elevated stress (β = 0.459, p< 0.001) and negatively associated with family (β = –0.339), friend (β =–0.239), and support from other significant persons (e.g. a close friend or mentor) (β = –0.199), as well as active (β = –0.345) and emotional coping (β = –0.221; all p< 0.001). Family support fully mediated the associations between ACEs and both depression (β = 0.042, p = 0.002) and anxiety (β = 0.063, p < 0.001), and partially mediated the association with stress (β = 0.217, p< 0.001). Active coping similarly fully mediated the ACE–depression (β = 0.017, p = 0.007) and ACE–anxiety (β = 0.025, p = 0.003) associations, while partially mediating the ACE–stress link (β = 0.125, p < 0.001). Friend support and emotion-focused coping showed smaller but significant partial mediation effects for anxiety (β = 0.020, p = 0.041; β = 0.013, p = 0.013, respectively) and stress (β = 0.027, p= 0.021 for emotion-focused coping), while their indirect effects on depression were not statistically significant. Females had higher odds of all mental health problems. Psychological abuse was associated with worse mental health outcomes, and sexual abuse significantly increased stress risk.

Conclusion

Social support and active coping strategies buffer the effect of ACEs on mental health. Interventions should prioritize enhancing family support and promoting active coping, among adolescents.

Introduction

Adolescence is a critical developmental period marked by significant biological, psychological, and social transitions. During this phase, exposure to environmental adversities can profoundly influence emotional and psychological well-being [1]. Among these adversities, adverse childhood experiences (ACEs) encompassing a range of traumatic events such as emotional, physical, or sexual abuse; neglect; and household dysfunction, including parental mental illness, substance abuse, and domestic violence have emerged as significant determinants of later mental health outcomes [2, 3].

The burden of ACEs is particularly pronounced in low- and middle-income countries (LMICs), where structural inequities heighten vulnerability and limit access to mental health resources [4]. In Bangladesh, recent national estimates suggest that approximately 14% of children and adolescents suffer from psychiatric disorders, with over 90% failing to access formal mental health services [5]. This striking gap suggests the urgency of addressing adolescent mental health in resource constrained settings.

Although nationally representative data on ACE prevalence in Bangladesh remain limited, smaller-scale studies suggest high rates of adversity among Bangladeshi children and adolescents [6–9]. For example, more than 97% of children had experienced at least one form of psychological abuse during their lifetime, while neglect affected 78% of children over their lifetime and 58% in the past year [7]. Physical abuse was similarly pervasive, with lifetime exposure rates for one or more forms of physical punishment reaching 99%, and over 83% reporting three or more forms [6]. Among university students, 54.1% reported at least one ACE, with emotional abuse (27%), maternal violence (18.2%), and physical abuse (15.2%) being the most prevalent [9]. Notably, a study of child sexual abuse survivors in Bangladesh found that 21.3% had attempted suicide at least once [8]. These alarming statistics highlight the urgency of examining how ACEs shape adolescent mental health in Bangladesh and similar LMICs.

A robust body of research has documented the association between ACEs and a wide range of negative mental health outcomes, including depression, anxiety, and chronic stress [10–12]. Exposure to multiple adversities during development can disrupt neurobiological functioning, impair emotional regulation, and increase physiological stress responses, placing adolescents at heightened risk for internalizing and externalizing behavioral problems [13, 14]. Identifying modifiable psychosocial factors that can buffer these effects is essential for developing effective prevention strategies.

Importantly, sociocultural factors in Bangladesh, such as hierarchical family structures, collectivist norms, and pervasive stigma toward mental illness may influence both the likelihood of ACE exposure and the expression or reporting of emotional distress. Gender norms also shape coping responses, with female adolescents often facing more restrictions on emotional expression and autonomy [15]. These contextual influences highlight the relevance of examining culturally embedded protective mechanisms, such as family support and active coping strategies.

Social support has been identified as a critical protective factor that promotes resilience among youth exposed to adversity [16, 17]. Supportive relationships with family members, peers, and community figures can mitigate psychological distress and foster emotional regulation [18, 19]. Evidence suggests that perceived social support mediates the relationship between ACEs and mental health outcomes, offering a potential target for preventive interventions [19–22].

Coping strategies, defined as cognitive and behavioral responses to stress, play a pivotal role in shaping mental health trajectories following adversity [23, 24]. Adaptive coping strategies, such as active problem-solving and emotion-focused coping, have been associated with lower levels of depression and anxiety, while maladaptive coping strategies, such as avoidance, can exacerbate psychological distress [17, 25, 26]. Emerging evidence supports the role of coping strategies as mediators linking ACEs to mental health outcomes, highlighting their relevance for intervention development [27].

Guided by the Transactional Model of Stress and Coping [28], the present study proposes a comprehensive mediation model in which ACEs influence adolescent mental health outcomes (depression, anxiety, and stress) both directly and indirectly through coping strategies and perceived social support. This framework emphasizes the dynamic interaction between environmental stressors and psychosocial resources, offering insights into actionable targets for prevention.

While extensive research has examined ACEs and mental health in high-income settings, there is a notable gap in LMICs such as Bangladesh, where cultural, social, and economic factors may shape both exposure to adversity and coping processes. Few studies have simultaneously examined the dual mediating roles of coping strategies and social support in adolescent mental health. Therefore, the present study aims to investigate (i) the associations between ACEs coping strategies, social support, and mental health outcomes (depression, anxiety, and stress) among Bangladeshi adolescents; and (ii) examine the mediating roles of coping strategies (active, maladaptive, and emotion-focused) and perceived social support (from family, friends, and others) in the relationship between ACEs and mental health outcomes. By addressing these gaps, the present study offers novel insights that can inform culturally relevant, school-based preventive strategies aimed at fostering resilience.

Methods

Study procedure and participants

A cross-sectional study was conducted among adolescents who were high school graduates. Data were collected at Jahangirnagar University, Dhaka, Bangladesh where participants came to attend the university entrance test in February 2025. During this exam period, the research team approached the test-taking students who resided in the university dormitories to invite them to participate in the study. Participants were recruited using a non-probability (convenience) sampling technique. While the sample included students from a wide range of geographic regions and schools across the country, it primarily represents adolescents with higher educational aspirations and may not be representative of the general adolescent population in Bangladesh. Participation was voluntary, and all students received detailed information regarding the study objectives and procedures before completing the semi-structured questionnaire. Of the 1,163 students who completed the survey, 1,139 responses were retained after excluding incomplete questionnaires.

Measures

Demographic variables

Participants provided information on their age, gender, and permanent residence.

Adverse childhood experience

ACEs were assessed using a standardized questionnaire informed by prior studies [2, 4]. The assessment covered several domains of abuse and household dysfunction. Psychological abuse was evaluated through items addressing verbal insults or actions instilling fear of physical harm by a parent or household adult. Physical abuse involved instances of being pushed, grabbed, slapped, or hit hard enough to cause marks or injuries. Sexual abuse focused on incidents of being touched or fondled inappropriately, or being coerced to touch someone else’s body sexually.

Household dysfunction was assessed through indicators of mental illness, maternal exposure to violence, criminal behavior, and exposure to substance use within the household. Mental illness included the presence of a depressed or mentally ill household member or instances of attempted suicide by a household member. Maternal violence was measured through items detailing physical abuse toward the mother, such as pushing, slapping, or being threatened with weapons. Substance use was defined as living with anyone who was problematic drinker or alcoholic or who used street drugs. Criminal behavior was captured by identifying cases where a household member was incarcerated.

Respondents were classified as exposed to a category if they responded “yes” to one or more items within that category. The total ACE score (range: 0–8) was computed by summing the number of endorsed categories, reflecting the extent of adversity experienced.

Depression and anxiety (PHQ-4)

The four item Patient Health Questionnaire (PHQ-4) was used to assess depressive and anxiety symptoms [29] based on the previous two weeks. It includes two items (i.e., how often have you been bothered by: little interest or pleasure in doing things?”) of depression (PHQ-2) and two items (how often have you been bothered by: feeling nervous, anxious, or on edge?”) of anxiety (GAD-2). Responses were recorded to on a 4-point Likert scale (0 = not at all to 3 = nearly every day). Scores for each subscale ranged from 0 to 6, with a score ≥ 3 indicating a positive screen for depression or anxiety [30]. The reliability of the scale is 0.77.

Perceived stress (PSS-4)

Self-perceived stress was assessed using the four-item Perceived Stress Scale (PSS-4) [31]. The scale includes two positively worded items (e.g., ‘In the last month, how often have you felt that you were unable to control the important things in your life’?) and two negatively worded items (e.g., ‘In the last month, how often have you felt confident about your ability to handle your personal problems’?). Each item is rated on a 5-point Likert scale (0 = never to 4 = very often) with the range of 0 to 16. A score of ≥ 6 indicates a high stress level in the study [32]. The reliability of the scale is 0.77.

Coping strategies (Mini-COPE)

Coping strategies were assessed using a modified 12-item Mini-COPE scale, adapted from the Brief COPE inventory. Items were grouped into three categories: active coping (4 items), maladaptive coping (4 items), and emotional coping (4 items). Each item was rated on a 4-point Likert scale (1 = I haven’t been doing this at all to 4 = I have been doing this a lot), with subscale scores ranging from 4 to 16. Higher scores indicated greater use of the respective coping style [33]. The reliability of the scale is 0.77.

Perceived social support (MSPSS scale)

The Multidimensional Scale of Perceived Social Support (MSPSS) was used to measure the perceived social support [34]. The MSPSS was assessed by three sources of social support: family support (4 items), friend support (4 items), and significant support by others (4 items). The 12-item scale was rated on a 7-point Likert scale (1 = very strongly disagree to 7 = very strongly agree). The total score of the scale was calculated by summing the items. The higher mean score indicates the higher levels of perceived social support. To obtain the subscales scores, each subscale score was summed up and divided by 4 to calculate the mean score for the respective dimension. In this approach the mean score ranging from 1 to 2.9 is considered as low support; a score of 3 to 5 considered moderate support; and a score from 5.1 to 7 is considered high support. The reliability of the scale is 0.90.

Ethics statement

All study procedures complied with the ethical standards of the relevant national and institutional committees, as well as the Declaration of Helsinki (1975, revised 2013). Ethical approval was obtained from the CHINTA Research Bangladesh Ethics Committee [Reference: chinta/2025/02(5)]. Written informed consent was obtained from all participants after providing information regarding the study objectives, procedures, potential risks, and voluntary nature of participation. Data were stored securely and used exclusively for research purposes. No incentives were provided.

Statistical analysis

Prior to conducting mediation analyses, we examined the bivariate relationships among the key study variables using Pearson’s correlation coefficients. The variables included ACEs, three coping strategies (active coping, maladaptive coping, and emotional coping), three types of social support (family, friend, and support from significant other persons), and three psychological outcomes (depression, anxiety, and stress). Pearson’s r values along with significance levels were computed using the psych package in R. The correlation matrix revealed several statistically significant associations.

To explore these relationships further, we conducted two separate parallel multiple mediation analyses using structural equation modeling (SEM) in the lavaan package in R. The first mediation model examined whether different types of perceived social support—family, friend, and support from other significant persons (e.g. a close friend or mentor) mediated the effect of ACEs on psychological outcomes. ACEs were modeled as the independent variable, and depression, anxiety, and stress served as separate dependent variables. All three forms of support were specified as parallel mediators. Bootstrapping with 1,000 samples was used to estimate confidence intervals for indirect effects. Covariances among the mediators were included to account for shared variance. The results provided estimates of the direct effects of ACEs on psychological outcomes, indirect effects through each type of social support, and total effects.

A second parallel mediation model was then conducted separately to examine the role of coping strategies as mediators. In this model, active coping, maladaptive coping, and emotional coping were specified as mediators of the relationship between ACEs and each of the psychological outcomes. The decision to run the two mediation models independently was made to reduce model complexity and avoid multicollinearity, as perceived social support and coping strategies may be conceptually and statistically interrelated. The model was similarly estimated using bootstrap procedures and included covariances between emotional coping and the other two coping styles to account for intercorrelations. All three outcome variables- depression, anxiety, and stress were allowed to correlate within the SEM model, given their high conceptual and empirical interrelations. Model fit was assessed using standard fit indices including the Comparative Fit Index (CFI) ≥ 0.95, Tucker-Lewis Index (TLI) ≥ 0.95, Root Mean Square Error of Approximation (RMSEA) ≤ 0.06, and Standardized Root Mean Square Residual (SRMR) ≤ 0.08 [35].

To examine the relationship between ACEs and mental health outcomes, we conducted a series of logistic regression analyses. For these analyses, depression and anxiety were treated as binary outcomes based on the PHQ-2 and GAD-2 subscales of the PHQ-4, with a cut-off score of ≥ 3 indicating a positive screen for each condition. This threshold is widely used and has been validated as an effective marker for identifying probable cases of depression and anxiety in both clinical and community settings [30]. Stress was similarly dichotomized using a PSS-4 score of ≥ 6, consistent with prior research identifying this threshold as indicative of elevated stress [32]. Depression, anxiety, and stress served as the dependent variables, while different types of childhood exposure to abuse (psychological, physical, and sexual), gender, and household adversities were modeled as independent variables. All odds ratios (OR) were adjusted for gender to account for potential confounding effects. The predictive strength of each model was assessed using Nagelkerke R² values, which indicated the proportion of variance explained by the predictors.

Results

Characteristics of the participants

The mean age of the participants was 19.15 ± 1.06. Among the participants, 50.6% were male and 56.7% were from rural areas. Approximately 42.67% of participants had at least one ACE [mean(sd) = 0.87(1.35)]. Regarding psychological symptoms, 39% had depression, 58% had anxiety, and 60.2% showed high stress.

Correlations between the observed variables

Table 1 presents the correlations between observed variables, revealing the relationships among ACE, coping strategies (active, maladaptive, emotion coping), social support dimensions (friend, family, support from other significant persons), and psychosocial outcomes (depression, anxiety, stress).

Table 1.

Correlations between the observed variables

Variables	1	2	3	4	5	6	7	8	9	10
1. ACE	–
2. Active Coping	−0.183*	–
3. Maladaptive Coping	0.001	0.002	–
4. Emotion Coping	−0.096*	0.263*	0.148*	–
5. Friend Support	−0.200*	0.220*	0.009	0.346*	–
6. Family Support	−0.340*	0.332*	−0.007	0.258*	0.489*	–
7. Support from other significant persons	−0.126*	0.180*	0.038	0.289*	0.363*	0.311*	–
8. Depression	0.044	−0.120*	0.243*	−0.036	−0.074*	−0.130*	0.014	–
9. Anxiety	0.048	−0.146*	0.310*	−0.095*	−0.141*	−0.178*	−0.029	0.510*	–
10. Stress	0.159*	−0.279*	0.286*	−0.125*	−0.176*	−0.272*	−0.105*	0.455*	0.614*	–

*p < 0.01

As expected, ACE shows significant negative correlations with active coping (r = −0.183, p < 0.05), emotion coping (r = −0.096, p < 0.05), and all dimensions of social support like friend support (r = −0.200, p < 0.05), family support (r = −0.340, p < 0.05), and support from other significant persons (r = −0.126, p < 0.05). ACE is positively correlated with stress (r = 0.159, p < 0.05), but its correlations with depression (r = 0.044, p = 0.14) and anxiety (r = 0.048, p = 0.10) were non-significant.

Among coping strategies, active coping is positively correlated with emotion coping (r = 0.263, p < 0.05) and all social support measures, particularly family support (r = 0.332, p < 0.05). Conversely, maladaptive coping shows weaker correlations overall, though it is significantly positively related to depression (r = 0.243, p < 0.05), anxiety (r = 0.310, p < 0.05), and stress (r = 0.286, p < 0.05).Emotion coping demonstrates moderate positive correlations with social support dimensions like friend support (r = 0.346, p < 0.05), family support (r = 0.258, p < 0.05), and support from other significant persons (r = 0.289, p < 0.05), suggesting its association with emotional resourcefulness and interpersonal networks. However, its associations with psychosocial outcomes are mixed, showing non-significant correlations with depression and weaker negative correlations with anxiety (r = −0.095, p < 0.05) and stress (r = −0.125, p < 0.05).Finally, the relationships among psychosocial outcomes are notably strong and positive. Depression correlates significantly with anxiety (r = 0.510, p < 0.05) and stress (r = 0.455, p < 0.05), while anxiety and stress show the strongest correlation (r = 0.614, p < 0.05), indicating the interconnected nature of these mental health challenges.

This pattern of correlations underscores the multifaceted impact of ACE on coping mechanisms, support systems, and mental health. Active coping strategies and robust social support appear to buffer against psychosocial distress, while maladaptive coping amplifies adverse outcomes. These findings provide a foundational basis for understanding the mediating roles of coping and social support in the ACE-mental health relationship.

Standardized effect of aces on mental health through social support

The structural equation model examining the mediating role of social support in the association between ACEs and mental health outcomes revealed several significant pathways (Fig. 1). ACEs had strong negative direct effects on family support (β = −0.339, 95% CI [−0.402, −0.275], p < 0.001), friend support (β = −0.239, 95% CI [−0.314, −0.172], p < 0.001), and support from other significant persons (β = −0.199, 95% CI [−0.288, −0.120], p < 0.001). Among the direct effects of ACEs on mental health, only stress showed a statistically significant positive association (β = 0.210, 95% CI [0.017, 0.395], p = 0.030), while the effects on depression and anxiety were non-significant. These results indicate that the associations between ACEs and both depression and anxiety were fully mediated by social support, in particular family support. Whereas the association with stress was only partially mediated by family support. Notably, indirect effects through family support were significant for all three mental health outcomes, including depression (β = 0.042, p = 0.002), anxiety (β = 0.063, p < 0.001), and stress (β = 0.217, p < 0.001). Friend support mediated the association between ACEs and anxiety (β = 0.020, p = 0.041), but its indirect effects on depression and stress were not significant. Support from other significant persons showed a marginally significant negative indirect effect on depression (β = −0.008, p = 0.060) but no significant effects on anxiety or stress. Overall, the total effect of ACEs on stress was substantial and significant (β = 0.459, 95% CI [0.276, 0.638], p < 0.001), highlighting stress as the most robust mental health outcome influenced by ACEs through diminished social support. The model fit indices indicated excellent fit [χ²(df) = 1694.257 (21), p < 0.001; CFI = 0.99, TLI = 0.99, RMSEA < 0.001, SRMR < 0.001] (Table 2).

Fig. 1.

Conceptualize structural equation modeling between ACE and mental health through social support and coping behavior. [Note: Mediators and outcome variables were correlated within each model.]

Table 2.

Standardized effect of aces on mental health through social support

Effect Type	Path/outcome	Beta (β)	95% CI	p-value
Direct Effects
ACE → Family Support	Family Support	−0.339	[−0.402, −0.275]	< 0.001
ACE → Friend Support	Friend Support	−0.239	[−0.314, −0.172]	< 0.001
ACE → Support from other significant persons	Support from other significant persons	−0.199	[−0.288, −0.120]	< 0.001
ACE → Depression	Depression	−0.000	[−0.071, 0.068]	0.999
ACE → Anxiety	Anxiety	−0.019	[−0.101, 0.055]	0.627
ACE → Stress	Stress	0.210	[0.017, 0.395]	0.030
Indirect Effects
ACE → Family Support → Depression	Depression	0.042	[0.019, 0.071]	0.002
ACE → Friend Support → Depression	Depression	0.006	[−0.007, 0.019]	0.374
ACE → Support from other significant persons → Depression	Depression	−0.008	[−0.016, −0.001]	0.060
ACE → Family Support → Anxiety	Anxiety	0.063	[0.031, 0.099]	< 0.001
ACE → Friend Support → Anxiety	Anxiety	0.020	[0.002, 0.041]	0.041
ACE → Support from other significant persons → Anxiety	Anxiety	−0.007	[−0.018, 0.002]	0.148
ACE → Family Support → Stress	Stress	0.217	[0.138, 0.316]	< 0.001
ACE → Friend Support → Stress	Stress	0.029	[−0.013, 0.073]	0.179
ACE → Support from other significant persons → Stress	Stress	0.003	[−0.020, 0.028]	0.776
Total Effects
ACE → Depression	Depression	0.040	[−0.031, 0.105]	0.235
ACE → Anxiety	Anxiety	0.057	[−0.020, 0.129]	0.122
ACE → Stress	Stress	0.459	[0.276, 0.638]	< 0.001

Standardized effect of aces on mental health through coping strategies

The second model evaluated coping strategies as mediators between ACEs and mental health outcomes (Fig. 1). ACEs were significantly negatively associated with active coping (β = −0.345, 95% CI [−0.462, −0.219], p < 0.001) and emotion-focused coping (β = −0.221, 95% CI [−0.348, −0.092], p = 0.001), while no significant effect was found on maladaptive coping. Among direct effects, only the association between ACEs and stress reached statistical significance (β = 0.307, 95% CI [0.125, 0.488], p = 0.001); effects on depression and anxiety remained non-significant. These findings indicate that the associations between ACEs and both depression and anxiety were fully mediated by coping strategies, particularly active coping. In contrast, the association between ACEs and stress was partially mediated, as both the direct and indirect effects were significant. In terms of indirect effects, active coping significantly mediated the relationships between ACEs and depression (β = 0.017, p = 0.007), anxiety (β = 0.025, p = 0.003), and stress (β = 0.125, p < 0.001). Emotion-focused coping also showed a small but significant mediation effect on anxiety (β = 0.013, p = 0.013) and stress (β = 0.027, p = 0.021), but not on depression. Maladaptive coping did not serve as a significant mediator for any outcomes, and ACEs were not significantly associated with maladaptive coping. This suggests that, in this sample, ACEs were not linked to increased reliance on maladaptive coping strategies. The total effect of ACEs on stress remained substantial and significant (β = 0.459, 95% CI [0.276, 0.638], p < 0.001), highlighting the importance of coping strategies particularly active coping in understanding the psychological burden of ACEs. The model also demonstrated excellent fit [χ²(df) = 1364.911 (21), p < 0.001; CFI = 0.99, TLI = 0.99, RMSEA < 0.001, SRMR = 0.001] (Table 3).

Table 3.

Standardized effect of aces on mental health through coping strategies

Effect Type	Path/outcome	Beta (β)	95% CI	p-value
Direct Effects
ACE → Active Coping	Active Coping	−0.345	[−0.462, −0.219]	< 0.001
ACE → Maladaptive Coping	Maladaptive Coping	0.002	[−0.149, 0.150]	0.981
ACE → Emotion Coping	Emotion Coping	−0.221	[−0.348, −0.092]	0.001
ACE → Depression	Depression	0.018	[−0.047 0.082]	0.563
ACE → Anxiety	Anxiety	0.019	[−0.057, 0.094]	0.611
ACE → Stress	Stress	0.307	[0.125, 0.488]	0.001
Indirect Effects
ACE → Active Coping → Depression	Depression	0.017	[0.006, 0.031]	0.007
ACE → Maladaptive Coping → Depression	Depression	0.000	[−0.014, 0.013]	0.981
ACE → Emotion Coping → Depression	Depression	0.004	[−0.001, 0.010]	0.217
ACE → Active Coping → Anxiety	Anxiety	0.025	[0.010, 0.042]	0.003
ACE → Maladaptive Coping → Anxiety	Anxiety	0.000	[−0.022, 0.024]	0.981
ACE → Emotion Coping → Anxiety	Anxiety	0.013	[0.004, 0.024]	0.013
ACE → Active Coping → Stress	Stress	0.125	[0.074, 0.185]	< 0.001
ACE → Maladaptive Coping → Stress	Stress	0.001	[−0.051, 0.052]	0.981
ACE → Emotion Coping → Stress	Stress	0.027	[0.006, 0.052]	0.021
Total Effects
ACE → Depression	Depression	0.040	[−0.031, 0.105]	0.233
ACE → Anxiety	Anxiety	0.057	[−0.020, 0.129]	0.123
ACE → Stress	Stress	0.459	[0.276, 0.638]	< 0.001

Relationship between aces and mental health problems

Table 4 presents the associations between various forms of ACEs and mental health problems, specifically depression, anxiety, and stress treated as binary variables. The regression models, adjusted for gender, explained 4.9%, 5.2%, and 6.8% of the variance in these outcomes, respectively. These values suggest that while ACEs were significantly associated with psychological outcomes, the overall variance explained was modest.

Table 4.

Relationship between ACEs and mental health problems

Childhood exposure to abuse	Depression (Nagelkerke R2= 4.9%)			Anxiety (Nagelkerke R2= 5.2%)					Stress (Nagelkerke R2= 6.8%)
			95% CI				95% CI				95% CI
	Sig.	OR	Lower	Upper	Sig.	OR	Lower	Upper	Sig.	OR	Lower	Upper
Gender	<.001	1.82	1.42	2.33	<.001	1.94	1.52	2.49	<.001	1.90	1.48	2.44
Psychological	.002	1.61	1.18	2.19	.046	1.37	1.00	1.88	.001	1.76	1.27	2.45
Physical	.706	1.09	.67	1.78	.963	.98	.60	1.61	.680	.89	.53	1.49
Sexual	.081	1.41	.95	2.09	.325	1.22	.81	1.84	.004	1.89	1.22	2.95
Household mental illness	.286	.80	.54	1.19	.251	.80	.54	1.17	.590	.89	.60	1.32
Mother was treated violently	.184	.72	.45	1.16	.433	1.20	.75	1.93	.638	1.12	.69	1.83
Criminal behavior in the household	.456	.79	.43	1.45	.050	.55	.31	.99	.541	.83	.46	1.50
Substance user in the household	.573	1.14	.71	1.83	.779	.93	.59	1.48	.379	.81	.50	1.29
Constant	<.001	.41			.588	.95			.519	.94

Odds ratios (OR) adjusted for gender

Reference category was “no” for all childhood exposure to abuse categories and “male” for gender

Gender was a significant factor across all outcomes, with females showing increased odds for depression (OR = 1.82; 95% CI: 1.42–2.33; p < 0.001), anxiety (OR = 1.94; 95% CI: 1.52–2.49; p < 0.001), and stress (OR = 1.90; 95% CI: 1.48–2.44). Psychological abuse was significantly associated with higher odds of depression (OR = 1.61; 95% CI: 1.18–2.19; p = 0.002), anxiety (OR = 1.37; 95% CI: 1.00–1.88; p = 0.046), and stress (OR = 1.76; 95% CI: 1.27–2.45; p = 0.001). Sexual abuse was significantly associated with stress (OR = 1.89; 95% CI: 1.22–2.95; p = 0.004), although it did not reach statistical significance for anxiety and depression. Other forms of ACEs, such as physical abuse, mental illness in the household, mother being treated violently, and household substance use and criminal behavior, did not demonstrate statistically significant associations with the mental health outcomes.

Discussion

The present study aimed to investigate the associations between ACEs and mental health problems among Bangladeshi adolescents and the mediating roles of social support and coping strategies between these relationships. The findings highlight the complex relationships between ACEs, coping strategies, social support, and mental health outcomes. ACEs were significantly associated with diminished social support across all dimensions (friend, family, and support from other significant persons), which in turn mediated their impact on depression, anxiety, and stress emphasizing the pivotal role of family support as a modifiable protective factor in adolescent psychopathology. Active coping emerged as a key mediator, linking ACEs to all mental health outcomes, while emotion-focused coping mediated anxiety and stress. Notably, maladaptive coping did not serve as a mediator. Among mental health outcomes, stress was the most robustly influenced, both directly and indirectly, by ACEs. Gender differences further revealed higher odds of depression, anxiety, and stress among females, with psychological and sexual abuse being significant factors of these outcomes. These findings contribute to the growing body of developmental psychopathology literature by identifying culturally relevant psychosocial pathways through which early adversity affects adolescent mental health in a low-resource context.

Consistent with some existing literature, the present study found that ACEs were significantly associated with greater stress among adolescents, but not with depression or anxiety. This aligns with research suggesting that early adversity often triggers heightened stress responses that may precede or exacerbate other forms of internalizing symptoms [14]. This pattern may reflect a developmental cascade, whereby chronic exposure to stress impairs regulatory systems before manifesting as clinical disorders. The consistency between our results and parts of the existing literature may be attributed to common mechanisms, such as heightened physiological reactivity, dysregulation of the hypothalamic–pituitary–adrenal (HPA) axis, and chronic activation of stress pathways following exposure to early adversity [36]. However, the lack of significant associations between ACEs and depression or anxiety in the present study may reflect differences in sample characteristics, cultural perceptions of emotional expression, measurement tools, or the developmental timing of symptom emergence. Previous study using data from the Canadian Health Survey reported significant association between childhood maltreatment and depression and anxiety, further reflecting the differences in sample characteristics (high school graduates vs. general population), sampling design (convenience vs. multistage stratified), measurement tools (PHQ-4 vs., WHO-CIDI) and cultural factors influencing symptom expression and reporting [37]. In particular, underreporting of emotional symptoms in settings with high mental health stigma, such as Bangladesh may account for weaker associations observed in our study.

In some sociocultural contexts, such as Bangladesh, adolescents may normalize emotional distress or underreport symptoms like depression and anxiety due to stigma and limited mental health literacy [38, 39]. Moreover, cultural coping norms that emphasize endurance and familial obligations might delay the expression or recognition of internalizing symptoms, potentially explaining the observed pattern of results.

The present study also found that ACEs were significantly associated with diminished perceived social support across all dimensions such as support from friends, family, and other significant individuals as well as with lower levels of active and emotional coping strategies. These findings are consistent with previous research demonstrating that exposure to early adversity can severely disrupt the development of interpersonal trust and relational skills, leading to lower levels of social support [17, 40, 41]. Similarly, research by Evans et al. (2013) found that cumulative childhood adversity predicts lower perceived social support in adolescence and adulthood, suggesting that early relational disruptions have lasting consequences [20].

Our findings regarding the association between ACEs and diminished active and emotional coping strategies align with studies suggesting that early adversities compromise the development of active emotional regulation and problem-solving skills [17, 25, 42]. Adolescents who grow up in chaotic, neglectful, or abusive environments often have fewer opportunities to learn and practice effective coping mechanisms, resulting in greater reliance on maladaptive or avoidant coping styles [43]. The similarities between our results and previous research may be explained by the shared underlying mechanisms, such as alterations in neurodevelopment (e.g., impaired prefrontal cortex functioning) and chronic stress exposure that undermines coping system development [36].

The present study also revealed important mediation effects, highlighting the mechanisms through which ACEs influence adolescent mental health. Specifically, family support significantly mediated the relationship between ACEs and depression, anxiety, and stress, while friend support mediated the relationship between ACEs and anxiety. These findings align with previous studies emphasizing the protective role of social support in mitigating the adverse psychological effects of early life stressors [20, 37, 44]. In particular, family support has been consistently identified as a critical buffer against the development of internalizing problems among adolescents exposed to adversity, likely due to the foundational role that familial relationships play in emotional security and resilience [17, 45]. This protective function may be particularly salient in the Bangladeshi context, where collectivist cultural norms and strong intergenerational family ties position the family as the primary source of emotional and instrumental support. Adolescents who perceive their families as understanding and supportive may be better equipped to regulate emotional responses to early adversity, thereby reducing vulnerability to internalizing symptoms such as depression.

Our results also mirror findings from a previous study indicating friend support becomes particularly salient in buffering anxiety symptoms during adolescence, a developmental period characterized by increased reliance on peer networks [45, 46]. Peer relationships are particularly important for navigating academic and social stressors and having trusted friends to share concerns may reduce anxiety and daily stress, even if it does not directly impact deeper emotional outcomes such as depression.

In addition, active coping strategies were found to mediate the relationship between ACEs and depression, anxiety, and stress, while emotional coping mediated the relationship between ACEs and anxiety and stress. These findings are in line with prior research suggesting that adolescents exposed to ACEs often experience disruptions in developing effective coping mechanisms [42, 43]. Similar findings have been reported in a study among Canadian general population, suggesting positive coping skills mediate the relationship between childhood abuse and depression and anxiety [37]. Adaptive coping strategies, such as active coping, have been shown to promote psychological resilience by enabling individuals to engage with stressors constructively, thereby reducing the likelihood of developing emotional problems [47]. The ability to actively address or reframe stressors, and process emotions, may help adolescents mitigate anxiety and stress. In contrast, depression may be more strongly linked to chronic feelings of helplessness or lack of perceived support, rather than specific coping styles. Emotional coping, while traditionally considered less effective than active coping, can still serve as a protective factor by facilitating emotional processing, particularly for anxiety and stress symptoms [42]. This suggests that emotion-focused strategies may have adaptive value in high-stress, low-control environments like those experienced by many adolescents in Bangladesh.

Notably, maladaptive coping did not mediate the relationship between ACEs and any mental health outcomes, nor was it significantly associated with ACE exposure. This finding is theoretically relevant, as it suggests that Bangladeshi adolescents exposed to childhood adversity may not necessarily adopt maladaptive coping styles such as avoidance or disengagement. It is possible that cultural or contextual factors, such as social expectations for emotional control or limited opportunities for behavioral disengagement, may attenuate the development or expression of these coping styles in this population. These findings emphasize the value of strengthening both familial and peer support networks and promoting active coping skills in school-based interventions to mitigate the psychological burden of ACEs among adolescents.

These findings are consistent with studies in high-income countries, where social support and coping behaviour have also been shown to buffer the effects of childhood adversity on mental health [37, 42, 43, 48]. This consistency underscores the universal relevance of these psychosocial resources across diverse sociocultural settings. However, the current study extends this literature by demonstrating these pathways in a South Asian context, where cultural norms, family structure, and mental health stigma may influence both the expression of adversity and the utilization of support systems.

The present study further revealed notable gender differences in mental health outcomes, with females exhibiting higher odds of experiencing depression, anxiety, and stress compared to males. Consistent with prior research, female adolescents appear more vulnerable to internalizing symptoms following exposure to adverse experiences [49, 50]. Several factors may contribute to this gender disparity. Biological differences, such as heightened hypothalamic-pituitary-adrenal (HPA) axis reactivity to stress in females may increase their vulnerability to emotional dysregulation [51]. Girls may experience unique psychosocial stressors, including greater exposure to emotional abuse, early caregiving responsibilities, and restricted autonomy, which can increase vulnerability to psychological distress. Additionally, societal norms in Bangladesh may encourage emotional expression more in females than males, potentially increasing the likelihood of symptom reporting among girls. These findings reinforce the need for gender-sensitive mental health strategies that consider the specific stressors, needs, and help-seeking patterns of adolescent girls.

In addition, psychological abuse was found to be significantly associated with higher odds of depression, anxiety, and stress, while sexual abuse was significantly associated with stress. These findings are aligned with previous studies demonstrating that psychological maltreatment often involving verbal aggression, emotional manipulation, and humiliation has profound and lasting impacts on adolescents’ emotional well-being [52, 53]. Psychological abuse, although less visible than physical forms of abuse, directly undermines self-esteem, trust, and emotional regulation capacities, increasing vulnerability to a wide range of psychiatric symptoms [54]. In Bangladesh, where discussions of mental health and emotional mistreatment are often stigmatized or minimized, psychological abuse may be both more prevalent and less likely to be recognized or reported, which may explain its particularly strong association with internalizing symptoms in this sample.

The significant association between sexual abuse and stress in our study is similarly supported by previous findings showing that experiences of sexual victimization are powerful predictors of chronic stress and trauma-related disorders [55, 56]. However, in contrast to some research that identified strong links between sexual abuse and depression or anxiety [57, 58], our findings showed a more specific association with stress. This variation could be attributed to cultural and contextual factors in Bangladesh, where stigma and silence surrounding sexual abuse may lead victims to internalize their experiences differently, manifesting more prominently psychological stress responses rather than clinically diagnosed depressive or anxiety disorders [59, 60]. It is also possible that the chronicity, severity, and timing of abuse, as well as available support systems, moderate these relationships. For instance, individuals who lack supportive environments may be more likely to experience ongoing stress symptoms following sexual abuse rather than develop overt depression or anxiety [61]. Overall, these findings advance conceptual understanding by delineating how modifiable psychosocial processes like social support and coping mediate the ACE-mental health link in a resource-constrained and culturally unique adolescent population.

Although our regression models identified significant associations between ACEs and mental health outcomes, they explained only a small proportion of variance in psychological distress (Nagelkerke R² = 4.9–6.8%). This indicates that additional factors, such as current life stressors, social context, or biological vulnerabilities likely contribute to adolescent mental health and were not captured in the present analysis. Moreover, while we adjusted for gender due to its strong association with mental health outcomes, we did not include other covariates such as age or residence status. Given the narrow age range and sampling context, their exclusion is unlikely to bias results substantially, though the potential influence of unmeasured confounders cannot be ruled out.

Strength, limitations, and future directions

This study offers several notable strengths. First, it is among the few studies to examine the complex associations between ACEs, coping strategies, social support, and mental health outcomes among adolescents in Bangladesh in a low-resource and understudied context. Second, the study used standardized, validated measures for assessing ACEs, depression, anxiety, stress, coping styles, and social support, enhancing the reliability and comparability of findings with existing research. Third, the large sample size and geographic diversity of participants enhance the relevance of findings for adolescents in similar educational and socio-cultural contexts. Additionally, the exploration of multiple mediators (family support, friend support, active and emotional coping) provided a nuanced understanding of potential pathways of adolescence resilience and risk in the face of adversity.

However, some limitations should be acknowledged. The cross-sectional design precludes any causal inferences between ACEs, coping, social support, and mental health outcomes. Second, reliance on self-report data introduces potential bias, including recall bias and underreporting of sensitive experiences such as abuse due to social desirability or stigma. Third, while the study utilized a relatively large and geographically diverse sample, participants were all university entrance examinees. This academic context may exclude more marginalized or out-of-school adolescents and may reflect a subgroup with higher resilience, social support, or educational access. As such, the generalizability of the findings should be limited to similar populations and interpreted with caution in broader adolescent contexts. Another limitation concerns the operationalization of ACEs as a unidimensional cumulative score, which may obscure important differences in the type, timing, and chronicity of adverse experiences. Treating all ACEs as equally weighted fails to capture whether certain experiences (e.g., early vs. late adversity, chronic vs. acute exposure) have differential effects on the development of coping strategies and social support systems. Future research should explore these dimensions using more nuanced approaches such as typological or longitudinal methods. Additionally, we were unable to adjust for potentially important contextual covariates such as socioeconomic status, parental education, and school access due to unavailable data. These factors are known to influence both ACE exposure and adolescent mental health, and their omission may have introduced residual confounding or inflated the observed mediating effects of social support and coping. Future studies should include these variables to provide a more comprehensive understanding of adolescent mental health risk pathways.

To address the limitation of cross-sectional data, future research should employ longitudinal designs to better capture the long-term effects of ACEs on adolescent and adult mental health outcomes, and to test causal mediation models. Incorporating multi-informant reports (e.g., from parents, teachers) and objective records where possible could help reduce biases associated with self-report data. Moreover, future studies could explore potential protective factors, such as resilience, religious coping, and community engagement, that may buffer the impact of ACEs in Bangladeshi adolescents. Given the observed gender differences, gender-sensitive longitudinal investigations would provide further insight into the developmental pathways linking early adversity to psychopathology.

Practical implications

The findings of this study have important practical implications for mental health prevention and intervention efforts in LMICs. Interventions targeting adolescents exposed to ACEs should prioritize strengthening family functioning and peer support, both of which emerged as crucial protective factors. Schools and community programs can play a pivotal role by adopting trauma-informed practices that identify and support at-risk adolescents.

School-based mental health programs, such as group counseling, peer support networks, and psychoeducation on coping, may help mitigate emotional distress and build resilience. Given the observed deficits in active coping among adolescents exposed to adversity, programs that explicitly teach problem-solving, cognitive restructuring, and emotional regulation skills are warranted. These can be delivered through low-cost, culturally adapted modules within existing educational frameworks.

From a policy perspective, the results underscore the need for adolescent-friendly mental health services, child protection mechanisms, and anti-stigma campaigns around abuse and psychological distress. Integrating ACE awareness into teacher training, school curricula, and community outreach initiatives could facilitate early identification and intervention. Investment in early prevention, mental health literacy, and trauma-informed care practices in schools and communities can help reduce the long-term burden of childhood adversity.

Conclusion

This study contributes to a growing body of evidence highlighting the profound impact of ACEs on adolescent mental health, emphasizing the critical roles of social support and coping strategies in mediating these relationships. Our findings revealed that diminished family and friend support, along with less use of active and emotional coping strategies, partially explained the link between ACEs and heightened symptoms of depression, anxiety, and stress. Moreover, gender differences and the specific influence of psychological and sexual abuse underscore the need for targeted interventions. These results call for urgent efforts to strengthen social support systems, promote active coping skills, and develop culturally sensitive, gender-responsive mental health interventions for adolescents in Bangladesh. Future longitudinal research is essential to unravel the causal pathways and to inform comprehensive prevention and intervention strategies aimed at mitigating the long-term consequences of childhood adversity. 

Authors’ contributions

FAM conceived and designed the study, conducted data analysis, interpreted the results, drafted the initial manuscript, and led subsequent revisions. AAH contributed to project administration, data collection, data analysis, and interpretation of findings. MMA and PS critically reviewed and edited the manuscript. MAM contributed to the study design and provided oversight, as well as reviewing and editing the manuscript. All authors read and approved the final version of the manuscript.

Funding

Dr Almerab is currently receiving funding support from the Princess Nourah Bint Abdulrahman University Researchers Supporting Project (number PNURSP2025R563), Princess Nourah Bint Abdulrahman University, Riyadh, Saudi Arabia.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethical approval and consent to participate

All study procedures complied with the ethical standards of the relevant national and institutional committees, as well as the Declaration of Helsinki (1975, revised 2013). Ethical approval was obtained from the CHINTA Research Bangladesh Ethics Committee [Reference: chinta/2025/02(5)]. Written informed consent was obtained from all participants after providing information regarding the study objectives, procedures, potential risks, and voluntary nature of participation. Data were stored securely and used exclusively for research purposes. No incentives were provided.

Consent for publication

Not required.

Competing interests

The authors declare no competing interests.