TABLE 1.
Study characteristics and key research findings analyses (community samples)
| Study (author, year; country) | Population | Sample size (N); female %; mean age | Study design | Measurement of child adversity; definition of childhood adversity | Psychosocial mediating variables; measurement | Psychosocial moderating variables; measurement | Mental health outcomes; measurement | Type of mediation analysis; potential confounders or covariates considered (yes/no) | Mediation/moderation results/key findings (pathway total/partial mediation, direct effect [DE], indirect effect [IE], % total effect mediated) |
|---|---|---|---|---|---|---|---|---|---|
| Cantave et al. (2019), Canada | Community sample part of project aiming to explore the biosocial roots of aggression. | 156; 0%; 24.10 | Cross‐sectional | CTQ‐SF; child maltreatment | Task‐oriented, emotion‐oriented and avoidance‐oriented coping strategies; the Coping Inventory of Stressful Situations (CISS; Endler & Parker, 1994) | Task‐oriented, emotion‐oriented & avoidance‐oriented coping strategies; the Coping Inventory of Stressful Situations (CISS; Endler & Parker, 1994) | Depression; Beck Depression Inventory‐II (BDI‐II; Beck et al., 1996) | Bootstrapping (BC; 95% CI), PROCESS Macro (Hayes, 2013);No | Significant IE of maltreatment on depression via emotion‐oriented coping strategies (b = 1.02 [SE = 0.50], BC 95% CI [0.24–2.20]). IE accounted for 20% of the total effect. Partial mediation.No IE found for task‐oriented and avoidance‐oriented coping strategies.Avoidance‐oriented and emotion‐oriented coping strategies did not moderate the maltreatment‐depression relationship.Task‐oriented coping strategies moderated the maltreatment‐depressive symptoms association (r 2 = 0.02, b = −0.23, t(148) = 2.08, p = 0.04]. 0.04]. |
| Crow et al. (2014), USA | Community sample recruited from the general medical gynaecological clinics | 3,902; 68,9%; 30.34 | Cross‐sectional | CTQ‐SF;Childhood trauma | Emotion dysregulation; Emotion Dysregulation Scale, short version (EDS; Powers et al., 2015) | None tested | Depression; Beck Depression Inventory‐II (BDI‐II; Beck et al., 1996) | Bootstrapping (BC; 95% CI), PROCESS Macro (Hayes, 2008);No | Significant IE (a x b = 0.525, 95% CI [0.48–0.55]) of childhood emotional abuse on depression through emotion dysregulation.Emotional abuse stronger predictor of adult depression. |
| Evans, Steel, and DiLillo (2013), USA | Community sample recruited from marriage licence records. | 372; 50%; 26.59 | Cross‐sectional | CTQ; child maltreatment | None tested | Social support;Perceived Social Support Index (PSS; Procidano & Heller, 1983) | Trauma symptoms;Trauma Symptom Inventory (TSI; Briere, 1995) | Moderation analysis. Two‐step hierarchical regression model, Baron and Kenny (1986) approach;No | Perceived social support (PSS) from family predicted moderation of the relationship between physical abuse (R 2 = 0.09, p = 0.007), emotional abuse (R 2 = 0.13, p = 0.009), emotional neglect (R 2 = 0.12, p = 0.008) and trauma.PSS from family did not moderate the child maltreatment‐ trauma symptoms for men. PSS from friends did not moderate the child maltreatment‐trauma link for both men and women. |
| Fitzgerald and Gallus (2020), USA | Community sample. Data taken from MIDUS study (midlife development in the United States. | 798; 49,1%; 57.52 | Longitudinal | CTQ; childhood maltreatment | Emotional support from family, friends and romantic partner questionnaires. Items specific to MIDUS study. | None tested | Depression; Centre for Epidemiologic Studies Depression (CES‐D; Radloff, 1977)Social Anxiety; Liebowitz Social Anxiety Scale (LSAS; Fresco et al., 2001) | Multiple mediator model. SEM, path analysis;Yes | DE of childhood maltreatment on depression (ß = 0.22, p < 0.05) and social anxiety (ß = 0.10, p < 0.05).Significant IE of child maltreatment on depression via emotional support from family (b = 0.09; 95% CI [0.045–0.146]) and romantic partner (b = 0.02; 95% CI [0.005–0.048]) but not friends (b = 0.02; 95% CI [−.002 to 0.040]). 20.9% of variance in depression.Significant IE of child maltreatment on social anxiety through emotional support from friends (b = 0.04; 95% CI [0.019–0.066]) and romantic partners (b = 0.03; 95% CI [0.008–0.048]). 7.3% of variance in social anxiety of variance in social anxiety. |
| Hayward et al. (2020), USA | Community sample recruited through amazons mechanical Turk. | 382; 48.2%; 35.60 | Cross‐sectional | CTQ‐SF; childhood adversity | Self‐concept clarity; Self‐Concept Clarity Scale (Campbell et al., 1996)Intolerance of uncertainty; Intolerance of Uncertainty Scale‐12 (IUS‐12; Carleton et al., 2007) | None tested | Depression; Depression subscale of the Depression Anxiety Stress Scale (DASS‐21; Lovibond & Lovibond, 1995)Generalized anxiety;GAD‐7 Questionnaire (Spitzer et al., 2006)OCD symptoms; Obsessive–Compulsive Inventory‐Revised (OCI‐R) scale (Foa et al., 2002)Social anxiety; Social Phobia Inventory (SPIN; Connor et al., 2000) | Parallel and serial mediation models; SEM;No | Significant IE of child adversity through SCC on:Depression (b = 0.07; 95% CI [0.05–0.10, p = <0.001);GAD (b = 0.04; 95% CI [0.02–0.05, p = <0.001);OCD C (b = 0.09; 95% CI [0.05–0.10], p = <0.001);Social anxiety (b = 0.13; 95% CI [0.07–0.19, p = <0.001).Both SCC and IOU mediated the link between child adversity and mental health outcomes.54% of the variance in depression, 55% in GAD, 44% in OCD, and 54% in social anxiety.In the model child adversity: Childhood trauma (0.94) and risk families (0.90). 2 IV. |
| Gomes Jardim et al. (2019), Brazil | Community sample‐ elderly population. | 260; 76.9%; 72.2 | Cross‐sectional | CTQ‐SF; childhood maltreatment | Personality factors: NEO‐Five Factor Inventory (NEO‐FFI; Costa and McCrae, 1995) | None tested | Depression; Mini International Neuropsychiatric Interview 5.0 plus Portuguese version (MINI plus) (Amorim, 2000) | SEM;No | Significant IE of childhood maltreatment on depression via neuroticism (b = 0.24, SE = 0.040, p < 0.001) and extraversion (b = 0.09, SE = 0.022, p < 0.001). Total mediation effect.Non‐significant effect of openness (b = 0.02, SE = 0.012, p = 0.091). Significant IE of CTQ score on depression via agreeableness (b = 0.03, SE = 0.017, p = 0.047) and conscientiousness (b = 0.05, SE = 0.016, p = 0.005). Partial mediation effect. |
| Kim et al. (2017), South Korea | Community sample recruited through local newspapers and posters | 1,027; 68.6%; 45.1 | Cross‐sectional | K‐CTQ (Korean version); childhood maltreatment | Rumination;Ruminative Response Scale (RRS; Nolen‐Hoeksema, 1991) | None tested | Depression; Beck Depression Inventory‐II (BDI‐II; Beck et al., 1996)Anxiety;SAI subscale of State Anxiety Inventory (SAI; Spielberger et al., 1983) | SEM; bootstrapping. Sobel test;Yes | Significant DE (b = 0.29, CI [0.003–0.49], p = 0.01) and IE (b = 0.38, CI [0.25–0.56], p = 0.01) of childhood maltreatment on mood through rumination.Total effect: 0.67IE: 0.38Significant mediating role of rumination (Sobel test: 4.19). |
| Klumparendt et al. (2019), Germany | Community sample recruited via PsyWeb. | 1,027; 68.6%; 45.1 | Cross‐sectional | CTQ‐SF (German version); childhood maltreatment | Emotion regulation;Difficulties in Emotion Regulation Scale (DERS; Gratz & Roemer, 2004). German versionAttachment;Experiences in Close Relationship Scale (ECR; Brennan et al., 1998). German versionAttributional style;Depressive Attributions Questionnaire (DAQ; Kleim, et al., 2011). German versionPTSD;PTSD‐Checklist for DSM‐5 (PCL‐5; Weathers et al., 2017) | None tested | Depression; Patient Health Questionnaire (PHQ‐8; Kroenke, Spitzer & Williams, 2001)German version | Multiple mediator model. SEM;No | Significant total IE of childhood maltreatment on depression through mediators (b = 0.32; p < 0.001) and not significant DE of CM on depression in total model (b = 0.002, p = 0.957). Total mediation.Emotion regulation, depressogenic attributional style and PTSD symptom severity mediated the relationship between childhood maltreatment and depression.Emotion regulation: strongest IE of limited access to emotion regulation strategies (b = 0.09, 95% CI [0.042–0.132], p < 0.001) and lack of emotional clarity (b = 0.04, 95% CI [0.017–0.067], p < 0.01).Significant IE through PTSD (b = 0.11, 95% CI [0.72–0.137], p < 0.001).Significant IE through depressogenic attribution style (b = 0.04, 95% CI [0.11–0.72], p = <0.01).No IE for attachment anxiety or avoidance |
| Kogan et al. (2021), USA | Community sample recruited from rural counties. Data from the African American Mens project. | 505; 0%; 21.85 (T1 collection) and 23.12 (T2 collection | Longitudinal | ACE questionnaire;Childhood maltreatment | Defensive schemas; Street Code measure (Stewart & Simons, 2006)Experiences in Close Relationships Scale (Wei et al., & Vogel, 2007)Cynical views of relationships (Simons et al., 2012).Social developmental factors measured by Network of Relationships Inventory (Furman & Buhrmester, 1985Multidimensional Measure of Religious Involvement (Levin et al., 1995)Contextual stressors measured by:Schedule of Racist Events (Landrine & Klonoff, 1996),Economic distress (Cho & Kogan, 2016)The community crime (Sampson et al., 1997). | None psychosocial moderator tested.Moderation analysis included OXTR DNAm | Depression 10‐item version of the Center for Epidemiologic Studies Depression Scale (CES‐D; Björgvinsson, et al., 2013) | Path analysis. Mplus. Bootstrapping (BC; 95% CI);No | ACEs predicted young adult contextual stress (b = 0.39, p < 0.001), which in turn forecast increases in defensive/hostile schemas (b = 0.25, p < 0.05). Defensive/hostile schemas predicted increases in social developmental risk factors (b = 0.15, p < 0.05), which were a proximal antecedent of depressive symptomatology (b = 0.10, p < 0.05) and substance abuse (b = 0.24, p < 0.001).No DE of ACEs on relational schemas.Concurrent stressors mediated the link between ACEs relational schemas.IE linking contextual factors to social developmental factors via defensive relational schemas only when levels of OCTR methylation were high (b = 0.13, 95% CI [0.013–0.242]).OCTR‐DNA methylation = moderator.ACEs affect mental health outcomes indirectly through contextual contemporary factors. |
| McQuaid et al. (2015), Canada | Community sample of first nations recruited from Aborigincal community/health centres. | 225; 75.11%; 35.8 | Cross‐sectional | A 20‐item modified version of the CTQ;Childhood trauma | Unsupportive social interactions; Unsupportive Social Interactions Inventory (USII: Ingram et al., 2001)Perceived discrimination; Perceived Ethnic Discrimination Questionnaire (Contrada et al., 2001) | Ingroup and outgroup unsupport;Unsupportive Social Interactions Inventory (USII: Ingram et al., 2001) | Depression; Beck Depression Inventory short form (BDI‐S; Beck & Beck, 1972) | Single mediation model and moderated mediation model.Multiple mediation. Bootstrapping (BC; 95% CI), PROCESS Macro (Hayes, 2008);Yes | IE of childhood trauma on depression through perceived discrimination. Partial mediation. Significant DE of childhood trauma on depression (b 0.08, p < 0.01).Moderated mediation analyses showed that the intervening role of discrimination was stronger when levels of outgroup unsupport were higher. Ingroup unspport did not moderate the mediated relationship (p = 0.30).Multiple mediation analyses: Emotion focused coping mediated the relationship between childhood trauma and depression (95% CI [0.03–0.11]. The path between emotion‐focused coping and depression was moderated by both ingroup and outgroup unsupport. |
| Mishra and Marceau (2020), USA | Community sample. Data taken from MIDUS (Brim, Ryff, & Kessler, 2004 | 1,053; 55%; 55.79 | Longitudinal | CTQ;Childhood maltreatment | Perceived stress; Perceived Stress Scale (Cohen et al., 1983) | None tested | Depression;the Recent Care of Common Mental Disorder in the United States: Prevalence and Conformance with Evidence‐Based Recommendations (Wang et al., 2000) | Path models. Mplus 7.4;Yes | IE: The effect of sexual abuse on depression partially explained by perceived stress levels in middle life (b = 0.04, p = 0.012).The effect of physical and emotional maltreatment on depression fully mediated by both perceived stress (b = 0.04, p = 0.004) (and cortisol). |
| Stevens et al. (2013), USA | Community sample recruited from an outpatient gynaecological clinic. | 139; 100%; 28.46 | Cross‐sectional | CTQ;Childhood abuse | Emotion Regulation; The Difficulties With Emotion Regulation Scale (DERS; Gratz & Roemer, 2004)Social support;Social Supports Provision Scale (SSPS; Cutrona & Russell, 1987)Interpersonal violence;Conflict Tactics Scale (CTS; Straus, 1979) | None tested | Ptsd;PTSD Symptom Scale–Self Reported (PSS‐SR; Foa et al. 1993) | Multiple mediation model. Bootstrapping (BC; 95% CI) (Preacher & Hayes, 2008);Yes | Significant IE of childhood trauma on PTSD through emotion regulation (b = 0.13, CI 95% [0.07–0.21]).Significant IE of child abuse on PTSD though interpersonal violence & social support.Emotion regulation mediates the relationship between child abuse and trauma. Effect significant after adjusting for social support and interpersonal violence.Social support and interpersonal violence mediates the relationship between child abuse and trauma. The effect of emotion regulation was stronger. |
| van Assche et al. (2020), Belgium | Community sample. Older adults | 81; 64%; 74.90 | Cross‐sectional | CTQ;Childhood trauma | Attachment;Experiences in Close Relationships – Revised (ECR‐R; Fraley et al., 2000) | None tested | Anxiety;Anxiety subscale of the Hospital Anxiety and Depression Scale (HADS‐A; Zigmond & Snaith, 1983)Depression;Geriatric Depression Scale (GD; Yesavage et al., 1982) | Bootstrapping (BC; 95% CI) (Preacher & Hayes, 2004);No | Significant IE of emotional neglect on late life anxiety (b = 0.4. 95% CI [0.00–0.15], p = 0.03)And depression (b = 0.05, 95% CI [0.00–0.17], p = 0.04) through attachment anxiety.No DE found between emotional neglect and depression or anxiety. |
| Wong et al. (2019), USA | Community sample recruited via amazons mechanical Turk. | 308; 45.12%; 35.49 | Cross‐sectional | ACE questionnaireAdverse childhood experiences | Self‐concept clarity;Self‐Concept Clarity (SCC; Campbell et al., 1996)Self‐esteem;Self‐Esteem scale (Rosenberg, 1965) | None tested | Depression;Beck Depression Inventory‐II (BDI‐II; Beck et al., 1996)Perceived stress; Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983). | Bootstrapping (BC; 95% CI) (Hayes, 2013). Mplus 7. Single and dual mediation model (Muthén & Muthén, 1998–2012);Yes | Significant IE of ACEs on depression through self‐concept clarity (SCC) (b = 0.10, 95% CI [0.04–0.16] p = 0.001, r 2 = 0.38) and perceived stress (b = 0.11, 95% CI [0.04–0.18], p = 0.001, r 2 = 0.41). 27% of variance in depression, 31% of variance in stress.Significant IE of ACEs on depression (b = 0.16, 95% CI [0.09–0.23], κ 2 = 0.18)43% of total effect) and perceived stress (b = 0.14, p = <0.001, 95% CI [0.07–0.20], κ 2 = 0.15, 39% of total effect) via self‐esteem.Self‐esteem had larger effect sizes compared to self‐concept clarity. |