Childhood Trauma Meta-Analysis Study Group including, Erika, Kuzminskaite, Anouk, W Gathier, Pim, Cuijpers, Brenda W J, H Penninx, Robert, T Ammerman, Eva-Lotta, Brakemeier, Sanne, Bruijniks, Carletto, Sara, Trisha, Chakrabarty, Katie, Douglas, Boadie, W Dunlop, Moritz, Elsaesser, Frank, Euteneuer, Anne, Guhn, Elizabeth, D Handley, Erkki, Heinonen, Marcus J, H Huibers, Andrea, Jobst, Gary, R Johnson, Daniel, N Klein, Johannes, Kopf-Beck, Lotte, Lemmens, Xiao-Wen, Lu, Somaia, Mohamed, Atsuo, Nakagawa, Satoshi, Okada, Winfried, Rief, Leonardo, Tozzi, Madhukar, H Trivedi, Suzanne van Bronswijk, Patricia van Oppen, Sidney, Zisook, Ingo, Zobel, Christiaan, H Vinkers, Psychiatry, APH - Mental Health, Amsterdam Neuroscience - Mood, Anxiety, Psychosis, Stress & Sleep, Amsterdam Neuroscience - Complex Trait Genetics, Amsterdam Neuroscience - Compulsivity, Impulsivity & Attention, Anatomy and neurosciences, and APH - Digital Health
Background: Childhood trauma is a common and potent risk factor for developing major depressive disorder in adulthood, associated with earlier onset, more chronic or recurrent symptoms, and greater probability of having comorbidities. Some studies indicate that evidence-based pharmacotherapies and psychotherapies for adult depression might be less efficacious in patients with a history of childhood trauma than patients without childhood trauma, but findings are inconsistent. Therefore, we examined whether individuals with major depressive disorder, including chronic forms of depression, and a reported history of childhood trauma, had more severe depressive symptoms before treatment, had more unfavourable treatment outcomes following active treatments, and were less likely to benefit from active treatments relative to a control condition, compared with individuals with depression without childhood trauma. Methods: We did a comprehensive meta-analysis (PROSPERO CRD42020220139). Study selection combined the search of bibliographical databases (PubMed, PsycINFO, and Embase) from Nov 21, 2013, to March 16, 2020, and full-text randomised clinical trials (RCTs) identified from several sources (1966 up to 2016–19) to identify articles in English. RCTs and open trials comparing the efficacy or effectiveness of evidence-based pharmacotherapy, psychotherapy, or combination intervention for adult patients with depressive disorders and the presence or absence of childhood trauma were included. Two independent researchers extracted study characteristics. Group data for effect-size calculations were requested from study authors. The primary outcome was depression severity change from baseline to the end of the acute treatment phase, expressed as standardised effect size (Hedges' g). Meta-analyses were done using random-effects models. Findings: From 10 505 publications, 54 trials met the inclusion criteria, of which 29 (20 RCTs and nine open trials) contributed data of a maximum of 6830 participants (age range 18–85 years, male and female individuals and specific ethnicity data unavailable). More than half (4268 [62%] of 6830) of patients with major depressive disorder reported a history of childhood trauma. Despite having more severe depression at baseline (g=0·202, 95% CI 0·145 to 0·258, I2=0%), patients with childhood trauma benefitted from active treatment similarly to patients without childhood trauma history (treatment effect difference between groups g=0·016, –0·094 to 0·125, I2=44·3%), with no significant difference in active treatment effects (vs control condition) between individuals with and without childhood trauma (childhood trauma g=0·605, 0·294 to 0·916, I2=58·0%; no childhood trauma g=0·178, –0·195 to 0·552, I2=67·5%; between-group difference p=0·051), and similar dropout rates (risk ratio 1·063, 0·945 to 1·195, I2=0%). Findings did not significantly differ by childhood trauma type, study design, depression diagnosis, assessment method of childhood trauma, study quality, year, or treatment type or length, but differed by country (North American studies showed larger treatment effects for patients with childhood trauma; false discovery rate corrected p=0·0080). Most studies had a moderate to high risk of bias (21 [72%] of 29), but the sensitivity analysis in low-bias studies yielded similar findings to when all studies were included. Interpretation: Contrary to previous studies, we found evidence that the symptoms of patients with major depressive disorder and childhood trauma significantly improve after pharmacological and psychotherapeutic treatments, notwithstanding their higher severity of depressive symptoms. Evidence-based psychotherapy and pharmacotherapy should be offered to patients with major depressive disorder regardless of childhood trauma status. Funding: None.