Psychological and pharmacological interventions for posttraumatic stress disorder and comorbid mental health problems following complex traumatic events: Systematic review and component network meta-analysis

Abstract

Background: Complex traumatic events associated with armed conflict, forcible displacement, childhood sexual abuse, and domestic violence are increasingly prevalent. People exposed to complex traumatic events are at risk of not only posttraumatic stress disorder (PTSD) but also other mental health comorbidities. Whereas evidence-based psychological and pharmacological treatments are effective for single-event PTSD, it is not known if people who have experienced complex traumatic events can benefit and tolerate these commonly available treatments. Furthermore, it is not known which components of psychological interventions are most effective for managing PTSD in this population. We performed a systematic review and component network meta-analysis to assess the effectiveness of psychological and pharmacological interventions for managing mental health problems in people exposed to complex traumatic events.

Methods and findings: We searched CINAHL, Cochrane Central Register of Controlled Trials, EMBASE, International Pharmaceutical Abstracts, MEDLINE, Published International Literature on Traumatic Stress, PsycINFO, and Science Citation Index for randomised controlled trials (RCTs) and non-RCTs of psychological and pharmacological treatments for PTSD symptoms in people exposed to complex traumatic events, published up to 25 October 2019. We adopted a nondiagnostic approach and included studies of adults who have experienced complex trauma. Complex-trauma subgroups included veterans; childhood sexual abuse; war-affected; refugees; and domestic violence. The primary outcome was reduction in PTSD symptoms. Secondary outcomes were depressive and anxiety symptoms, quality of life, sleep quality, and positive and negative affect. We included 116 studies, of which 50 were conducted in hospital settings, 24 were delivered in community settings, seven were delivered in military clinics for veterans or active military personnel, five were conducted in refugee camps, four used remote delivery via web-based or telephone platforms, four were conducted in specialist trauma clinics, two were delivered in home settings, and two were delivered in primary care clinics; clinical setting was not reported in 17 studies. Ninety-four RCTs, for a total of 6,158 participants, were included in meta-analyses across the primary and secondary outcomes; 18 RCTs for a total of 933 participants were included in the component network meta-analysis. The mean age of participants in the included RCTs was 42.6 ± 9.3 years, and 42% were male. Nine non-RCTs were included. The mean age of participants in the non-RCTs was 40.6 ± 9.4 years, and 47% were male. The average length of follow-up across all included studies at posttreatment for the primary outcome was 11.5 weeks. The pairwise meta-analysis showed that psychological interventions reduce PTSD symptoms more than inactive control (k = 46; n = 3,389; standardised mean difference [SMD] = -0.82, 95% confidence interval [CI] -1.02 to -0.63) and active control (k-9; n = 662; SMD = -0.35, 95% CI -0.56 to -0.14) at posttreatment and also compared with inactive control at 6-month follow-up (k = 10; n = 738; SMD = -0.45, 95% CI -0.82 to -0.08). Psychological interventions reduced depressive symptoms (k = 31; n = 2,075; SMD = -0.87, 95% CI -1.11 to -0.63; I2 = 82.7%, p = 0.000) and anxiety (k = 15; n = 1,395; SMD = -1.03, 95% CI -1.44 to -0.61; p = 0.000) at posttreatment compared with inactive control. Sleep quality was significantly improved at posttreatment by psychological interventions compared with inactive control (k = 3; n = 111; SMD = -1.00, 95% CI -1.49 to -0.51; p = 0.245). There were no significant differences between psychological interventions and inactive control group at posttreatment for quality of life (k = 6; n = 401; SMD = 0.33, 95% CI -0.01 to 0.66; p = 0.021). Antipsychotic medicine (k = 5; n = 364; SMD = -0.45; -0.85 to -0.05; p = 0.085) and prazosin (k = 3; n = 110; SMD = -0.52; -1.03 to -0.02; p = 0.182) were effective in reducing PTSD symptoms. Phase-based psychological interventions that included skills-based strategies along with trauma-focused strategies were the most promising interventions for emotional dysregulation and interpersonal problems. Compared with pharmacological interventions, we observed that psychological interventions were associated with greater reductions in PTSD and depression symptoms and improved sleep quality. Sensitivity analysis showed that psychological interventions were acceptable with lower dropout, even in studies rated at low risk of attrition bias. Trauma-focused psychological interventions were superior to non-trauma-focused interventions across trauma subgroups for PTSD symptoms, but effects among veterans and war-affected populations were significantly reduced. The network meta-analysis showed that multicomponent interventions that included cognitive restructuring and imaginal exposure were the most effective for reducing PTSD symptoms (k = 17; n = 1,077; mean difference = -37.95, 95% CI -60.84 to -15.16). Our use of a non-diagnostic inclusion strategy may have overlooked certain complex-trauma populations with severe and enduring mental health comorbidities. Additionally, the relative contribution of skills-based intervention components was not feasibly evaluated in the network meta-analysis.

Conclusions: In this systematic review and meta-analysis, we observed that trauma-focused psychological interventions are effective for managing mental health problems and comorbidities in people exposed to complex trauma. Multicomponent interventions, which can include phase-based approaches, were the most effective treatment package for managing PTSD in complex trauma. Establishing optimal ways to deliver multicomponent psychological interventions for people exposed to complex traumatic events is a research and clinical priority.

Fig 1. PRISMA flow diagram.

Fig 2. Any psychological treatment for PTSD symptoms versus control at posttreatment across all populations.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box, the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the effect size in each study. CI, confidence interval; Cont_N, number in control group; ES, effect size; Int_N, number in intervention group; PTSD, posttraumatic stress disorder.

Fig 3. Psychological treatments for PTSD symptoms by intervention category versus control at posttreatment across all populations.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the ES in each study. CBT, cognitive behavioural therapy; CI, confidence interval; Cont_N, number in control group; EMDR, eye movement desensitisation and reprocessing therapy, ES, effect size; Int_N, number in intervention group; IPT, interpersonal therapy; PTSD, posttraumatic stress disorder.

Fig 4. Antipsychotics versus placebo for PTSD symptoms at posttreatment.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box, the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the effect size in each study. CI, confidence interval; PTSD, posttraumatic stress disorder; SMD, standardised mean difference.

Fig 5. Prazosin versus placebo for PTSD symptoms at posttreatment.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box, the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the effect size in each study. CI, confidence interval; PTSD, posttraumatic stress disorder; SMD, standardised mean difference.

Fig 6. Psychological treatments for PTSD symptoms by intervention category versus control at posttreatment in veterans.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box, the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the effect size in each study. CBT, cognitive behavioural therapy; CI, confidence interval; Cont_N, number in control group; EMDR, eye movement desensitisation and reprocessing therapy; Int_N, number in intervention group; PTSD, posttraumatic stress disorder; SMD, standardised mean difference.

Fig 7. Psychological treatments for PTSD symptoms by intervention category versus control at posttreatment in refugee populations.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box, the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the effect size in each study. CBT, cognitive behavioural therapy; CI, confidence interval; Cont_N, number in control group; EMDR, eye movement desensitisation and reprocessing therapy; Int_N, number in intervention group; PTSD, posttraumatic stress disorder; SMD, standardised mean difference.

Fig 8. Trauma-focused CBT for PTSD symptoms versus control at posttreatment in war-affected populations.

The size of the grey box reflects how much weight each study received in the meta-analysis (i.e., the larger the box, the more this study contributed to the pooled effect represented by the blue diamond). Black bars represent the 95% CI for the ES in each study. CBT, cognitive behavioural therapy; CI, confidence interval; Cont_N, number in control group; ES, effect size; Int_N, number in intervention group; PTSD, posttraumatic stress disorder.

Fig 9. Network diagram for all combinations of components extracted from included studies (edge thickness weighted by inverse variance).

AC, active control; C, cognitive restructuring; IE, imaginal exposure; IV, in vivo exposure; M, mindfulness; MU, multidimensional; PE, psychoeducation; R, relaxation; S, support; VR, virtual reality exposure; WL, waitlist.