The worldwide lifetime prevalence of post-traumatic stress disorder (PTSD) ranges between 3.9% in the general population and 5.6% in those exposed to trauma (Koenen et al., 2017). Not only is PTSD associated with poor quality of life and functional impairment (Alonso et al., 2004), but it is also associated with several mental and physical health conditions, such as depression, anxiety, substance abuse and cardiovascular and metabolic disease (Kessler et al., 1995; Ahmadi et al., 2011). A previous Mental Elf blog demonstrated that PTSD is a significant and potentially modifiable risk factor for dementia (Kaul, 2021).
Currently, the most used treatments for PTSD in adults are Trauma-Focused Cognitive Behavioural Therapy (TF-CBT) and Eye Movement Desensitisation and Reprocessing (EMDR). Previous meta-analyses have provided the most evidence for TF-CBT and EMDR. However, the lack of high-quality research examining other psychological treatments, such as combined somatic/cognitive therapies, has often led to their exclusion from meta-analyses, therefore making it difficult to draw concrete conclusions. Furthermore, the literature examining the relative effectiveness of PTSD treatments is limited, as traditional pairwise meta-analysis does not allow for the relative effectiveness across all treatments to be assessed.
In turn, Mavranezouli and colleagues (2020a) set out to fill this gap by examining the relative effectiveness of psychological treatments for adult PTSD using network meta-analysis (NMA). They aimed to utilise their findings to support the updating of NICE guidelines for PTSD.
The authors searched five electronic databases (MEDLINE, Embase, PsycINFO, CINAHL and the Cochrane Library) using medical subject headings (MeSH), free-text terms and a study design filter.
For studies to be included in the systematic review, they had to meet the following criteria:
- RCTs examining psychological, psychosocial, and other non-pharmacological interventions targeting clinically significant post-traumatic stress symptoms
- Adults with either a PTSD diagnosis (as defined by the DSM, ICD or similar) or clinically significant-PTSD symptoms (as indicated by baseline scores above cut-off on a validated scale more than one month after the traumatic event).
Studies were excluded from the systematic review if they:
- Looked at participants with adjustment disorders, traumatic grief, psychosis as a coexisting condition and learning disabilities
- Looked at women with PTSD during pregnancy or in the first year following childbirth
- Looked at adults in contact with the criminal justice system.
The network meta-analysis (NMA) considered two main outcomes: PTSD change scores and remission. Based on the availability of data for the two outcomes of interest, the authors conducted three separate NMAs looking at:
- PTSD symptom change scores between baseline and treatment endpoint
- PTSD symptom change scores between baseline and 1 to 4 month follow-up
- Remission at treatment endpoint.
The authors conducted Bayesian network meta-analyses, wherein the different interventions included are compared using both direct comparisons of interventions within RCTs and indirect comparisons across trials based on a common comparator, such as a waitlist.
The network meta-analysis included 90 trials, 6,560 participants and 20 interventions (including two inactive controls).
Risk of bias assessment
- The Cochrane risk of bias tool was used to assess risk in trials included in the NMA. All trials were found to be at a high risk of bias due to lack of participant and provider masking. Most studies were also found to be at risk of selection and reporting bias.
PTSD symptom change scores between baseline and treatment endpoint
- This NMA was formed by 71 RCTs, 4,700 participants and 19 interventions
- When comparing to waitlist, Eye Movement Desensitisation and Reprocessing (EMDR) was found to be the most effective across interventions with a large evidence base (N>100)
- Only psychoeducation had an inconclusive effect
- However, there was no evidence of differential effects between most treatments except:
- EDMR and counselling (mean SMD −1.34, 95% CrI −2.19 to −0.49) and
- Trauma-focused cognitive behavioural therapy (TF-CBT) and counselling (mean SMD −0.73, 95% CrI −1.37 to −0.09).
PTSD symptom change scores between baseline and 1 to 4 month follow-up
- This NMA was formed by 28 RCTs, 2315 participants and 15 interventions
- Across interventions with a large evidence base (N>100), only EMDR (mean SMD −1.12, 95% CrI −1.94 to −0.27) and TF-CBT (mean SMD −0.73, 95% CrI −1.23 to −0.25) were found to be significantly more effective than a waitlist control at 1-4 month follow up. There was no evidence of differential effects between the two.
Remission at treatment endpoint
- This NMA was formed by 34 studies, 2,249 participants and 16 interventions
- There were 151 arms assessing 19 interventions on a total of 4,700 participants
- Similar to the post-treatment efficacy findings, EMDR was found to be the most effective at improving remission rates across interventions with a large evidence base (N>100)
- Once again, comparisons between active treatments only found differences between:
- EDMR and counselling (mean LOR 2.04, 95% CrI 0.37 to 3.79) and
- TF-CBT and counselling (mean LOR 1.12, 95% CrI 0.12 to 2.15).
- This review found that EDMR (Eye Movement Desensitisation and Reprocessing) and TF-CBT (Trauma-Focused Cognitive Behavioural Therapy) were the most effective in reducing symptoms and improving remission rates for PTSD in adults
- EDMR and TF-CBT were the only interventions found to be significantly more effective than a waitlist control at 1 to 4 month follow up
- Combined somatic/cognitive therapies, self-help, non-TF-CBT, SSRIs and counselling were also effective at reducing PTSD symptoms at post-treatment
- Self-help with support and counselling were effective in improving rates of post-treatment remission
- Comparisons between active treatments only found counselling to be less effective than EMDR and TF-CBT.
Strengths and limitations
This study had several clear strengths, including:
- This is the first network meta-analysis (NMA) of psychological treatments for adult PTSD conducted with the intention of informing clinical guidance
- The use of NMA techniques enables both direct and indirect comparisons between interventions, which allows us to investigate relative effectiveness
- Included studies examining changes in post-treatment PTSD symptoms were comparable across interventions, as evidenced by inconsistency checks
- The findings of this study are consistent with other published reviews endorsing TF-CBT and EMDR as the most effective treatments for adult PTSD
- Transparency and adherence to open science. For example, the authors made large amounts of information (e.g. search strategy and details of statistical analysis) available within their supplemental materials.
However, it was also limited in many ways:
- Between-trial heterogeneity was high across all NMAs. The authors state that this is likely due to heterogeneity across the populations assessed within the trials, variability of interventions within each treatment node of analysis and difference across study settings. In turn, it may be inaccurate to draw conclusions based on different types of trials
- There was evidence of inconsistency in studies included in the NMA investigating post-treatment remission, and therefore conclusions made based on this NMA should be drawn with caution
- All TF-CBT interventions were considered as a single node, and therefore some nuance may be lost as to how effective different types of TF-CBT may be
- It is important to interpret these results with caution, as included trials were at risk of bias (particularly selection and reporting bias) and NMAs are limited by the quality of the trials included within them
- The study included limited evidence on the longer-term effectiveness of psychological interventions in increasing remission rates.
It is also important to note that this study set out to support the updating of national guidance for PTSD (NICE, 2018), which may have served as a conflict of interest as the authors may have felt a need to provide further evidence supporting the original NICE guidelines for PTSD. Several authors also disclosed conflicts of interest, such as links to Cognitive Therapy for PTSD (a form of TF-CBT), NICE and EMDR trials. These were stated clearly in the paper which therefore allows the reader to accurately assess the validity of the results.
Implications for practice
Where can research go from here?
Future research should attempt to:
- Identify which interventions are effective for complex PTSD in particular, as this study did not make the distinction between PTSD and complex PTSD
- Conduct further RCTs investigating the effectiveness of EMDR, as conclusions drawn within this study are based on a limited evidence base (11 RCTs)
- Further investigate the effectiveness of other interventions, such as non-TF-CBT and supported self-help
- Distinguish between the different types of TF-CBT, as these were combined into one broad category within this study
- Include longer follow-up time points to assess the long-term effectiveness of psychological interventions for PTSD.
What are the clinical implications of these findings?
- These findings support the use of EMDR and TF-CBT, which are the most commonly used interventions for PTSD in adults
- Supported self-help, combined somatic/cognitive therapies and non-TF-CBT should be considered as alternative treatment options
- Counselling appears to be less effective than other treatments, and therefore should only be offered if other more effective treatments are unavailable
- These findings supported the updating of national guidance for PTSD (NICE, 2018)
- The results of this study also furthered an economic analysis which assessed the cost-effectiveness of psychological interventions for adults with PTSD, where EMDR appeared to be the most cost-effective (Mavranezouli et al., 2020b).
Statement of interests
Mavranezouli I, Megnin-Viggars O, Daly C. et al (2020a). Psychological treatments for post-truamatic stress disorder in adults: a network meta-analysis. Psychological Medicine 1-14. https://doi.org/10.1017/S0033291720000070
Ahmadi N, Hajsadeghi F, Mirshkarlo HB et al. (2011). Post-traumatic stress disorder, coronary atherosclerosis, and mortality. American Journal of Cardiology, 108, 29–33.
Alonso J, Angermeyer MC, Bernert S. et al (2004). Disability and quality of life impact of mental disorders in Europe: Results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatrica Scandinavica 109, Supplementum 420, 38–46.
Kaul A. (2021). New review suggests that PTSD may be a modifiable risk factor for dementia. The Mental Elf, August 2021
Kessler RC, Sonnega A, Bromet E. et al. (1994). Posttraumatic stress disorder in the National Comorbidity Survey. Archives of General Psychiatry, 52, 1048–1060.
Koenen KC, Ratanatharathorn A, Ng L. et al. (2017). Posttraumatic stress disorder in the World Mental Health Surveys. Psychological Medicine, 47, 2260–2274.
National Institute for Health and Care Excellence [NICE](2018). Post-traumatic Stress Disorder (NICE Guideline 116).
Mavranezouli I, Megnin-Viggars O, Grey N. et al (2020b). Cost-effectiveness of psychological treatments for post-traumatic stress disorder in adults. PLoS ONE 15 (4): e0232245. https://doi.org/10.1371/journal. pone.0232245