Mind-body exercise interventions for prevention of post-traumatic stress disorder in trauma-exposed populations: a systematic review and meta-analysis

Abstract

Objective

Mind-body exercise (MBE) interventions, such as yoga, are increasingly recognised as an adjunct treatment for trauma-related mental disorders but less is known about their efficacy as a preventative intervention. We aimed to systematically review if, and what type of, MBE interventions are effective at preventing the development of post-traumatic stress disorder (PTSD) or acute stress disorder (ASD) in trauma-exposed populations.

Design

Systematic review and meta-analysis.

Methods

A systematic search of MEDLINE, PsycINFO, EMBASE and CENTRAL databases was conducted to identify controlled trials of MBE interventions aimed at preventing the development of PTSD or ASD in high-risk populations. Risk of bias was assessed using the revised Cochrane risk-of-bias and ROBINS-I tools. Pooled effect sizes using Hedges’ g and 95% CIs were calculated using random effects modelling for the main meta-analysis and planned subgroup and sensitivity analyses.

Results

Six studies (N analysed=399) were included in the final meta-analysis. Overall, there was a small effect for MBE interventions in preventing the development of PTSD (g=−0.25, 95% CI −0.56 to 0.06) among those with previous or ongoing exposure to trauma. Although a prespecified subgroup analyses comparing the different types of MBE intervention were conducted, meaningful conclusions could not be drawn due to the small number of studies. None of the included studies assessed ASD symptoms.

Conclusion

Limited evidence was found for MBE interventions in reducing PTSD symptomology in the short term. Findings must be interpreted with caution due to the small number of studies and possible publication bias.

PROSPERO registration number

CRD42020180375

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This is the first systematic review of mind-body exercise interventions as a possible means of prevention for post-traumatic stress disorder.

• We performed a detailed systematic strategy to search for the best-quality evidence of effectiveness in mind-body exercise interventions and accessed the methodological rigour of each included study.

• We completed a priori planned subanalyses to provide further insight into the effectiveness of different types of mind-body exercise interventions.

• There were a limited number of studies meeting the inclusion criteria and relatively small sample sizes across some of the included studies.

• Further high-quality and longitudinal research on mind-body exercise, particularly on its application to high-risk populations, should be conducted to establish the efficacy of these findings.

Introduction

It is estimated that up to 70% of individuals will encounter a traumatic experience during their lifetime.^{1 2} Although most people tend to recover without professional intervention, approximately 3%–7% develop post-traumatic stress disorder (PTSD).^{3 4} PTSD describes persistent stress reactions that occur beyond 1 month after a traumatic event, while severe acute stress reactions may also present in the initial month after trauma exposure, termed acute stress disorder (ASD). To date, both PTSD and ASD remain challenging to treat. Global guidelines recommend trauma-focused psychotherapies as first-line treatments for PTSD and ASD due to its largest and strongest evidence base.⁵ However, despite the efficacy of trauma-focused psychotherapy in treating trauma-related disorders, drop-out rates are considerably high and varying treatment responses, including non-response, partial response or relapse following initial response remain an issue.^6–8 Another significant barrier to care is low rates of help-seeking, with more than two-thirds of people suffering from PTSD symptoms not seeking care.^{9 10} Among the most common barriers for not seeking help include stigma, not recognising symptoms as problematic, not knowing where to seek care, and limited accessibility.^{11 12} To address these barriers, researchers and public health authorities have recommended the need for novel interventions aimed at pre-emptive or preventative strategies to mitigate the development of PTSD.^13–15

Investment in preventative interventions for PTSD has increased in recent years. Despite this, there is a substantial absence of high-quality evidence for preventative interventions. A recent systematic review of pharmacological and psychosocial interventions delivered pre and postincident found overall low-quality evidence for preventing the onset of PTSD.¹⁶ The results indicated that interventions with a trauma-focus appeared to have better outcomes and some support was found for an indicated prevention approach, which refers to interventions delivered to those displaying early symptoms of PTSD, but not yet meeting diagnostic criteria for PTSD.¹⁷ Little evidence was found for selective approaches that target populations at risk, or universal approaches, which target the entire population. While the review found some emerging evidence for team cohesion training, attention bias training and eye movement desensitisation and reprocessing, to date, there is still insufficient evidence for any of these interventions to be strongly recommended for prevention of PTSD.^18–20

Research suggests targeting key modifiable mechanisms known to increase the risk for PTSD as a promising approach for PTSD prevention.²¹ These include emotional regulation, extinction learning, contextual processing as well as autonomic nervous system functioning, all of which have been previously implicated in the onset and persistence of post-traumatic symptoms.^22–25 Evidence suggests that exposure to trauma may lead to dysfunctions in these mechanisms even among those not meeting criteria for PTSD.²⁶ Trauma-exposed individuals are also likely to suffer from impairments in physical health. PTSD and post-traumatic stress symptoms more generally have been associated with greater frequency and severity of pain as well as cardiorespiratory, gastrointestinal and generalised physical complaints.²⁷ A recent meta-analysis found that trauma-exposed individuals were 2.7 times more likely to experience functional somatic syndrome, which was also strongly associated with PTSD.²⁸ Theories suggest that this high comorbidity between PTSD symptoms and pain may be a result of mutually maintaining or shared vulnerability factors such as attentional bias towards threat, anxiety sensitivity, reminders of pain-related trauma and behavioural avoidance.^{29 30} While psychotherapies for PTSD have the ability to modify cognitive and emotional mechanisms,^{31 32} and appear to reduce the frequency of physical health symptoms, to date, there is no evidence for these therapies in reducing discomfort from somatic symptoms.³³ In addition to the further investigation needed to establish the precise underlying mechanisms of PTSD and physical health, there is also a need for better integration of both physical and mental health strategies to effectively address the burden of post-traumatic stress.

In the past two decades, there has been increasing interest in complementary and integrative health approaches for PTSD, such as mind-body exercise (MBE) interventions. MBE interventions, also known as active mind-body movement therapies, can be defined as interventions that integrate both psychological and physical health strategies, often through the combination of breathing techniques, controlled physical movement and mindfulness practices simultaneously.³⁴ Mindfulness is historically a Buddhist practice used to cultivate awareness on the present moment experience and to attend to that experience in a non-judgemental manner, while concurrently disengaging from arising beliefs, thoughts and emotions.³⁵ One popular type of MBE is yoga, a practice that includes mindful physical movement, sustained bodily postures, diaphragmatic breathing techniques and meditation. Another example of MBE is tai chi, a low-impact martial arts practice with a similar emphasis on slow and mindful movement as well as deep breathing.³⁶ In 2017, yoga was the most used integrative health approach among US adults and continues to gain global popularity and acceptability.^37–39 Reviews have shown preliminary effectiveness for MBE interventions in treating PTSD and has been reported as a safe and well-tolerated treatment.^40–43 Despite this, MBE is yet to be recommended as a stand-alone treatment due to the limited number of high-quality controlled trials.⁵ Nonetheless, MBE is recognised as a promising emerging intervention⁴² and is increasingly offered in clinical populations as an adjunct or supplementary treatment for mental disorders, including PTSD.⁴⁴

While the evidence base for MBE interventions as a treatment approach is yet to be established, research suggests that MBE may be an ideal candidate as a preventative intervention for PTSD because of its acceptability, ability to actively target cognitive and emotional processing as well as somatic and autonomic nervous system functioning.^{45 46} A recent brain imaging study on PTSD patients found that participants who were treated with mindfulness-based stress reduction (MBSR), a broader intervention programme that includes yoga and meditation, had increased activation in brain regions involved in both cognitive and emotional processing⁴⁷ and that have been previously implicated in the pathophysiology of PTSD.⁴⁸ Previous reviews have also demonstrated preliminary evidence for yoga in improving regulation in the sympathetic nervous system and hypothalamic–pituitary–adrenal axis,⁴⁹ reduced inflammatory biomarkers⁵⁰ as well as improved symptom severity for somatisation disorders.⁵¹ Researchers further suggest that the emphasis on mindful somatic processing in MBE enables individuals to identify and tolerate physical sensations,⁴⁶ which could be particularly empowering for traumatised individuals experiencing difficulty regulating their level of arousal or experiencing sensation, also referred to as ‘emotional numbness’.⁵²

The available evidence among those with established symptoms of PTSD, taken together with reports of high satisfaction rates,^{53 54} suggests that MBE may be a promising approach for selective or indicated prevention, which applies to those at risk or with some symptoms, but do not yet meet diagnostic criteria for PTSD or ASD.^{17 55} To date, there is no consensus regarding the efficacy of MBE interventions in preventing the development of trauma-related disorders. Thus, this review aims to systematically examine if, and what type of MBE interventions are effective in preventing the development of PTSD or ASD in trauma-exposed populations. To the best of our knowledge, this is the first systematic review and meta-analysis that examines the evidence for MBE interventions that prevent the development of PTSD or ASD in trauma-exposed populations.

Methods

Search strategy

We conducted a literature search of four relevant databases (MEDLINE, Embase, PsycINFO and CENTRAL) from inception up to 14 March 2023. Sets of search terms related to mind-body practices, PTSD or ASD and controlled trials were combined (see online supplemental table S1 for full search strategy). Reference lists of eligible studies were also manually screened to identify any additional relevant studies. We registered our protocol with PROSPERO prior to starting the systematic review and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Statement guidelines for reporting.

Eligibility criteria

The inclusion criteria were as follows: (a) participants were drawn from adult samples with previous or ongoing exposure to actual or threatened death, serious injury or sexual violence, as defined by the DSM-5 PTSD Criterion A⁵⁶ through interpersonal trauma (eg, interpersonal violence) or non-interpersonal trauma (eg, accidents, disasters, war); (b) study participants were not selected on the basis of already established PTSD or ASD symptom levels; (c) studies were controlled trials, either RCTs or non-RCTs; (d) studies investigated the effects of an MBE intervention on PTSD or ASD symptoms; (e) symptoms of PTSD or ASD were assessed using a validated measure; (f) studies were original research articles written in English language and published in peer-reviewed journals.

To be included, MBE interventions had to combine both active physical movement with mindfulness practices simultaneously, such as yoga or tai chi. Interventions based on massage, aromatherapy, acupuncture and progressive muscle relaxation were not included because these are passive therapies that do not include both attentional control or mindful components and physical movement. Similarly, interventions that involved only mindfulness were excluded. However, interventions did not have to be exclusively MBE. That is, if the intervention incorporated other elements in addition to MBE or were part of a broader intervention such as MBSR, then these were eligible for inclusion.

As the current review examined MBE interventions as a preventative intervention, studies were excluded if the sample of participants was diagnosed or met criteria for clinical level symptoms for PTSD or ASD or were in treatment for PTSD or ASD. Studies were further excluded if participants were below the age of 18 years old. Provided that all the other inclusion criteria were met, no restrictions were placed on any comparative or control interventions.

Eligibility of studies was determined by two independent researchers (LT and JS or RL), who screened each individual title and abstract from the database search and reviewed the full text of potentially eligible studies. Any disagreements were discussed and resolved through discussion and consultation with a third researcher (SBH). Referencing software (Endnote V.X9) was used to facilitate data management, for the removal of duplicates, and for finding the full texts of articles.

Data extraction

The first author (LT) designed a data extraction sheet, which included study authors, year of publication, population, sample size at baseline, study design, description of intervention (and control group), timing of exposure, duration and frequency of intervention, study assessment timepoints and outcome of interest (scale) and results. Missing or ambiguous data were clarified by contacting the relevant study authors. As studies assessed PTSD symptoms using various psychometric instruments, the effect size measure selected was the standardised mean difference (Hedges’ g).^57–59

Risk of bias assessment

The Revised Cochrane Risk-of-Bias tool⁶⁰ was used to evaluate the risk of bias in included randomised trials. Each of the five bias domains was scored as high, some concerns or low levels of bias risk. The ROBINS-I tool⁶¹ was used to assess non-RCTs and scored against seven bias domains. The overall risk of bias response options was low, moderate, serious, critical or no information to base judgement. Studies assessed as critical were recommended to be excluded from analysis. Risk of bias appraisal was done independently by two reviewers (LT and JS or RL) and consensus was resolved through discussion.

Data-analysis

Pre and postdata based on an analysis stated to be by intention-to-treat (ITT) were used for the primary meta-analysis. If data from the ITT analysis were not reported, this was obtained by contacting the study authors. The reported sample sizes of each group, pre and post means and SD from each study were extracted. If a study comprised multiple arms, where there was two intervention arms of the same type, we combined both intervention groups into a single group using the recommended formulae for combining summary statistics across two groups described in the Cochrane Handbook for Systematic Reviews of Interventions.⁵⁹ We then calculated a pooled standardised mean difference (Hedges’ g) and a CI for that difference. To determine the SD for the pre–post difference, we assumed a within-person correlation of 0.5. This was based on a previous study showing general consistency of correlations in trials with repeated assessment of patient-reported outcomes, which found a mean correlation of 0.5.⁶² Effect sizes that resulted in a negative direction indicated that the intervention group had superior effects compared with the control group. A prespecified subgroup analysis examining the different types of mind-body exercise was also conducted. As considerable diversity in methodology was expected across studies, we calculated pooled effect size estimates using a random-effects model with the DerSimonian‐Laird method of estimation. To assess heterogeneity, Cochran’s Q as well as the I² statistic and CI were calculated.⁶³ Publication bias was assessed through visual inspection of funnel plots for asymmetry and the Egger’s test. Data analyses were performed using Comprehensive Meta-Analysis software V.4.⁶⁴

Patient and public involvement

This research was conducted without patient or public involvement.

Results

The database search identified a total of 1321 titles following the removal of duplicates (figure 1). The title and abstract of each were examined independently by two researchers who identified 124 potentially relevant articles. High inter-rater reliability was achieved (Kappa=0.96). Following full-text independent appraisal, six articles—four RCTs^65–68 and two non-RCTs^{69 70}—met full criteria for inclusion. The primary causes for exclusion were that studies did not use an MBE intervention (n=51) or study participants were diagnosed or met criteria for clinical level symptoms of PTSD (n=25). The risk of bias assessment for all four randomised controlled trials was identified as having some concerns. None was found to be of high risk (online supplemental figure S1). The ROBINS-I assessment for the two non-randomised controlled trials identified one study with low risk of bias,⁷⁰ another with moderate risk.⁶⁹ None was found to be of critical risk of bias; hence, all six studies were included in the final meta-analyses.

Figure 1.

Flow diagram of screening and selection process. MBE, mind-body exercise; NRCT, non-randomised controlled trial; PTSD, post-traumatic stress disorder; RCT, randomised controlled trial.

The final six studies (N analysed=399) involved populations with previous or ongoing exposure to trauma (table 1). One study involved a sample of survivors with previous interpersonal violence.⁶⁵ Three studies involved populations with non-interpersonal trauma, which included male veterans with previous service in Afghanistan or Iraq,⁶⁶ women veterans with combat exposure,⁶⁸ earthquake survivors⁷⁰ and a mixed sample of active-duty military personnel with ongoing exposure and military veterans with previous exposure.⁶⁹ The final study comprised of junior medical doctors with ongoing exposure through their occupation, although more than half of the study participants also reported previous interpersonal trauma.⁶⁷ Apart from the study on earthquake survivors who had experienced the traumatic event 8 months prior to the intervention, the remaining five studies did not report the extent or timing of trauma exposure.

Table 1.

Characteristics of included studies

Study	Population	Sample size at baseline (N)	Study design	Intervention type; control group	Timing of exposure	Duration; frequency of intervention	Study assessment timepoints	Outcome of interest (scale); effectiveness
Kelly & Garland1	Female survivors of interpersonal violence (IPV)	45 (Number analysed=39)	RCT	Trauma informed Mindfulness Based Stress Reduction (TI-MBSR) (n=24) Wait-list control (n=21)	Unknown	8 weeks	T1: Pre T2: Post	PCL-C: Intervention group reported significantly greater reductions in post-traumatic stress than the waitlist control group.
Rice et al69	Mixed sample of active-duty military and veteran volunteers	124	NRCT	In person MBSR (n=49) Virtual MBSR (n=44) Wait-list control (n-41)	Unknown	8 weeks; 1×2 hours weekly	T1: Pre T2: Post	PCL-M: No significant difference between groups.
Saban et al68	Women military veterans with at least one risk factor for cardiovascular disease	136	RCT	MBSR (n=75) Group-based health promotion education programme (Active control condition) (n=61)	Unknown	8 weeks; 1×2.5 hours weekly for both groups	T1: Pre T2: Mid-way through intervention (4 weeks) T3: Post T4: 6 months post	PCL-C: Intervention group showed a significantly steeper decline in symptoms compared with the active control condition from T1 to T4.
Seppala et al66	Male U.S. military veterans with service in Afghanistan or Iraq	21	RCT	Sudarshan Kriya yoga (n=11) Waitlist-control (n=10)	Unknown	7 days; 3 hours daily	T1: Prior T2: 1-week post T3: 1-month post T4: 1 year	PCL-M: Intervention group showed significantly fewer symptoms over time.
Taylor et al67	Junior medical doctors	21	RCT	Personalised yoga (n=11) Group fitness+homework (control) (n=10)	Unknown	8 weeks; at least 1x per week	T1: Prior T2: Post	PCL-5 with LEC-5 and DSM-5 Criterion A: Pre-post completer analysis was reported, which suggested no change in scores between groups. ITT data were obtained from study authors
Thordardottir et al70	Earthquake survivors in southern Iceland	64 (Number analysed=58)	NRCT	Hatha yoga (n=31) Wait-list control (n=35)	8 months prior to intervention	6 weeks; 2×60 min weekly	T1: Pre T2: Post	PDS: No significant difference between groups.

LEC-5, Life Events Checklist for DSM-5; MBSR, mindfulness-based stress reduction; NRCT, non-randomised controlled trial; PCL-5, PTSD Checklist for DSM-5; PCL-C, PTSD Checklist Civilian Version; PCL-M, PTSD Checklist Military Version; PDS, Post-traumatic Diagnostic Scale; RCT, randomised controlled trial.

All six studies involved the use of yoga. Three studies involved yoga as a stand-alone intervention,^{66 67 70} each with varying styles of yoga, and three studies involved MBSR, a broader intervention that includes meditation, yoga and facilitated group discussions.^{65 68 69} One MBSR study included trauma-focused group discussions,⁶⁵ while the other two MBSR studies were not trauma-focused.^{68 69} The duration of intervention delivery mostly ranged from 6 to 8 weeks,57 65 67 70 with the exception of one study that was delivered over 7 consecutive days.⁶⁶ All interventions were delivered in a group format except for one, which was delivered one-on-one.⁶⁷ One study⁶⁹ comprised of two intervention arms, in-person MBSR and virtual MBSR. Although the virtual MBSR group used individual devices, they were still able to participate in real time with the other group members. All other interventions were delivered in-person and compared with a waitlist control group, apart from one study that used group fitness as a control group,⁶⁷ and another study that used an active control condition, which involved a group-based health promotion education programme.⁶⁸

Effects of MBE interventions on PTSD symptoms

All included studies assessed PTSD symptoms, and none assessed ASD symptoms. Three studies did not report data from the ITT analysis and study authors were contacted to obtain the relevant imputed data. However, we were only able to obtain data from one study,⁶⁷ resulting in the available case analysis data extracted for the other two studies.^{65 70} Overall, there was a small effect of MBE interventions on PTSD symptoms (g=−0.25; 95% CI: −0.56 to 0.06; figure 2). The 95% prediction interval ranged from −1.06 to 0.56. Cochran’s Q was 9.40 and the I² was 46.8% with the 95% CI ranging from 0% to 77%. However, as the analysis included less than 10 studies, these estimates of heterogeneity may not be accurate.^{59 71–73} Visual inspection of the funnel plot revealed some asymmetry, and results of the Egger’s test were statistically significant (p=0.044), which may indicate the possibility of publication bias or could also reflect the relationship between sample size and effect size (online supplemental figure S2). Given the small number of studies, the power of the test was low and must be interpreted with caution. The main analysis was rerun to exclude two studies where the available case analysis pre and postdata was reported instead of the ITT data, where similar results were obtained (g=−0.20, 95% CI −0.59 to 0.19). Although all six studies involved the use of yoga, we performed a prespecified subgroup analyses for type of MBE intervention. We compared yoga as a stand-alone intervention to the broader MBSR intervention that included yoga. The effect estimates were relatively larger for yoga as a stand-alone intervention (g=−0.34, CI −0.79 to 0.10) compared with MBSR (g=−0.17, CI −0.59 to 0.26) (figure 2).

Figure 2.

Forest plot of the effect of mind-body exercise interventions on symptoms of post-traumatic stress disorder.

Discussion

Mind-body exercise is widely practised and increasingly offered clinically as an adjunct or supplementary treatment for those with trauma-related disorders.⁴⁴ MBE may also be a suitable means to, at least partially, prevent trauma-related psychopathology in high-risk trauma-exposed populations. As far as we are aware, this is the first systematic review and meta-analysis that examines the evidence for MBE interventions to prevent the development of trauma-related disorders, including ASD and PTSD. Although we did not find any studies assessing ASD symptoms, our review found overall limited evidence for MBE interventions in reducing the likelihood of developing PTSD in the short-term among populations with previous or ongoing exposure to trauma. However, these results should be considered tentative as the limited studies identified meant that caution must be applied to the interpretation of these results.

Although the subgroup analyses on type of MBE interventions showed a relatively larger effect for yoga as a stand-alone intervention compared with yoga as part of the broader MBSR intervention, MBSR studies comprised larger sample sizes with lower variability and likely provided more reliable estimates than stand-alone yoga, which had smaller sample sizes and more variability. Furthermore, the small number of studies identified meant that meaningful comparisons could not be concluded. This points to an important need for future research to conduct dismantling studies to isolate the effects of various components within MBE interventions. Yet, it is worth noting that while both MBSR and stand-alone yoga interventions include both mindfulness meditation and yoga practices, the key distinguisher of MBSR is its standardised protocol implemented by trained and certified instructors.⁷⁴ Stand-alone yoga interventions, on the other hand, typically encompass a wide range of practices and styles, and instructor experience and training can vary widely.⁴⁵ Such variability can make standardisation difficult, which would subsequently make designing and conducting higher quality trials with large sample sizes more challenging. Future research on stand-alone yoga should consider the use of standardisation, such as guidelines for instructor training and utilising a curriculum or protocol to ensure consistency across different settings, while allowing some degree of flexibility to meet the needs of participants. Achieving this balance could help facilitate the design of more robust future trials.

Key strengths of this review include our detailed systematic search strategy, the independent risk of bias assessment, the inclusion of studies with controlled groups for comparison and our subgroup analyses by type of MBE intervention. Furthermore, while there have been previous reviews of MBE on trauma-exposed populations,40 41 43 75 76 these have included participants with diagnosed or clinical levels of PTSD, and none has examined the evidence for MBE interventions as a possible means of prevention. Despite these major strengths, our study has several important limitations. First, given the small number of studies identified, there is limited certainty on the precision of effect estimates for MBE interventions. There are also some concerns related to the risk of bias and possible publication bias. Second, the use of self-report measures for PTSD means that our findings are limited to individual perception of symptoms rather than clinician administration of diagnostic outcomes. Third, use of passive waitlist-control groups in most of the identified studies may have also introduced some level of expectancy bias, where participants may have had prior expectations of the study conditions, which may have subsequently influenced the results. Only two of the identified studies used an active-control comparison to limit the possible influence of this effect. While the use of an active comparison can be challenging to implement and dependent on resource constraints, future studies using active comparison control groups are needed to advance research into the efficacy of MBE interventions. Some commonly used comparisons include exercise, relaxation or meditation.⁷⁷ However, the decision on what active comparison to use should ultimately depend on the mechanism of effect being investigated. It is also not known whether the size of effect may have been moderated by type and frequency of trauma, duration of intervention as well as the different control conditions used. More rigorous and high-quality RCTs with long-term follow-up assessments and active control comparisons are required to establish MBE as an effective preventative intervention strategy. Finally, although the studies assessed mind-body strategies to prevent PTSD, it is important to note that the current review did not identify any studies of individuals prior to trauma exposure. Thus, interventions delivered following trauma may not necessarily be functioning at a truly preventative or pre-emptive stage as they may not impact mechanisms before post-traumatic psychopathological processes have commenced. Furthermore, the timing and level of exposure were not reported in the identified studies other than in one study of earthquake survivors. It is possible that some individuals may have had past PTSD, in which case, these interventions would not be considered as ‘true’ prevention. However, as the identified studies selected samples based on trauma exposure and not symptom level, it is plausible that MBE may be effective for trauma-exposed populations with unknown levels of PTSD symptoms. There is also emerging evidence to suggest that the trajectory of PTSD is not linear,⁷⁸ and these MBE interventions may have possibly served as selected or indicated prevention in those with delayed-onset PTSD, where occurrence develops more than 6 months after a traumatic event.⁵⁶ Indeed, a recent meta-analysis found that veterans and other professional groups were almost two times as likely to develop delayed-onset PTSD compared with non-professional survivor groups.⁷⁹ This finding highlights the importance of PTSD prevention research to include evaluations of interventions beyond the 3-month timepoint after a traumatic event, particularly among military cohorts and high-risk occupational groups.

Despite the small number of studies identified, it was timely and important to conduct a rigorous and up-to-date synthesis of the currently available evidence given the rising popularity of MBE interventions being offered in healthcare and community settings. Unsurprisingly, only six studies were identified that met the inclusion criteria, with many excluded studies involving participants diagnosed, or with clinical level symptoms of PTSD. Prevention of PTSD is a challenging area of research to conduct compared with treatment interventions and remains under-researched despite its importance.^{15 16} Nevertheless, current guidelines on the prevention and treatment of PTSD consider MBE interventions a potential priority candidate for further research due to its emerging evidence.⁸⁰ Therefore, while our findings were limited to a small number of studies, we believe that the meta-analysis and subsequent sensitivity analyses were warranted. Our study also reveals several benefits that are relevant to both trauma-exposed populations and targeted public health prevention strategies. First, our review did not find any adverse events for any of the identified interventions, which is in line with previous reviews.40–43 76 81 82 This lends support that MBE could be safely offered to supplement existing evidence-based strategies. Second, our findings add the very limited pool of studies on preventative interventions for PTSD and show for the potential for MBE in reducing the risk of PTSD psychopathology, at least in the short-term.

Overall, while there may be potential for MBE interventions in preventing the development of PTSD, caution should be applied in the interpretation of results. Further research is required to disentangle how each stand-alone component of MBE contributes to PTSD symptomatology. More high-quality and longitudinal research on MBE is also needed, particularly on its application to high-risk populations to establish the feasibility and efficacy of these findings.

Data availability statement

No data are available.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.