https://orcid.org/0000-0003-0017-8151
Figures
Abstract
Background
First responders are faced with stressful and traumatic events in their work that may affect their psychological health. The current review examined the effectiveness of psychological interventions to treat posttraumatic stress disorder (PTSD), anxiety, depression, stress and burnout in first responders.
Methods
Four databases were searched to identify controlled studies that examined the efficacy of psychological interventions to reduce PTSD symptoms (primary outcome) in first responders (including firefighters, police/law enforcement officers, search and rescue personnel, emergency and paramedics teams). Secondary outcomes were anxiety, depression, burnout, and stress.
Results
15 studies were identified, including 10 studies that measured PTSD, 7 studies for anxiety, 10 studies for depression, 7 studies for stress and 1 for burnout. Interventions were associated with a significant reduction in PTSD (SDM = -0.86; 95% CI = -1.34 –- 0.39), depression (SDM = -0.63; 95% CI = -0.94 –-0.32), and anxiety (SDM = -0.38; 95% CI = -0.71 –-0.05) but not stress (SDM = -0.13; 95% CI = -0.51–0.25). CBT-based and clinician-delivered interventions were associated with significantly greater reductions in PTSD than other types of interventions and non-clinician interventions, but no differences were found for depression. There was evidence of moderate to high risk of bias across all studies.
Citation: Alshahrani KM, Johnson J, Prudenzi A, O’Connor DB (2022) The effectiveness of psychological interventions for reducing PTSD and psychological distress in first responders: A systematic review and meta-analysis. PLoS ONE 17(8): e0272732. https://doi.org/10.1371/journal.pone.0272732
Editor: Saeed Ahmed, Rutland Regional Medical Center, UNITED STATES
Received: October 22, 2021; Accepted: July 25, 2022; Published: August 24, 2022
Copyright: © 2022 Alshahrani et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The term first responders has been used extensively in the literature to refer to a range of professions and occupations such as firefighters, police officers and paramedics [1–3]. These professions are considered by many as the ‘traditional’ first responders [4, 5]. However, there has been debate in the scientific and medical literature about which groups are classified as first responders [4]. For example, a number of studies have included 911 operators or dispatchers, correctional workers, nurses, physicians, public safety personnel and have drawn a distinction between ‘traditional’ and ‘untraditional’ first responders [4, 6]. Other studies have investigated risk workers which contain traditional and untraditional first responders, including public safety personnel and frontline healthcare professionals [7]. Nevertheless, recognising these definitional differences, the current research was interested in focussing on traditional first responders who respond to accidents or disasters in the early phases to protect and preserve life, property, and the environment; they include police officers, firefighters, search and rescue personnel, and emergency and paramedic teams [3].
Although the specific roles of different groups of first responders varies, they all face multiple potentially psychologically traumatic incidents in their daily work which puts them at heightened risk of experiencing mental health difficulties and disorders including posttraumatic stress disorder (PTSD), depression, anxiety and burnout [8–10]. In particular, first responders are more frequently exposed to potentially psychologically traumatic events than the general population, sometimes causing them to become ‘secondary traumatic victims’ [11–13]. Recent reviews have estimated the prevalence of PTSD and other psychological disorders among these populations. Petrie et al. [14] found prevalence rates in paramedics were 11% for PTSD, 15% for depression, 15% for anxiety, and 27% for general psychological distress. Other reviews have reported prevalence rates ranging between 6.4% to 57% for firefighters, and 5.8% to 19.6% for police officers [15–17]. As such, there is a pressing need to better understand the mental health of first responders and to identify efficient and appropriate interventions that are suitable for delivery to these groups. Due to the nature of their work, there is also a particular need to understand the effectiveness of interventions for helping to reduce PTSD symptoms in first responders.
In the last two decades, five systematic reviews have attempted to synthesize the effectiveness of interventions for reducing psychological distress among first responders, especially PTSD [2, 18–21]. Smith and Robert [18] identified a range of studies that tested interventions to reduce stress and PTSD among emergency ambulance personnel, and they found that all included articles (10 studies) had a lack of quality due to their methodological flaws such as, self-selection of groups and inadequate timing of the interventions. Haugen and his colleagues [2] reviewed papers that treated PTSD in first responders, and they found 17 articles but only two randomized controlled trials (RCTs). Alden et al. [19] evaluated the effectiveness of interventions targeting posttraumatic symptoms in first responders. They identified 21 studies and 8 of them were RCTs. They found that trauma focused therapies can be effective for first responders, but faced some limitations such as, the small sample sizes of most included studies, and few investigations with any control condition. Winders et al. [20] examined the prevention and treatment of psychiatric symptoms in first responders and how effectiveness was related to the income level of the country. They found, in 25 eligible studies, that most interventions were effective for preventing and treating psychological illness in first responders in all three levels of income countries. Another relevant review was conducted by Anderson et al. [2020] in frontline healthcare and public safety personnel. They reported inconsistent evidence regarding the effectiveness of organizational peer-support and crisis-focused psychological interventions for reducing PTSD symptoms but highlighted the need for further studies. However, a recent meta-analysis was published in 2021 [22] that examined the effectiveness of psychotherapy for posttraumatic stress injuries among public safety personnel. They found, by identifying 8 included studies, these interventions were associated with significant reductions in PTSD, anxiety, and depression symptoms.
Overall, three limitations are evident in the existing literature into interventions for psychological distress among first responders. First, controversy about the importance and effectiveness of interventions remains. While some studies and reviews have suggested that interventions are highly effective, others have reported no significant benefit of interventions [21, 23, 24]. Second, there appears to be a lack of high-quality intervention studies, and as such it is hard to draw firm conclusions based on the results of any study in isolation. Three, it is possible that some types of interventions are more effective than others, however, no studies or reviews have carefully investigated these potential differential effects. Cognitive Behaviour Therapy (CBT) and Eye Movement Desensitization and Reprocessing (EMDR) have been found to be effective in treating mental health disorders in both general clinical populations [11] and first responders [25], but there is a great deal of debate about the possible benefits of other interventions such as critical incident stress debriefing (CISD) [12]. In addition, previous meta-analyses have not compared the effectiveness of clinician delivered versus non-clinician delivered interventions or individual-based versus group-based interventions. In order to address these limitations, it was decided that a systematic review and meta-analysis were needed to assess the effectiveness of interventions used to treat PTSD and other aspects of mental health in first responders. The moderating effects of gender and age were also examined given that previous research has found that females and older workers have reported higher rates of PTSD compared to males and younger workers, respectively [26]. In addition, a number of important other methodological moderating factors were also explored (i.e., risk of bias, sample size, number of sessions and length of interventions).
Aims and objectives
The first aim of the current review was to synthesise the evidence for the effectiveness of psychological interventions for improving mental health symptoms in first responders. Our primary outcome of interest was PTSD symptoms; secondary outcomes were depression, anxiety, stress, and burnout. Our second aim was to compare the effectiveness of different types and formats of interventions (CBT versus other interventions, clinician delivered versus non- clinician delivered interventions & individual-based versus group-based interventions) for reducing symptoms of PTSD, depression, anxiety, stress, and burnout.
Method
This systematic review was registered in advance on PROSPERO, after specifying the inclusion criteria and analytical methods: CRD4MASKED. https://www.crd.york.ac.uk/prospero/display_record.php?ID=MASKED. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines that can be found in S1 Appendix.
Search strategy
Four electronic databases were searched: EMBASE, PsycInfo, CINAHL, and Cochrane Register of Controlled Trials- from inception to June 2nd 2019, updated to September 8th 2021. The search strategy was performed by using three key blocks of terms: “interventions”, “first responders” and “posttraumatic stress disorder” or “depression” or “anxiety” or “burnout” or “stress”. Also, MeSH and keyword terms were examined based on terms usually used within previous reviews in the fields of first responders and PTSD (see eMethods.1 in S1 Appendix) [26–30].
Eligibility criteria
Eligibility criteria were:
- Population: Studies that looked at first responders who work at the sites of critical incidents including police officers, firefighters, search and rescue personnel, and emergency and paramedic teams. We included any studies focused on these groups, including those with and without any pre-existing mental health diagnoses.
- Intervention: Studies with an individual or group psychological intervention designed to reduce PTSD, anxiety, depression, burnout, and/or stress symptoms that were delivered by registered clinicians (e.g., psychologists or psychiatrists) and non-clinicians (e.g., experiences police officers or supervisors).
- Comparison: Any type of control group (e.g., no intervention, alternative intervention, or wait list) was included.
- Outcomes: The primary outcome was PTSD symptoms as rated by an observer (e.g., doctor or researcher) using validated scales such as PTSD checklist [31] or self-reported PTSD symptoms measured using validated scales such as, PTSD symptoms scale self-report PSS-SR [32]. Secondary outcomes were stress, anxiety, depression, and burnout also measured via observer ratings or self-report, using validated scales (e.g., the Hospital Anxiety and Depression scales HADS [33], for anxiety and depression and the Maslach Burnout Inventory MBI, [34] for burnout).
- Design: Studies that used randomized controlled trial (RCTs) designs or controlled before-after designs (CBAs), as defined in the Cochrane handbook, were included [35].
- Context: Studies that investigated any treatment effects of psychological interventions in the first responders that were published in the English language were included.
Exclusion criteria
We excluded studies conducted on other types of first responders, such as in-hospital health care workers (e.g., emergency room doctors and nurses) or military first responders (e.g., peacekeepers and soldiers). Also, all studies that used different types of interventions, such as physical or pharmacological interventions rather than psychological interventions were excluded, as were non-English language studies and gray literature.
Study selection
All search results from each database were exported to Endnote version X8.2 (Clarivate Analytics, Philadelphia, United State) and all duplicates were removed. The study selection consisted of two stages. In the first stage, titles and abstracts were screened based on the above criteria by one author (KA). A second author (JJ) checked and reviewed 10% of included studies. To estimate the level of agreement we calculated the Kappa score which indicated good agreement (k = 0.845). In the second stage all full texts were screened by one reviewer (KA) with a second independent reviewer (AP) screening 20% of full texts, with 100% agreement.
Data extraction
We used Excel 2016 [Microsoft Inc, Washington] to organize a data extraction form. We extracted quantitative data for the meta-analysis on a separate Excel file. All data extraction was undertaken by KA. The first 10% of eligible articles were independently extracted by AP to check for agreement. Any discrepancies were resolved by discussion. The following descriptive information was extracted from eligible studies:
- Study: country, research design, recruitment method and content of the control condition.
- Participants: sample size, age, gender, discipline, setting.
- Intervention: content of the intervention, measurement time points.
- Outcomes: PTSD, stress, anxiety, depression and burnout.
Risk of bias (quality) assessment
We used the Effective Practice and Organisation of Care (EPOC) risk of bias tool [36] to carry out a critical assessment, as it is suitable for use across all different types of intervention designs, as reported in Cochrane handbook [35]. The tool comprises nine standardized criteria, each one rated on a three-point scale (0 = low risk, 1 = unclear risk, and 2 = high risk). Studies that obtained low-risk score have been considered across at least six of nine criteria to be less susceptible to risk of bias.
Data analysis
Comprehensive Meta-Analysis software version 3 (CMA) [37] was used to calculate random effect sizes and other meta-analyses processes. The main meta-analysis examined the effectiveness of psychological interventions for improving PTSD in first responders. Secondary meta-analyses examined anxiety, depression, and stress interventions in same population.
We calculated the effect sizes using standardised difference of means (SDM) and associated 95% confidence intervals for PTSD, anxiety, depression, and stress outcomes in all included studies. We used post-treatment means, standard deviations (SD), and sample sizes to calculate the effect sizes between treatments and control groups. When means and SD values were not provided in the original studies, we used the available sample sizes and p-values to impute missing SDM (following the procedures outlined by Borenstein and colleagues (see Borenstein & Rothstein, 2013). For studies that had two types of interventions compared to a waiting list control group, we divided the sample size of the control group in half in each type to reduce the variance related with each effect size and to avoid double counting of participants [38]. If studies collected more than one follow-up assessment point, we used the first assessment point following the intervention.
For comparison groups, we identified the most common type of intervention used for each outcome and compared with all other interventions. If there was no common intervention for an outcome, then the outcome was not included in subsequent analyses. In addition, we also compared between interventions according to whether they were provided by clinicians versus non-clinicians (providers) and whether they were group-based versus individual-based interventions (format).
To account for heterogeneity, a random effects model was used in all analyses and assessed with I2 statistic [35]. I2 values indicate 25% as low, 50% as moderate, and 75% as high heterogeneity. In addition, between-study heterogeneity was assessed by Cochran’s Q-statistic (measure of weighted squared deviations) [39]. The Cochran Q test (significance level, p < 0.05) was used to compare between types of interventions. Specifically, the Qbetween statistic was used to examine if the SDM was significantly different between groups of interventions. Moreover, meta-regressions were used to examine the impact of any potential moderating variables on the overall effect size for different outcomes. Meta-regression follows similar principles to regression or multiple regression in primary studies, except that in meta-analyses the variables are at the level of the study and not the participant [38]. Six moderating variables were investigated (risk of bias, gender, mean age, sample size, number of sessions, and total length of interventions by minutes). Sensitivity analyses were performed to: 1) examine the balance of the results among studies with a lower risk of bias rating and 2) to investigate the potential impact of outliers. Publication bias was calculated with Egger’s regression test [39], and further explored using funnel plots and Duval and Tweedie’s trim and fill analysis [40].
Results
A total of 4572 papers were identified by our searches. In addition, 13 papers were discovered by manual reference list scanning (see Fig 1). We removed duplicated articles, and then screened the title and abstracts of 3344 studies. 80 out of these were screened as full text. 20 papers were eligible to include in the systematic review and meta-analysis [25, 41–58]. For articles which were eligible for the systematic review but which did not report the necessary outcome data for meta-analysis, we contacted the authors and co-authors, to request the relevant data. Five papers [25, 42, 46, 47, 59] were not included in the meta-analysis because either the authors no longer had access to the data or the authors did not respond. The total number of studies included was 15 [41, 44, 45, 50–56, 58, 59].
aSome studies were excluded for more than one reason so the total studies do not equal 66.
Characteristics of the studies and participants
Across the 15 papers, the total sample size was 928 (mean age = 39.5 years, 64.8% males). 10 studies measured PTSD, 7 studies measured anxiety, 10 studies measured depression, and 7 studies assessed stress. One study evaluated burnout (this study was not meta-analysed due to there being an insufficient number of studies for meta-analysis). In terms of participants, 6 studies were conducted with police officers, 3 studies were conducted with firefighters, 2 studies were conducted with ambulance personnel, and 4 studies were conducted in a mix of disaster workers that included police officers, firefighters, and paramedics. 4 out of 15 studies were conducted in the USA, 3 in Australia, 3 in the Netherlands and one study each in England, Saudi Arabia, Iran, Thailand and Mexico.
Characteristics of the interventions
A range of different types of interventions was used. Cognitive behaviour therapy (CBT) was the most frequent with 4 different types of CBT used in 3 studies, then critical incident stress debriefing (CISD) and resilience training were used in 3 studies each, while eye movement desensitization and reprocessing (EMDR) was used in 2 studies, and 1 study each used brief eclectic psychotherapy, eclectic group counselling, written emotional expression, and mental agility and psychological strength. One study had more than one intervention that used 2 types of cognitive behaviour therapy (the long type (CBT-L) and brief type (CBT-B)) compared to a waiting list control group [52]. The mean number of treatment sessions was 7.53, the longest individual session lasted 2.5 hours and the shortest one lasted 15 minutes (see Table 1). In terms of providers, two types of intervention deliverers were identified: clinician (8 studies) versus non-clinicians (6 studies), and one study [50] was not clear because there was no more details about providers or delivered type of interventions. In addition, interventions were delivered in an individual (8 studies) or group format (6 studies), and one study was not clear.
Risk of bias characteristics
The risk of bias was calculated using the EPOC 9 criteria. 8 studies were considered low risk, 6 studies were scored as moderate risk, and 1 study was high risk (see eFigure 1 in the S1 Appendix). In addition, in 6 of the studies the comparison or control condition did not include any intervention component. Future research should endeavour to include an active control condition. Lack of blinding across studies was also a potential source of high risk of bias given that many studies did not use blinding or it was not reported clearly in the papers. Similarly, there were some concerns related to lack of allocation concealment and missing data.
Main and secondary meta-analysis
The main meta-analysis results found psychological interventions led to a significant reduction in PTSD and depression at p<0.01 level (10 comparisons: SDM = -0.86; 95% CI = -1.34 –-0.39; I2 = 88.97%; Fig 2), (10 comparisons: SDM = -0.63; 95% CI = -0.94 –-0.32; I2 = 72.48%; Fig 3) respectively, and a significant reduction in anxiety symptoms at the p<0.05 level (7 comparisons: SDM = -0.38; 95% CI = -0.71–0.05; I2 = 67.12%; Fig 4) but no significant reduction for stress (7 comparisons: SDM = -0.13; 95% CI = -0.51–0.25; I2 = 71.67%; Fig 5). As I2 values shown above, the heterogeneity between the studies in the PTSD analysis was high indicating the need to consider the role of potential moderators. Therefore, the following 6 moderators were considered: risk of bias, gender, mean age, sample size, session number and the length of treatments (See Table 3).
Comparing between interventions
Cognitive behaviour therapy was the most common intervention (4 comparisons from 3 studies) used to treat PTSD, anxiety and depression. However, across the 7 anxiety studies, CBT was used only once, therefore, we were only able to compare CBT with all other interventions for the outcomes of PTSD and depression. We also compared interventions delivered by clinicians (8 studies) against those delivered by non-clinicians (6 studies) and the interventions that were delivered individually (8 studies) versus those provided in a group format (6 studies) for PTSD and depression outcomes. In terms of comparing between CBT and other interventions, the results showed that there were statistically significant differences between the standardised difference in means between the CBT and all other interventions for PTSD symptoms, such that CBT was more effective (Q = 5.74; p = 0.01), but this was not found for depression, where the difference was non-significant (Q = 3.44; p = 0.06). Interventions delivered by clinicians were associated with greater reductions in PTSD compared with those delivered by non-clinicians (Q = 7.59; p< 0.001), but interventions delivered individually were not associated with greater reductions in PTSD than group interventions (Q = 3.47; p = 0.06). For depression, however, there were no differences between clinician compared with non-clinician delivery (Q = 1.78; p = 0.18), or individual compared with group format (Q = 2.46; p = 0.11). See Table 2.
Tests of moderation
The moderate risk (n = 6) and high risk (n = 1) studies were combined and compared with the low risk (n = 8) studies using meta-regression. The results showed that there was moderating effect of risk of bias for only PTSD (Q = 5.05; p = 0.01), such that studies with low risk was reported to have higher SDMs. However, there were no moderating effect of risk of bias for anxiety (Q = 0.59; p = 0.44), depression (Q = 1.13; p = 0.28) and stress outcomes (Q = 0.41; p = 0.81) (see Table 3). We also examined the moderating effects of gender, mean age, sample size, number of intervention sessions and the total number of minutes of interventions (Table 3). For PTSD, only the number of sessions was found to be significant moderators (Q = 14.6; p = 0.001), such that studies with more intervention sessions was found to have larger SDMs. For anxiety, only mean age was found to be a significant moderator on the outcome, such that studies with a higher mean age had a larger SDM (Q = 10.4; p = 0.01). For depression and stress there were no significant moderators. Moreover, gender was found to have no moderating effects on any of the outcomes.
Publication bias
Egger’s regression coefficients found evidence of publication bias in the depression studies (intercept = -3.97; 95% CI, -6.07 –-1.34; p = 0.003), but not for PTSD (intercept = -4.57; 95% CI, -8.64 –-0.51; p = -0.031), anxiety (intercept = -2.29; 95% CI, -6.14–1.55; p = 0.186) and stress (intercept = 7.36; 95% CI, -6.58–21.3; p = 0.230) (See e-figures 2–5 in S1 Appendix). However, a funnel plot suggested that missing studies were reported in only the right side of the mean for PTSD and depression (4 studies in each), and 3 studies in anxiety. Duval and Tweedie’s trim and fill analysis [40] found a small reduction in the SMD after missing studies for PTSD, depression, and anxiety were imputed (SDM = -0.21; CI, -0.34 –-0,08), (SDM = -0.26; 95% CI, -0.39 –-0.12), and (SDM = -0.07; CI, -0.22–0.07) respectively.
Sensitivity analyses
When we used a leave-one-out method [35], the finding for PTSD and depression varied (between -0.73 and -0.98 for PTSD and between -0.53 and -0.71 for depression), but both outcomes remained significant, suggesting the findings were not driven by any single study. In contrast, there were two studies in the anxiety analysis [48, 60] which when removed in isolation led the analyses to become statistically non-significant (SDM = -0.304; p = 0.07) and (SDM = -0.214; p = 0.08) respectively. Similarly, there was one study in the stress analysis [45] which when removed led to a statistically significant finding (SDM = -0.133; p = 0.50).
Discussion
The main aim of the present systematic review and meta-analysis was to synthesise the evidence about the effectiveness of psychological interventions for improving mental health symptoms in first responders. A secondary aim was to compare the effectiveness of different types and format of interventions (CBT versus other interventions, clinician delivered versus non- clinician delivered interventions & individual-based versus group-based interventions) included in the review. The results of the meta-analysis showed that psychological interventions were associated with a significant reduction in PTSD symptoms, depression, and anxiety, but not stress. Moreover, the analyses also revealed that CBT was significantly more effective than other interventions for PTSD only, not for depression symptoms. Interventions delivered by clinicians were found to be more effective for PTSD outcomes but not for depression outcomes.
Subgroup analyses also indicated that interventions were more effective in studies that were classified as having low risk of bias contrasted to the moderate-to-high risk of bias studies. The number of sessions was found to be a significant moderator for interventions for PTSD, such that more sessions were associated with greater intervention effectiveness. For anxiety outcomes, only the mean age was found to be a significant moderator of intervention effectiveness, such that higher mean age was associated with greater intervention effectiveness. For depression and stress, no significant moderating variables were found.
The main finding of this meta-analytic review is in line with most previous systematic reviews [2, 18–22] that psychological interventions are effective for reducing PTSD symptoms among first responders. Haugen et al. [2] found large significant treatment effects in two RCT designs included in their review. While Alden et al. [19] found, in 8 RCTs out of 21 studies included in their review, that psychological interventions were effective for treating PTSD symptoms in first responders. Similarly, with additional other types of first responders (e.g., military soldiers, nurses, and doctors), Winders et al. [20] found 13 out of 18 studies that evaluated psychological treatments reported a positive impact of interventions for preventing and treating psychiatric symptoms in first responders. Furthermore, in a review of studies in frontline healthcare and public safety personnel, Anderson et al. [21] found some mixed evidence for the effectiveness of organizational peer-support and crisis-focused psychological interventions. Similarly, Bahji et al. [22], in their meta-analysis found that the psychological interventions reduced PTSD symptoms in public safety personnel. Therefore, taken together, there is clear evidence from multiple reviews that psychological interventions are effective in helping to treat PTSD in first responders.
However, the current review differs in three important ways from previous reviews. Firstly, this is the first meta-analysis to compare between the effectiveness of intervention types. Secondly, unlike other reviews [2, 18, 21] our meta-analysis focused on only two types of study designs (RCTs and controlled before-after designs), thereby, providing more robust evidence of the effectiveness of psychological interventions in this context. Thirdly, our review covered a range of mental health outcomes in addition to PTSD.
There are several similarities between our current review and a recently published meta-analysis [22]. However, our present review differs to this in four key ways:(1) Bahji et al. [22] focused on ‘public safety personnel’, while this meta-analysis investigated first responders using the ‘traditional’ definition; (2) Bahji’s et al. [22] limited their searches from 2008 to 2019 only, while we searched from database inception to 2021, which explains the larger number of studies in this meta-analysis (15 studies compared with 8 studies); (3) Unlike [22], the present meta-analysis specifically compared between three different elements of interventions: CBT versus other interventions, clinicians versus non-clinicians, and individual versus group interventions; (4) Unlike [22], in the current meta-analysis, we also examined the effects of 6 moderators which were gender, mean age, sample size, number of intervention sessions and the total number of minutes of interventions by using meta-regression analysis.
Moreover, our review was the first to test whether there was any difference in the effectiveness of different types of psychological interventions in first responders, and the first to report that CBT was significantly more effective than other types of psychological interventions for reducing PTSD only, not for depression. This pattern of findings is consistent with findings from previous meta-analyses in non-first responder groups. Researchers have reported that several psychological treatments are effective for reducing PTSD in adults [54, 61]. For example, two meta-analytical reviews [62, 63] reported that trauma-focused cognitive therapy (TF-CBT) and EMDR were the most effective psychological therapies for treating PTSD in adults. While another recent meta-analysis [64] found that interpersonal psychotherapy was effective for reducing PTSD. However, most included studies in those meta-analyses were adults who were ‘first victims’ (e.g., sexual assaults, car accidents, refugees) who often suffered from one traumatic event. This is different to first responders who can be regarded as ‘second victims’, and who experience multiple traumas in their daily work. Therefore, the nature of their work makes them different from other populations due to the barriers they suffered such as stigma, fear of losing their job, and not knowing where to get help [65]. In terms of intervention providers, our findings suggested interventions delivered by clinicians were significantly more effective for reducing PTSD symptoms compared to those delivered by non-clinicians. However, there were no significant differences between individual versus group delivery or clinician versus non-clinician delivery for depression outcomes. Therefore, taken together, the current findings suggest, where possible, that future interventions targeting PTSD symptoms ought to be delivered by clinicians.
We also found that the number of sessions was a significant moderator in PTSD studies such that psychological interventions were more effective in studies with a higher number of treatment sessions. This finding is consistent with some previous reviews [66, 67]. Another significant moderator in studies with anxiety outcomes was age such that psychological interventions were more effective in studies with participants with higher mean age. This finding may be explained by the fact that symptoms of anxiety have been found to be higher in older compared to younger [68].
Surprisingly, the current systematic review and meta-analysis only identified a small number of RCT studies conducted in paramedics. It is now established that the prevalence of PTSD is higher in paramedics than other first responders. Therefore, there is an urgent need to investigate the effectiveness of psychological interventions in paramedics group. Yet, we only found 2 studies in paramedics compared with 6 in police officers, 3 in firefighters, and 4 in mixed disaster workers. Also, there is a clear lack of studies that treat burnout among the same population, and this is concerning as the burnout prevalence rate among paramedics has been found to range from 16% to 56% [69]. Therefore, our findings suggest that future research ought to conduct further, higher quality RCTs aimed at improving mental health outcomes in paramedics (as well as other first responders).
All included studies were from 8 different countries, and most of them (11 out of 15 studies) were from the USA, Australia, England, and the Netherlands, which limits the generalisability of the current findings. Differences between developing and developed countries in terms of prevalence of PTSD have been reported [70]. Also, there are differences in the type and availability of psychological interventions for PTSD and mental health generally in developed and developing countries. Therefore, it would be useful for future research to conduct comparative studies between these countries in terms of the types and frequency of traumatic events first responders are exposed to, the available supports they use and the effectiveness of any existing interventions.
Strengths and limitations
The present systematic review and meta-analysis has a number of strengths. The design of this systematic review followed the guidance of the Cochrane handbook for systematic reviews of interventions [35] and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [71]. The protocol of our review was registered on the PROSPERO database. Also, the current meta-analysis focused on RCT designs, which have numerous benefits such as providing strong empirical evidence of treatment’s efficacy, randomisation of the participants to the intervention and control groups and minimisation of the allocation bias [72].
Nevertheless, we are aware that there are also some limitations that ought to be acknowledged. The findings are based on a relatively small number of studies and there was a high level of heterogeneity between studies due to differences in the included studies, such as being conducted in developed versus developing countries, outcome measures, sample sizes, and length of interventions. Even though the review attempted to account for the heterogeneity by conducting subgroup analysis and meta-regressions, the variance was still high. Another limitation was that many of the studies have relatively short follow-up periods, therefore, we are unable to draw firm conclusions about the longer-term effects of interventions [73]. This is important here because first responders are vulnerable to multiple traumas in their work, even after receiving treatment, and follow-up assessment helps determine whether treatment gains can be maintained when exposed to various traumatic events [19]. Furthermore, overall, there were seven studies that were scored as having moderate or high risk of bias total scores, and six studies with no intervention in control groups that may lead to the blinding bias Therefore, future studies should ensure that participants are blinded, include active control conditions, make sure that missing data are addressed adequately and allocation concealment is utilised to help reduce risk of bias. Also, the publication bias analyses identified that there were some missing studies in the right-hand side of the funnel plots for PTSD, depression, and anxiety. A number of factors may account for this potential publication bias. For example, it might be because of the relatively small number of studies conducted in this area and/or that many of the studies are underpowered due to small sample sizes leading to small effects. Therefore, researchers should endeavour to prioritise these types of studies and ensure that future studies are adequately powered, preregistered and embrace open research practices as much as possible [74, 75]. It might also be fruitful for future research to explore the effectiveness of psychological interventions on other stress-related physiological outcomes (e.g., cortisol levels; cf., [76]), as well as psychological outcomes. We also recognise that we did not include other types of first responders (e.g., emergency room personnel; military personnel) and recommend that further research reviews whether interventions may help reduce mental health problems in these groups.
Conclusion
The current systematic review and meta-analysis found that psychological interventions were effective for reducing PTSD, depression, and anxiety symptoms but not for stress in first responders. Further research is needed to estimate the effectiveness of other types of interventions with high quality RCT designs over longer periods of follow-up.
References
- 1. Bricker LRC, Petermann TN, Hines M, Sands J. The Legal Definitions of “First Responder”. NCHRP Research Results Digest. 2013;(385).
- 2. Haugen PT, Evces M, Weiss DS. Treating posttraumatic stress disorder in first responders: A systematic review. Clinical Psychology Review. 2012;32(5):370–80. pmid:22561967
- 3. Prati G, Pietrantoni L. The relation of perceived and received social support to mental health among first responders: a meta‐analytic review. Journal of Community Psychology. 2010;38(3):403–17.
- 4. Benedek DM, Fullerton C, Ursano RJ. First responders: mental health consequences of natural and human-made disasters for public health and public safety workers. Annu Rev Public Health. 2007;28:55–68. pmid:17367284
- 5. Greinacher A, Derezza-Greeven C, Herzog W, Nikendei C. Secondary traumatization in first responders: a systematic review. European journal of psychotraumatology. 2019;10(1):1562840. pmid:30719236
- 6. Carleton RN, Afifi TO, Taillieu T, Turner S, Krakauer R, Anderson GS, et al. Exposures to potentially traumatic events among public safety personnel in Canada. Canadian Journal of Behavioural Science/Revue canadienne des sciences du comportement. 2019;51(1):37.
- 7. Di Nota PM, Bahji A, Groll D, Carleton RN, Anderson GS. Proactive psychological programs designed to mitigate posttraumatic stress injuries among at-risk workers: a systematic review and meta-analysis. Systematic reviews. 2021;10(1):1–21.
- 8.
Sterud T, Hem E, Ekeberg Ø, Lau B. Health problems and help-seeking in a nationwide sample of operational Norwegian ambulance personnel. 2008 [cited 2018 Sep 6]; http://www.biomedcentral.com/1471-2458/8/3
- 9. Bentley MA, Crawford J Mac, Wilkins JR, Fernandez AR, Studnek JR. An assessment of depression, anxiety, and stress among nationally certified EMS professionals. Prehospital Emergency Care. 2013;17(3):330–8. pmid:23414106
- 10. Basinska B. A., & Wiciak I. Fatigue and professional burnout in police officers and firefighters. Internal Security. 2012;4(2).
- 11.
American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-5). 5th ed. Arlington, VA: American Psychiatric Publishing; 2013.
- 12. Jacobs J, Horne-Moyer HL, Jones R. The effectiveness of critical incident stress debriefing with primary and secondary trauma victims. International Journal of Emergency Mental Health [Internet]. 2004; Available from: http://psycnet.apa.org/record/2004-14117-002 pmid:15131998
- 13. Jones S. Describing the Mental Health Profile of First Responders: A Systematic Review. 2017;(529).
- 14. Petrie K, Milligan-Saville J, Gayed A, Deady M, Phelps A, Dell L, et al. Prevalence of PTSD and common mental disorders amongst ambulance personnel: a systematic review and meta-analysis. Social Psychiatry and Psychiatric Epidemiology [Internet]. 2018 Sep 5 [cited 2019 Jan 23];53(9):897–909. Available from: http://link.springer.com/10.1007/s00127-018-1539-5 pmid:29869691
- 15. Wagner S. L., White N., Fyfe T., Matthews L. R., Randall C., Regehr C., et al. Systematic review of posttraumatic stress disorder in police officers following routine work-related critical incident exposure. American Journal of Industrial Medicine. 2020;63(7):600–15. pmid:32419181
- 16. Carleton RN, Afifi TO, Turner S, Taillieu T, Duranceau S, LeBouthillier DM, et al. Mental disorder symptoms among public safety personnel in Canada. The Canadian Journal of Psychiatry. 2018;63(1):54–64. pmid:28845686
- 17. Syed S, Ashwick R, Schlosser M, Jones R, Rowe S, Billings J. Global prevalence and risk factors for mental health problems in police personnel: a systematic review and meta-analysis. Occupational and environmental medicine. 2020;77(11):737–47. pmid:32439827
- 18. Smith A, Roberts K. Interventions for post-traumatic stress disorder and psychological distress in emergency ambulance personnel: a review of the literature. Emergency Medicine Journal. 2003;20(1):75–8. pmid:12533381
- 19. Alden LE, Matthews LR, Wagner S, Fyfe T, Randall C, Regehr C, et al. Systematic literature review of psychological interventions for first responders. Work & Stress. 2020;1–23.
- 20.
Winders WT, Bustamante ND, Garbern SC, Bills C, Coker A, Trehan I, et al. Establishing the Effectiveness of Interventions Provided to First Responders to Prevent and/or Treat Mental Health Effects of Response to a Disaster: A Systematic Review. Disaster Medicine and Public Health Preparedness. 2020;
- 21. Anderson GS, Di Nota PM, Groll D, Carleton RN. Peer support and crisis-focused psychological interventions designed to mitigate post-traumatic stress injuries among public safety and frontline healthcare personnel: a systematic review. International journal of environmental research and public health. 2020;17(20):7645. pmid:33092146
- 22.
Bahji A, Di Nota PM, Groll D, Carleton RN, Anderson GS. Psychological interventions for posttraumatic stress injuries among public safety personnel: a systematic review and meta-analysis. 2021.
- 23. Roberts NP, Kitchiner NJ, Kenardy J, & Bisson JI. Cochrane Database of Systematic Reviews Multiple session early psychological interventions for the prevention of post-traumatic stress disorder (Review). Cochrane Database of Systematic Review [Internet]. 2010 [cited 2018 Sep 7];4. Available from: www.cochranelibrary.com
- 24. Kleim B, Westphal M. Mental Health in First Responders: A Review and Recommendation for Prevention and Intervention Strategies. Traumatology. 2011;17(4):17–24.
- 25. Ignacio J, Amaya C, Givaudan M, Alaide M. EMDR Individual Protocol for Paraprofessional Use: A Randomized Controlled Trial With First Responders. 2013;7(2):24–5. Available from: https://www.ingentaconnect.com/contentone/springer/emdr/2013/00000007/00000002/art00001?crawler=true
- 26. Berger W, Coutinho ESF, Figueira I, Marques-Portella C, Luz MP, Neylan TC, et al. Rescuers at risk: a systematic review and meta-regression analysis of the worldwide current prevalence and correlates of PTSD in rescue workers. Social Psychiatry and Psychiatric Epidemiology [Internet]. 2012 Jun 18 [cited 2019 Jan 23];47(6):1001–11. Available from: http://link.springer.com/10.1007/s00127-011-0408-2 pmid:21681455
- 27. Kaplan JB, Bergman AL, Christopher M, Bowen S, Hunsinger M. Role of Resilience in Mindfulness Training for First Responders. Mindfulness. 2017;8(5):1373–80. pmid:34484459
- 28. Jones R, Gruer L, Gilchrist G, Seymour A, Black M, Oliver J. Recent contact with health and social services by drug misusers in Glasgow who died of a fatal overdose in 1999. Addiction [Internet]. 2002 Dec;97(12):1517–22. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=106862104&site=ehost-live pmid:12472635
- 29. Stergiopoulos E, Cimo A, Cheng C, Bonato S, Dewa CS. Interventions to improve work outcomes in work-related PTSD: a systematic review. BMC Public Health [Internet]. 2011 Jan;11(1):838. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=104604806&site=ehost-live pmid:22040066
- 30. Sterud T, Ekeberg Ø, Hem E. Health status in the ambulance services: a systematic review. BMC Health Services Research. 2006;6(1):82. pmid:16817949
- 31. Blanchard EB, Jones-alexander J, Buckley TC, Forneris CA. Pergamon Psychometric Properties of the PTSD Checklist (PCL). 1996;34(8).
- 32. Foa EB, Riggs DS, Constance V. Reliability and Validity of a Brief Instrument for Assessing Post-naumatic Stress Disorder. 1993;6(4):459–73.
- 33. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. 1983;(6):361–70.
- 34. Maslach C, Jackson SE. The measurement of experienced burnout*. 1981;2:99–113.
- 35.
Higgins J, Green S. Chapter 22: Overview of reviews. Cochrane handbook for systematic reviews of interventions. Cochrane Database of Systematic Reviews [Internet]. 2008;187–235. http://onlinelibrary.wiley.com/doi/10.1002/9780470712184.fmatter/summary
- 36.
EPOC. EPOC. Suggested Risk of Bias Criteria for EPOC Review. 2014;2–5.
- 37.
Borenstein, M., & Rothstein H. Comprehensive Meta-Analysis software. Biostat, Englewood. NJ; 2013.
- 38. Michie S, Abraham C, Whittington C, John McAteer ASG. Effective techniques in healthy eating and physical activity interventions: A meta- regression. Health Psychology. 2009;28(6):690. pmid:19916637
- 39. Cochran WG. The combination of estimates from different experiments. Biometrics. 1954;10(1):101–29.
- 40.
Egger M, Smith GD, Schneider M, Minder C. Papers Bias in meta-analysis detected by a simple, graphical test [Internet]. 1997 [cited 2018 Aug 29]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2127453/pdf/9310563.pdf
- 41. Duval S, Tweedie R. Trim and fill: a simple funnel‐plot–based method of testing and adjusting for publication bias in meta‐analysis. Biometrics. 2000;56(2):455–63. pmid:10877304
- 42. Alghamdi M, Hunt N, Thomas S. The effectiveness of Narrative Exposure Therapy with traumatised firefighters in Saudi Arabia: A randomized controlled study. Behaviour Research and Therapy [Internet]. 2015;66:64–71. Available from: pmid:25701801
- 43. Backman L, Arnetz BB, Levin D, Lublin A. Psychophysiological effects of mental imaging training for police trainees. Stress Medicine. 1997;13(1):43–8.
- 44. Ireland Malouff J. M., & Byrne B. M. The efficacy of written emotional expression in the reduction of psychological distress in police officers. International Journal of Police Science & Management. 2007;9(4):303–11.
- 45. Macnab A, Sun C, Lowe J, A. M, C. S, Macnab A, et al. Randomized, controlled trial of three levels of critical incident stress intervention. Prehospital and disaster medicine [Internet]. 2003;18(4):367–71. Available from: http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=med4&NEWS=N&AN=15310050%5Cnhttp://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=emed6&NEWS=N&AN=15310050 pmid:15310050
- 46. Ramey SL, Perkhounkova Y, Hein M, Chung S, Franke WD, Anderson AA. Building Resilience in an Urban Police Department. Journal of Occupational & Environmental Medicine [Internet]. 2016 Aug;58(8):796–804. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=117119276&site=ehost-live
- 47. Sarason IG, Johnson JH, Berberich JP, Siegel JM. Helping Police Officers to Cope with Stress: A Cognitive-Behavioral Approach. American Journal of Community Psychology. 1979;7(6):593–603. pmid:517465
- 48. Shipley P, Baranski J V. Police officer performance under stress: A pilot study on the effects of visuo-motor behavior rehearsal. International Journal of Stress Management. 2002;9(2):71–80.
- 49. Skeffington PM, Rees CS, Mazzucchelli TG, Kane RT. The primary prevention of PTSD in firefighters: Preliminary results of an RCT with 12-month follow-up. PLoS ONE [Internet]. 2016;11(7):1–22. Available from: http://dx.doi.org/10.1371/journal.pone.0155873 pmid:27382968
- 50. Tuckey MR, Scott JE. Group critical incident stress debriefing with emergency services personnel: a randomized controlled trial. Anxiety, Stress & Coping [Internet]. 2014 Jan 2;27(1):38–54. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=103998896&site=ehost-live pmid:23799773
- 51. Wee DF, Mills DM, Koehler G. The effects of critical incident stress debriefing (CISD) on emergency medical services personnel folloing the Los Angeles civil disturbance. International Journal of Emergency Mental Health. 1990;1:33–7.
- 52. Behnammoghadam M, Kheramine S, Zoladl M, Cooper RZ, Shahini S. Effect of eye movement desensitization and reprocessing (EMDR) on severity of stress in emergency medical technicians. Psychology Research and Behavior Management. 2019;12:289–96. pmid:31118845
- 53. Bryant RA, Kenny L, Rawson N, Cahill C, Joscelyne A, Garber B, et al. Efficacy of exposure-based cognitive behaviour therapy for post-traumatic stress disorder in emergency service personnel: a randomised clinical trial. Psychological medicine. 2019;49(9):1565–73. pmid:30149825
- 54. Carlier IVE, Voerman AE, Gersons BPR. The influence of occupational debriefing on post-traumatic stress symptomatology in traumatized police officers. British Journal of Medical Psychology. 2000;73(1):87–98. pmid:10759053
- 55. Chongruksa D, Parinyapol P, Sawatsri S, Pansomboon C. Efficacy of eclectic group counseling in addressing stress among Thai police officers in terrorist situations…[corrected][published erratum appears in COUNS PSYCHOL Q 2012 September; 25(3): 343]. Counselling Psychology Quarterly [Internet]. 2012 Mar;25(1):83–96. Available from: http://search.ebscohost.com/login.aspx?direct=true&db=cin20&AN=104538002&site=ehost-live
- 56. Christopher MS, Hunsinger M, Goerling LRJ, Bowen S, Rogers BS, Gross CR, et al. Mindfulness-based resilience training to reduce health risk, stress reactivity, and aggression among law enforcement officers: A feasibility and preliminary efficacy trial. Psychiatry Research. 2018;264(March):104–15.
- 57. Difede JA, Malta LS, Best S, Henn-Haase C, Metzler T, Bryant R, et al. A randomized controlled clinical treatment trial for World Trade Center attack-related PTSD in disaster workers. Journal of Nervous and Mental Disease. 2007;195(10):861–5. pmid:18043528
- 58. Doctor RS, Cutris D, Isaacs G. Psychiatric morbidity in policemen and the effect of brief psychotherapeutic intervention: A pilot study. Stress Medicine. 1994;
- 59. Gersons BPR, Carlier IVE, Lamberts RD, Van Der Kolk BA. Randomized clinical trial of brief eclectic psychotherapy for police officers with posttraumatic stress disorder. Journal of Traumatic Stress. 2000;13(2):333–47. pmid:10838679
- 60. Wild J, El-Salahi S, Degli Esposti M, Thew GR. Evaluating the effectiveness of a group-based resilience intervention versus psychoeducation for emergency responders in England: A randomised controlled trial. PloS one. 2020;15(11):e0241704. pmid:33180798
- 61. Benish SG, Imel ZE, Wampold BE. The relative efficacy of bona fide psychotherapies for treating post-traumatic stress disorder: A meta-analysis of direct comparisons. Clinical Psychology Review. 2008;28(5):746–58. pmid:18055080
- 62. Cusack K, Jonas DE, Forneris CA, Wines C, Sonis J, Middleton JC, et al. Psychological treatments for adults with posttraumatic stress disorder: A systematic review and meta-analysis. Clinical Psychology Review. 2016;43:128–41. pmid:26574151
- 63. Bisson JI, Ehlers A, Matthews R, Pilling S, Richards D, Turner S. Psychological treatments for chronic post-traumatic stress disorder. British Journal of Psychiatry [Internet]. 2007;190(02):97–104. Available from: https://www.cambridge.org/core/product/identifier/S0007125000233485/type/journal_article
- 64. Watts B V., Schnurr PP, Mayo L, Young-Xu Y, Weeks WB, Friedman MJ. Meta-analysis of the efficacy of treatments for posttraumatic stress disorder. Journal of Clinical Psychiatry. 2013;74(6):541–50. pmid:23842024
- 65. Althobaiti S, Kazantzis N, Ofori-Asenso R, Romero L, Fisher J, Mills KE, et al. Efficacy of interpersonal psychotherapy for post-traumatic stress disorder: A systematic review and meta-analysis. Journal of affective disorders. 2020;264:286–94. pmid:32056763
- 66. Haugen PT, McCrillis AM, Smid GE, Nijdam MJ. Mental health stigma and barriers to mental health care for first responders: A systematic review and meta-analysis. Journal of Psychiatric Research [Internet]. 2017;94:218–29. Available from: http://dx.doi.org/10.1016/j.jpsychires.2017.08.001 pmid:28800529
- 67. Hansen NB, Lambert MJ, Forman EM. The psychotherapy dose-response effect and its implications for treatment delivery services. Clinical Psychology: science and practice. 2002;9(3):329.
- 68. Lambert JE, Alhassoon OM. Trauma-focused therapy for refugees: Meta-analytic findings. Journal of counseling psychology. 2015;62(1):28. pmid:25485547
- 69. Wolitzky‐Taylor KB, Castriotta N, Lenze EJ, Stanley MA, Craske MG. Anxiety disorders in older adults: a comprehensive review. Depression and anxiety. 2010;27(2):190–211. pmid:20099273
- 70. Reardon M, Abrahams R, Thyer L, Simpson P. Review article: Prevalence of burnout in paramedics: A systematic review of prevalence studies. EMA—Emergency Medicine Australasia. 2020;32(2):182–9. pmid:32067408
- 71. Yatham S., Sivathasan S., Yoon R., da Silva T. L., & Ravindran A V. Depression, anxiety, and post-traumatic stress disorder among youth in low and middle income countries: A review of prevalence and treatment interventions. Asian Journal of Psychiatry. 2018;38:78–91. pmid:29117922
- 72. Moher D, Liberati A, Tetzlaff J, Altman DG. Academia and Clinic Annals of Internal Medicine Preferred Reporting Items for Systematic Reviews and Meta-Analyses: Annals of Internal Medicine. 2009;151(4):264–9.
- 73. Levin KA. Study design VII. Randomised controlled trials. Evidence-Based Dentistry. 2007;8(1):22–3. pmid:17380181
- 74. Johnson J, Panagioti M. Interventions to improve the breaking of bad or difficult news by physicians, medical students, and interns/residents: A systematic review and meta-analysis. Vol. 93, Academic Medicine. 2018. 1400–1412 p. pmid:29877913
- 75. Bosnjak M, Fiebach CJ, Mellor D, Mueller S, O’Connor DB, Oswald FL, et al. A template for preregistration of quantitative research in psychology: Report of the joint psychological societies preregistration task force. American Psychologist. 2021. pmid:34807636
- 76. O’Connor DB. Leonardo da Vinci, preregistration and the architecture of science: Towards a more open and transparent research culture. Health Psychology Bulletin. 2021;5(1):39–45.