Search results

Figure 1 shows the PRISMA flow chart for the searches.

Figure 1

PRISMA flow chart. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

From 8412 screened articles, 179 full texts were reviewed for eligibility and 84 studies were identified for inclusion and data extraction. The publication dates of the reviews ranged from 1998 to 2022. The 84 reviews were classified into categories of interventions to provide a framework for evidence synthesis (see table 1).

Reviews of pharmacological interventions

Eight systematic reviews evaluated pharmacological interventions on the development of (diagnosis) of PTSD following injury (online supplemental table 1).33–40 Reviews either included any pharmacological interventions (eg, glucocorticoids, opioids, antidepressants, beta-blockers, gabapentinoids, non-steroidal anti-inflammatory drug (NSAIDS), paracetamol, antihistamines and benzodiazepines)33 35 38–40 or specific agents (eg, benzodiazepines36 or propranolol37). The clinicians delivering the interventions were rarely specified. Data on the timing of administration of the medication following trauma ranged from time of admission to the emergency department to up to 36 days following traumatic injury. The setting of the delivery of the interventions was reported as emergency departments. Assessed risk was not an inclusion criterion of any of the reviews. The distress outcomes measured were mainly diagnosis of/severity of symptoms of PTSD or acute stress disorder (ASD) and anxiety and depression.

Five reviews reported there was low-quality evidence that hydrocortisone was found efficacious in preventing PTSD.33–35 38 40 Other pharmacological agents, including benzodiazepines, were not found to be effective in preventing PTSD.33–35 37–40 Moreover, it was reported that there are possible harmful effects of using benzodiazepines in this setting.36

Reviews of psychological interventions

There were 44 reviews of psychological interventions which reported on distress-related outcomes in the time following injury (online supplemental table 1).33–35 41–82 Anxiety and depression were most commonly reported outcomes, however, reviews also included distress-related outcomes including PTSD symptoms and diagnosis, fear avoidance, kinesophobia, pain coping, pain acceptance, pain-related distress. The intervention duration ranged from single sessions to weekly sessions over 12 months. The settings varied widely and included hospital outpatients, home visits, emergency departments, general practice and inpatient settings.

A large proportion of the reviews (N=15) reported the distress-related outcomes of interventions employing a variety of psychological strategies including acceptance and commitment therapy, cognitive–behavioural therapy (CBT), behavioural therapy, goal setting and self-management strategies.34 44 50 62 64–72 75 82 In these reviews, which included a heterogeneous mix of psychological interventions, the reported benefits for distress-related outcomes were mixed and the reported quality of evidence was variable (ranged from low to high or otherwise not assessed).

The results of reviews that focused on specific psychological interventions were more consistent. Three reviews evaluated CBT following injury,41–43 and all reviews reported CBT to improve distress-related outcomes although, overall the quality of evidence was described as low. One review evaluated behavioural therapy.44 Three reviews evaluated mindfulness interventions following injury, and all reviews reported no evidence for improvements in distress-related outcomes.45–47 Two reviews evaluated neuro/biofeedback interventions following injury and both reported moderate to high-level evidence for these interventions being beneficial for distress-related outcomes.48 49 goal-setting following injury50 and music51 with all reporting benefits for distress-related outcomes. The quality of the included studies within these three reviews was not assessed.

Seven reviews evaluated educational interventions.52–58 The reported benefits of these interventions for distress-related outcomes were mixed and the quality of the included studies was reported to vary from low to high and some reviews did not assess quality. Two reviews investigated the efficacy of single-session debriefing interventions, and neither found this approach to be beneficial for distress-related outcomes including PTSD diagnosis.59 60 The authors reported these studies to be of variable quality. By contrast, reviews of interventions to prevent PTSD which included multiple treatment sessions reported improvements in anxiety and depression.33 73 76–80 Five reviews had inclusion criteria requiring studies to have included participants on the basis of identified risk or analysed these studies separately from studies that did not have such inclusion criteria.41 42 73 77 78 The remaining reviews did not specify that studies include participants based on risk screening measures. Three of these reviews synthesised and compared results from studies that had used risk screening tools and studies that had not used risk screening tools to identify participants for inclusion.42 73 77 These reviews reported that the use of risk screening tools was associated with better outcomes following intervention when compared with interventions applied to anyone, regardless of risk profile. Similarly, the remaining two reviews that had inclusion criteria for studies to include participants on the basis of higher-risk profiles reported that interventions in this population were effective.41 78

Psychologists were the most commonly listed clinician, however, the clinicians listed in the delivery of the interventions also included doctors, nurses, counsellors, researchers, social workers, case managers, occupational therapists and physiotherapists. Four reviews focused specifically on cognitive and/or behaviour-based approaches that were delivered by non-psychologists.50 61 64 82 The populations for these four reviews were largely those who would ordinarily present to a physiotherapist or other MSK specialist (eg, occupational therapists and athletic trainers) following injury, such as athletes or people with neck pain following a motor vehicle accident. These interventions were delivered in hospital clinics (inpatient and outpatient), community clinics, homes, emergency departments, research departments and allied health clinics.

Reviews of exercise-based interventions

Eight reviews evaluated exercise-based interventions in the time following injury and reported distress-related outcomes (online supplemental table 1).83–90 The distress outcomes reported in the reviews included self-efficacy, fear avoidance, anxiety and depression. One review evaluated yoga, reporting potential benefits for distress-related outcomes.83 Two reviews investigated hydrotherapy, also reporting potential benefits for distress-related outcomes.84 85 The remaining five reviews included studies evaluating different types of exercise-based interventions (home exercise programmes, weight training, supervised exercise sessions, therapeutic recreation, adventurous training and sports and outdoor activity training) and the distress-related outcomes reported in these reviews were mixed.86–90

The delivery of the interventions included face-to-face supervised exercise, telehealth-delivered, class-based delivery and prescribed home exercises. The settings varied widely and included rehabilitation centres, physiotherapy clinics, hospital outpatients and inpatient settings. Several reviews included data on the duration and frequency of the exercise interventions and reported a large range from 15 min sessions to 3-hour sessions and a frequency that ranged from one session per week to seven sessions per week. The total duration of interventions ranged from 1 week to 12 months. Risk screening tools were not reported to be used routinely and assessed risk was not an inclusion criterion of studies for any reviews.

The reviews of mixed exercise interventions indicated improvements in self-efficacy, fear avoidance, anxiety, stress and PTSD symptoms, however, the quality of the evidence was described as low or was not assessed.86–90

Most of the reviews did not include information on the clinicians delivering the interventions. However, several described physiotherapists, occupational therapists, trained non-professionals, exercise scientists, personal trainers, nurses and chiropractors facilitating the interventions.85 86 88

Reviews of multimodal interventions

For the purpose of this paper, multimodal interventions refer to interventions that have been delivered simultaneously by more than one clinical discipline. Six reviews evaluated studies of interventions which included more than one therapeutic modality (online supplemental table 1).91–96 The settings included rehabilitation hospitals, outpatients departments, inpatient settings and community rehabilitation settings. Risk screening tools were not reported to be used routinely and assessed risk was not an inclusion criterion of studies for any reviews.

The distress outcomes investigated were varied. For those who focused on rehabilitation, fear avoidance, self-efficacy and fear of falling were often investigated. Other variables included PTSD diagnosis, anxiety, depression, stress and helplessness. Many reviews included interventions delivered in the acute phase but found that distress outcomes were not reported. For example, in the review by Chudyk et al, of the 55 included studies, only 5 reported on distress outcomes.91

It was reported that there was mixed evidence for the benefits of multidisciplinary or multimodal rehabilitation interventions, although the quality of evidence reported across reviews was variable.91–96

Clinicians delivering the interventions were frequently not specified in these reviews, especially those focused on rehabilitation settings. For those reviews that did specify who was delivering the interventions, clinicians included general practitioners or other doctors, physiotherapists, osteopaths, chiropractors, nurses, nurse practitioners, researchers, rehabilitation physicians, social workers and naprapaths (Scandinavian manual therapist). The modes of delivery for all interventions in this category included face-to-face clinical care (individual or group based), telehealth, pamphlets and online education. The duration of the interventions ranged from single telephone contact to 2-year case management interventions with a broad range of duration and frequency of contact seen between interventions.

Reviews that focused on physical/manual therapy interventions

Nine reviews investigated physical interventions (online supplemental table 1).97–105 There was one review on each of the following: local heat application,97 massage,98 physiotherapy99 and McKenzie method.100 Two reviews focused on transcutaneous electrical nerve stimulation (TENS),101 102 two investigated mixed conservative strategies,103 104 (defined as any form of non-invasive procedure such as education, exercise, manual therapy or electrotherapy), and there was one review on spinal manipulation.105 The settings included inpatient and outpatient hospital settings, allied health clinics and rehabilitation settings. Risk screening tools were not reported to be used routinely and assessed risk was not an inclusion criterion of studies for any reviews.

For the two studies that reviewed the use of TENS, one reported that there was evidence which had a low risk of bias to suggest that TENS can reduce pain-related anxiety.102 The other reported that of the few studies that reported on psychological outcomes, no effect was demonstrated on anxiety or depression, although the quality of evidence was reportedly low.101 The reviews that focused on massage therapy98 and spinal manipulation105 both reported that interventions had a positive effect on psychological outcomes including anxiety and that the quality of evidence was variable. The reviews on conservative interventions103 104 and local heat application97 reported that conclusions could not be drawn about the effectiveness of interventions either due to low quality or insufficient data for the population of interest. A review of McKenzie method reported that moderate to high quality evidence suggested that the interventions had an effect on improving pain self-efficacy, depression, psychological distress and fear avoidance beliefs.100 One review evaluated physiotherapy interventions for subacute and chronic low back pain and reported only one trial that included psychological outcomes. This trial found physiotherapy to improve fear avoidance beliefs, with the quality of evidence of all studies included described as variable.99

The interventions were reported as being delivered face to face on an individual basis or were not specified. The frequency of delivery of interventions ranged from single session to 98 sessions and the duration of individual sessions varied significantly according to the type of intervention. The reviews in this category reported on distress outcomes including anxiety, depression, back pain beliefs and fear avoidance beliefs. Many of the reviews aimed to capture information on distress outcomes but found that included studies did not report on these. For example the review by Rampazo et al captured 30 studies but found only one study reported on depression.101

Clinicians delivering the intervention were listed as physiotherapists, chiropractors, massage therapists, counsellors or were frequently not specified in the review.

Reviews that focused on workplace interventions

This category included three reviews of interventions delivered in the workplace (see online supplemental table 1).106–108 The types of interventions in the reviews included those where the setting of the delivery of the intervention was the workplace. Interventions were exercise interventions, ergonomic interventions, education and relaxation interventions. These reviews were primarily interested in distress outcomes pertaining to physical function or return to work, but they all included or aimed to include psychological outcomes including anxiety, depression and fear avoidance beliefs. Risk screening was not described in any of the reviews and the duration of the interventions ranged from single education sessions to permanently implemented ergonomic changes.

Clinicians delivering the interventions were reported as being physiotherapists, psychologists, researchers, job coaches or were not specified. The duration and frequency of interventions were largely not reported but ranged from single-session education interventions to permanently implemented ergonomic interventions.

One review found there was not enough high-quality data to draw conclusions about the effects of the interventions on psychological outcomes.106 One review, with evidence described as moderate to high quality, reported that workplace interventions can be effective in improving fear avoidance beliefs and other psychological outcomes.107

Reviews that investigated virtual reality

Two reviews investigated the use of virtual reality interventions. The quality of the evidence was reported to be low or not reported.109 110 The types of interventions included exercise and movement, exposure, education and relaxation augmented by virtual reality. The distress outcomes investigated were anxiety, depression, fear avoidance, PTSD symptoms and driving phobia. Clinicians delivering the interventions included physiotherapists, researchers and psychologists or were not described. The use of risk screening tools was not described. The duration of the sessions ranged from several minutes up to 120 min and the number of sessions ranged from 1 to 25. Risk screening was not described and the treatment settings listed included hospitals and were otherwise not described.

The authors of one review reported that there were some clinically significant improvements seen for fear avoidance.109 In the other article, the authors reported that the interventions may be feasible and acceptable but no conclusions regarding efficacy were able to be drawn.110

Reviews that investigated emergency department interventions

This category captured one review that explored interventions for low back pain in the emergency department. The types of interventions were not comprehensively listed but a brief overview included medication, physiotherapy and prehospital transport interventions. Information about the duration of the interventions was not provided. The setting was emergency departments and anxiety was the only outcome measure reported. The review did not assess the quality of articles and reported that interventions in the prehospital setting may be helpful in reducing anxiety.111 The clinicians delivering the intervention were not reported. The use of risk screening tools and the duration of interventions was not described.

Reviews that investigated occupational therapy interventions

This category of interventions captured two reviews, both with large numbers of included studies.112 113 They types of interventions fell under the umbrella of occupational therapy but were reported to include activity alterations, ergonomic interventions, informational interventions and interventions to improve utilisation of services such as social services and employment services. The inclusion criteria for both reviews were that the interventions were delivered by occupational therapy practitioners. While a broad range of types of interventions and the outcome measures taken were listed in the reviews, a synthesis of outcomes pertaining to distress was not reported in one review,112 and improvements for mental health outcomes were reported in the other review.113 The use of risk screening tools and the settings were not described. Duration of the intervention was described in one review as ranging from two thirty minute sessions weekly up to 2.5-hour sessions 5 days a week for an unspecified number of weeks. The quality of the included articles was not reported.

Reviews that investigated follow-up interventions after discharge

This category of interventions included three reviews that specifically explored the time following discharge from the intensive care unit (ICU), and therefore, included populations with injury.114–116 The variety of interventions included psychological, education, physiotherapy, exercise and counselling interventions. The settings included outpatients departments, hospitals and home visits. The duration of the interventions ranged from 6 weeks to 3 months in duration or were not well described. The clinicians were reported to include nurses, doctors, psychologists or were not reported. The use of risk screening tools was not described. All three reviews reported low quality or insufficient evidence for the included studies. Two of the studies114 115 reported that the interventions could be helpful to improve psychological outcomes.