Organised trauma systems and designated trauma centres for improving outcomes in injured patients

Abstract

Rationale

Trauma systems have become the standard of care in high‐income countries, but remain uncommon in low‐ and middle‐income countries. High‐quality evidence of effectiveness is needed to advocate for the development of trauma systems in low‐ and middle‐income countries, where the burden of injury is highest.

Objectives

To assess the benefits and harms of organised trauma systems and designated trauma centres compared with usual care in injured patients.

Search methods

We searched CENTRAL, MEDLINE, Embase, ClinicalTrials.gov, and WHO ICTRP on 16 December 2023. We also searched grey literature, checked reference lists of included studies, and contacted the authors of relevant studies.

Eligibility criteria

We included randomised controlled trials, non‐randomised trials, controlled before‐after studies, and interrupted time series studies comparing organised trauma systems or designated trauma centres with usual care. We planned to include patients who had had major trauma (i.e. Injury Severity Score greater than 15), but made a post‐hoc decision to include patients regardless of injury severity. Studies were considered for inclusion regardless of date, language, or publication status.

Outcomes

The critical outcomes were patient outcomes (such as mortality, survival, and recovery), and adverse effects. Important outcomes were utilisation and access to trauma care services, quality of care provided, equity, and knowledge about trauma care services. Studies only reported patient outcomes (mortality, survival); there were no reports on adverse effects, utilisation and access to services, quality of care, equity, and knowledge about trauma care services.

Risk of bias

We used the Cochrane RoB 1 tool and guidance from the Cochrane Effective Practice and Organisation of Care (EPOC) group to evaluate individual studies.

Synthesis methods

Two review authors independently selected studies for inclusion, extracted data, and assessed risk of bias and certainty of evidence using GRADE. We could not perform a meta‐analysis due to substantial clinical heterogeneity across studies. We re‐analysed data from individual studies so they could be presented in a standardised format as relative effect, change in level, and change in slope. We summarised findings using a narrative synthesis.

Included studies

There were four interrupted time series studies (157,111 participants). Two studies (131,220 participants) compared organised trauma systems to usual care and two studies (25,891 participants) compared designated trauma centres to usual care. Two studies were conducted in the US, one in the UK, and one in Norway.

Synthesis of results

It is very uncertain whether organised trauma systems reduce mortality compared to usual care because the certainty of the evidence was very low (2 studies, 131,220 participants). One study (20,357 participants, follow‐up 6 years) reported reduced mortality with an organised trauma system intervention and a relative effect of −30% (change in level −2.02, 95% confidence interval (CI) −3.17 to −0.86; change in slope −0.10, 95% CI −0.53 to 0.33). A second study (110,863 participants, follow‐up 9 years) reported improved survival with intervention and a relative effect of 14% (change in level 0.13, 95% CI −0.50 to 0.76; an increase of 0.13 survival cases per quarter; change in slope 0.08, 95% CI 0.01 to 0.15).

No studies reported data on adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma care services.

It is very uncertain whether designated trauma centres reduce mortality compared to usual care because the certainty of the evidence was very low (2 studies, 25,891 participants). One study (7247 participants) reported a decrease in mortality for all patients (relative effect −42%; change in level −3.60, 95% CI −11.26 to 4.07; change in slope −0.51, 95% CI −2.26 to 1.23), patients with Injury Severity Score 15 to 24 (relative effect −44%; change in level −8.80, 95% CI −29.61 to 12.00; change in slope −0.96, 95% CI −5.69 to 3.77), and patients with Injury Severity Score greater than 24 (relative effect −55%; change in level −17.87, 95% CI −47.12 to 11.37; change in slope −1.90, 95% CI −8.55 to 4.75). A second study (18,644 participants, follow‐up 11 years) reported reduced mortality for adults (relative effect −67%; change in level −17.52, 95% CI −42.27 to 7.23; change in slope −2.09, 95% CI −6.22 to 2.04) and children (relative effect −84%; change in level −18.56, 95% CI −30.11 to −7.01; change in slope −1.12, 95% CI −3.04 to 0.80).

No studies reported data on adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma care services.

Authors' conclusions

The available evidence is currently insufficient to quantify the implications and impact of organised trauma systems and designated trauma centres on clinical practice. This is primarily due to a lack of studies with high methodological rigour for assessing the effects of clinical interventions, as well as the absence of reporting on important outcomes for determining their effectiveness.

Future research could provide evidence by utilising observational studies with high methodological rigour when randomised trials are not feasible; and focus on collecting important outcomes such as the utilisation, access, and quality of care provided, and knowledge about trauma care services.

Funding

This Cochrane review had no dedicated funding.

Registration

Protocol available via doi.org/10.1002/14651858.CD012500.

Plain language summary

Do trauma‐dedicated hospital networks and trauma‐dedicated hospitals improve outcomes for people with traumatic injuries?

Key messages

• Due to a lack of robust evidence, we could not determine the benefits and unwanted effects of hospital networks (organised trauma systems) or individual hospitals (designated trauma centres) that are specialised in the care of people with traumatic injuries.

• This research calls for improvement in designing studies and reporting important measures such as death and survival, unwanted effects, use or access to trauma care services, quality of the care provided, providing care to everyone who needs it, and informing everyone about the care available.

What are organised trauma systems, designated trauma centres, and usual care?

An organised trauma system is an entire service designed to provide care to injured people (for example, from road traffic collisions, weapons, and burns). It covers emergency and initial care at the location of the injury, transportation to the hospital, determining which hospital each patient is taken to, and care beyond their hospital stay. Trauma systems also include recovery of costs, overseas training, research, and injury prevention. Designated trauma centres are hospitals dedicated to treating people with traumatic injuries, and they form part of a trauma system network in a specific geographical location.

Trauma systems and trauma centres have been reported to offer a cost‐effective approach to managing people with traumatic injuries in countries with a high standard of living. Evaluating its effectiveness by well‐designed research is particularly important because this would support their use in resource‐limited (poor) settings where the burden of injury is highest.

Usual care refers to general hospitals and healthcare systems that are not specifically dedicated to treating people with traumatic injuries. In regions or geographical locations where trauma systems have not been established, usual care is typically the standard approach.

What did we want to find out?

We wanted to find out if organised trauma systems and designated trauma centres are:

• better than usual care for avoiding deaths and improving a person's recovery;

• associated with unwanted effects.

We specifically wanted to find their impact on:

• healthcare‐related unwanted effects

• use and access to health care

• quality of the care provided

• whether care was provided to everyone who needed it, and

• whether people knew what care was available.

What did we do?

We searched for studies that investigated the effect of organised trauma systems and designated trauma centres compared to usual care. We were interested in deaths, unwanted effects, use or access to trauma care services, the quality of the care provided, providing care to everyone who needs it, and informing everyone about the care available.

We compared and summarised the results of the studies, and rated our confidence in the evidence based on factors such as study methods, and how precise and reliable the results were.

What did we find?

We found four studies involving 157,111 people that evaluated the effectiveness of organised trauma systems or designated trauma centres. However, none of these studies were conducted in resource‐limited countries.

We are not confident in the evidence because:

• there are insufficient studies to draw robust conclusions;

• the evidence did not comprehensively address all that we were interested in;

• some studies did not report data on what we were interested in;

• the studies varied in terms of the populations studied and the methods of delivering the care.

What are the limitations of the evidence?

Our confidence in the evidence is very low, and the results of further research could differ from the results of this review. The studies were not well designed, results were very inconsistent across the different studies, and they did not report all that we were interested in.

How up to date is this evidence?

The evidence is up to date to 16 December 2023.

Summary of findings

Summary of findings 1. Organised trauma system compared to usual care for improving outcomes in injured patients.

Patient or population: injured patients Setting: trauma system geographical catchment area constituting 'trauma system region' Intervention: organised trauma system Comparison: usual care
Outcomes	Impact	№ of participants (studies)	GRADE	Comments
Patient outcomes (mortality assessed at 6 years' follow‐up; survival rate assessed at 9 years' follow‐up)	Mortality Relative effect −30%; change in level −2.02 (95% CI −3.17 to −0.86); change in slope −0.10 (95% CI −0.53 to 0.33)	131,220 (2 studies: He 2016; Moran 2018)	⊕⊝⊝⊝ Very lowa,b,c	Data analysed at first quarter following intervention. Outcomes reported using different scales; hence data could not be pooled.
	Survival rate Relative effect 14%; change in level 0.13 (95% CI −0.50 to 0.76); change in slope 0.08 (95% CI 0.01 to 0.15)
Adverse effects	—	—	—	No studies reported this outcome.
Utilisation and access to trauma care services	—	—	—	No studies reported this outcome.
Quality of care provided	—	—	—	No studies reported this outcome.
Knowledge about trauma service provision	—	—	—	No studies reported this outcome.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The initial confidence in the estimate of effect for non‐randomised evidence (i.e. interrupted time series studies) is low confidence and we further downgraded confidence due to inconsistency and imprecision and upgraded due to plausible confounding.
^aInitially downgraded owing to their non‐randomised design.
^bDowngraded due to serious inconsistency and serious imprecision.
^cUpgraded due to plausible confounding. A trauma system is likely to admit sicker patients than usual hospitals, therefore confounding against detection of observed effect.

Summary of findings 2. Designated trauma centre compared to usual care for improving outcomes in injured patients.

Patient or participants: injured patients Setting: trauma hospital where all trauma patients are sent/treated Intervention: designated trauma centre Comparison: 'usual care'
Outcomes	Impact	№ of participants (studies)	GRADE	Comments
Patient outcomes (mortality, assessed at 11 years' follow‐up for all participants, and 6 years' follow‐up for paediatric participants)	All patients Relative effect −42%; change in level −3.60 (95% CI −11.26 to 4.07); change in slope −0.51 (95% CI −2.26 to 1.23)	25,891 (2 studies: Groven 2011; Rotondo 2009)	⊕⊝⊝⊝ Very lowa,b,c	Data analysed at first quarter following intervention. Mortality outcomes were reported using different scales (based on age or injury severity); hence data could not be pooled. Mortality in adults was also assessed at 6 years' follow‐up: relative effect −67%; change in level −17.52 (95% CI −42.27 to 7.23); change in slope −2.09 (95% CI −6.22 to 2.04).
	Participants with Injury Severity Score 15–24 Relative effect −44%; change in level −8.80 (95% CI −29.61 to 12.00); change in slope −0.96 (95% CI −5.69 to 3.77)
	Participants with Injury Severity Score > 24 Relative effect −55%; change in level −17.87 (95% CI −47.12 to 11.37); change in slope −1.90 (95% CI −8.55 to 4.75)
	Paediatric participants Relative effect −84%; change in level −18.56 (95% CI −30.11 to −7.01); change in slope −1.12 (95% CI −3.04 to 0.80)
Adverse effects	—	—	—	No studies reported this outcome.
Utilisation and access to trauma care services	—	—	—	No studies reported this outcome.
Quality of care provided	—	—	—	No studies reported this outcome.
Knowledge about trauma service provision	—	—	—	No studies reported this outcome.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The initial confidence in the estimate of effect for non‐randomised evidence (i.e. interrupted time series studies) is low confidence and we further downgraded confidence due to inconsistency and imprecision and upgraded due to plausible confounding.
^aInitially downgraded owing to their non‐randomised design.
^bDowngraded due to serious inconsistency and serious imprecision.
^cUpgraded due to plausible confounding. A trauma system is likely to admit sicker patients than usual hospitals, therefore confounding against detection of observed effect.

Background

Almost five million people die from injury worldwide every year [1], and many others survive but have significant long‐term disability. Trauma is estimated to account for the loss of 180 million disability‐adjusted life years annually [2], with around 90% of the burden occurring in low‐ and middle‐income countries [3]. Road traffic injuries result in an overall economic loss of 167.8 billion US dollars worldwide [4]. In response, trauma care has evolved over recent decades and culminated in the implementation of organised trauma systems in many healthcare jurisdictions [5, 6]. There is emerging evidence that such strategies may effectively reduce the burden of trauma and organised trauma systems are rapidly becoming the accepted standard in many high‐income countries [7, 8, 9, 10, 11].

Despite low‐ and middle‐income countries carrying most of the trauma burden [3, 12, 13], few have implemented formal trauma systems. In their guideline documents on 'Prehospital Trauma Care', the World Health Organization recommended developing resource‐appropriate trauma systems [14, 15]. Researchers have also based recommendations on the feasibility of providing care given resource limitations, prehospital capabilities, and structural challenges in low‐ and middle‐income countries (e.g. poor road infrastructure) [12]. Although individual reports suggest that systemised trauma care can improve resource planning, organisation, and patient outcomes [16], this review sought to synthesise the available evidence.

Description of the condition

Trauma describes a spectrum of physical injuries that arise from mechanisms as varied as low‐energy falls, road traffic collisions, weapons, and burns. These are often categorised as intentional (e.g. suicide attempt, assault), unintentional and blunt (e.g. fall), or penetrating (e.g. gunshot wound). They can range from minor (e.g. ankle sprain) to fatal (e.g. decapitation).

Description of the intervention and how it might work

Organised trauma systems are a planned approach to providing trauma care, which may include prehospital care, emergency care, surgical care, critical care, and rehabilitation services. An organised trauma system stipulates how patients move between providers, who provides care, where and when it is available, and how the system recovers costs. Additionally, organised trauma systems oversee education and training, research, injury prevention, and advocacy initiatives.

Organised trauma systems triage patients to the most appropriate level of care according to their needs. Trauma systems may be centralised within a specified geographical area (i.e. regionalised) to facilitate co‐ordinated transport. Designated trauma centres are specialist hospitals that provide a higher level of emergency care for patients with injuries. Trauma centres may be categorised depending on case volume, range of available services, staffing requirements, educational/research priorities, and injury prevention initiatives. For example, in the US, a level I trauma centre offers a comprehensive range of clinical services and provide leadership in teaching, research, and injury prevention and control. Level II centres initiate definitive care for all casualties but with less research capacity and fewer services than at level I. Level III centres provide assessment, stabilisation, and basic emergency operations. Level IV centres only provide essential trauma life support [14, 17, 18]. Additionally, an organised trauma system may incorporate elements of injury prevention, such as the promotion of helmet‐wearing among cyclists, appropriate seat belt use, education against alcohol and other illicit substance use among drivers and cyclists, and firearm safety [17, 19, 20].

An organised trauma system may improve the outcome of severely injured patients by identifying those who require co‐ordinated and multidisciplinary care and promptly transporting them to an appropriate level of care. This system may increase skills and expertise among providers through high patient volume and concentration of resources. Regionalised trauma systems and trauma centres provide leadership and organisation of trauma care to the designated population [21]. Trauma systems often support prevention programmes to reduce the burden of injury, reduce barriers to care, improve the quality of care provided, and lead to efficient resource utilisation. The trauma system may also strengthen the provider workforce by supporting lower levels of care (e.g. level III and IV), offering a co‐ordinated referral mechanism within the system, and improving community health. Trauma system centralisation may minimise delays by providing appropriate treatment at the scene and rapid transportation while continuing necessary treatment. Despite these benefits, the centralisation of healthcare systems risks de‐skilling lower facilities and delaying patient treatment if they are bypassing other hospitals on the way to a designated trauma centre [22]. The models available in high‐income countries are costly and may adversely affect healthcare provision for other services in low‐ and middle‐income countries [14].

Why it is important to do this review

Reports describing reductions in the burden of injury and improvements in injury care after creating a trauma system have typically employed weak study designs for assessing intervention effectiveness (e.g. uncontrolled before‐after studies) [9, 10]. The lack of support for trauma system development in low‐ and middle‐income countries may be, in part, due to the lack of evidence of their effectiveness in low‐resource settings. Furthermore, good evidence will assist replication of the success of trauma systems in the US to other regions [7, 8, 23].

Because of the enormous burden of trauma globally and the high cost of establishing trauma systems, it is essential to assess their effectiveness. Doing so is particularly important for low‐ and middle‐income countries, given the more significant burden of trauma and critical financial restraints. A quality evaluation of trauma system effectiveness is likely to inform health policy and resource allocation decisions and ultimately improve care for injured patients.

Objectives

To assess the benefits and harms of organised trauma systems and designated trauma centres compared with usual care in injured patients.

Methods

We followed the Methodological Expectations for Cochrane Intervention Reviews when conducting the review [24], and PRISMA 2020 for reporting our findings [25].

Criteria for considering studies for this review

Types of studies

We planned to include the following study designs.

• Randomised controlled trials and non‐randomised trials with at least two intervention sites and two control sites.

• Controlled before‐after studies with at least two intervention sites and two control sites.

• Interrupted time series studies that defined point of time when the intervention occurred and had a minimum of three points before and after the intervention.

There were no restrictions on publication status, language, or date of publication.

Types of participants

We included both injured patients and healthcare professionals providing care to those patients. We did not include patients with fragility fractures or those that were not admitted to hospital. We had planned to only include healthcare professionals providing care to, and patients who had, major trauma (i.e. Injury Severity Score of 15 or greater). However, we made a post hoc decision to broaden the Injury Severity Score inclusion threshold and include all patients regardless of Injury Severity Score as others may also be affected by the interventions.

Types of interventions

The interventions of interest were establishing an organised trauma system or a designated trauma centre (or both) compared to non‐trauma system care or no‐trauma centre care (i.e. current ordinary standard care for most low‐ and middle‐income countries). We defined an organised trauma system as a preplanned approach to providing the spectrum of trauma services. Such trauma services include, but are not limited to, injury prevention, timely transport from the scene of the injury to the trauma care facility, availability of trauma care providers and services, and rehabilitation. Trauma systems require a presence of trauma centres often categorised as levels I to IV according to their service capacities [17, 26]. Our two comparisons were:

• organised trauma system compared to usual care, and

• designated trauma centre compared to usual care.

Outcome measures

Critical outcomes

• Patient outcomes: mortality, morbidity, measures of recovery such as good recovery, and disability

• Adverse effects

Important outcomes

• Utilisation and access to trauma care services: volume of trauma patients, bed occupancy, length of hospital stay, appropriateness of trauma procedures, resource utilisation representing economic outcomes, patients' waiting time to access trauma services, injury‐appropriate service time, ambulance service call volume

• Quality of care provided: adherence to standards of trauma care with tangible patient benefit (e.g. trauma care audit filters [17, 27, 28, 29, 30])

• Equity: timely access to trauma care and differential effects of outcomes across advantaged and disadvantaged populations

• Knowledge about trauma care services: healthcare provider and consumer knowledge regarding injury prevention, healthcare provider knowledge/skills regarding standards of injury care, and performance in trauma moulage scenarios

Search methods for identification of studies

We conducted the search with the advice and assistance of the Cochrane Effective Practice and Organisation of Care (EPOC) Group (Supplementary material 1).

Electronic searches

The Cochrane EPOC Group Information Specialist developed the search strategies in consultation with the review authors. We searched the Cochrane Database of Systematic Reviews (CDSR) and the Database of Abstracts of Reviews of Effects (DARE) for primary studies included in related systematic reviews.

We searched the following databases on 16 December 2023.

• Cochrane Central Register of Controlled Trials (CENTRAL, 2023, Issue 12) including the Cochrane EPOC Group's Specialised Register

• MEDLINE (from 1946) In‐Process and other non‐indexed citations, OvidSP

• Embase (from 1974), OvidSP

• National Health Service Economic Evaluation Database (NHS EED)

We did not search Cumulative Index to Nursing and Allied Health Literature (CINAHL) following expert advice from the EPOC search specialist that CINAHL would not retrieve relevant results due to its focus.

We used two methodology search filters to limit retrieval to appropriate study designs: a modified version of the Cochrane Highly Sensitive Search Strategy (sensitivity and precision maximising version 2008 revision [31]) to identify randomised trials [32], and a Cochrane EPOC Group methodology filter to identify non‐randomised trial designs.

Searching other resources

Grey literature

We conducted a grey literature search up to 16 December 2023 to identify studies not indexed in the databases listed above and via the sources listed below.

• OpenGrey (http://www.opengrey.eu)

• Grey Literature Report (New York Academy of Medicine; http://www.greylit.org).

Trial registries

We searched the following trial registries up to 16 December 2023.

• World Health Organization International Clinical Trials Registry Platform (ICTRP) (http://www.who.int/ictrp/en)

• ClinicalTrials.gov, from the US National Institutes of Health (NIH) (http://clinicaltrials.gov)

We searched the African Online Journals website up to 1 February 2023.

We contacted authors of relevant studies and reviews to clarify the published reports and seek unpublished data; and researchers with relevant expertise in the review topic.

We checked the reference lists of all included studies. We screened individual journals and conference proceedings, and cross‐checked references of the included studies and relevant systematic reviews.

We have provided appendices for all strategies used, including a list of sources screened and relevant reviews/primary studies reviewed (Supplementary material 1).

Data collection and analysis

We downloaded all titles and abstracts retrieved from the search to a reference management database and used Covidence [33] to remove duplicates and streamline review production. We reanalysed selected studies according to EPOC recommendations (Supplementary material 5). In potential studies that had missing data, we attempted to obtain them from the primary authors by email.

Selection of studies

Two of three review authors (MM, TH, and DM) independently screened titles and abstracts for inclusion. They resolved any disagreements by discussion and another review author (TH). We coded all studies as either 'retrieve' (i.e. eligible or potentially eligible) or 'do not retrieve'. We retrieved the full‐text reports for the former group. Two review authors independently screened the full‐text articles and identified studies for inclusion. They resolved disagreements through discussion. We identified and recorded reasons for exclusions.

Data extraction and management

We used a standard data collection form based on a Cochrane protocol [34]. Two review authors (MM and TH) independently extracted data, and resolved disagreements by discussion. We extracted the following study characteristics.

• Methods: study design, study centres and location, study setting, withdrawals, date of the study, and follow‐up

• Participants: number, mean age, age range, gender, severity of the condition, diagnostic criteria, inclusion criteria, exclusion criteria, and other relevant characteristics

• Interventions: intervention components, comparison, and fidelity assessment

Further, we used the Template for Intervention Description and Replication (TIDieR) to describe the interventions [35].

• Brief name of the intervention

• Description of rationale, theory, or goal of essential elements of the intervention

• Description of physical or informational material used for the intervention

• Description of each of the procedures, activities, or processes used in the intervention, including enabling or supporting activities

• Description of providers (background, expertise, training)

• Modes of delivery of the intervention (face‐to‐face, telephone, Internet)

• Description of location(s) infrastructure or relevant features of where intervention occurred

• Description of frequency, intensity, and duration of the intervention

• If intervention was tailored or adapted, description of why, when, and how

• If intervention was modified, description of the changes.

• How well they planned their intervention (if they assessed fidelity or adherence, description of how, by whom, and whether strategies for maintaining fidelity were employed)

• How well the intervention was carried out, and the description of how the intervention adhered to the plan

• Outcomes: main and other outcomes specified and collected, and time points reported

For economic effects reported by studies focusing on resource utilisation only, we planned to use the Cochrane and Campbell Economic Methods Group guidance [36], which includes the following criteria; however, we did not identify any studies that reported economic outcomes.

• Is the chosen time horizon appropriate to include relevant costs and consequences?

• Is the actual perspective chosen appropriate?

• Are all important and relevant costs for each alternative identified?

• Are all costs measured appropriately in physical units?

• Are all important variables, whose values are uncertain, appropriately subjected to sensitivity analysis?

• Do the conclusions follow from the data reported?

• Does the study discuss the generalisability of the results to other settings and patient/client groups?

• Does the article indicate that there is no potential conflict of interest of study researcher(s) and funder(s)?

• Are ethical and distributional issues discussed appropriately?

• Other: trial funding, notable conflicts of interest of trial authors, and ethical approval.

Risk of bias assessment in included studies

Two review authors (MM and TH) independently assessed the risk of bias in each study by using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions [32] and guidance from the EPOC Group [37]. A third review author (BS) was available to arbitrate if necessary.

For randomised controlled trials, non‐randomised trials, and controlled before‐after studies, we planned to assess the risk of bias according to the following domains.

• Random sequence generation

• Allocation concealment

• Blinding of participants and personnel

• Blinding the outcome assessment

• Incomplete outcome data

• Selective outcome reporting

• Baseline outcome measurement

• Baseline characteristics

• Other bias

For interrupted time series studies, we assessed the risk of bias according to the following domains.

• If the intervention was independent of other changes

• If the shape of the intervention effect was prespecified

• If the intervention was unlikely to affect data collection

• If knowledge of the intervention was adequately prevented

• If incomplete outcome data were adequately addressed

• If the study was free from selective outcome reporting

• If the study was free from other risks of bias

We judged each potential source of bias as high, low, or unclear, and justified judgement in the risk of bias tables. We summarised our decision on the risk of bias across different studies for each of the domains listed. We planned to assess blinding separately for different key outcomes (e.g. objective outcomes such as mortality, and self‐reported outcomes such as knowledge). However, the studies only reported objective data. Where information on the risk of bias related to unpublished data or correspondence with a trial author, we noted this in the risk of bias table. When considering treatment effects, we took into account the risk of bias for the studies that contributed to that outcome.

Measures of treatment effect

We planned to estimate the effect of organised trauma systems and designated trauma centres using the following.

• Risk ratio (RRs), adjusting for baseline differences for dichotomous data, with the appropriate associated 95% confidence interval (CI).

• Mean difference (MD) when studies used the same scale or standardised mean difference (SMD) when studies used different scales for continuous data, with 95% CI.

We reported measurements of treatment effect using the same scale (i.e. quality of life, disability scales).

Measurement of treatment effects for randomised controlled trials, non‐randomised trials, and controlled before‐after studies

We planned to extract the intervention effect estimate reported for outcomes in the included studies, the P value, 95% CI, and the method used in their calculation. For dichotomous outcomes, we planned to use RRs and, for continuous outcomes, MDs or SMDs. However, we included no randomised controlled trials, non‐randomised trials, or controlled before‐after studies in this review.

Measurement of treatment effects for interrupted time series studies

For interrupted time series studies, we reported outcome results as changes in level and slope. When analysis and reporting were not appropriate, we re‐analysed the data using segmented regression and followed the Cochrane EPOC Group guidelines [38, 39].

Unit of analysis issues

We planned to perform analysis at the same level as the allocation of the intervention and controls to avoid unit of analysis errors in the studies with cluster designs. In the event of a unit of analysis error and insufficient information to allow re‐analysis, we planned to contact the study authors. If this was unsuccessful, we planned not to report the CI and P value and record a 'unit of analysis error'. However, this was not necessary as none of the included studies used cluster designs.

Where a trial had multiple arms, we planned to include only the relevant arms. If two comparisons needed to be entered into the same meta‐analysis, we planned to halve the control group to avoid double‐counting.

Dealing with missing data

We planned to note when data were missing at random and proceed with analysis. When data were not missing at random, we contacted the study authors for additional information. We used electronic methods to retrieve data from graphs [40]. Although we planned to undertake a sensitivity analysis to estimate the impact of missing data, this proved unnecessary because there were insufficient studies to undertake a meta‐analysis.

Reporting bias assessment

We planned to compare the studies plotted in a matrix to indicate unreported outcomes. We also planned to search protocols, abstracts, and trial registries to compare listed outcomes with those reported in published studies. However, we only compared the methods with the results sections of the included studies.

Synthesis methods

We planned to pool data from studies judged clinically homogeneous using Review Manager [41]. We planned to undertake meta‐analysis only if this would be meaningful (i.e. if the treatments, participants, and the underlying clinical question were similar enough for pooling to make sense). If we encountered skewed data, we planned to note that the data were skewed and consider the implication of this.

We followed EPOC recommendations for the analysis and reporting of the interrupted time series studies by presenting outcomes along two dimensions [42]: change in level and change in slope. Change in level is an immediate effect of the intervention measured by the difference between the fitted value for the first postintervention time point and the predicted outcome at the same point. Change in slope expresses the longer‐term effects of the intervention. When studies did not report interrupted time series analyses according to EPOC recommendations, we requested additional data from the study authors and performed appropriate re‐analyses.

We were unable to perform a meta‐analysis due to clinical heterogeneity across studies, including different scales for reporting outcomes that precluded data pooling. We followed expert advice on rating the certainty of the evidence [43].

We did not plan to conduct a full economic analysis given the anticipated scale of heterogeneity between studies. Instead, we planned to provide a narrative summary of economic results. However, this was not possible as the included studies did not report economic outcomes.

Investigation of heterogeneity and subgroup analysis

We planned to investigate heterogeneity by visual inspection of forest plots and the Chi² test and report results of meta‐analyses where there was no substantial heterogeneity. We had planned to perform subgroup analyses for the following prespecified subgroups if there was substantial heterogeneity and an adequate number of included studies (i.e. more than 10).

• High‐income country settings versus low‐ and middle‐income country settings

• Adults versus children

The rationale for the first subgroup analysis was that high‐income country and low‐ to middle‐income country settings differ in both human and physical resources. Such differences will almost certainly result in heterogeneity and require separate analyses. The rationale for the adults versus children subgroup analysis was that patients of different ages have particular needs and, therefore, may require different resources, which could lead to different outcomes. For both subgroup analyses, we planned to assess critical outcomes and had considered applying a test of interaction to evaluate statistically significant differences between subgroups. However, we could not perform these subgroup analyses because of the limited number of included studies.

Equity‐related assessment

Since a regional trauma system is fundamentally inclusive by being a public service that aims to triage and admit patients according to their needs, and at the same time strive to ensure the highest quality of care possible, implementing a trauma system can be viewed as a measure to ensure equitable services.

Sensitivity analysis

We planned to conduct sensitivity analyses to investigate the robustness of the treatment effect on patient outcomes using multiple imputation methods under the following circumstances.

• There were 'data not missing at random', and if efforts to obtain additional information from study authors were unsuccessful.

• When we performed re‐analysis (e.g. in a cluster randomised trial where the intracluster correlation coefficient was not considered initially) for checking the stability of our results.

These were not possible due to the limited number of included studies.

Certainty of the evidence assessment

We used methods and recommendations described in Chapter 12 of the Cochrane Handbook [32], the Cochrane EPOC Group worksheets [44], and GRADEpro GDT software [45].

We created two summary of findings tables for our two comparisons (organised trauma system compared to usual care, and designated trauma centre compared to usual care) with the following outcomes.

• Patient outcomes (mortality, survival rate)

• Adverse effects

• Utilisation and access to trauma care services

• Quality of care provided

• Knowledge about trauma service provision by healthcare providers or consumers.

We made a post hoc decision to include mortality and survival rate under patient outcomes because the protocol did not specify which 'patient outcomes' would be included, and most studies reported mortality.

Two review authors evaluated the certainty of evidence (high, moderate, low, and very low) using GRADE domains (risk of bias, inconsistency, indirectness, imprecision, publication bias, plausible confounding, and large effect) in relation to the outcomes [44, 46].

We rated the certainty of the evidence based on EPOC guidance [47] as described below.

• High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

• Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

• Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

We explained all decisions to downgrade or upgrade the certainty of the evidence using footnotes to aid readers' understanding of the review where needed. As it was not possible to perform a meta‐analysis, we presented results in a narrative summary.

Consumer involvement

Consumers were not involved in this review due to limited resources, although the review authors did use core outcome sets for the review's outcomes, which were developed with consumer involvement.

Results

Description of studies

The characteristics of included, excluded, and ongoing studies are shown in Table 3; Supplementary material 2; Supplementary material 3; and Supplementary material 4.

1. Overview of included studies and syntheses.

Study ID	Study designs	Country/location	Intervention, population, follow‐up, and comparison	Outcomes
Groven 2011	Interrupted time series	Oslo University Hospital, Ulleval (OUH‐U)	Intervention: designation of a level I trauma centre, serving > 50% of the Norwegian population, admitting approximately 1300 trauma patients annually. 40% of admitted participants had an Injury Severity Score ≥ 15. Study followed 7247 participants over 6 years (2002–2008), including 2582 participants during pre‐intervention and 4665 postintervention. Comparison: outcomes before and after the establishment of the designated trauma centre	Mortality
He 2016	Interrupted time series	Cuyahoga County, Ohio, USA	Intervention: development of a regionalised trauma system. Comprising 2 Level I trauma centres, 4 Level II trauma centres, and 7 non‐trauma hospitals, and implemented a co‐ordinated prehospital care system. Follow‐up of 20,357 participants aged > 15 years from 2006 to 2012, including 11,122 participants pre‐intervention and 9235 postintervention. Comparison: outcomes before and after the development of designated trauma system.	Mortality
Moran 2018	Interrupted time series	Hospitals under the Trauma and Research Network (TARN), England	Intervention: development of trauma system, reconfiguration of all hospitals and resources to form a comprehensive trauma network, and major trauma centres, prehospital systems in England. Follow‐up of 110,863 severely injured participants. Comparison: outcomes before and after the establishment of the designated trauma centre.	Survival rate
Rotondo 2009	Interrupted time series	Greenville, North Carolina, USA	Intervention: designation of a level I trauma centre housed under a university hospital, improved staffing, training, and laying down clinical protocols. Follow‐up of 7587 (pre‐intervention) and 11,057 (postintervention) trauma patients recorded in the hospital registry. Comparison: before designation of the trauma centre (1994–1999) and after intervention (2000–2005).	Mortality

Results of the search

In the search on 16 December 2023, we identified 18,934 studies. After excluding duplicates, we screened 10,935 studies and we excluded 10,819 irrelevant studies. We conducted a full‐text screen of 116 records, excluded 110 records with reasons, and listed two as ongoing studies. We included four studies in the review. See Figure 1.

1.

Included studies

Study design and setting

The review included four interrupted time series studies (157,111 participants). Two studies (131,220 participants) compared organised trauma systems to usual care (Table 4), and two studies (25,891 participants) compared designated trauma centres to usual care (Table 5). We contacted the authors of one study and obtained additional data (Moran 2018 [48]). All included studies were undertaken in high‐income countries, specifically the US (He 2016 [49]; Rotondo 2009 [50]), the UK (Moran 2018), and Norway (Groven 2011 [51]).

2. Trauma systems compared to usual care.

Reference	Outcome	Relative effects %	Change in slope (95% CI)	Change in the level (95% CI)
He 2016	Mortality	−30	−0.10 (−0.53 to 0.33)	−2.02 (−3.17 to 0.86)
Moran 2018	Survival rate	14	0.08 (0.01 to 0.15)	0.13 (−0.50 to 0.76)

CI: confidence interval.

3. Designated trauma centres compared to usual care.

Study	Subgroup	Relative effect (%)	Change in slope (95% CI)	Change in level (95% CI)
Groven 2011	All participants	−42	−0.51 (−2.26 to 1.23)	−3.60 (−11.26 to 4.07)
	ISS 15–24	−44	−0.96 (−5.69 to 3.77)	−8.80 (−29.61 to 12.00)
	ISS > 24	−55	−1.90 (−8.55 to 4.75)	−17.87 (−47.12 to 11.37)
Rotondo 2009	Adults	−67	−2.09 (−6.22 to 2.04)	−17.52 (−42.27 to 7.23)
	Children	−84	−1.12 (−3.04 to 0.80)	−18.56 (−30.11 to −7.01)

CI: confidence interval; ISS: Injury Severity Score.

Participants

The included studies reported data from 157,111 participants (range 7247 to 110,863) recruited from individual trauma centre registries (Groven 2011; He 2016; Rotondo 2009), or national trauma databases (Moran 2018). The studies reported on data from participants from between 1994 (Rotondo 2009) and 2017 (Moran 2018). The mean age of participants ranged between 33 and 59 years. None of the studies reported the proportion of children, although one study excluded children under 15 years (He 2016). Most participants were male (between 61% and 73%) across the three studies that reported this demographic characteristic (Groven 2011; Moran 2018; Rotondo 2009). Median Injury Severity Score ranged from 6 (He 2016) to 16 (Moran 2018).

One study excluded older adults (age greater than 65 years) with isolated femoral neck fractures (Moran 2018), one excluded people who arrived from outside the trauma network (He 2016), and one study excluded people who died on or within 30 minutes of arrival (Groven 2011). Rotondo 2009 did not fully explain its exclusion criteria.

Interventions

Studies evaluated interventions that fell naturally into two distinct categories: establishing organised trauma systems (He 2016; Moran 2018), and designated trauma centres (Groven 2011; Rotondo 2009).

Trauma system regionalisation involves forming a local network for managing, transporting, and caring for injured patients. The trauma system includes the designation of specific trauma hospitals and interventions (e.g. transfer and care protocols) to work together to deliver trauma care smoothly. In one study, authors reported data from the regionalisation of trauma services across England (Moran 2018). The second study described a regional trauma network formed by establishing a field triage protocol, trauma care protocols, and the creation of injury‐prevention programmes (He 2016).

Trauma centre designation involves reconfiguring individual hospitals to help them specialise in the care of injured patients. One such hospital elected a trauma director, instituted trauma policies and guidelines, developed a trauma‐team activation protocol, and established referral mechanisms (Groven 2011). In one study, the health agency added a considerable financial investment to the academic hospital to support the incomes of the core staff and institute clinical care guidelines (Rotondo 2009).

Outcomes

All four studies reported patient outcomes: three reported mortality (Groven 2011; He 2016; Rotondo 2009), and one reported survival rate (Moran 2018).

No studies reported adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma care services.

Funding sources

One study reported that they did not receive any funding (Moran 2018). The remaining three studies did not reveal their funding sources (Groven 2011; He 2016; Rotondo 2009).

Excluded studies

We excluded 110 studies with reasons. The main reason for exclusion was ineligible study design. Most studies were of interrupted time series design; however, they did not meet EPOC inclusion criteria (i.e. have a defined point of time when the intervention occurred and a minimum of three time points before and after intervention delivery) [39]. See Supplementary material 3.

Ongoing studies

Two potentially eligible studies are still recruiting and have been classified as ongoing studies. See Supplementary material 4.

Risk of bias in included studies

See Risk of bias assessment in included studies and the risk of bias summary in Figure 2.

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Incomplete outcome data are adequately addressed

All studies were at low risk of bias since the interrupted time series study designs ensure consistent data collection throughout the study period, hence there were no incomplete outcome data.

Selective outcome reporting

All studies were at low risk of bias because studies reported outcomes according to their published methods, and the prespecified outcomes were consistently reported before and after the intervention.

Knowledge of intervention is adequately prevented

All studies were at low risk of bias because participants were unlikely to have known about the intervention. While healthcare providers may have been aware of changes associated with the intervention, the interrupted time series study design evaluates the changes associated with the intervention in the entire healthcare system across the two comparison periods. This is unlikely to impose any risk of bias between the study groups.

Intervention is unlikely to affect data collection

All studies were at low risk of bias because data source and data collection methods were similar before and after the intervention.

Intervention is independent of other changes

All studies were at low risk of bias because interrupted time series design accounts for secular changes.

Shape of intervention effect is prespecified

All studies were at low risk of bias because the point of analysis was the point of intervention.

Other bias

All studies were at high risk of bias; quasi‐experimental study designs have a high risk of bias due to a lack of randomisation.

Synthesis of results

Organised trauma systems compared to usual care for improving outcomes in injured patients

See Table 1.

Patient outcomes

Mortality

It is very uncertain whether organised trauma systems reduce mortality compared to usual care (relative effect at first quarter following intervention delivery −30%; change in level −2.02, 95% CI −3.17 to −0.86; change in slope −0.10, 95% CI −0.53 to 0.33; 1 interrupted time series study, 20,357 participants, 6 years' follow‐up; very low‐certainty evidence; He 2016). We downgraded the certainty of the evidence three levels due to serious risk of bias, serious inconsistency, and serious imprecision, and upgraded certainty due to plausible confounding as trauma systems were likely to admit sicker patients than usual hospitals.

Survival rate

It is very uncertain whether organised trauma systems reduce survival rate compared to usual care (relative effect at first quarter following intervention delivery 14%; change in level 0.13, 95% CI −0.50 to 0.76; change in slope 0.08, 95% CI 0.01 to 0.1; 1 interrupted time series study, 110,863 participants; 9 years' follow‐up; very low‐certainty evidence; Moran 2018). We downgraded the certainty of the evidence three levels due to serious risk of bias, serious inconsistency, and serious imprecision, and upgraded certainty due to plausible confounding as trauma systems were likely to admit sicker patients than usual hospitals.

Other outcomes

No studies reported adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma service provision.

Designated trauma centre compared to usual care for improving outcomes in injured patients

See Table 2.

Patient outcomes

Mortality

Two interrupted time series studies (25,891 participants) assessed mortality (Groven 2011; Rotondo 2009). Data could not be combined because they used different scales for reporting outcomes.

It is very uncertain whether designated trauma centres reduce mortality compared to usual care. Groven 2011 (7247 participants) reported reduced mortality for all participants treated in a designated trauma centre (relative effect −42%; change in level −3.60, 95% CI −11.26 to 4.07; change in slope −0.51, 95% CI −2.26 to 1.23), participants with Injury Severity Score 15 to 24 (relative effect −44%; change in level −8.80, 95% CI −29.61 to 12.00; change in slope −0.96, 95% CI −5.69 to 3.77), and participants with Injury Severity Score greater than 24 (relative effect −55%; change in level −17.87, 95% CI −47.12 to 11.37; change in slope −1.90, 95% CI −8.55 to 4.75) (very low‐certainty evidence). Similarly, Rotondo 2009 (18,644 participants) reported reduced mortality for adults (relative effect −67%; change in level −17.52, 95% CI −42.27 to 7.23; change in slope −2.09, 95% CI −6.22 to 2.04) and children (relative effect −84%, change in level −18.56, 95% CI −30.11 to −7.01; change in slope −1.12, 95% CI −3.04 to 0.80) (very low‐certainty evidence). We downgraded the certainty of the evidence three levels due to serious risk of bias, serious inconsistency, and serious imprecision, and upgraded certainty due to plausible confounding as designated trauma centres were likely to admit sicker patients than usual hospitals.

Other outcomes

No studies reported adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma service provision.

Equity assessment

There were no specific data on this variable; however, the nature of participants dictates equity in participant inclusion.

Reporting biases

All four included studies reported outcomes according to their published methods. However, none of the studies registered or published their protocols in advance and so it is unknown whether they are affected by selective reporting of outcomes.

Discussion

This study found insufficient evidence to support the effectiveness of trauma systems or designated trauma centres compared with usual care. Despite screening over 18,000 studies (Figure 1), we included only four quasi‐experimental studies, all of which were from high‐income countries. Organised trauma systems and designated trauma centres have recently gained prominence as a potential 'gold standard' approach for managing injured patients in high‐income countries, owing to their capacity to consolidate and efficiently manage resources, and enhance skills development [17, 52]. Trauma systems, functioning as public services with autonomous financing systems akin to many other emergency services, may offer equitable solutions in healthcare provision. Various studies have suggested that trauma systems could help mitigate inequalities and provide cost‐effective management of patients in resource‐limited settings such as low‐ and middle‐income countries [1, 14, 15]; however, these claims need to be supported by high‐quality evidence.

Despite expert opinion and reports of effectiveness in some observational studies [26, 53, 54, 55], this review found no good evidence of their effectiveness. The four included studies were limited in terms of outcomes reporting, focusing primarily on patient outcomes such as mortality and survival rates. Important outcomes such as healthcare‐related adverse effects, utilisation and access to trauma care services, quality of care provided, equity, and knowledge of healthcare providers and consumers regarding trauma service provision were not reported.

Several studies assessing trauma systems in resource‐limited settings did not meet our eligibility criteria (see Supplementary material 3). Since the implementation of trauma systems may mitigate inequalities in injury management, the absence of robust studies in these settings represents a significant drawback. This trend is likely to persist unless there is a shift in the research designs addressing this subject in trauma care and emergency medicine.

These research results serve as a warning to policymakers, funders, and researchers regarding deficiencies in the current evidence base for trauma systems and centres, particularly in low‐ and middle‐income settings. Furthermore, the available research underscores the need for the inclusion of a wide range of outcomes that can describe the effectiveness of clinical interventions beyond mortality, encompassing measures of healthcare‐related adverse effects, utilisation and access to trauma care services, quality of care provided, equity, and knowledge about trauma care services among healthcare providers and consumers.

Summary of main results

We included four interrupted time series studies that evaluated organised trauma systems and designated trauma centres compared to usual care.

It is very uncertain whether organised trauma systems reduce mortality or improve survival compared to usual care (2 interrupted time series studies, 131,220 participants; follow‐up 9 to 11 years; very low‐uncertainty evidence; He 2016; Moran 2018). We did not identify any studies that reported adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma care services.

It is very uncertain whether designated trauma centres reduce mortality compared to usual care (2 interrupted time series studies, 25,891 participants; follow‐up 6 to 11 years; very low‐certainty evidence; Groven 2011; Rotondo 2009). We did not identify any studies that reported adverse effects, utilisation and access to trauma care services, quality of care provided, equity, or knowledge about trauma care services.

Limitations of the evidence included in the review

The certainty of the evidence for the effectiveness of the trauma system and designated trauma centres was very low. Our systematic and comprehensive search identified no randomised controlled trials and the best available evidence was limited to observational studies that have a high risk of bias. We initially downgraded studies due to serious risk of bias, serious imprecision, and serious inconsistency. Although all studies were upgraded by one level because of plausible confounding (e.g. patients admitted to trauma systems or trauma centres are sicker than their counterparts in the usual care hospitals), the certainty of the evidence remained very low. Other limitations of the body of evidence from plausible bias may relate to patients who are too sick or die in a non‐designated centre; insufficient data records; and changes in coding, transfer, and admission criteria.

Studies included in this review excluded people who arrived at hospitals dead or who died within 30 minutes of arrival. There was no effect due to withdrawals since all studies were retrospective (Supplementary material 2).

We could not pool data or conduct meta‐analyses because of the low number of included studies, and the different scales used to report outcomes between studies. It is important to note that for some of the results, the CIs were wide and covered a positive and negative effect of the intervention; therefore, interpreting results as an increase or decrease in change of level and slope could appear misleading. However, changes in level and slope account for seasonality, trends, and autocorrelation. In contrast, limited statistical power in interrupted time series design may relate to the failure to control the distribution and the number of data points before and after the intervention, and the use of control series. Caution is needed when interpreting these results.

Limitations of the review processes

We conducted a comprehensive literature search to identify eligible studies reporting on the effectiveness of organised trauma systems or designated trauma centres. We performed re‐analyses of all included studies according to EPOC guidance. We attempted to contact corresponding authors to request missing information or additional data for potentially eligible studies. However, obtaining additional data was unsuccessful for several reasons. First, many authors did not maintain databases for such a long time, or some had moved from their institutions or had retired. In some studies, there was no email contact information available, preventing us from reaching the authoring team. Therefore, we excluded 19 studies due to failure to obtain additional data (Supplementary material 3).

It is possible that the strict eligibility criteria recommended by Cochrane preferentially excluded studies from low‐ and middle‐income countries, and it is noteworthy that all four included studies were from high‐income countries. The absence of included studies from low‐ and middle‐income countries may limit the generalisability of this review to those settings.

Agreements and disagreements with other studies or reviews

We found six reviews reporting on the effects of trauma systems [26, 53, 54, 55, 56, 57]. Two were narrative reviews [26, 53], and four were systematic reviews [54, 55, 56, 57]. Of these, five reviews reported benefits of trauma systems in terms of mortality [26, 53, 54, 55, 56]. One review reported the impact of trauma care systems in low‐ and middle‐income countries and highlighted deficiencies in training, research, and funding [57]. Most reviews did not assess the risk of bias and certainty of the evidence, failing to convey that the weak study designs used do not provide a reliable indication of the likely intervention effect [26, 53, 54, 56, 57]. In contrast, one review employed a rigorous methodology and assessed risk of bias and the quality of evidence in primary studies, but included observational studies in their review [55]. Similar to the other five, this review did not report outcomes for assessing the effectiveness of interventions (such as adverse effects or harms, utilisation of services, access, equity, quality of care provided, etc.). Moreover, it found low‐ or very low‐quality evidence for the assessed outcomes and could not provide recommendations based on GRADE criteria.

The current review differs by including rigorous primary research and focusing on important outcomes that accurately measure the effectiveness of healthcare interventions, such as patient outcomes, adverse effects or harms, utilisation and access to trauma care services, quality of care provided, equity, and knowledge about trauma care services. It also limits the inclusion criteria to randomised controlled trials, non‐randomised trials, and quasi‐experimental observational studies, which are appropriate for assessing causal‐effect relationships. As a result, this review contributes to the existing body of evidence by highlighting the lack of appropriate research designs (such as non‐randomised trials and quasi‐experimental research), and important outcomes that can assess the effectiveness of trauma systems, which is an essential factor for ensuring a thorough evaluation, particularly in low‐ and middle‐income countries where trauma systems are most needed.

Authors' conclusions

Implications for practice

The available evidence is currently insufficient to quantify the implications and impact of organised trauma systems and designated trauma centres on clinical practice. This is primarily due to a lack of studies with high methodological rigour for assessing the effects of clinical interventions, as well as the absence of reporting on important outcomes for determining their effectiveness.

Implications for research

The findings of this review emphasise the need for better designed primary research for evaluating the effectiveness of trauma systems interventions. Although it may be difficult to subject these interventions to randomised controlled trials, future research could consider using quasi‐experimental designs such as interrupted time series and controlled before‐after studies. Quasi‐experimental designs provide greater rigour than other observational designs, and, therefore, if undertaken in low‐ and middle‐income settings, they will be valuable in providing much‐needed evidence for the trauma system interventions that we are addressing in this review. Future research could focus on collecting important outcomes such as the utilisation and access to trauma care services, quality of care provided, and both consumer and provider knowledge regarding trauma system service provision. The findings of this review expose fundamental gaps and methodological deficiencies in the available evidence base for interventions in trauma care.

Supporting Information

Supplementary materials are available with the online version of this article: 10.1002/14651858.CD012500.

Supplementary materials are published alongside the article and contain additional data and information that support or enhance the article. Supplementary materials may not be subject to the same editorial scrutiny as the content of the article and Cochrane has not copyedited, typeset or proofread these materials. The material in these sections has been supplied by the author(s) for publication under a Licence for Publication and the author(s) are solely responsible for the material. Cochrane accordingly gives no representations or warranties of any kind in relation to, and accepts no liability for any reliance on or use of, such material.

Supplementary material 1 Search strategies

Supplementary material 2 Characteristics of included studies

Supplementary material 3 Characteristics of excluded studies

Supplementary material 4 Characteristics of ongoing studies

Supplementary material 5 Data package

Supplementary material 6 GRADE evidence profiles

New

Additional information

Contributions of authors

MM: writing the original draft, methodology, writing review and editing.

BS: conceptualisation, writing the original draft, methodology.

TH: conceptualisation, methodology, writing review and editing.

JH: data curation, writing review and editing.

AR: conceptualisation, writing the original draft.

RG: conceptualisation, writing the original draft.

JP: conceptualisation, writing the original draft.

DOC: supervision, writing review and editing.

DM: methodology, writing review and editing.

Declarations of interest

MM: none.

BS: none.

TH: none.

JH: none.

AR: none.

RG: none.

JP: none.

DOC: none.

DM: none.

Sources of support

Internal sources

• None, Other

  No source of support

External sources

• The Australian Satellite of the Effective Practice and Organisation of Care (EPOC) Group received funding from the National Health and Medical Research Council., Australia

  This study was supported by Cochrane EPOC group.

Registration and protocol

Protocol (2017) doi.org/10.1002/14651858.CD012500

Data, code and other materials

As part of the published Cochrane Review, the following is made available for download for users of the Cochrane Library: full search strategies for each database (Supplementary material 1); full citations of each unique report for all studies included (Supplementary material 2), ongoing (Supplementary material 4), or excluded at the full‐text screen (Supplementary material 3), in the final review; study data, including study information, study arms, and study results or test data (Supplementary material 5); analysis data, including overall estimates and settings, subgroup estimates, and individual data rows (Supplementary material 5); and GRADE evidence profiles (Supplementary material 6). Appropriate permissions have been obtained for such use. Analyses and data management were conducted within Cochrane's authoring tool, RevMan, using the inbuilt computation methods. The following was used to generate analyses outside of RevMan: Stata 17: StataCorp. 2021. Stata Statistical Software: Release 18. College Station, TX: StataCorp LLC. Template data extraction forms are available from the authors on reasonable request.

History

Protocol first published: Issue 1, 2017