Ambulance diversion and its use as an ED overcrowding mitigation strategy: Does it work? A scoping review

Abstract

Objective

Emergency department (ED) overcrowding is a worldwide issue with significant negative consequences, including increased patient mortality. Ambulance diversion (AD) is sometimes used as an intervention to momentarily relieve overcrowded EDs, however, jury is still out about the negative consequences both for emergency medical services (EMS) who are required to divert to an alternative destination, and for patients whose care is delayed. Additionally, there is no operational guidance to best operationalize AD. The objective of this scoping review was to collate and organize the peer-reviewed published literature on the effects of both diversion and diversion aversion measures, on emergency medical services (EMS) and patient outcomes.

Method

A systematic, comprehensive search was conducted in various databases to identify relevant studies. Medline, Embase, CINAHL, Psychinfo, Cochrane and ClinicalTrials.gov databases were searched. Online ACEP and NAEMSP portals were also searched. Included studies discussed AD in the setting of ED overcrowding that reported either EMS or patient outcomes. The effects of interventions implemented to reduce AD were also reported. Two independent reviewers screened the articles and consensus was reached when disagreements arose.

Results

Out of 10,061 identified records, 95 papers meeting the inclusion criteria contributed to the results. 51 were observational, 16 simulation, 15 interventional, 10 descriptive, 2 systematic reviews and 1 mixed method. 12 articles reported negative EMS outcomes compared to only 2 neutral or positive EMS outcomes. 19 articles reported negative patient outcomes, whereas 9 reported neutral or positive outcomes. 34 articles reporting on intervention attempts to reduce diversion found overall positive results with diversion aversion. Only 7 articles studied the qualitative effects of diversion.

Conclusion

There is no conclusive evidence on the effects of AD on EMS and patient outcomes. 31 articles reported negative EMS or patient outcomes with 11 articles reporting neutral or positive outcomes. Measures to reduce or avoid diversion, however, showed overall positive trend in the results when diversion was averted. More research to ascertain accurate effects with standardised criteria for outcomes is required. Qualitative outcomes were also not well reported and further research should be conducted to determine the psychological impact on both staff and patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12245-025-00933-0.

Introduction

Emergency department (ED) overcrowding is the situation where the need for emergency services exceeds available resources for patient care in the emergency department, hospital, or both [1]. This is a global health threat that is increasingly encountered in many countries [2, 3]. Crowding has been shown to increase the risk of adverse outcomes and patient mortality as well as reduce the quality of care delivery in the form of increased transport and treatment delays [2, 3]. Impaired access to healthcare in the form of ambulance diversion and increased patient waiting times resulting in patient elopement have also been reported [3]. Potential financial impacts secondary to increased reattendances and hospitalisations have also resulted in higher healthcare costs [3]. For the ED staff, crowding has also been identified as a reason for reduced job satisfaction and staff reduction [4].

Several interventions to combat ED overcrowding have been proposed but with mixed results. These include measures applied at the ED level to increase resources and manpower or those targeted at increasing disposition alternatives. Hospital wide practices to improve work processes and flow efficiency as well as the use of overcrowding scores and simulation modelling have also been utilised for operational research [2, 3]. One such overcrowding measure commonly mentioned as a temporary solution is the practice of ambulance diversion (AD) or ambulance bypass. Once thought to be a logical solution to ED overcrowding, emerging literature has begun to question its benefits. Given the extensive literature available on AD, a deep dive into the effects of AD will be crucial to determine its feasibility of continuing such a practice.

AD is a widely implemented strategy in emergency healthcare, yet its definition remains ambiguous and inconsistently applied across settings. AD was first proposed by Lagoe and Jastremski in 1990 as a strategy to reduce ED overcrowding [5]. One of the few available definitions was provided by the National Association of Emergency Medical Services Physicians (NAEMSP) which defines diversion as a “situation in which an ambulance is forced to seek an alternate hospital destination other than that to which it would normally transport a patient, because the closest appropriate facility has declared that it is unable to accept patients as a result of a lack of normally available resouces” [6]. The American College of Emergency Physicians (ACEP) have also developed guidelines for AD when “hospital resources, including emergency services, may occasionally be overwhelmed and may not be able to provide optimal patient care” [7]. At its core, AD occurs when incoming ambulances are redirected from the nearest ED to other facilities due to a closure of the former. However, its practical application often extends beyond these scenarios. In some instances, AD may occur owing to patient or family requests for care at alternate facilities, or a hospital’s lack of ability to care for specific types of patients, further complicating the definition [8].

There were an estimated 501,000 instances of diversion within the USA in 2003 [9]. A more recent investigation from the National Hospital Ambulatory Medical Care Survey conducted in 2009 in the United States revealed that 33% of ED visits occurred in EDs that reported that they had gone on AD at some time during the previous year [10]. In Perth, total AD hours was also reported to increase 74% from 2001 to 2002 [11]. The rationale driving diversion is to facilitate load distribution to hospitals with greater capability to receive patients and hence more equipped to provide medical care to patients during crowding. The reverse can also be true when large EDs are overwhelmed, patients are diverted to underutilised EDs during such overcrowding periods.

Recent evidence, however, has raised concerns regarding the potential downstream effects on emergency medical services (EMS) resource availability and delayed provision of care to patients [7]. Pham et al. performed the most recent systematic review on AD in 2006, which revealed increased EMS transport times but no increase in patient mortality [12]. Nevertheless, the authors acknowledged the paucity of evidence on the impact of AD on EMS resource utilisation and patient-centred outcomes. Comprehensive knowledge of the impact of diversion both on the EMS system and in terms of patient effects is essential to guide policy decisions on the continued adoption of diversion.

As AD is increasingly adopted as an overcrowding mitigation measure, literature also reports interventions to avoid AD with concomitant variations in the protocols aimed at reducing the duration and consequences of prolonged diversion. Certain communities have reported success in the implementation of a complete ban. Knowledge of the results of these strategies will provide reference to other EDs to consider changes at a systemic level.

To our knowledge, there are no recent reports that cover all relevant aspects of AD. These studies focused on either EMS or patient outcomes. The psychological effects on the EMS crew have been the most neglected part. A consolidated approach to all the outcomes will provide a holistic understanding of the impact of diversion. Similarly, a compilation of measures adopted to prevent diversion will also provide an indirect understanding of how diversion avoidance can potentially impact both outcomes. This scoping review was therefore undertaken with the aim of collating the existing evidence in the context of ED overcrowding, with a specific focus on those impacts.

Methods

Selection criteria

Eligible articles: 1) considered AD in the setting of ED overcrowding and 2) reported either EMS or patient outcomes. Both prospective and retrospective studies including those with quantitative and qualitative designs, were eligible for inclusion. Excluded articles included opinion pieces, letters, editorials, or commentaries. Studies that examined diversion in the setting of disaster of mass casualties, studies that focused on only non-land transport or economic effects were excluded. Studies that specifically examined only rural EDs were also excluded to ensure the homogeneity of characteristics and issues of interest.

Search strategy

We performed a comprehensive search of the medical literature to identify all relevant studies in consultation with a librarian. The search terms were refined and performed on PubMed, EMBASE, CINAHL, PsychInfo, The Cochrane Library (See Supplementary, Additional File 1 Description of Search Strategy and Terms). Ongoing trials were also searched through clinicaltrials.gov website. A search of the abstracts via the NAEMSP and the ACEP online portals were also performed. Finally, a hand-search of references of all identified articles was also carried out. The search was carried out from inception to 31 August 2023. Non-English language papers were excluded.

Study selection

Two reviewers independently performed the initial eligibility screening. Full text articles from potentially relevant articles were then identified. Discordant decisions during the initial and full text screenings were resolved through consensus. Data extraction was performed via the same methodology.

Data collection process

We developed a data extraction form (See Supplementary, Additional File 2 Data Extraction Form) and refined it accordingly. Information recorded included study specific information: author, year of publication, year of data collection if available, study objective, study design, study setting, sample size if available, and year of publication. We represented the outcomes of the study in terms of EMS-related outcomes, patient related outcomes or outcomes based on interventions. Each article had 2 authors (MOJH, BJWL, MABMZ) independently extracting the data items. The 3rd author checked the extracted data and disagreements were then resolved by discussion.

Protocol

This scoping review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (See Supplementary, Additional File 3 PRISMA checklist). No formal protocol was registered on online platforms. No critical appraisal was conducted.

Results

The PRISMA flow chart in Fig. 1 provides a visual representation of the search process. A total of 10,061 articles were identified through the searches via PubMed, EMBASE, CINAHL, PsychInfo, The Cochrane Library and clinicaltrials.gov and non-database sources (American College of Emergency Physician, National Association of EMS Physicians portals). After removal of duplicates and full text eligibility screening, 83 articles were identified. 12 additional articles were selected via handsearching which resulted in a total of 95 articles included in this review.

Fig. 1.

PRISMA flow diagram

AD definition

The lack of a universal and standardised definition for AD results in significant variability in policy, practice, and research. Establishing a common definition is essential to ensure consistent application and to better evaluate its impact on patient care and system-wide efficiency. In reference to the few available definitions and factors available, in Fig. 2, we present a conceptual framework whereby crucial concepts of AD can be incorporated into a holistic definition for complete appreciation of the factors at play. The interpretation involves a common condition resulting in movement during a certain time point due to various push factors in a sequential fashion. A possible definition would therefore be a situation where an ambulance must bypass the nearest appropriate emergency care facility before arrival at scene due to overcrowding.

Fig. 2.

Organisation of the concepts used to define AD in the literature

Study characteristics

An overview of the characteristics of the articles included in the review is presented in Table 1 while a summary of geographic distribution is provided in Fig. 3. Of the 95 studies identified, most were observational. 16 simulation and 15 interventional studies were identified, with 1 study being a combination of observational and simulation. 2 review articles fulfilled the inclusion criteria and were included in this scoping review. Figure 3 shows that a large majority of the articles identified were from the United States, followed by Canada and Australia.

Table 1.

Characteristics of included studies

Author, Year	Study duration	Setting	Type of study	EDs | Patients or Trips	Objectives	Conclusion
Impact of Diversion
Lagoe and Jastremski, 1990 [5]	Mar– Aug 1989	Urban	Observational	4| -	Demonstrate effects of diversion plan between 4 urban hospitals to 4 neighbouring small city hospitals	Cross-county AD activation resulted in shorter diversion hours and boarding time
Neely et al., 1994 [13]	Sept 1991-Jan 1992	NA	Observational	14| 481	Examine impact of AD due to hospital constraints	AD resulted in increase transport times and distances
					Patient identification suitable for AD
Redelmeier et al., 1994 [14]	1 Jan 1986—31 Dec 1989	Urban	Observational	13| 153,167	Examines prevalence, demographic, risk factors and impact of AD	AD effective in reducing mortality rate but increases scene and transport time
Silka et al., 2001 [15]	Aug 1997—Oct 1997	Urban	Observational	-| 2534	Examine reasons and impact of advanced life support (ALS) diversion	ALS AD associated with increase in total prehospital and patient transfer timings
Schull et al., 2003 [16]	1 Jan 1998—31 Dec 1999	NA	Observational	-| 11,400	Determine impact of ED gridlock (simultaneous AD at multiple hospitals) on prehospital delays for chest pain patients	ED gridlock associated with increased transport and out-ofhospital interval
Begley et al., 2004 [17]	28 Jul 1999–26 Oct 2001	Urban	Observational	7| 18 888	Examine impact of AD on trauma mortality (in Houston EMS area)	Possible association between ED diversion and death rates, particularly among the most severe trauma patients transferred from lower-level hospitals
Cone, 2004 [18]	1 Feb to 31 May 1997	NA	Observational	20| 3609	Determine impact of AD on prehospital delays for chest pain patients	AD associated with delays in out-of- hospital ambulance transport for chest pain patients during gridlock
	1 Feb to 31 May 1999
Eckstein and Chan, 2004 [19]	NA	Urban	Observational	59| 21 240	Determine effect of ED crowding on paramedic ambulance availability	Increase in ambulance out-ofservice time and AD, secondary to ED crowding, increases waiting time for patient transfer to open ED gurney
Fatovich, 2005 [11]	2001—2002	MA	Observational	1| 51 885	Determine impact of AD on mortality rate	No significant difference in mortality rate between prediversion and diversion period
Burt et al., 2006 [9]	2003	Mixed	Observational	405| 5913	Estimate frequency and reasons for AD	Diversion is higher in medium/large EDs
					Estimate impact of AD on delayed ED care	Ambulance patients have a higher average patient care time than ED patients
Millin et al., 2006 [20]	NA	NA	Observational	NA	Examine impact of AD on EMS and trauma patients management	AD increases EMS transport times but not mortality rate
Pham et al., 2006 [12]	1 Jan 1965—13 April 2006	Urban, suburban, and rural	Review	NA	Systematic review to summarise impact of AD and diversion reduction policies	Stopping diversion decreases turnaround time by 60 min
						High diversion is associated with increase transport interval but decrease on-scene interval
						Diversion is not associated with mortality
						Periods of moderate and high diversion associated with increase in time to thrombolysis in patients with AMI
						Relative paucity of studies on the effects of AD
American College of Emergency Physicians ACEP), 2007 [1]	NA	Urban, rural	Observational	NA	Summary of status of access to emergency medical care and future concerns	AD studies found association with increase time to drug therapy for acute myocardial infarctions
Carter and Grierson, 2007 [21]	2001–2002	Urban	Observational	6| -	Determine impact of AD on EMS resources	Ambulance diversion had no negative effect on EMS resources
McCarthy et al., 2007 [22]	1 Jan– 31 Dec 2000	NA	Observational	15| 128 165	Determine likelihood of reroute to different facility during AD	The likelihood of reroute during AD decreases in time whenever hospital spent on alert
					Identify factors associated with diversion	Each increase in ED visits decreases in likelihood of reroute
Walker-Cilio et 2008 [23]	NA	NA	Descriptive	14 EMS providers	Identify perception of AD by EMS providers and impact on management	Barriers are placed when diversion/critical care diversion is enacted but does not affect the comfort of EMTs bringing trauma patients for specialized care
Crosse et al., 2009 [24]	2003—2006	Urban	Descriptive	NA	Examine indicators and factors contributing to ED crowding	Increased AD would lead to catastrophic delays in treatment, death rates, boarding and wait times
Handel, 2009 [25]	2009	Urban and rural	Observational	284| -	Measure implementation of crowding solutions and effects	Hospitals over capacity have the highest hours/month on diversion and higher percentage of LWBS
Shenoi et al., 2009 [26]	15 Aug 2002—31 Dec 2004	Urban	Observational	23| 63 780 pts	Determine statistical association between AD and paediatric mortality	Bivariate analysis revealed no association between AD and paediatric mortality
Yankovic et al., 2010 [88]	2 Jan 1999—31 Dec 2000	NA	Observational	58| -	Determine impact of AD on clinical outcomes of critically ill patients	AD associated with increased mortality and myocardial fatalities
Cooney et al., 2011 [8]	NA	NA	Descriptive	NA	Resource document for NAEMSP position statement on AD and ED offload time	Diversion was associated with ED crowding and increase in transport and treatment times
Handel et al., 2011 [27]	1 Jan– 31 Dec 2009	Urban	Observational	9| 43 000—101 000	Examine variations in ED crowding during AD	Marked variation in ED workload rates and AD occurrence during 24-h period
Shen and Hsia, 2011 [28]	Jan 2000—Nov 2006	Urban	Observational	149| 13 860	Examine impact of AD on mortality rates among patients with AMI	Longer AD led to higher mortality rate
Hing and Bhuiya, 2012 [10]	2003—2009	Urban, rural	Observational	-| 30 904	Describe trend in wait times in ED and variation with ED crowding	Wait time in the ED with boarding and diversion was longer
Hsia et al., 2013 [29]	2007	NA	Observational	202| 3 368 527	Determine impact of AD on bounce back admissions	AD not associated with increased bounce back admission
Sun et al., 2013 [30]	2007	Mixed	Observational	187| 995 379	Assess outcomes of ED crowding	ED crowding associated with higher odds of mortality and increased LOS
Batt, 2014 [31]	2007—2008	Urban	Mixed	-| 22,770	Determine impact of ED boarding on hospital revenue	ED boarding leads to potential losses in hospital revenue
					Determine factors to better manage non-ED demand to reduce boarding	ED boarding leads to unfilled patient need (measured by AD and LWBS)
Geiderman et al., 2015 [32]	NA	Urban, suburban, and rural	Descriptive	NA	Examine causes and impact of AD	AD associated with transport delay for critically ill patients
Shen and Hsia, 2015 [33]	2001—2011	Urban	Observational	-| 30,904	Impact of AD on access to technology, treatment and health outcomes for Medicare patients with AMI	AD associated with increased mortality
						High level diversion decreased access to cardiac technology
Hsia et al., 2017 [34]	2001—2011	Observational	-| 91,263	-| 91,263	Impact of ED crowding on blacks compared to whites	Diversion in black-serving EDs have higher diversion and mortality rates compared to non-black serving EDs
						Black patients have higher mortality rate compared to whites
Findings for EDOC/AD
Todd Taylor, 2000 [35]	NA	NA	Descriptive	NA	Summarise recommendations for best practices for EDOC and AD	Provides recommendations for best practices for EDOC and AD
Frank, 2001 [36]	NA	NA	Observational	NA	Summarise issues of ED crowding in US and steps taken	Provides summary of various intervention and strategies in managing EDOC
Lagoe, 2002 [37]	Jan 1996—Mar 2001	Urban	Observational	4| -	Identify impact of AD	AD increasingly occurring throughout the year due to staff and resource limitations
						AD can be employed to reduce numbers of incoming transports
Schneider et al., 2003 [38]	Mar 2001	NA	Observational	89| -	Assess degree of crowding in EDs	Demonstrates that EDs throughout the United States are severely crowded with 11% of EDs on AD
Warden, 2003 [39]	1 Jan 1996—31 Dec 1999	Urban	Descriptive	-| 81,313	Evaluate amount of AD and potential predictive factors	AD increased significantly over time and associated only with decrease in available hospital beds
Eckstein et al., 2005 [19]	NA	NA	Descriptive	NA	Overview of issues, causes and mitigating recommendations of EDOC	Impact of EDOC includes AD and significant offload delay
Fatovich et al., 2005 [40]	2001—2002	Urban	Observational	3| 259 580	Systematic evaluation of access block, EDOC and AD	Access block increases diversion hours and shows positive correlation with AD
Epstein and Tian, 2006 [41]	2003	Urban	Observational	-| 46 164	Develop and validate ED work score to quantify crowding and staff workload	An ED work score to predict AD successfully developed and internally validated
American College of Emergency Physicians (ACEP), 2009 [1]	1997—2007	Urban	Observational	NA	Explore available evidence of public health impact of EDOC and boarding	More diversion time in urban hospital
						AD results in longer transport times and increases mortality and morbidity
Queensland Health, 2012 [42]	Oct 2011—May 2012	Urban	Descriptive	14| -	Outline findings from Metropolitan ED Access Initiative	Data demonstrates increasing trend on bypass
						Increased offload was associated with ramping
Hsia et al., 2012 [43]	2007	Urban and rural	Observational	202| 7 148 713	Examine prevalence of AD in hospitals serving large minority populations	Hospitals serving high proportions of minority populations have higher risk of experiencing diversion
Kahn et al., 2014 [44]	Jan 2000—Jul 2008	Suburban	Observational	24| -	Describe characteristics associated with AD	Hospitals providing specialty services were more likely to have higher diversion rates
Kingswell et al., 2017 [45]	1983—2015	NA	Review	NA	Synthesise literature pertaining to ambulance ramping	Ramping is problematic and AD was associated with a significant increase in duration of ambulance ramping
Hsia et al., 2018 [46]	2005—2012	NA	Observational	208| -	Compares inpatient versus ED volume as driver of AD	Inpatient and ED volumes associated with high probability of ED going on diversion
Gross et al., 2023 [47]	NA	NA	Descriptive	NA	Synthesises literature pertaining to EDOC	EDOC is a complex problem and solutions require coordinated efforts across the health care delivery system
						Expanding ED space has not been shown to improve throughput metrics of ED LOS or AD
						Diversion strategies studied did not attempt to quantify reduction in ED recidivism rates
Impact of Measures to Manage AD
Kelen et al., 2001 [48]	5 Dec 2000—19 Mar 2001	Urban	Observational	1| 1589	Determine impact of ED managed acute care unit (ED-ACU) on EDOC and AD	Decreased diversion rate and LWBS
Scheulen et al., 2001 [49]	1996	Urban, suburban, and rural	Observational	23| -	Examine impact of AD policies on patient flow and geographic areas	ED diversion policy has some limited effect in preventing further patient volume in urban and suburban areas but has virtually no impact in rural areas
Lagoe et al., 2003 [50]	Jan 2001—June 2002	Urban	Observational	4| -	Examine effects of implementation of procedures to reduce AD	Reduced AD time in all hospitals and LOS
Schull et al., 2003 [51]	Jan– Dec 1999	NA	Observational	1| 37 999	Study relationship between physician, nursing and patient factors on AD	Admitted patients in the ED are important determinants of AD
						Nurse hours and most emergency physicians are not determinants of AD
						Reducing the volume of walk-in patients is unlikely to lessen the use of diversion
Geer and Smith, 2004 [52]	June 2001—June 2002	Urban	Descriptive	1| 19 000	Effects of implementation of measures to reduce AD	The admission centre was effective in reducing diversion hours and overall LOS
Sedlak SK and Roberts A, 2004 [53]	2002—2004	NA	Interventional	1 ED	Identify improvement processes to overcome EDOC	Reduced diversion hours and significantly decreased LWBS and LOS
Vilke et al., 2004 [54]	NA	Urban	Interventional	2| -	Determine effects on neighbouring facility if hospitals stayed off diversion	Reciprocating effects can be decreased with one institution’s commitment to avoid diversion, thus decreasing the need for diversion at a neighbouring facility
Vilke et al., 2004 [55]	Oct 2001– Sep 2003	Urban, suburban, and rural	Interventional	21| 253,766	Impact of community intervention to reduce AD	Decreased diversion hours and risk of patients being diverted
McConnell et al., 2005 [56]	7 Aug 2001—6 Aug 2003	Urban	Observational	-| 43 183 (before)/40 672 (after)	Determine impact of increasing adult ICU beds on ED LOS and AD	Expansion of ICU effective in decreasing diversion hours in and ICU LOS
Sprivulis et al., 2005 [57]	1 Jan 2003—31 Dec 2003	Urban	Observational	7| 263 155	Examine relationship between AD and low acuity patient attendances to EDs	Reducing low acuity patient is not effective in reducing AD
Sprivulis and Gerrard, 2005 [58]	30 Jun 2002—4 Jan 2004	NA	Observational	8| -	Determine impact of emergency management internet portal, with pre-emptive ambulance distribution guidelines, on AD	The ED System Viewer was reduced diversion hours but increased offload delays
Burt and McCaig, 2006 [59]	2003—2004	Urban and rural	Observational	4500| -	Present estimates of structure and process characteristics of EDs and treatment capacity	Diversion more likely in metropolitan hospitals than non-metropolitan areas
						Metropolitan areas had longer waiting time and treatment times
Patel et al., 2006 [60]	Jan 2001– Sep 2003	Urban and rural	Interventional	14| -	Determine impact of region wide programme implementation on AD	Reduced diversion hours significantly despite increased
						ED admissions, inpatient census and city population
Shah et al., 2006 [61]	Jul 2003– Aug 2003	Urban, suburban, and rural	Interventional	2| 2708	Determine impact of physician-directed ambulance destination control program on AD	Reduced overall EMS diversion hours and significantly reduced diversion hours in community teaching hospital
Yancer et al., 2006 [62]	2003—2005	NA	Observational	1| -	Evaluate effectiveness of hospital initiatives to decrease AD	Decreased diversion hours, number of patients boarding and LOS
Asamoah et al., 2007 [63]	Sept 1 2004—Feb 2006	Urban, suburban, and rural	Interventional	10| 73 295	Determine impact of novel diversion protocol on diversion hours and drop-off times	Novel diversion protocol reduced diversion hours significantly but was not able to reduce offload time and turnaround time
Han et al., 2007 [64]	Nov 20,024– Oct 2005	Urban	Observational	-| 40,978	Examine effects of ED expansion on AD	ED expansion was not significantly associated with AD
						ED expansion increased diversion hours but decreased LWBS
Holroyd et al., 2007 [65]	9 Dec 2005—9 Feb 2006	Urban	Interventional	-| 2831	Evaluate impact of triage liaison physician (TLP)	Reduced LWBS and LOS
Khaleghi et al., 2007 [66]	26 Nov– 16 Dec 2001	NA	Observational	5| 2111	Evaluate impact of minimising bypass on individual ED	Minimising bypass trial was successful in reducing AD hours
Howell et al., 2008 [67]	Nov 2006—Feb 2007	Urban	Interventional	1| 33,721	Determine impact of active bed management on AD hours and ED throughput times	Reduced diversion hrs and total throughput times
Borders et al., 2009 [68]	2003—2007	Urban and rural	Observational	31| 9.8–10.4 million	Provides overview of AD in US and summary of work on California ED Diversion Project	Small reduction in mortality rate
Han et al., 2010 [69]	11 May– 9 Sep 2005	Urban	Observational	10| 17,265	Determine impact of physician triage on ED LOS, LWBS and AD	Reduced diversion hours, treat and release turnaround time and boarding time but found increase in EMS transport time
Mcleod et al., 2010 [70]	Apr 2008– Sep 2009	Urban	Interventional	4| 102 000—107 000	Evalute impact of Regional Emergency Patient Access and Coordination (REPAC) system on AD	Decreased ED LOS and LWBS but increased boarding time
Castillo et al., 2011 [71]	1 Sep 2006–31 Aug 2008	NA	Interventional	11| -	Assess impact of California ED Diversion project to decrease AD	Reduced diversion time (strategy 3 lowest percentage of adverse patients help to prevent delay of medical treatment)
Friedman et al., 2011 [72]	18 Sep– 15 Oct 2006	Urban	Observational	9| -	Study effects of no diversion trial for 2 weeks	Reduced total diversion hours and decreasing trend in diversion hours across all regions
Patel & Vinson, 2012 [73]	Jan 2006– Dec 2009	Urban and rural	Interventional	17| -	Determine impact of tight diversion criteria on AD	Reduced response time and LOS for admitted patients Reduced total AD hours significantly
Watase et al., 2012 [74]	Sep 2007– Sep 2009	Urban	Observational	1| -	Examine effect of ED only full capacity protocol on AD	Discharge LOS, admission rate, LWBS rate, and full capacity for ICU and ward were significantly associated with AD
Burke et al., 2013 [75]	Jan 2008– Dec 2009	Urban	Observational	9| -	Explore impact of AD ban	Decreased turnaround time and decreased LOS of admitted and discharged patients
Rathlev et al., 2013 [76]	Jan 2008– Mar 2009	NA	Observational	7| 12,185	Compare impact before and after implementation of no diversion	Reduced LWBS and LOS in all hospitals
Howard et al., 2014 [77]	1 Jul 2009–31 Jul 2010	Urban	Interventional	6| -	Determine effects of no diversion trial	Enhanced relationships, decreased tension resulting from diversion patterns, and created a more patient centered approach to prehospital care
Sayah et al., 2014 [78]	Jan 2005– Dec 2011	Urban	Observational	1| 30 000	Assess impact of process improvement project	Reduced diversion hours to 0 and decreased LWBS, LOS and door to provider time
Willard et al., 2017 [79]	Feb 2013– Jun 2014	Urban	Interventional	1| 43,179	Examine effectiveness of full capacity protocol in reducing EDOC	Reduced LWBS and AD, while accommodating a significant increase in ED volume and increased hospital admission rates
Hanchate et al., 2020 [80]	2007—2012	Urban	Observational	-| 361 006	Examine impact of AD ban on patients transported to ED	Associated with decreased proportion transported to the ED
Bains et al., 2021 [81]	16 Apr 2020–2 Feb 2021	Urban	Observational	11| 56 684	Determine impact of Centralised Ambulance Destination Determination (CAD-D) program	Program did not decrease AD rate
Simulation Modelling
Litvak et al., 2001 [82]	NA	NA	Simulation	2| 8000	Determine effects of ED Divert Model on hospital demand and capacity	The most significant driver of ED diversions in the two hospitals studied is the lack of sufficient inpatient capacity
Hagtvedt et al., 2009 [83]	NA	NA	Simulation	NA	Identify method and impact (centralised vs decentralised) to reduce AD	Centralised form of routing is required to reduce AD
Ramirez et al., 2009 [84]	NA	NA	Simulation	-| 55,863	Analyse impact of diversion policies using discrete event simulation	Discrete-event stimulation model was able to reduce diversion hours
Nafarrate et al., 2010 [85]	NA	NA	Simulation	NA	Analyse impact of AD on ED performance using discrete event simulation	Used discrete event simulation model (bi-criteria analysis) to evaluate the trade-off between time spent on diversion and average waiting time for patients in the ED
Deo and Gurvich, 2011 [86]	NA	NA	Simulation	NA	Investigate impact of decentralised decision making using queueing network model on AD	Decentralised decision making explain the lack of pooling benefits
Ramirez-Nafarrate et al., 2011 [87]	NA	NA	Simulation	NA	Study effect of single factor AD policy (simulation optimization approach)	Single-Factor AD policies outperforms no diversion and produce better results than a simple AD policy
Ramirez-Nafarrate et al., 2012 [88]	NA	NA	Simulation	NA	Study effect of AD policies obtained by MDP formulation on average waiting time of patients	Significant improvement in average waiting time spent by patients in ED
Delgado et al., 2013 [89]	1966—2012	Urban, suburban, rural	Simulation (Systematic review)	NA	Systematic review of simulation model studies	Cooperative strategies were more effective in reducing EMS resources
						Triggering AD based on no. of pts boarding or no. of pts in waiting room offered the best balance between accessibility and wait times
Lee, 2014 [90]	NA	NA	Simulation	9| -	Investigate impact of hospital selection policies on response time	Proposed novel hospital selection policy significantly and robustly outperforms other policies
Kao et al., 2015 [91]	NA	NA	Simulation	6| 33 281	Evaluate impact of ambulance and patient diversion strategies on crowdedness and importance of regional coordination via computer simulation	The impact of diversion diverting ambulance transported patients is slightly more effective than low-acuity ambulance transported patients
Lin et al., 2015 [88]	NA	NA	Simulation	NA	Evaluate effectiveness of AD strategies on EDOC via computer simulation	The all-patients-will-be-diverted strategy results in lower average patient waiting time for service, lower adverse patients and lower
Nezamoddini and Khasawneh, 2016 [92]	NA	NA	Simulation	NA	Investigate effects of multiple transfer strategy using modelling	Modelling of hospital capacity allocation via transferring patients is an effective way to decrease the number of patients waiting
Pförringer et al., 2018 [93]	Feb 2013– Mar 2017	NA	Simulation	40| 536 399	Study effect of discrete agent based simulation to reduce AD	Effective solutions against crowding require common policies to limit closure status periods based on quantitative thresholds
Baek et al., 2020 [94]	2017	Urban	Simulation	10| -	Evaluate effectiveness of centralised AD model based on rolling horizon optimisation framework	Policy yields an efficient centralised AD management strategy
Li et al., 2021 [95]	NA	Urban and suburban	Simulation	NA	Develop ambulance destination policy to mitigate offload delay	Optimal policies can significantly reduce ambulance offload delay, time to bed for pts and out-of-service time for paramedics at the expense of increased ambulances travel distances
Hou et al., 2022 [96]	2014—2016	NA	Simulation	NA	Explore decongestion interventions using simulation model	Diverting a small number of ambulances seems to be more effective and efficient in congestion reduction compared to other approaches

Fig. 3.

Geographical distribution of studies

Distribution of articles

With reference to Fig. 4, majority of the articles emerged from the USA with the first article published on AD in 1990. Comparatively, the first relevant article from Asia was available only in 2015. A slight reduction in the number of articles after the first systematic review was published in 2006 was also noticeable [12].

Fig. 4.

Geo-spatial distribution of included studies in the scoping review addressing AD

Studies focusing on the effect of AD on EMS outcomes primarily address the issues of key time intervals. Tables 2 and 3 shows the impact of AD on EMS outcomes with 12 articles reporting negative EMS outcomes [13–16, 18–20, 22, 32, 42, 45, 97]. The most reported worsened timing was increased transport interval times. AD frequently results in longer transport intervals with 6 articles highlighting transport delays ranging from 1.7–7 minutes [13, 14, 16, 18, 20, 22]. This increased travel time is secondary to the need for transport to more distant hospitals. Multiple studies also reported overall increase in prehospital times which can create a cascading effect and result in delayed system response intervals. Prolonged offload times, or “ramping”, were identified as a significant issue in 3 articles with instances exceeding 30 min. This can further compromise on ambulance resources and contribute to reduced availability to respond to further calls.

Table 2.

Summary of results of EMS outcomes

Outcomes of Interest	Effects	Reported mechanism	Results
Negative EMS Outcomes (12 articles)
Prehospital/Out of hospital interval	Worsens timings	Special consideration top reason for diversion - Delay from 7.9 mins to 10.3 mins [15] Gridlock - Delay from 45.7 mins to 47.5 mins [16]	Total time delay of between 1.8-3 mins [15, 16]
		-	Increase prehospital time by 3.1% (44.8 mins to 46.2 mins) [18]
System Response Interval	Worsens timings	Female sex, older age & presence of ALS crew predicted longer intervals– delay system response interval from 9.7 to 10.8 mins (11.3%)	Delay by 1-1.1 mins [14, 18]
Scene Time	Worsens scene times	-	Delay by 9% (1 min, mean 17 mins, range 6-34 mins) [14]
Transport Interval	Worsens transport timings	Hospital diversion patients need 7 more mins (average 3 miles further) [13] Gridlock - delay from 15 mins to 17.4 mins [16] Time spent on red or yellow alert and higher annual ED volume results in higher likelihood of reroute– delay by 3.8 mins [22] Diverting past local hospital to trauma centre results - delayof 1.7-5 mins [20] Associated with transport delays [32]	Delay of 1.7-7 mins [13, 14, 16, 18, 20, 22]
Turnover time/Patient transfer interval	Worsens timings	Delay from 33.4 mins to 36.4 mins for diverted ambulances [15]	Increase by 2.4 mins [15]
Offload/Drop off Time	Worsens timings	Offload delay due to unavailable stretcher/ED gurney [19, 97] 21,240 incidents ambulances waiting longer than expected offload time of 15 mins [97] 200% increase in offload delays of > 30 mins [42] Increase ramping duration by 12 mins (longest duration of ramping 523mins) [45]	≥10% patients waited > 1 hr [19] Offload delay of 2 mins to >30mins [42, 45, 97]
Neutral/Positive EMS Outcomes (2 articles)
System Response Interval	No difference	-	No significant difference in response time between control and diversion period [21]
Scene Time	Improve timings	-	Decrease on scene times by 8.2% [18]
Transport interval	No difference	-	No significant difference in transport time between control and diversion period [21]
Turnover time/Patient transfer interval	No difference	-	No significant difference in turnaround time between control and diversion period [21]

Table 3.

Summary of results of patient outcomes

Outcomes of Interest	Effects	Reported mechanism	Results	Type of Studies
Negative Patient Outcomes (19 articles)
Left without being seen (LWBS)	Worsens LWBS rates	Highest median % of patients LWBS at EDs with overcapacity	1-3% of patients [25]	Observational
Wait Times	Worsens waiting time	Metropolitan EDs had higher diversion with greater waiting times and treatment times as compared to non-metropolitan areas [9, 59] 20% of patients in metropolitan EDs waited > 1 hr to see a physician compared to 1% in nonmetropolitan EDs [59]	12.8% of metropolitan EDs had average waiting times > 1 hr compared to <1% of nonmetropolitan EDs [9, 59] Worsens wait times [24]	Observational Observational
Boarding	Worsen boarding situation (prolonged boarding time and number of boarding patients)	- - Access block ED occupancy showing strong positive correlation with AD [11]	73% (57/78 EDs) boarding ≥2 inpatients with 22% of patients in ED awaiting inpatient beds [38] AD increased with no. of admitted patients boarded in the ED (6.2mins/pt) [51] Worsens boarding by 11-16% [11, 27] Worsensboarding [24, 31]	Observational Observational Observational Observational Observational, Simulation
Length of Stay (LOS)	Worsens LOS	Patients admitted on days with high ED crowding experienced longer hospital LOS [30] Worsen patients-LOS-hr on days with 10hrs or more diversion vs on days with 5-10 hrs of diversion patients-LOS-hr [27]	Worsens by 0.8% [30] 4.7 patients-LOS-hr vs 4.1 patients-LOS-hr [27]	Observational Observational
Mortality	Worsens mortality rates	Worsens mortality by 14.2% during gridlock [98] Worsens 1 year mortality up to 35% when AD occurs >12hrs [33] - - Association with death rates in Houston trauma hospitals Admission on days with >5 hrs of AD compared to 0hr [30] In areas with high diversion, blacks have 2.88% higher 99 day mortality and 3.09% higher 1 year mortality [34]	Increase by 9.8% - 35% [33, 98] Worsens mortality [99] Increase mortality among patients with AMI [28] Increase by 3.4-11% [17] Worsens odds of inpatient mortality by 6% [30] Worsens mortality rates among blacks compared to whites in areas with high diversion (2.88% higher 99 day mortality, 3.09% higher 1 year mortality) [34]	Observational Observational Observational Observational Observational Observational
Other relevant outcomes	Worsens access to cardiac technology Treatment Times No. of hospital beds Increases time to drugs Minority population	Worsens access to cardiac care intensive unit (−2.56%), access to catheterization lab (−2.67%), access to CABG (−2.3%) 50% of patients in non-metropolitan areas with lower diversion spent <90mins in treatment area compared to 20% of patients in metropolitan area [59] Decrease in number of hospital beds was associated with increased total AD - -	Worsens access [33] Prolonged treatment times [59] Decrease hospital bed availability results in increased AD [39] Association with increase time to drug therapy in AMI [100] Higher risk of AD in hospitals serving high minority populations [43]	Observational  Observational Observational Observational Interventional
Neutral/Positive Patient Outcomes (9 articles)
Left without being seen	No difference	-	No difference [44, 72]	Interventional Observational
Boarding	Decrease number of boarders	-	Decrease by 37% [5]	Observational
Admission	No difference	- -	No difference [66] No association with bounce-back admissions [29]	Observational Observational
Mortality	No difference	Bypass to trauma hospital does not add to mortality [20] Mortality rate lower, increasing AD not accompanied with increase transport deaths (presumably because critically ill patients not diverted)	No difference [20, 26, 40] Decrease mortality by 1.8% [14]	Observational Observational

In contrast, only 2 articles reported neutral or positive EMS outcomes arising from diversion [18, 21]. The studies identified instances where AD did not significantly impact system response, transport times, or patient turnover times. One study highlighted an improvement with decrease on scene times by 8.2% [18]. This reduction was possibly due to operational efficiencies, although the underlying reasons were not fully explored.

Negative patient outcomes were reported in 19 articles [9, 11, 17, 24, 25, 27, 28, 30, 31, 33, 34, 38, 39, 43, 51, 59, 98–100] identified with increased mortality being the most reported negative outcome. A total of 7 articles reported worsened mortality rates by up to 35% in diversions lasting for more than 12 hours [17, 28, 30, 33, 34, 98, 99]. Literature also highlights racial disparities with increased mortality among black patients in areas with high diversion due to delayed access to critical care services like catheterization labs and intensive care [34].

7 studies reported neutral outcomes [20, 26, 29, 40, 44, 66, 72] with 2 studies finding positive impact of AD [5, 14]. Despite diversion, some studies reported no significant difference in mortality rates or admission rates. Positive outcomes were noted in the form of decreased mortality and boarding situation [5, 14].

A significant number of articles were written on the effect of interventions implemented for diversion avoidance as shown in Fig. 5. The effects of mitigating measures for AD are categorised based on the input, throughput, and output model, each addressing different aspects of ED operations. Input focused strategies primarily aims to control the inflow of patients to the ED via various modes (walk-ins, brought in by ambulances, referred by primary care) (See Supplementary, Additional file 4 Table 4a Summary of effects of input measures). These include diversion bans and load redistribution measures, which shift ambulance traffic to less burdened facilities. Methods looking at diversion reduction or ban was the most common form of intervention attempted with total of 8 articles reporting various combination [45, 55, 60, 72, 73, 75, 76, 80]. The implementation of a complete diversion ban in various studies led to a significant reduction in diversion hours and improved overall patient outcomes, such as reducing the number of patients left without being seen. 17 articles focused on reducing inflow into ED which accounted for majority of interventions implemented [49, 54, 55, 57, 58, 60, 61, 63, 66, 70, 72, 73, 75–77, 80, 81].

Fig. 5.

Summary of articles on diversion mitigating measures classified based on impact of ED flow. Specific measures implemented to avoid diversion are classified based on the impact of measures on the input-throughput-output model of ED operations

Throughput measures target improvements within the ED to enhance patient onboarding speed and efficiency [47, 53, 64, 65, 69, 74] (See Supplementary, Additional file 5 Table 4b Summary of effects of throughput measures). This includes strategies such as deploying physicians at triage and expanding ED capacity. While some articles noted that physician led triage could reduce diversion hours, others found no significant changes in diversion outcomes. Nevertheless, interventions like expanding ED space or adding acute care units contributed to decreased diversion hours and enhanced patient flow within the ED [47, 64].

Output measures focus on increasing the ED’s ability to discharge patients or transfer them to other hospital units, such as intensive care units (ICUs) [48, 52, 67, 79] (See Supplementary, Additional file 6 Table 4c Summary of effects of output measures). Successful strategies, like implementing active bed management and full capacity protocols, significantly reduced diversion hours and improved patient length of stay (LOS) [67, 79]. However, some output focused interventions, such as those targeting hospital-wide practices, reported trade-offs like increased LOS in specific areas while still reducing diversion durations. 5 articles adopted a more holistic approach by looking at how to improve the gaps of both throughput and output concurrently [36, 50, 62, 68, 71, 78] (See Supplementary, Additional file 7 Table 4 d Summary of effects of mixed throughput and output measures). All managed to demonstrate success in reducing AD occurrences.

Gaps in literature

A notable gap identified in the literature is the lack of qualitative research that explores the experiences of EMS personnel, hospital staff and patients affected by diversion. Only 7 articles studied the qualitative impact of diversion [23, 45, 52, 62, 72, 76, 78].

As part of our data extraction, we initially sought to gather diversion triggers as a subtheme of our review. However, we found significant heterogeneity in the literature with many articles utilising subjective markers. Only broad themes emerged with the recurring issue of ED crowding being the top criterion compelling hospitals to adopt diversion [9, 10, 15, 53, 56, 57, 61, 65, 76, 86, 100].

Discussion

Conceptual framework for diversion

The lack of breakdown of crucial time points from dispatch of ambulances to arrival at hospitals was a challenge identified. Dissecting major time points at which AD can occur is vital for identifying operational inefficiencies and implementing targeted interventions. In Fig. 6, we illustrate 4 critical junctures where AD can occur: prior to arriving at the scene, prior to departure from the scene, en-route to the hospital and after reaching the hospital. Corresponding negative implications for EMS, patient, and EMS crew outcomes based on review results are also presented. Negative EMS outcomes was seen with increased transport times (between departure from scene to arrival at hospital) as well as with prolonged offload and turnover time (upon arrival at hospital). Negative patient outcomes occur in the form of increased patient anxiety during transport (departure from scene to arrival at hospital) and lack of access to time-sensitive cardiac intervention (upon arrival to hospital). The negative EMS crew outcomes are qualitative in nature and occur from the point of arrival at scene to arrival at hospital. Adopting this structured approach will provide clarity on how diversion impacts emergency care and guide the development of standardised, evidence-based policies to manage AD effectively.

Fig. 6.

Conceptual flow diagram of the known effects of AD on EMS, patient, and EMS crew outcomes at key time points

Distribution of studies

With reference to Fig. 4, the first systematic review was published in 2006 and comprehensively assessed the effects of AD on EMS transport times and patient outcomes [12]. While it found increased EMS times, there was no significant impact on patient mortality. This may have contributed to the perception that the critical issue surrounding diversion had been addressed, reducing the impetus for subsequent studies. Furthermore, researchers may have redirected their focus and priorities toward examining alternative strategies for managing overcrowding beyond diversion.

In our local setting, diversion is also a practice that has been adopted to provide temporary relief during overcrowding. The search conducted found no relevant articles emerging from our nation on the effects of AD. Geographical and interhospital proximity differences should be considered when considering their relevance to our small nation. Nevertheless, the results of the review help to shed some light on the potential outcomes and hence the need to be measured and cautious when adopting diversion.

EMS and patient outcomes

The discussion of our findings on AD reveals several critical insights into its impacts on both EMS and patient outcomes. A major limitation that needs to be mentioned is the lack of uniformity of measurement of patient related AD outcome measures, which makes reporting of patient related outcomes extremely challenging. In addition, the reported data are inferential association with diversion hours, either reporting on aggregate outcomes at the ED or hospital level which may falsely inflate the negative impacts. In the United States, for example, EMS or hospitals are not required to report on patients who were diverted [101]. A recommendation to consider would be to require EMS and hospitals to identify patients who are diverted and their related outcomes, rather than time periods of AD.

Most of the evidence reviewed indicates that AD is associated with adverse outcomes for EMS, particularly in terms of increased transport times and offload delays. Multiple studies have demonstrated that diversion often requires longer travel distances, leading to delays of 1.7–7 min during patient transport [13, 14, 16]. These delays can have a compounding effect, impacting the availability of EMS resources for subsequent calls​. Additionally, diversion contributes to extended offload times, further reducing the EMS capacity to respond to new emergencies, which could exacerbate regional response times and delay care delivery [19, 97]. While majority of studies reported adverse impacts of AD on EMS efficiency, a few instances showed neutral or slightly positive outcomes, likely influenced by specific operational contexts or efficiencies. The overall trend highlights the importance of carefully managing and limiting the use of diversion to avoid the systemic delays it can introduce.

The impact of AD on patient outcomes mirrors the concerning trends seen in EMS performance but demonstrates a more mixed picture. The review found that diversion can increase the likelihood of adverse patient outcomes, including higher mortality rates, particularly when diversions last for extended periods [33]. Several articles highlighted that when diversion exceeds 12 h, mortality rates can rise by up to 35%, with racial disparities evident in the increased mortality among Black patients due to delayed access to critical care services like catheterization labs [34]​. However, a comparable number of studies also reported neutral or positive outcomes, such as reduced mortality when critically ill patients are redirected to specialized trauma centers [20]. These findings underscore the complexity of the relationship between diversion and patient outcomes, suggesting that the context and management of diversion play a significant role in its impact.

AD mitigation strategies

The review also emphasized the effectiveness of strategies designed to reduce or eliminate AD. Diversion bans and coordinated regional efforts, such as those implemented in San Diego and Sacramento, resulted in substantial reductions in diversion hours and improvements in overall EMS and patient flow [54, 60]​. Our review revealed that diversion reduction and bans are not only possible but also result in beneficial outcomes. The expansion of hospital capacity, both in terms of ED space and ICU availability, was identified as a key strategy for improving patient throughput and reducing the need for diversion [56, 64]. In services or areas where the AD ban may be difficult to implement, having a clear trigger for diversion would ensure a more equitable distribution of resources and loads. Additionally, interventions such as deploying physicians at triage or implementing full-capacity protocols within hospitals have been shown to decrease the number of diversion hours, illustrating that internal hospital management can significantly impact the frequency and duration of diversion [65, 74]​. The use of technology in the form of simulation modelling or work score predictors has also been studied in literature [41, 93, 94]. Such simulation has been utilised to study effects of various diversion policies as well as attempt to identify triggers [82, 84, 85, 88, 89, 95]. Integrating more health technology solutions is needed to improve coordination and patient care delivery.

Future research

The paucity of qualitative data is a prominent feature from the review results. While some studies have collected data on patients’ satisfaction and opinions of EMS staff on diversion [23, 45, 52, 62, 72, 76, 78], the broader psychological impacts of diversion on EMS crews– such as stress and burnout– remain underexplored. Understanding these qualitative aspects could inform better support mechanisms for staff and improve patient care experiences during diversion periods.

Although the ACEP has provided overarching guidelines for EDs to activate diversion, no international objective criteria are adopted or available. This made both extraction of data and drawing conclusion from the sources on reasons for diversion extremely challenging. Further studies looking at standardised criteria or consensus to activate diversion will be helpful for policy makers to consider adopting.

The significant variability in study design and outcome measures complicates comparisons across studies. Future research should focus on addressing these gaps by conducting prospective studies that use standardized criteria for evaluating the effects of AD. It would be beneficial to develop and test frameworks that incorporate both quantitative outcomes, such as time intervals and patient mortality, and qualitative data, such as staff feedback and patient experiences. Additionally, exploring the long-term impacts of diversion reduction strategies and their integration with health technology, such as predictive models for patient flow, could help refine best practices in managing EDOC​.

Limitations

Our scoping review has several limitations. First, only English articles were included, which may exclude crucial articles in foreign languages.

Secondly, many studies are observational and retrospective in nature, which limits the ability to draw definitive conclusions about causality. Simulation modelling have also been employed on the basis existing data, but it is similarly difficult to determine the real world impact [83, 86, 87, 90–92, 96, 102]. This highlights the need for more standardised research methodologies and outcome metrics in future studies.

Thirdly, as the intent of our study was to perform a scoping review, no risk of bias assessment or meta-analysis was performed.

Fourthly, in our design of the search strategy, we acknowledge that it may not have been able to comprehensively include all qualitative studies available.

Fifthly, as mentioned earlier, many included studies lacked granular detail on the specific reasons (e.g. capacity vs specialty need) for individual diversion instances, limiting deeper analysis.

Finally, most of the articles were from the USA which has diverse EMS and healthcare landscapes even within regions. This may affect generalisation to developing countries or countries that do not share a similar land geography or healthcare practices. Nevertheless, the included studies focused mainly on urban prehospital settings and hence developed countries may face similar issues and challenges. Given the large number of articles screened, the review should also have been comprehensive enough to provide an accurate picture of diversion outcomes.

Conclusion

In conclusion, while AD can serve as a temporary solution to relieve ED overcrowding, there is no conclusive evidence on its effects. Nevertheless, its associated risks to both EMS efficiency and patient outcomes suggest that it should be employed with caution. Successful interventions to reduce or eliminate diversion have demonstrated positive outcomes, emphasizing the importance of coordinated regional efforts and hospital capacity expansion. Further assessment of the qualitative impact of AD on affected staff is also imperative to appreciate its psychological impact. A universally standardised AD definition should also be adopted to provide greater clarity to drive research and policy implementation. Moving forward, a multifaceted approach that includes policy development, standardization of research methodologies, and greater attention to the experiences of affected stakeholders will be crucial in optimizing the use of AD as an emergency management strategy.

Abbreviations

AD

Ambulance diversion

ED

Emergency department

EMS

Emergency medical services

LOS

Length of stay

LWBS

Left without being seen

Authors’ contributions

The authors confirm contribution to the paper as follows: Study conception and design: MOJH, DMRH, MEHO, FJS; Conduct of search: MOJH, FJS; Screening of records and full text articles: MOJH, BJWL, MABMZ; Design of relevant figures: BJWL, IJSY, LLLT; Draft manuscript preparation: MOJH; Review and editing: FJS. All authors revised the paper and approved the final version of the manuscript.

Funding

Authors received no external funding.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not Applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.