Scuba diving fatalities in Australia 2001 to 2013: Chain of events.

INTRODUCTION: We aimed to identify the possible chain of events leading to fatal scuba diving incidents in Australia from 2001-2013 to inform appropriate countermeasures. METHODS: The National Coronial Information System was searched to identify scuba diving-related deaths from 2001-2013, inclusive. Coronial findings, witness and police reports, medical histories and autopsies, toxicology and equipment reports were scrutinised. These were analysed for predisposing factors, triggers, disabling agents, disabling injuries and causes of death using a validated template. RESULTS: There were 126 known scuba diving fatalities and 189 predisposing factors were identified, the major being health conditions (59; 47%), organisational/training/experience/skills issues (46; 37%), planning shortcomings (29; 23%) and equipment inadequacies (24; 19%). The 138 suspected triggers included environmental (68; 54%), exertion (23; 18%) and gas supply problems (15; 12%) among others. The 121 identified disabling agents included medical-related (48; 38%), ascent-related (21; 17%), poor buoyancy control (18; 14%), gas supply (17; 13%), environmental (13; 10%) and equipment (4; 3%). The main disabling injuries were asphyxia (37%), cardiac (25%) and cerebral arterial gas embolism/pulmonary barotrauma (15%). CONCLUSIONS: Chronic medical conditions, predominantly cardiac-related, are a major contributor to diving incidents. Divers with such conditions and/or older divers should undergo thorough fitness-to-dive assessments. Appropriate local knowledge, planning and monitoring are important to minimise the potential for incidents triggered by adverse environmental conditions, most of which involve inexperienced divers. Chain of events analysis should increase understanding of diving incidents and has the potential to reduce morbidity and mortality in divers.

Workplace chemical and toxin exposures reported to a Poisons Information Centre: a diverse range causing variable morbidity.

OBJECTIVE: The aim of this study is to determine the period prevalence, nature and causes of workplace chemical and toxin exposures reported to the Victorian Poisons Information Centre (VPIC). PATIENTS AND METHODS: All cases classified as 'workplace: acute' when entered into the VPIC database (June 2005-December 2013) were analysed. Data were collected on patient sex, the nature of the chemical or toxin, route of exposure and season. RESULTS: Overall, 4928 cases were extracted. Exposures to men (71.5% of calls) differed from women (P<0.001), with most exposures relating to industry/trade substances (23.7%) and cleaners/bleaches/detergents (36.9%), respectively. Ocular (33.2%), inhalational (27.7%) and dermal (22.1%) exposures were most common. Exposures were most common in Spring and most seasonal variation was found for veterinary/animal, agricultural/plant and household categories (P<0.05). In all, 3445 (69.9%) cases had symptoms related to their exposure at the time of the call. However, the proportion of symptomatic cases within the major substance categories differed significantly (P<0.001). Chemicals associated with the most symptoms were cleaners/bleaches/detergents, industrial/trade substances and acids. CONCLUSION: Mild-moderately important workplace exposures are common. Significant variations exist between the sexes and seasons. Poisons Information Centres may play a role in ongoing surveillance of chemical and toxin exposures and a minimum exposure dataset is recommended.

Chain of events analysis for a scuba diving fatality.

INTRODUCTION: A scuba diving fatality usually involves a series of related events culminating in death. Several studies have utilised a chain of events-type analysis (CEA) to isolate and better understand the accident sequence in order to facilitate the creation of relevant countermeasures. The aim of this research was to further develop and better define a process for performing a CEA to reduce potential subjectivity and increase consistency between analysts. METHODOLOGY: To develop more comprehensive and better-defined criteria, existing criteria were modified and a template was created and tested using a CEA. Modifications comprised addition of a category for pre-disposing factors, expansion of criteria for the triggers and disabling agents present during the incident, and more specific inclusion criteria to better encompass a dataset of 56 fatalities. Four investigators (raters) used both the previous criteria and this template, in randomly assigned order, to examine a sample of 13 scuba diver deaths. Individual results were scored against the group consensus for the CEA. Raters' agreement consistency was compared using the Index of Concordance and intra-class correlation coefficients (ICC). RESULTS: The template is presented. The index of concordance between the raters increased from 62% (194/312) using the previous criteria to 82% (257/312) with use of this template indicating a substantially higher inter-rater agreement when allocating criteria. The agreement in scoring with and without template use was also quantified by ICC which were generally graded as low, illustrating a substantial change in consistency of scoring before and after template use. CONCLUSION: The template for a CEA for a scuba diving fatality improves consistency of interpretation between users and may improve comparability of diving fatality reports.

Diving with pre-existing medical conditions.

INTRODUCTION: This is the second report based on a survey of Divers Alert Network Asia-Pacific (DAN AP) members who dive with cardiovascular and respiratory conditions and diabetes. It examines the medical management of the divers' conditions, any diving modifications used to mitigate the risk and outcomes. METHODOLOGY: An online cross-sectional survey was sent to 833 divers who had declared a targeted medical condition when applying for DAN AP membership between July 2009 and August 2013. RESULTS: Two-hundred-and-sixty-eight respondents (32%) provided sufficient information on their conditions to be included in the analyses. These included ischaemic heart disease (31), arrhythmias (20), cardiac septal defects (31), other cardiac conditions (10), hypertension (127), diabetes (25), asthma (40) and pneumothorax (5). Forty-nine per cent had sought specialist diving medical advice about their condition and 23% reported modifying their diving practices to mitigate their risk. The cohort had completed 183,069 career dives, 57,822 of these since being diagnosed with their medical condition. There were 27 individuals who reported having decompression illness (25 of whom were subsequently diagnosed with a persistent foramen ovale), and two individuals who experienced an arrhythmia during diving. CONCLUSIONS: Some DAN AP members are diving with medical conditions which could potentially impact the safety of their diving. A minority modified their diving practices to mitigate the risk of their condition and approximately half sought specialist diving medical advice. The incidence of diving-related problems precipitated by known and managed pre-existing health conditions seems low but further studies of larger cohorts and incorporating fatality data would be necessary to confirm this. These results are limited by the 32% response rate and potential for bias towards selection of those most careful with their health.

Informed consent in emergency care research: An oxymoron?

Emergency care needs to be underpinned by the highest quality evidence. However, research involving critically ill patients in the emergency setting has unique ethical, logistical and regulatory issues. Informed consent is a well-established principle in conventional research. In this article, we discuss informed consent as it pertains to the difficulties of research in the emergency setting. Alternatives to informed consent are discussed. Human research ethics committees require a greater understanding of consent issues in emergency care research for Australia to remain competitive internationally.

The demographics and diving behaviour of DAN Asia-Pacific members with and without pre-existing medical conditions.

INTRODUCTION: This report examines Diver Alert Network Asia-Pacific (DAN AP) members with and without cardiac or respiratory conditions, diabetes or hypertension and compares their demographics, health and diving activities. METHODOLOGY: Two online cross-sectional surveys of DAN AP members were conducted. The first sought information from 833 divers who applied for membership between July 2009 and August 2013 and who had declared the targeted medical conditions. The second, conducted between December 2014 and April 2015, was sent to 9,927 current members with known email addresses. The groups were compared for age, gender, body mass index, fitness, smoking and diving qualifications, history, currency and practices. RESULTS: Of 343 (41%) respondents to the first survey, 267 (32%) provided sufficient information for inclusion. Of 1,786 (18%) respondents to the second survey, 1,437 (15%) had no targeted medical condition and were included in the analysis. Those with medical conditions were on average 4.7 years older (P 〈 0.001); more overweight or obese (68% versus 57%, P = 0.001); took more medications (57% vs. 29%, P 〈 0.001), smoked less (4% vs. 7%, P = 0.02) and did less repetitive diving (median 75 vs 90, P 〈 0.001). Other diving demographics were similar. CONCLUSIONS: A substantial number of people are diving with medical conditions and there is a need to better understand the associated risks. Divers need to be well-educated about the potential impact such conditions may have on diving safety and should monitor their health status, especially as they age.

Estimating the risk of a scuba diving fatality in Australia.

INTRODUCTION: There are few data available on which to estimate the risk of death for Australian divers. This report estimates the risk of a scuba diving fatality for Australian residents, international tourists diving in Queensland, and clients of a large Victorian dive operator. METHODOLOGY: Numerators for the estimates were obtained from the Divers Alert Network Asia-Pacific dive fatality database. Denominators were derived from three sources: Participation in Exercise, Recreation and Sport Surveys, 2001-2010 (Australian resident diving activity data); Tourism Research Australia surveys of international visitors to Queensland 2006-2014 and a dive operator in Victoria 2007-2014. Annual fatality rates (AFR) and 95% confidence intervals (95% CI) were calculated using an exact binomial test. RESULTS: Estimated AFRs were: 0.48 (0.37-0.59) deaths per 100,000 dives, or 8.73 (6.85-10.96) deaths per 100,000 divers for Australian residents; 0.12 (0.05-0.25) deaths per 100,000 dives, or 0.46 (0.20-0.91) deaths per 100,000 divers for international visitors to Queensland; and 1.64 (0.20-5.93) deaths per 100,000 dives for the dive operator in Victoria. On a per diver basis, Australian residents are estimated to be almost twenty times more likely to die whilst scuba diving than are international visitors to Queensland, or to lower than fourfold on a per dive basis. On a per dive basis, divers in Victoria are fourteen times more likely to die than are Queensland international tourists. CONCLUSIONS: Although some of the estimates are based on potentially unreliable denominator data extrapolated from surveys, the diving fatality rates in Australia appear to vary by State, being considerably lower in Queensland than in Victoria. These estimates are similar to or lower than comparable overseas estimates, although reliability of all such measurements varies with study size and accuracy of the data available.

Management of acute agitation in Hong Kong and comparisons with Australasia.

BACKGROUND: Little is known about the use of sedation drugs for the management of acute agitation in Hong Kong's Accident and Emergency Departments (AEDs) and how it compares with Australasian practice. OBJECTIVE: The aim of this study was to determine drug preferences, clinicians' perceived confidence in management, barriers/gaps in training and perceived usefulness of existing clinical practice guidelines (CPGs) in Hong Kong. METHOD: A validated questionnaire was used, with case vignettes typical of patients presenting to AEDs with acute agitation. The questionnaire was distributed by hand to all trainees and fellows of the Hong Kong College of Emergency Medicine (HKCEM). Two reminders were sent. RESULTS: Of 483 HKCEM members, 280 (58.0% [95% CI 53.5-62.3]) responded. For monotherapy, 46.8% (95% CI 41.0-52.6) of respondents chose haloperidol to manage the undifferentiated patient, followed by midazolam (33.9%, 95% CI 28.6-39.7) and diazepam (13.9%, 95% CI 10.4-18.5). Most respondents (83.6%, 95% CI 78.8-87.5) would not administer combination therapy. Respondents were confident in managing agitation overall. The lack of local/institutional CPGs (55.7%, 95% CI 49.9-61.4) was perceived as an important barrier. Institutional guidelines were considered the most useful CPGs (66.4%, 95% CI 60.7-71.7). Most respondents (72.9%, 95% CI 67.4-77.7) perceived a HKCEM endorsed CPG would be useful. CONCLUSION: Haloperidol and benzodiazepines are frequently used as monotherapy for the management of acute agitation in Hong Kong's AEDs. Management in Hong Kong differs from Australasian practice in that combination therapy is less common and clinicians' choice of sedation drugs are less variable overall. Results suggest that future work on CPG development and training regarding the safe use of combination therapy would be well received.

Off-label and unlicenced medicine administration to paediatric emergency department patients.

OBJECTIVE: To determine the prevalence and nature of off-label and unlicenced (off-label/unlicenced) medicine administration to paediatric ED patients. METHODS: We undertook a retrospective, observational study in six EDs (July 2011 to June 2012, inclusive). Patients, aged 0-17 years, who were administered a medicine in the ED were included. At each site, 50 eligible patients were randomly selected each month of the study period. An explicit review of each patient's records was undertaken. Medicines were classified as on or off-label/unlicenced according to categories of use approved by the Therapeutic Goods Administration. RESULTS: There were 3343 patients enrolled (56.5% men, mean ± SD age 6.7 ± 5.4 years). Of the 6786 medicine doses administered, 2072 (30.5%, 95% CI 29.4-31.7%) were off-label/unlicenced. The off-label/unlicenced doses were administered to 1213 (36.3%, 95% CI 34.7-37.9%) patients. Patients administered an off-label/unlicenced medicine were younger than those who were not (P < 0.01). Salbutamol, ondansetron, ipratropium, fentanyl and oxycodone were the medicines most commonly administered off-label. In 910 (44.0%) cases, the dose/frequency was not approved; in 592 (28.6%), there was an unapproved indication for treatment; in 158 (7.6%), the medicine was administered via an unapproved route; in 154 (7.4%) the medicine was not approved for the weight or age; and in 74 (3.5%) an unlicenced product was administered. The remaining cases had combinations of reasons. CONCLUSIONS: Off-label/unlicenced medicine administration is common. A registry of commonly used off-label medicines is recommended in which the safety and efficacy of their off-label use have been demonstrated by published evidence.

Measuring the effectiveness of a revised clinical practice guideline for the pre-hospital management of supraventricular tachycardia.

OBJECTIVE: To evaluate the effect of changes to the pre-hospital management of patients with supraventricular tachycardia (SVT) following intervention with a revised Clinical Practice Guideline (CPG). The major CPG revisions were removal of verapamil, addition of adenosine and an emphasis on Valsalva manoeuvre. METHOD: We undertook a retrospective case study using data collected by paramedics. All adult patients attended by paramedics from the periods 14 February 2012 to 14 September 2012 (old CPG) and 14 February 2013 to 14 September 2013 (revised CPG) were included. Patients were excluded if SVT was not recorded during initial assessment on a hardcopy ECG. Management guided by the old and revised CPGs was compared: reversion effectiveness, elements of therapy associated with reversion effectiveness and adverse events. Logistic regression determined patient factors significantly associated with reversion. RESULTS: Patients were predominantly women, aged approximately 57 years old and most lived in the Victorian metropolitan region. Vagal manoeuvre use and effectiveness decreased in the post-intervention group. Fewer patients in the post-intervention group (141/420, 33.6%) remained in SVT on arrival at hospital compared with the pre-intervention group (205/403, 50.8%). Initial heart rate >170/min and longer scene time were 2.6 and 1.05 times more likely to result in reversion, respectively. CONCLUSION: The revised CPG improved pre-hospital SVT reversion success. This expansion of practice has not demonstrated improvements to utilisation or effectiveness of the Valsalva manoeuvre. Adenosine is effective and safe for pre-hospital use.

Prehospital management of supraventricular tachycardia in Victoria, Australia: epidemiology and effectiveness of therapies.

OBJECTIVE: The present study examined patient demographics, characteristics and the effectiveness of current prehospital supraventricular tachycardia (SVT) management by ambulance paramedics in Victoria, Australia. METHOD: We conducted a retrospective study of all Victorian patients attended by paramedics between 14 February 2012 and 13 February 2013, where SVT was the initial or final diagnosis. Patients were excluded if SVT was not captured on ECG, incomplete data were recorded, or SVT began after initial assessment. Data were extracted from the VACIS(®) clinical data warehouse. Accuracy of paramedic SVT diagnosis was examined. RESULTS: Nine hundred and thirty-three patients were enrolled, including 882 (94.5%) adults and 564 (60.5%) women. Mean adult and paediatric (<18 years) patient ages were 57.5 (SD 18.1) and 10.0 (SD 4.5) years, respectively. Median ambulance response time was 11.0 (IQR 8.0) min. Paramedics correctly identified SVT in 119/123 (96.7%, 95% CI: 91.5, 99.0) of adult ECG strips examined. There were 273/882 (31.0%) patients who spontaneously reverted while in paramedic care. Valsalva manoeuvre was undertaken by 212/882 (24.0%) patients and reverted the SVT in 99/358 (27.7%) attempts. Verapamil was administered to 38/882 (4.3%) patients and reverted 33 (86.8%). Aramine was administered to 43/882 (4.9%) patients and 35 reverted following administration (81.4%). Synchronised cardioversion (70 J) reverted four patients at first attempt. Ultimately, 438 (49.7%) patients remained in SVT on arrival at hospital. CONCLUSION: Patient characteristics associated with SVT are more likely to be middle-aged women with a history of hypertension and hypercholesterolaemia. Therapies were underutilised leading to reduced clinical guideline effectiveness. Where therapies were instigated, reversion rates are greater than previously reported.

Clinical research priorities in emergency medicine: results of a consensus meeting and development of a weighting method for assessment of clinical research priorities.

INTRODUCTION: There is limited evidence regarding clinical research priorities in emergency medicine outside of some special interest groups. The ACEM Clinical Trials Group undertook a consensus meeting with the aim of developing a reproducible weighting matrix for assessing clinical research priorities. METHODS: A session at the ACEM annual scientific meeting was dedicated to this meeting. Results from a survey of the ACEM researcher database were presented, along with a proposed weighting matrix. After discussion and adjustment, consensus was achieved on the matrix. RESULTS: It was agreed that the following criteria be used in the matrix: research category and sub-category priority ranking from the ACEM researcher database survey, frequency of presentation of potentially eligible participants, the level of pre-existing evidence regarding the proposed research question and the likely clinical impact of the research. Each criterion was given a weighting, with clinical impact given the heaviest weighting. The weighting matrix was subsequently applied to the list of research questions that resulted from the researcher database survey and a list of research priorities determined. CONCLUSION: The weighting matrix allows reproducible comparison of research questions. The resultant list of research priorities will act as a guide for the ACEM Clinical Trials Group in determining future projects.

Introduction of an N-acetylcysteine weight-based dosing chart reduces prescription errors in the treatment of paracetamol poisoning.

OBJECTIVE: Under- or overdosing of N-acetylcysteine (NAC), when used to treat paracetamol toxicity, is associated with significant morbidity and mortality. This study evaluated the effect of a weight-based dosing chart (WBDC) introduced to decrease NAC prescription errors. METHODS: We undertook a pre- and post-intervention trial in a single ED. The intervention (the NAC WBDC) was introduced in January 2011 and publicised by posters and presentations at medical handovers and education sessions. ED staff were not aware that use of the WBDC was to be evaluated. Data were collected using a retrospective explicit medical record review by a single investigator. The study end-point was the proportion of NAC prescriptions with errors. RESULTS: The 81 and 42 patients enrolled in the pre- and post-intervention periods, respectively, did not differ in age, sex or weight (P > 0.05). Post-intervention, there were significant reductions in prescription errors of fluid type/volume (50.6% vs 4.8%, P < 0.001), NAC dosage (13.6% vs 0.0%, P = 0.01) and infusion rate (11.1% vs 0.0%, P = 0.03). The proportion of prescriptions with any errors also decreased (56.8% vs 14.3%, P < 0.001). However, there were no improvements in the documentation of patient weight (65.4% vs 64.3%, respectively, P = 0.90) or the proportion of incomplete prescriptions (4.9% vs 11.9%, P = 0.16). CONCLUSION: The introduction of a WBDC did not produce a clinically significant reduction in major NAC prescription error rates (as pre-defined in this study); however, there was a clear trend towards a reduction. The WBDC significantly reduced total and minor NAC prescription error rates.

Review article: A primer for clinical researchers in the emergency department: Part III: How to write a scientific paper.

In this series we address key topics for clinicians who conduct research as part of their work in the ED. Analysis of research data does not represent the completion of a project as the findings need to be communicated to clinicians and other researchers in the field. In this section, we describe how to write up clinical research data for publication in a peer-reviewed journal. We also describe the editorial and peer-review process.

National Coroners Information System: a valuable source of lessons for emergency medicine.

OBJECTIVE: To interrogate the National Coroners Information System (NCIS) to determine the recurrent themes among coroners' recommendations that aimed to increase the safety of ED care. METHODS: This was a retrospective analysis of NCIS closed cases, from Queensland, New South Wales, Tasmania, Victoria, Australian Capital Territory, South Australia and North Territory, entered since its inception in 2000. The keyword 'emergency department' returned 1645 cases, of which 180 were found to be relevant. The primary outcomes were the number and nature of cases where recommendations for improvements in ED care had been made and the recurrent themes of these recommendations that could inform education initiatives. RESULTS: Of the 180 cases, 108 (60.0%) were of deceased men and subject age ranged from 2 days to 91 years. The commonest causes of death were trauma (26.7%), infection (24.4%), cardiac events (15.0%) and poisoning (8.9%). No coronial recommendations were required in 19 cases. For the remainder, recommendation themes related to issues of risk management/medico-legal, diagnostic/therapeutic error, education, documentation/communication and re-presentation. The themes associated with the different doctor designations (consultant, registrar, resident/intern) were similar, although registrars and residents/interns tended towards more diagnostic/therapeutic errors. The themes associated with hospital type (referral, urban, regional/rural) were also similar. Although theme analysis is important, some individual cases were particularly instructive. CONCLUSION: The NCIS data theme analysis identifies important high-risk patients and presenting complaints. These should be incorporated into emergency physician training. EDs should review the coronial recommendations to ensure that, where possible, they have been adopted.

Simple clinical targets associated with a high level of patient satisfaction with their pain management.

INTRODUCTION: We aimed to determine factors that are significantly associated with a high level of patient satisfaction with their pain management. METHODS: We undertook an observational study in a large metropolitan ED. Adult patients with a triage pain score of ≥4 (0-10 scale) were enrolled. Data collected included demographics, presenting complaint, pain scores every 30 min, whether nurse-initiated analgesia was administered, the nature of the pain relief administered, time to first dose of analgesia, elements of staff communication and whether 'adequate analgesia' was provided (defined as a decrease in pain score to <4 and a decrease from the triage pain score of ≥2). The primary end-point, determined at follow up within 48 h, was the level of satisfaction with pain management (6-point scale: very unsatisfied-very satisfied). RESULTS: One hundred and sixty-seven (82.7%) of 202 enrolled patients were followed up - mean (SD) age 46.4 (18.3) years, 75 (44.9%) men. Eighty-one (48.5%) patients were very satisfied with their pain management. Only two clinical variables were significantly associated with a high level of satisfaction: receipt of 'adequate analgesia' (as defined) and specific communication regarding pain management. Forty-four (58.7%) versus 37 (40.2%) patients who did/did not receive 'adequate analgesia', respectively, were very satisfied (difference 18.5%, 95% CI 2.3-34.7, P= 0.027). Seventy-seven (53.9%) and four (16.7%) patients who were/were not advised by ED staff that their pain management was important, respectively, were very satisfied (difference 37.2%, 95% CI 17.7-56.6, P= 0.002). CONCLUSIONS: Our 'adequate analgesia' definition might provide a useful clinical target, which, combined with adequate communication, might help maximize patient satisfaction.

Injuries significantly associated with thoracic spine fractures: a case-control study.

OBJECTIVE: To determine injuries significantly associated with traumatic thoracic spine (T-spine) fractures METHODS: This was a case-control study undertaken in an adult trauma centre. Cases were patients admitted with a traumatic T-spine fracture between January 1999 and August 2007. Each case had two controls matched for sex, age (+/-5 years) and injury severity classification (major/minor). Data were collected from patient medical records and the trauma service database. Multivariate logistic regression was used to determine injuries significantly associated with T-spine fracture. RESULTS: Two hundred and sixty-one cases and 512 controls were enrolled. In both groups, mean age was 41 years and 70% of patients were male. Univariate analysis revealed a range of injuries that were significantly more common among the cases, especially cervical and lumbar spine injuries, sternal/scapular/clavicular/rib fractures, pneumo/haemothorax and pulmonary contusions (P < 0.01). Skull fractures and lower limb injuries were significantly more common among the controls (P < 0.01). Logistic regression analysis revealed that only cervical and lumbar spine injuries and rib fractures were positively associated with T-spine fracture (P < 0.001). Skull fractures and lower limb injuries were negatively associated with T-spine injury (P < 0.001). CONCLUSION: Cervical and lumbar spine injuries and rib fractures are significantly associated with T-spine fracture. The presence of these injuries should raise suspicion of concomitant T-spine injury.

Therapeutic errors among children in the community setting: nature, causes and outcomes.

AIM: This study aimed to determine the epidemiology of therapeutic errors among children in the community setting. METHODS: This was a prospective, observational study of 491 consecutive cases reported to the Victorian Poisons Information Centre, between January 2006 and March 2007. A total of 450 (91.7%) parents/carers were followed up by telephone approximately 48 h after the initial call. The main outcome measures were the nature, causes and outcomes of the errors and actions taken or recommendations given to avoid future errors. RESULTS: The majority of children (334, 68.0%, 95% confidence interval (CI) 63.7, 72.1) were aged

Review article: leaving the emergency department without being seen.

Patients who leave the ED without being seen (LWBS) are unlikely to be satisfied with the quality of the service provided and might be at risk from conditions that have not been assessed or treated. We therefore examined the available research literature to inform the following questions: (i) In patients who attend for ED care, what factors are associated with the decision to LWBS? (ii) In patients who attend for ED care, are there adverse health outcomes associated with the decision to LWBS? (iii) Which interventions have been used to try to reduce the number of patients who attend for ED care and LWBS? From the available literature, there was insufficient evidence to draw firm conclusions; however, the literature does suggest that patients who LWBS have conditions of lower urgency and lower acuity, are more likely to be male and younger, and are likely to identify prolonged waiting times as a central concern. LWBS patients generally have very low rates of subsequent admission, and reports of serious adverse events are rare. Many LWBS patients go on to seek alternative medical attention, and they might have higher rates of ongoing symptoms at follow-up. Further research is recommended to include comprehensive cohort or well-designed case-control studies. These studies should assess a wide range of related factors, including patient, hospital and other relevant factors. They should compare outcomes for groups of LWBS patients with those who wait and should include cross-sectoral data mapping to truly detect re-attendance and admission rates.