Background:Trauma accounts for over 60,000 deaths annually in South Africa and is also responsible for a high proportion of emergency centre (EC) visits. Up to 91% of trauma patients in the EC experience acute pain, underscoring a critical public health concern, particularly in low and middle-income countries (LMICS), where research indicates poor pain management and a paucity of data. This study aimed to describe trauma pain assessment and management practices in a busy Cape Town EC. Methods:This single-centre retrospective chart review in a high trauma burden EC collected a convenience sample over two weeks in 2024. Data were extracted from the medical records of adult trauma patients presenting to the EC. Results:A total of 234 patients were included, predominantly male (73.1%), with a median age of 33 (IQR 26-41). Only 32.9% (77) of the patients had their pain assessed and documented, largely only in the triage process, and no patient had their pain reassessed. Furthermore, only 42.3% (99) of patients received analgesia, including opioids administered to 48.5% (48), non-steroidal anti-inflammatories to 45.5% (45), paracetamol to 69.7% (69), and ketamine to 17.2% (17). The time from arrival to the administration of the first analgesia in the EC was documented in 83 patients, with a median time to analgesia of 375 min (IQR 152-611). Conclusion:This study demonstrates findings consistent with the global crisis of inadequate pain management. It reveals poor pain assessment and management, with prolonged waiting times for analgesia despite various indicators of potential pain severity, highlighting a crucial need for changes in pain management strategies. Training, awareness, and protocols to improve pain management are essential, along with considering nurse-led analgesia at triage.

Introduction: : An effective Emergency Medical Service (EMS) depends on a well-functioning Emergency Medical Dispatch (EMD), which influences response times, resource allocation, and patient outcomes. Competent EMD staff are essential for gathering information, making rapid decisions, dispatching resources, and providing life-saving pre-arrival instructions. While high-income countries use standardized computer-aided dispatch systems, many low- and middle-income countries (LMICs), including South Africa, lack such systems and rely on staff competencies. This scoping review sought to identify the competencies for EMD staff. Methods: : We searched the international literature using the keywords “Emergency Medical Service”; “EMS call centre”; “competencies” and their synonyms. Two reviewers independently reviewed the identified articles. The articles were tabled highlighting findings focused on EMD system requirements; EMD staff competency requirements; education and training for EMD staff. Results: : Among the 35 articles identified, five (14%) originated from LMICs with no articles specifically addressing the competency requirements for staff working in an EMD. A total of 16 (46%) studies highlighted the need for qualified staff (nurses, physicians and paramedics). The use of computer aided dispatch and Medical Priority Dispatch System in an EMD was identified in 13 (37%). The review of existing programs identified eight EMD programs emphasizing effective interview and communication skills, use of technology, medical terminology, first aid, stress management, resource allocation, effective reporting and feedback. Literature showed a renewed interest in EMD research within the last 5 years, particularly in LMICs where EMS are poorly developed. Discussion: : The scoping review highlighted a lack of evidence on the competencies and performance needs of EMD staff. While key skills such as information gathering, decision-making, and resource allocation are recognized, further research is needed to define core competencies and develop context-appropriate training, especially in LMICs where such evidence is scarce.

Introduction In South Africa (SA), Emergency Medicine (EM) has rapidly evolved from an unstructured service to a recognised speciality since 2004. Specialists in EM now lead many Emergency Departments (EDs) in SA, and the impact that the speciality has made is hard to measure against a resource-constrained system and growing healthcare needs. This study aims to explore these aspects through the perspectives of key stakeholders, providing insights into the speciality's progress and its future trajectory. Methods A qualitative study was used to examine the experiences and perceptions of local EM leaders. Semi-structured one-on-one interviews were conducted with a purposive convenience sample from hospitals across SA. An appreciative inquiry framework was used to explore positive aspects of EM alongside its challenges and analysed using inductive thematic analysis. Results Ten primary interviews, followed by a further three secondary interviews were conducted across four provinces of SA. Four primary themes emerged: EM has changed everything, systems in the chaos; the EM team goes the extra mile - the people in EM; the ED is the hub and heart of the hospital; and a team of EM specialists can innovate and change beyond the ED. Each theme encapsulated distinct perspectives on the achievements and challenges faced by EM professionals, highlighting the complex and evolving role of the speciality within SA healthcare. Conclusion EM has evolved into a crucial speciality, providing essential acute care, improving patient flow, and mitigating access barriers. The findings emphasise EM’s function as both a frontline responder and a system stabiliser, compensating for gaps in primary and inpatient care. Strong EM leadership, training and teamwork are central to the accomplishments, although many challenges remain. Strengthening EM leadership, integrating EM into broader healthcare strategies, and fostering multidisciplinary collaboration are key to ensuring the speciality’s continued growth and impact. Lessons from SA’s experience offer valuable insights for other African systems.

Background:Snakebites have been recognised as a neglected tropical disease by the World Health Organization and remain a potentially preventable cause of morbidity around the world, particularly in Africa. South Africa (SA) has a well-documented prevalence of snakebites, and there has been a recent surge in attention on snakebite due to dwindling antivenom stocks. uMkhanyakude District in the far northeast of SA has one of the highest incidences of snakebite and uses more antivenom than elsewhere in SA, and the impact of antivenom shortages and the high prevalence of disease has not been assessed recently. Methods:A descriptive, retrospective, observational study was undertaken to describe victims of snakebites presenting from 1 September 2019 to 31 August 2022 to a district hospital, Mosvold Hospital. Data were manually extracted from patients' medical records. Information about demographics, clinical presentations, treatments and outcomes was collected and analysed. Results:A total of 155 snakebite cases presented, with an incidence rate of 58 snakebite cases per 100 000 people per year. Most patients were young, with a median age of 19 years (range 0-94 years), and most bites occurred outdoors (75/155, 48.4 %). Patients were most often bitten on the lower limbs (107/155, 69.0 %), and most presented with minimal swelling (117/155, 75.5 %). Antivenom was administered to 33 patients (33/155, 21.3 %), and 24 (24/33, 72.7 %) of those who received antivenom experienced some form of post-antivenom reaction. Three patients died during their hospital stay, resulting in a mortality rate of 1.2 deaths per 100 000 people per year. Antivenom shortages contributed to the outcomes of several patients. Conclusion:Cytotoxic snakebites are a not infrequent presentation, many requiring admission, antivenom and subsequent management. Emergency personnel need to be familiar with local protocols for snakebite management, and have the resources to manage severe envenomation, including access to antivenom. Expanding knowledge and education about snakebites provided to community members might also improve outcomes and prevent bites.

Introduction: Zimbabwe has experienced significant population growth, as well as a rise in noncommunicable diseases, impacting the demand for emergency care services. However, there is limited data on the state of Zimbabwe's emergency care systems to meet this growing need. This pilot study aimed to assess the capacity of Emergency Departments at three major referral hospitals in Harare, Zimbabwe. Methodology: A cross-sectional descriptive study was conducted using the World Health Organization’s Hospital Emergency Assessment Tool (HEAT). Key participants from each hospital were interviewed in 2023 to gather data on facility characteristics, human resources, diagnostic services, and signal functions. Results: Three large hospitals offering round-the-clock emergency care services were included. A core of dedicated doctors and nurses manned the Emergency Department in all three facilities, yet none had specific training in emergency medicine. There were various significant challenges to the operation of emergency departments, including access to water for one, and two with limited capacity to perform emergency diagnostic tests as they lacked the necessary equipment and consumables. One hospital reported a sufficient set of emergency protocols and guidelines, while the other two hospitals exhibited varying levels of availability in these areas. Two facilities reported adequate capacity in performing >85 % of signal functions. Common challenges included the absence of emergency-trained staff, lack of a formal triage system, and equipment deficiencies, all of which constrained their ability to deliver certain signal functions. Conclusion: This assessment revealed a spectrum of service availabilities across the three hospitals, with common barriers identified. To strengthen the country's emergency care system, efforts should be focused on improving consumable supplies, prioritising training initiatives in emergency medicine to build a skilled workforce and implementing standardised protocols and triage systems across facilities. Continuous monitoring and evaluation of implemented interventions are necessary to ensure sustainable improvements in emergency care delivery.

As new space missions are being prepared, now is the time for accessible designs and approaches. In a workshop, we asked attendees to discuss the adjustments for people with disabilities in relation to the established barriers to human spaceflight. Potential challenges were grouped into medical, physiological, subsistence, and technical. These challenges and potential solutions will inform future space missions and the emerging and more diverse field of space tourism.

Introduction Emergency care (EC) is rapidly expanding across Africa, yet research output remains limited despite high regional need. The African Conference on Emergency Medicine (AfCEM) is the continent’s only dedicated EC scientific meeting, offering a unique platform to showcase and strengthen regional research. This analysis of abstracts from AfCEM 2024 aims to explore authorship patterns, research themes, and partnerships, highlighting progress and opportunities in African EC research. Method Three reviewers analysed the 217 accepted AfCEM 2024 abstracts by authorship, country involvement, partnerships, topic, and methodology. Findings were summarised descriptively and compared with data from previous AfCEM conferences to explore trends in emergency care research across the region. Results Of 217 abstracts submitted to AfCEM 2024, East Africa contributed the largest share (43 %), with marked underrepresentation from Central and North Africa. Authorship spanned 34 countries, with 70.5 % of authors and 74.7 % of first authors affiliated with African institutions. However, 54.3 % of abstracts included at least one HIC author, and these were over four times more likely to have five or more contributors. Supadel-sponsored or resident trainees accounted for 16 % of first authors. Most abstracts presented primary data (69.1 %) and used observational or descriptive methods (50.2 %), with 20.3 % employing qualitative or mixed approaches. Research largely focused on emergency presentations, highlighting a growing but uneven research landscape across the continent Conclusion This analysis highlights growing African leadership and thematic diversity in EC research, as showcased at AfCEM 2024. While collaborations with high-income countries remain common, African-led work is increasing. Research largely aligns with global priorities but remains predominantly observational, with limited patient-centred or francophone representation. AfCEM continues to reflect and shape the region’s research landscape. Ongoing efforts should focus on equitable partnerships, broader access, and inclusive, cross-disciplinary engagement.

INTRODUCTION: Accessible spaceflight may seem a distant concept. As part of a diverse European Space Agency funded Topical Team, we are working on the physiological feasibility of space missions being undertaken by people with physical disabilities. Here, the first activity of this team is presented in the form of key lessons learned from aviation to inform new work on space missions. DISCUSSION: The first lesson is agreeing on realistic expectations about impairments, their severity, and the possibility of flying independently. This is important in terms of astronaut recruitment and societal expectations. The second lesson relates to training and adjustments for people with disabilities. Flexibility is important while maintaining safety for everyone involved. The third lesson is about managing unconscious bias from the different stakeholders. We conclude by arguing that engagement with people from different backgrounds is essential for the success of the first space mission with people with physical disabilities. Miller-Smith MJ, Tucker N, Anderton R, Caplin N, Harridge SDR, Hodkinson P, Narici MV, Pollock RD, Possnig C, Rittweger J, Smith TG, Di Giulio I. Lessons for flying astronauts with disabilities drawn from experience in aviation . Aerosp Med Hum Perform. 2024; 95(9):716–719.

INTRODUCTION:During centrifuge-simulated suborbital spaceplane flights, launch and re-entry frequently cause visual symptoms, and G-induced loss of consciousness can occur. G-related effects may be more prominent during re-entry from microgravity on actual flights. A modified anti-G maneuver that does not involve a breath strain and is suitable for members of the public may be effective against these effects. METHODS:Recruited were 13 healthy subjects (age range 34-82 yr) who had experienced visual symptoms during centrifuge-simulated suborbital centrifuge profiles as part of a previous study. Onset and duration of greyout were recorded during an acceleration profile simulating spaceplane launch and re-entry in an upright seated position. The profile was undertaken twice: once in a relaxed state and once while undertaking anticipatory muscle tensing consisting of pre-tensing the leg and abdominal muscles (i.e., the muscle tensing component of a standard aircrew anti-G straining maneuver). RESULTS:Muscle tensing was well tolerated and prevented 100% of greyout on launch and 54% of greyout on re-entry, as well as delaying the onset of greyout when it did occur on re-entry. Combined with the previous study's data, this indicates an overall population incidence of greyout of ∼70% on launch, falling to zero with muscle tensing, and ∼80% on re-entry, falling to ∼37% with muscle tensing. DISCUSSION:Anticipatory pre-tensing of leg and abdominal muscles prevented greyout completely during the launch phase and partially during re-entry, and should be considered as part of routine suborbital spaceplane operations. Studies providing relevant data from actual flights are required. Pollock RD, Britton JK, Green NDC, Hendriksen D, Hodkinson PD, Anderton RA, Smith TG. Prevention of G-induced effects on vision and consciousness during simulated suborbital spaceflight. Aerosp Med Hum Perform. 2024; 95(12):897-901.

Long-duration spaceflights beyond low-Earth orbit, including missions to the Moon and Mars, pose significant health risks. Although biomedical approaches commonly appear in the literature, considering psychological and social factors alongside physiologic health offers a more holistic approach to astronaut care. Integrating the biopsychosocial (BPS) framework into medical planning addresses complex spaceflight challenges and aids in developing mitigation strategies. This review examined health risks associated with long-duration spaceflight within a BPS framework. Sources included governmental space agencies, academic textbooks, and relevant publications from multiple databases. Considering the National Aeronautics and Space Administration's Human Research Program's 5 main hazards, a conceptual model was developed to highlight the multifactorial BPS effects of spaceflight. In space, astronauts face unique environments and biological adaptations, including fluid shift, plasma volume loss, bone density loss, and muscle atrophy. Noise and the absence of natural light disrupt circadian rhythms, causing sleep disturbances and fatigue, which affect physical and mental health. Studies on crews in isolated and confined extreme environments reveal psychosocial challenges, including impaired mood and cognition, interpersonal tension, and miscommunication. International collaboration in spaceflight introduces differences in communication, problem solving, and social customs due to diverse cultural backgrounds. Upcoming long-distance missions likely will amplify these challenges. This review emphasizes BPS health considerations in long-duration spaceflight. It highlights the interplay among psychological, social, and biological factors, advocating for multidisciplinary teams and a holistic approach to astronaut health and mission planning and the potential added value of BPS perspectives in considering countermeasures.

Introduction: Previous studies deriving and validating triage scores for patients with suspected COVID-19 in Emergency Department settings have been conducted in high- or middle -income settings. We assessed eight triage scores' accuracy for death or organ support in patients with suspected COVID-19 in Sudan. Methods: We conducted an observational cohort study using Covid-19 registry data from eight emergency unit isolation centres in Khartoum State, Sudan. We assessed performance of eight triage scores including: PRIEST, LMIC-PRIEST, NEWS2, TEWS, the WHO algorithm, CRB-65, Quick COVID-19 Severity Index and PMEWS in suspected COVID-19. A composite primary outcome included death, ventilation or ICU admission. Results: In total 874 (33.84 %, 95 % CI:32.04 % to 35.69 %) of 2,583 patients died, required intubation/noninvasive ventilation or HDU/ICU admission . All risk -stratification scores assessed had worse estimated discrimination in this setting, compared to studies conducted in higher -income settings: C -statistic range for primary outcome: 0.56-0.64. At previously recommended thresholds NEWS2, PRIEST and LMIC-PRIEST had high estimated sensitivities (>= 0.95) for the primary outcome. However, the high baseline risk meant that lowrisk patients identified at these thresholds still had a between 8 % and 17 % risk of death, ventilation or ICU admission. Conclusion: None of the triage scores assessed demonstrated sufficient accuracy to be used clinically. This is likely due to differences in the health care system and population (23 % of patients died) compared to higher -income settings in which the scores were developed. Risk -stratification scores developed in this setting are needed to provide the necessary accuracy to aid triage of patients with suspected COVID-19.

AbstractNeck pain associated with helmet‐wear is an occupational health problem often observed in helicopter pilots and aircrew. Whether aircrew helmet wearing is associated with physiological and biomechanical differences between sexes is currently unknown. This study investigated neuromuscular activation patterns during different helmet‐wearing conditions. The helmet load was manipulated through a novel Helmet Balancing System (HBS) in healthy, non‐pilot male and female participants (n = 10 each, age 19–45 years) in two phases. Phase A assessed the acute effects of helmet‐wear on neck muscles activation during head movements. Phase B examined changes in muscle activity and cervical disc height after wearing a helmet for 45 min. In Phase A, muscle activity was similar between sexes in many movements, but it was higher in female participants when wearing a helmet than in males. The HBS reduced muscle activity in both sexes. In Phase B, female participants exhibited a greater level of muscular fatigue, and male participants’ cervical disc height was significantly decreased [5.7 (1.4) vs. 4.4 (1.5) mm, P < 0.001] after continuous wearing. Both sexes showed no significant change in muscle fatigue and disc height [male: 5.0 (1.3) vs. 5.2 (1.4) mm, P = 0.604] after applying HBS. These findings demonstrate sex‐specific physiological and biomechanical responses to wearing a helmet. They may indicate different postural and motor control strategies, associated with different neck pain aetiologies in male and female aircrew, the knowledge of which is important to reduce or prevent musculoskeletal injuries associated with helmet wearing.

Background: The COVID-19 Pandemic had profound effects on healthcare systems around the world. In South Africa, field hospitals, such as the Mitchell’s Plain Field Hospital, managed many COVID patients and deaths, largely without family presence. Communicating with families, preparing them for death and breaking bad news was a challenge for all staff. Aim: This study explores the experiences of healthcare professionals working in a COVID-19 field hospital, specifically around having to break the news of death remotely. Setting: A150-bed Mitchells Plain Field Hospital (MPFH) in Cape Town. Methods: A qualitative exploratory design was utilised using a semi-structured interview guide. Results: Four themes were identified: teamwork, breaking the news of death, communication and lessons learnt. The thread linking the themes was the importance of teamwork, the unpredictability of disease progression in breaking bad news and barriers to effective communication. Key lessons learnt included effective management and leadership. Many families had no access to digital technology and linguo-cultural barriers existed. Conclusion: We found that in the Mitchell’s Plain Field Hospital, communication challenges were exacerbated by the unpredictability of the illness and the impact of restrictions on families visiting in preparing them for bad news. We identified a need for training using different modalities, the importance of a multidisciplinary team approach and for palliative care guidelines to inform practice. Contribution: Breaking the news of death to the family is never easy for healthcare workers. This article unpacks some of the experiences in dealing with an extraordinary number of deaths by a newly formed team in the COVID era.

BACKGROUND: Unexplained physiological events (PE), possibly related to hypoxia and hyperventilation, are a concern for some air forces. Physiological monitoring could aid research into PEs, with measurement of arterial oxygen saturation (Spo₂) often suggested despite potential limitations in its use. Given similar physiological responses to hypoxia and hyperventilation, the present study characterized the cardiovascular and respiratory responses to each.METHODS: Ten healthy subjects were exposed to 55 mins of normobaric hypoxia simulating altitudes of 0, 8000, and 12,000 ft (0, 2438, and 3658 m) while breathing normally and voluntarily hyperventilating (doubling minute ventilation). Respiratory gas analysis and spirometry measured end-tidal gases (PETo₂ and PETco₂) and minute ventilation. Spo₂ was assessed using finger pulse oximetry. Mean arterial, systolic, and diastolic blood pressure were measured noninvasively. Cognitive impairment was assessed using the Stroop test.RESULTS: Voluntary hyperventilation resulted in a doubling of minute ventilation and lowered PETco₂, while altitude had no effect on these. PETo₂ and Spo₂ declined with increasing altitude. However, despite a significant drop in PETo₂ of 15.2 mmHg from 8000 to 12,000 ft, Spo₂ was similar when hyperventilating (94.7 ± 2.3% vs. 93.4 ± 4.3%, respectively). The only cardiovascular response was an increase in heart rate while hyperventilating. Altitude had no effect on cognitive impairment, but hyperventilation did.DISCUSSION: For many cardiovascular and respiratory variables, there is minimal difference in responses to hypoxia and hyperventilation, making these challenging to differentiate. Spo₂ is not a reliable marker of environmental hypoxia in the presence of hyperventilation and should not be used as such without additional monitoring of minute ventilation and end-tidal gases.Haddon A, Kanhai J, Nako O, Smith TG, Hodkinson PD, Pollock RD. Cardiorespiratory responses to voluntary hyperventilation during normobaric hypoxia. Aerosp Med Hum Perform. 2023; 94(2):59-65.

Introduction: Prehospital care in many low-and middle-income countries is underdeveloped and needs strengthening for improved outcomes. Where formal prehospital care systems are under development, integration of a layperson first responder programme may help improve access for those in need. The World Health Organization recently developed the Community First Aid Responder (CFAR) learning program in support of this system, providing that it may require adaptation to be contextually suitable and sustainably implemented at country level. This study assesses a pilot WHO CFAR course in Kinshasa, Democratic Republic of Congo, to inform future rollouts and related research.Methods: We conducted a 3-day in-person pilot CFAR training with 42 purposively selected community health workers. Data collection involved quantitative and qualitative phases. The first consisted of structured pre-and post-training surveys, and a course evaluation by participants. The second consisted of two focus group discussions involving purposively selected community health workers in one group, and a convenience sample of course instructors and organisers in the other. Perceptions regarding course content, perceived knowledge acquisition and self-confidence gain were analysed using descriptive statistics for the quantitative data and content analysis for qualitative data.Results: Course participants were predominantly male (76.3 %) with a median age of 42 years and most (80.5 %) had no prior first aid training. Most were satisfied that the learning objectives were reached, the logistics were adequate, and that the content and teaching language were appropriately tailored to local context. The majority (94.7 %) found the 3-day duration insufficient. There was a significant self-confidence gain regarding first aid skills (average 17.9 % in pre-to 95.3 % in post-training, p < 0.001). Favourable opinions on the course structure, content, logistics and teaching methods were noted.Conclusion: A CFAR course pilot was successfully conducted in Kinshasa. The course is appropriate for context and well received by participants. It can form a key component of developing prehospital care systems in resource-constrained settings.

INTRODUCTION: A flight trial was conducted to determine whether breathing 60% oxygen during high performance flight maneuvers using contemporary pilot flight equipment induces atelectasis and to explore whether cabin altitude had any influence on the extent of atelectasis identified.METHODS: On 2 separate days, 14 male aircrew flew as passengers at High [14,500-18,000 ft (4420-5486 m)] and Low [4000-6000 ft (1219-1829 m)] cabin pressure altitude in a Hawk T Mk1 aircraft breathing 60% oxygen. Sorties comprised 16 maneuvers at +5 Gz, each sustained for 30 s. Lung volumes (spirometry), basal lung volume (electrical impedance tomography, EIT), and peripheral oxygen saturation during transition from hyperoxia to hypoxia (pulmonary shunt fraction) were measured in the cockpit immediately before (Pre) and after (Post) flight.RESULTS: Forced inspiratory vital capacity (FIVC) was significantly lower Postflight after High (-0.24 L) and Low (-0.38 L) sorties, but recovered to Preflight values by the fourth repeat (FIVC4). EIT-derived measures of FIVC decreased after High (-3.3%) and Low (-4.4%) sorties but did not recover to baseline by FIVC4. FIVC reductions were attributable to decreased inspiratory capacity. Spo₂ was lower Postflight than Preflight in High and Low sorties.DISCUSSION: Breathing 60% oxygen during flight results in a 3.8-4.9% reduction in lung volume associated with a small decrease in blood oxygenation and an estimated pulmonary shunt of up to 5.7%. EIT measures suggest persisting airway closure despite repeated FIVC maneuvers. There was no meaningful influence of cabin pressure altitude. The operational consequence of the observed changes is likely to be small.Tank H, Kennedy G, Pollock R, Hodkinson P, Sheppard-Hickey R-A, Woolford J, Green NDC, Stevenson A. Cabin pressure altitude effect on acceleration atelectasis after agile flight breathing 60% oxygen. Aerosp Med Hum Perform. 2023; 94(1):3-10.

BackgroundUneven vaccination and less resilient health care systems mean hospitals in LMICs are at risk of being overwhelmed during periods of increased COVID-19 infection. Risk-scores proposed for rapid triage of need for admission from the emergency department (ED) have been developed in higher-income settings during initial waves of the pandemic.MethodsRoutinely collected data for public hospitals in the Western Cape, South Africa from the 27th August 2020 to 11th March 2022 were used to derive a cohort of 446,084 ED patients with suspected COVID-19. The primary outcome was death or ICU admission at 30 days. The cohort was divided into derivation and Omicron variant validation sets. We developed the LMIC-PRIEST score based on the coefficients from multivariable analysis in the derivation cohort and existing triage practices. We externally validated accuracy in the Omicron period and a UK cohort.ResultsWe analysed 305,564 derivation, 140,520 Omicron and 12,610 UK validation cases. Over 100 events per predictor parameter were modelled. Multivariable analyses identified eight predictor variables retained across models. We used these findings and clinical judgement to develop a score based on South African Triage Early Warning Scores and also included age, sex, oxygen saturation, inspired oxygen, diabetes and heart disease. The LMIC-PRIEST score achieved C-statistics: 0.82 (95% CI: 0.82 to 0.83) development cohort; 0.79 (95% CI: 0.78 to 0.80) Omicron cohort; and 0.79 (95% CI: 0.79 to 0.80) UK cohort. Differences in prevalence of outcomes led to imperfect calibration in external validation. However, use of the score at thresholds of three or less would allow identification of very low-risk patients (NPV ≥0.99) who could be rapidly discharged using information collected at initial assessment.ConclusionThe LMIC-PRIEST score shows good discrimination and high sensitivity at lower thresholds and can be used to rapidly identify low-risk patients in LMIC ED settings.

COVID-19 infection rates remain high in South Africa. Clinical prediction models may be helpful for rapid triage, and supporting clinical decision making, for patients with suspected COVID-19 infection. The Western Cape, South Africa, has integrated electronic health care data facilitating large-scale linked routine datasets. The aim of this study was to develop a machine learning model to predict adverse outcome in patients presenting with suspected COVID-19 suitable for use in a middle-income setting. A retrospective cohort study was conducted using linked, routine data, from patients presenting with suspected COVID-19 infection to public-sector emergency departments (EDs) in the Western Cape, South Africa between 27th August 2020 and 31 st October 2021. The primary outcome was death or critical care admission at 30 days. An XGBoost machine learning model was trained and internally tested using split-sample validation. External validation was performed in 3 test cohorts: Western Cape patients presenting during the Omicron COVID-19 wave, a UK cohort during the ancestral COVID-19 wave, and a Sudanese cohort during ancestral and Eta waves. A total of 282,051 cases were included in a complete case training dataset. The prevalence of 30-day adverse outcome was 4.0%. The most important features for predicting adverse outcome were the requirement for supplemental oxygen, peripheral oxygen saturations, level of consciousness and age. Internal validation using split-sample test data revealed excellent discrimination (C-statistic 0.91, 95% CI 0.90 to 0.91) and calibration (CITL of 1.05). The model achieved C-statistics of 0.84 (95% CI 0.84 to 0.85), 0.72 (95% CI 0.71 to 0.73), and 0.62, (95% CI 0.59 to 0.65) in the Omicron, UK, and Sudanese test cohorts. Results were materially unchanged in sensitivity analyses examining missing data. An XGBoost machine learning model achieved good discrimination and calibration in prediction of adverse outcome in patients presenting with suspected COVID19 to Western Cape EDs. Performance was reduced in temporal and geographical external validation.