Mark Holland

The growth of medical technology and subspecialization means that most medical care can no longer be provided by single doctors working alone. Notwithstanding its benefits, this transition of care from individuals to teams has made healthcare fragmented, with loss of continuity and potentially dangerous handovers , which can result in futile cycles of further enquiry before any effective intervention is delivered. These and other unforeseen consequences have, in our opinion, created major challenges for the care of acutely ill patients, which include reduced availability and slow access to acute care, and the impairment of the traditional bedside assessment and diagnostic processes due to faulty or missing information. To correct these issues, essential information, w hich harms patients if it is incorrect or unavailable, needs to be clearly defined, with systems put in place to ensure it is complete, current, correct, and immediately available. Furthermore, acute care would be focused more on patient needs if quality-of-care outcomes more explicitly measured what matters most to patients.

Aim: To determine the in-hospital mortality of eight physiological categories based on shock
index, pulse pressure and ROX index, and to compare each category according to admission
level of consciousness and National Early Warning Score.
Method: A non-interventional observational study of 122,262, unselected, adult emergency
admissions between 2014 and 2022.
Results: In-hospital mortality increases according to physiological category and whether the
admission NEWS was<3 or ≥3. For NEWS ≥3, patients were more likely to die when not alert.
Irrespective of total NEWS, patients with a low ROX index <22 are more likely to die.
Conclusion: Patients with the same NEWS value can have different physiological
derangements. Level of consciousness also provides greater insight than NEWS alone
regarding the risk of in-patient mortality.

Introduction Hospital attendances due to respiratory infection peak in winter, contributing to pressures within acute services. We assessed the prevalence of suspected respiratory infection within acute medical admissions during winter and evaluated performance against recommendations for initial assessment. Methods Data were collected through the Society for Acute Medicine (SAM) Benchmarking Audit, comprising a hospital-level survey and 24-hour patient-level data collection for unplanned acute medical attendances on 22 February 2024. Performance metrics assessed included those from the SAM's clinical quality indicators (CQI) for medical admissions, and British Thoracic Society (BTS) guidelines for community acquired pneumonia. Results Data were available for 4390 patients at 76 hospitals. Suspected respiratory infections accounted for 22.8% of all unplanned medical attendances; these patients were older (age ≥70 years: 58.2% vs 44.7%, p<0.001) and had higher National Early Warning Score 2 (NEWS2) scores (NEWS2 ≥3: 63.8% vs 23.8%, p<0.001) than those without respiratory infection; they were more likely to be assessed in the emergency department (80.8% vs 63.7%, p<0.001), and had lower rates of discharge without overnight admission (14.9% vs 35.9%, p<0.001). 71.0% of patients underwent a chest X-ray within 4 hours of arrival; 27.0% were reported within 12 hours. Antibiotics were administered ≥4 hours from arrival in 32.9%. Performance against these indicators varied between hospitals. Nine hospitals (12.7%) had a separate respiratory admission service; this was not associated with improved performance against SAM CQIs or BTS guidance. Conclusion Respiratory infections contribute significantly to acute medical attendances via the emergency department. There remains significant scope to improve key steps in initial assessment and management.

To evaluate the current provision of medical same day emergency care (SDEC) services within the UK, and the current utilisation of these pathways in the assessment of unplanned medical attendances.
Survey data was used from the Society for Acute Medicine Benchmarking Audit (SAMBA), including anonymised patient-level data collected annually using a day of care survey.
Hospitals accepting unplanned medical attendances within the UK, 2019-2023.
34 948 unplanned and 4342 planned attendances across 188 hospital sites.
29.8% of unplanned medical attendances received their initial medical assessment within SDEC services (2403 patients in SAMBA23), with the proportion increasing over time. 82.4% of patients assessed in SDEC services were discharged without overnight admission. Assessment in SDEC services was less likely in male patients, patients with frailty and older adults (all p<0.005).Selected operational standards for SDEC delivery, set by the Society for Acute Medicine, were met in 64%-91% of hospitals. Most hospitals (82%) accepted referrals from emergency department triage and 63% accepted referrals directly from the paramedic team. 38% of hospitals did not use a recognised selection criteria to identify suitable patients for SDEC and only 8% used a criteria designed to identify patients suitable for discharge. Overall, 34.7% of medical attendances discharged without overnight admission received their medical assessment in locations other than SDEC.
Medical SDEC provides assessment for one-third of patients seen through acute medicine services. Although the proportion of patients assessed within SDEC is increasing, further innovation and improvements are needed to ensure appropriate patients access this service.

Patients with an elevated admission National Early Warning Score (NEWS) are more likely to die while in hospital. However, it is not known if this increased mortality risk is the same for all diagnoses. The aim of this study was to determine and compare the increased risk of in-hospital mortality associated with an elevated NEWS and different primary discharge diagnoses in unselected emergency admissions to a UK university teaching hospital.BACKGROUNDPatients with an elevated admission National Early Warning Score (NEWS) are more likely to die while in hospital. However, it is not known if this increased mortality risk is the same for all diagnoses. The aim of this study was to determine and compare the increased risk of in-hospital mortality associated with an elevated NEWS and different primary discharge diagnoses in unselected emergency admissions to a UK university teaching hospital.A non-interventional observational study of 122,321 consecutive, unselected, adult patients with complete data admitted as an emergency between 2014 and 2022.METHODSA non-interventional observational study of 122,321 consecutive, unselected, adult patients with complete data admitted as an emergency between 2014 and 2022.The overall in-hospital mortality was 4.3%. Eighty diagnostic groupings accounted for 85.8% of all admissions and 89.4% of all in-hospital deaths. Depending on diagnostic grouping, the risk of mortality associated with an admission NEWS ≥ 3 ranged from 2.3- to 100-fold. For example, the in-hospital mortality of COPD patients increased from 1.9% for those with admission NEWS < 3 to 35.6% for those with NEWS ≥ 3, for chest pain mortality increased from 0.1 to 3.9%, and for patients with an opiate overdose from 0.2 to 7.7%. Conversely, for admission NEWS < 3, aspiration pneumonia and intracranial hemorrhage had in-hospital mortalities of 13.7% and 12.1%, respectively.RESULTSThe overall in-hospital mortality was 4.3%. Eighty diagnostic groupings accounted for 85.8% of all admissions and 89.4% of all in-hospital deaths. Depending on diagnostic grouping, the risk of mortality associated with an admission NEWS ≥ 3 ranged from 2.3- to 100-fold. For example, the in-hospital mortality of COPD patients increased from 1.9% for those with admission NEWS < 3 to 35.6% for those with NEWS ≥ 3, for chest pain mortality increased from 0.1 to 3.9%, and for patients with an opiate overdose from 0.2 to 7.7%. Conversely, for admission NEWS < 3, aspiration pneumonia and intracranial hemorrhage had in-hospital mortalities of 13.7% and 12.1%, respectively.There is enormous variation in the mortality risk associated with an increased admission NEWS in different commonly encountered diagnoses. Therefore, the mortality risk of some 'low risk' conditions can be dramatically increased if their admission NEWS is elevated, whereas some 'high risk' conditions are still likely to die even if their admission NEWS is low.DISCUSSIONThere is enormous variation in the mortality risk associated with an increased admission NEWS in different commonly encountered diagnoses. Therefore, the mortality risk of some 'low risk' conditions can be dramatically increased if their admission NEWS is elevated, whereas some 'high risk' conditions are still likely to die even if their admission NEWS is low.

Coma is a medical emergency that can challenge the diagnostic and management skills of any clinician. A systematic and logical approach is necessary to make the correct diagnosis, the broad diagnostic categories being neurological, metabolic, diffuse physiological dysfunction and functional. Even when the diagnosis is not immediately clear, appropriate measures to resuscitate, stabilize and support a comatose patient must be rapidly performed. The key components in the assessment and management of a patient, namely history, examination, investigation and treatment, are performed in parallel, not sequentially. Unless the cause of coma is immediately obvious and reversible, help from senior and critical care colleagues is necessary. In particular, senior help is needed to make difficult management decisions in patients with a poor prognosis.

Objectives Increasing operational pressures on emergency departments (ED) make it imperative to quickly and accurately identify patients requiring urgent clinical intervention. The widespread adoption of electronic health records (EHR) makes rich feature patient data sets more readily available. These large data stores lend themselves to use in modern machine learning (ML) models. This paper investigates the use of transformer-based models to identify critical deterioration in unplanned ED admissions, using free-text fields, such as triage notes, and tabular data, including early warning scores (EWS).

Design A retrospective ML study.

Setting A large ED in a UK university teaching hospital.

Methods We extracted rich feature sets of routine clinical data from the EHR and systematically measured the performance of tree- and transformer-based models for predicting patient mortality or admission to critical care within 24 hours of presentation to ED. We compared our proposed models to the National EWS (NEWS).

Results Models were trained on 174 393 admission records. We found that models including free-text triage notes outperform structured tabular data models, achieving an average precision of 0.92, compared with 0.75 for tree-based models and 0.12 for NEWS.

Conclusions Our findings suggests that machine learning models using free-text data have the potential to improve clinical decision-making in the ED; our techniques significantly reduce alert rate while detecting most high-risk patients missed by NEWS.

Performance within acute medicine services is impacted by ongoing pressures on acute care services. Data from the Society for Acute Medicine Benchmarking Audit 2023 (SAMBA23), was used to assess performance of acute medicine services compared to key clinical quality indicators, comparing performance by initial assessment location. Data was analysed for 8213 unplanned attendances across 161 hospitals. Comparing by initial assessment location, performance against the clinical quality indicators was unchanged from 2022. Only 29% of daytime arrivals assessed within the Emergency Department received consultant review within target times. Delays were seen in transfer between acute care locations. 29% of patients requiring admission were not admitted to the AMU. There is ongoing variation in acute medical service performance nationally, with significant delays in patient access to appropriate assessment locations.Performance within acute medicine services is impacted by ongoing pressures on acute care services. Data from the Society for Acute Medicine Benchmarking Audit 2023 (SAMBA23), was used to assess performance of acute medicine services compared to key clinical quality indicators, comparing performance by initial assessment location. Data was analysed for 8213 unplanned attendances across 161 hospitals. Comparing by initial assessment location, performance against the clinical quality indicators was unchanged from 2022. Only 29% of daytime arrivals assessed within the Emergency Department received consultant review within target times. Delays were seen in transfer between acute care locations. 29% of patients requiring admission were not admitted to the AMU. There is ongoing variation in acute medical service performance nationally, with significant delays in patient access to appropriate assessment locations.

Cardio-Renal-Metabolic (CaReMe) diseases, in the form of heart failure, chronic kidney disease and diabetes mellitus, justify prescription of multiple prognostically beneficial medications, specifically renin-angiotensin system inhibitors, mineralocorticoid receptor antagonists, and sodium-glucose co-transporter-2 inhibitors. Use of these medications is complicated by association with adverse effects, particularly acute kidney injury and hyperkalaemia. Balancing risk and benefit is a common dilemma in acute medicine, with increasingly frequent and complex treatment decisions. Physicians should contemplate adjustments to medications within the context of not just acute illness but also long-term benefit. In the setting of hyperkalaemia, potassium-binding medications can be utilised. At hospital discharge optimisation of therapy can be achieved through clear safety netting advice, scheduled biochemical follow-up, and planned clinical review

IntroductionOlder people living with frailty are at high risk of emergency hospital admission and often have complex care needs which may not be adequately met by conventional models of acute care. This has driven the introduction of adaptations to acute care pathways designed to improve outcomes in this patient group. The identification of differences in the organisational approach to frailty may highlight opportunities for quality improvement.MethodsThe Society for Acute Medicine Benchmarking audit is a national service evaluation which uses a single day-of-care methodology to record patient and organisational level data. All acute hospitals in the United Kingdom are eligible to participate. Emergency admissions referred to acute medical services between 00:00 and 23:59 on Thursday 23rd June 2022 were recorded. Information on the structure and operational design of acute frailty services was collected. The use of a validated frailty assessment tool, clinical frailty scale within the first 24 h of admission, assessment by an acute frailty service and clinical outcomes were reported in patients aged 70 year and above. A mixed effect generalised linear model was used to determine factors associated same-day discharge without overnight stay in patients with frailty.ResultsA total of 152 hospitals participated. There was significant heterogeneity in the operational design and staffing model of acute frailty services. The presence of an acute frailty unit was reported in 57 (42.2%) hospitals. The use of validated frailty assessment tools was reported in 117 (90.0%) hospitals, of which 107 (91.5%) used the clinical frailty scale. Patient-level data were recorded for 3604 patients aged 70 years and above. At the patient level, 1626 (45.1%) were assessed using a validated tool during the admission process. Assessment by acute frailty services was associated with an increased likelihood of same-day discharge (adjusted OR 1.55, 95%CI 1.03- 2.39).ConclusionThere is significant variation in the provision of acute frailty services. Frailty-related policies and services are common at the organisational level but implemented inconsistently at the patient level. Older people with frailty or geriatric syndromes assessed by acute frailty services were more likely to be discharged without the need for overnight bed-based admission.