Accuracy of pre-hospital trauma notification calls

What influences ambulance clinician decisions to pre-alert emergency departments: a qualitative exploration of pre-alert practice in UK ambulance services and emergency departments

BACKGROUND

Ambulance clinicians use pre-alerts to inform receiving hospitals of the imminent arrival of a time-critical patient considered to require immediate attention, enabling the receiving emergency department (ED) or other clinical area to prepare. Pre-alerts are key to ensuring immediate access to appropriate care, but unnecessary pre-alerts can divert resources from other patients and fuel 'pre-alert fatigue' among ED staff. This research aims to provide a better understanding of pre-alert decision-making practice.

METHODS

Semi-structured interviews were conducted with 34 ambulance clinicians from three ambulance services and 40 ED staff from six receiving EDs. Observation (162 hours) of responses to pre-alerts (n=143, call-to-handover) was also conducted in the six EDs. Interview transcripts and observation notes were imported into NVIVO and analysed using thematic analysis.

FINDINGS

Pre-alert decisions involve rapid assessment of clinical risk based on physiological observations, clinical judgement and perceived risk of deterioration, with reference to pre-alert guidance. Clinical experience (pattern recognition and intuition) and confidence helped ambulance clinicians to understand which patients required immediate ED care on arrival or were at highest risk of deterioration. Ambulance clinicians primarily learnt to pre-alert 'on the job' and via informal feedback mechanisms, including the ED response to previous pre-alerts. Availability and access to clinical decision support was variable, and clinicians balanced the use of guidance and protocols with concerns about retention of clinical judgement and autonomy. Differences in pre-alert criteria between ambulance services and EDs created difficulties in deciding whether to pre-alert and was particularly challenging for less experienced clinicians.

CONCLUSION

We identified potentially avoidable variation in decision-making, which has implications for patient care and emergency care resources, and can create tension between the services. Consistency in practice may be improved by greater standardisation of guidance and protocols, training and access to performance feedback and cross-service collaboration to minimise potential sources of tension.

Future Perspectives on Radiomics in Acute Liver Injury and Liver Trauma

Background: Acute liver injury occurs most frequently due to trauma, but it can also occur because of sepsis or drug-induced injury. This review aims to analyze artificial intelligence (AI)'s ability to detect and quantify liver injured areas in adults and pediatric patients. Methods: A literature analysis was performed on the PubMed Dataset. We selected original articles published from 2018 to 2023 and cohorts with ≥10 adults or pediatric patients. Results: Six studies counting 564 patients were collected, including 170 (30%) children and 394 adults. Four (66%) articles reported AI application after liver trauma, one (17%) after sepsis, and one (17%) due to chemotherapy. In five (83%) studies, Computed Tomography was performed, while in one (17%), FAST-UltraSound was performed. The studies reported a high diagnostic performance; in particular, three studies reported a specificity rate > 80%. Conclusions: Radiomics models seem reliable and applicable to clinical practice in patients affected by acute liver injury. Further studies are required to achieve larger validation cohorts.

Use of prehospital reverse shock index times Glasgow Coma Scale to identify children who require the most immediate trauma care

BACKGROUND

Appropriate prehospital trauma triage ensures transport of children to facilities that provide specialized trauma care. There are currently no objective and generalizable scoring tool for emergency medical services to facilitate such decisions. An abnormal reverse shock index times Glasgow Coma Scale (rSIG), which is calculated using readily available parameters, has been shown to be associated with severely injured children. This study sought to determine if rSIG could be used in the prehospital setting to identify injured children who require the highest levels of care.

METHODS

Patients (1-18 years old) transferred from the scene to a level 1 pediatric trauma center from 2010 to 2020 with complete prehospital and emergency department vital signs, and Glasgow Coma Scale (GCS) scores were included. Reverse shock index times GCS was calculated as previously described ((systolic blood pressure/heart rate) × GCS), and the following cutoffs were used: ≤13.1, ≤16.5, and ≤20.1 for 1- to 6-, 7- to 12-, and 13- to 18-year-old patients, respectively. Trauma activation level and clinical outcomes upon arrival to the pediatric trauma center were collected.

RESULTS

There were 247 patients included in the analysis; 66.0% (163) had an abnormal prehospital rSIG. Patients with an abnormal rSIG had a higher rate of highest-level trauma activation compared with those with a normal rSIG (38.7% vs. 20.2%, p = 0.013). Patients with an abnormal prehospital rSIG also had higher rates of intubation (28.8% vs. 9.52%, p < 0.001), intracranial pressure monitor (9.20 vs. 1.19%, p = 0.032), need for blood (19.6% vs. 8.33%, p = 0.034), laparotomy (7.98% vs. 1.19%, p = 0.039), and intensive care unit admission (54.6% vs. 40.5%, p = 0.049).

CONCLUSION

Reverse shock index times GCS may assist emergency medical service providers in early identification and triage of severely injured children. An abnormal rSIG in the emergency department is associated with higher rates of intubation, need for blood transfusion, intracranial pressure monitoring, laparotomy, and intensive care unit admission. Use of this metric may help to speed the identification, care, and treatment of any injured child.

LEVEL OF EVIDENCE

Prognostic and Epidemiological; Level IV.

Emergency Clinical Procedure Detection With Deep Learning

Information about a patient's state is critical for hospitals to provide timely care and treatment. Prior work on improving the information flow from emergency medical services (EMS) to hospitals demonstrated the potential of using automated algorithms to detect clinical procedures. However, prior work has not made effective use of video sources that might be available during patient care. In this paper we explore the use convolutional neural networks (CNNs) on raw video data to determine how well video data alone can automatically identify clinical procedures. We apply multiple deep learning models to this problem, with significant variation in results. Our findings indicate performance improvements compared to prior work, but also indicate a need for more training data to reach clinically deployable levels of success.