Cardiac Arrest Calls Research Articles

Building confidence and knowledge of scribes during cardiac arrest calls: an educational QI project

Building confidence and knowledge of scribes during cardiac arrest calls: an educational QI project

2317 Geriatrician-led perioperative services: an example of value-based care

Abstract Background Perioperative services must adapt to the needs of an increasingly older surgical population. Perioperative medicine for Older People undergoing Surgery (POPS) services integrate geriatric medicine teams into surgical pathways to provide quality and cost-effective medical care. This project aims to examine value-based outcomes (clinical and financial impact) of embedding a POPS service at a district general hospital. Methods Following a period of implementation on an acute Trauma and Orthopaedic (T&O) ward, a two-week pilot was undertaken. All emergency fragility fracture admissions aged over 65 years with Clinical Frailty Scores (CFS) of ≥5 were included. Patients with hip fractures were excluded. The POPS service provided medical consultation, medicine rationalisation, proactive treatment escalation planning and shared decision making, as well as leading multidisciplinary team meetings. Outcome metrics: geriatric medicine consults, medical emergency team (MET)/cardiac arrest calls, staff/patient satisfaction and clinical coding. The REDUCE trial cost calculator was used to estimate savings. Results 35 patients were included, mean age 84 years, mean CFS score 7. Ward MET calls and cardiac arrest calls were reduced from a weekly average of 2.5 to 0, and weekly referrals to geriatric medicine reduced from 3 to 0. Experience-based design surveys identified thematic improvements relating to leadership, communication, dignity and respect. Improved quality of documentation resulted in the comorbidity score tariff increasing from £3325 to £6096 per patient. For services introduced by POPS including Comprehensive Geriatric Assessment and delirium assessments, the REDUCE trial cost calculator estimated an additional saving of £2926 per patient totalling hospital savings of £2 million per year (for an estimated 700 patients per year). Conclusion Implementation of a POPS service at a district general hospital can lead to cost savings, improved patient and staff experience, and improved clinical outcomes within a sustainable workforce model.

Outline and validation of a new dispatcher-assisted cardiopulmonary resuscitation educational bundle using the Delphi method

AimDispatcher-assisted cardiopulmonary resuscitation (DA-CPR) is time-dependent. To date, evidence-based training programmes for dispatchers are lacking. This study aimed to reach expert consensus on an educational bundle content for dispatchers to provide DA-CPR using the Delphi method. MethodAn educational bundle was created by the Swedish Resuscitation Council consisting of three parts: e-learning on DA-CPR, basic life support training and audit of emergency out-of-hospital cardiac arrest calls. Thereafter, a two-round modified Delphi study was conducted between November 2022 and March 2023; 37 experts with broad clinical and/or scientific knowledge of DA-CPR were invited. In the first round, the experts participated in the e-learning module and answered a questionnaire with 13 closed and open questions, whereafter the e-learning part of the bundle was revised. In the second round, the revised e-learning part was evaluated using Likert scores (20 items). The predefined consensus level was set at 80%. ResultsDelphi rounds one and two were assessed by 20 and 18 of the invited experts, respectively. In round one, 18 experts (18 of 20, 90%) stated that they did not miss any content in the programme. In round two, the scale-level content validity index based on the average method (S-CVI/AVE, 0.99) and scale-level content validity index based on universal agreement (S-CVI/UA, 0.85) exceeded the threshold level of 80%. ConclusionExpert consensus on the educational bundle content was reached using the Delphi method. Further work is required to evaluate its effect in real-world out-of-hospital cardiac arrest calls.

Intensive care following in-hospital cardiac arrest / periarrest calls—experience from one Scottish hospital

BackgroundIn-hospital cardiac arrest/periarrest is a recognised trigger for consideration of admission to the intensive care unit (ICU). We aimed to investigate the rates of ICU admission following in-hospital cardiac arrest/periarrest, evaluate the outcomes of such patients and assess whether anticipatory care planning had taken place prior to the adult resuscitation team being called.MethodsAnalysis of all referrals to the ICU page-holder within our district general hospital is between 1st November 2018 and 31st May 2019. From this, the frequency of adult resuscitation team calls was determined. Case notes were then reviewed to determine details of the events, patient outcomes and the use of anticipatory care planning tools on wards.ResultsOf the 506 referrals to the ICU page-holder, 141 (27.9%) were adult resuscitation team calls (114 periarrests and 27 cardiac arrests). Twelve patients were excluded due to health records being unavailable. Admission rates to ICU were low — 17.4% for cardiac arrests (4/23 patients), 5.7% (6/106) following periarrest. The primary reason for not admitting to ICU was patients being “too well” at the time of review (78/129 — 60.5%). Prior to adult resuscitation team call, treatment escalation plans had been completed in 27.9% (36/129) with Do Not Attempt Cardiopulmonary Resuscitation (DNACPR) forms present in 15.5% of cases (20/129). Four cardiac arrest calls were made in the presence of a valid DNACPR form, frequently due to a lack of awareness of the patient’s resuscitation status.ConclusionsThis study highlights the significant workload for the ICU page-holder brought about by adult resuscitation team calls. There is a low admission rate from these calls, and, at the time of resuscitation team call, anticipatory planning is frequently either incomplete or poorly communicated. Addressing these issues requires a collaborative approach between ICU and non-ICU physicians and highlights the need for larger studies to develop scoring systems to aid objective admission decision-making.

Cardio pulmonary resuscitation (CPR) in the frail and those with multiple health conditions: Outcomes before and during the COVID pandemic

Coronavirus 2019 (COVID-19)-era resuscitation guidelines advised personal protective equipment before chest compressions and proactive advanced care planning. We investigated the impact of COVID-19 on cardiopulmonary resuscitation (CPR) outcomes according to scoring of frailty and of multiple health conditions. A retrospective single-centre analysis of clinical and electronic records for all adult cardiac arrest calls on wards between June 2020 and June 2021 was performed. Data were compared with a cohort pre-COVID (March 2017-March 2018). In total, 62 patients received CPR in 2020-21 compared with 113 in 2017-18. Similar rates of return of spontaneous circulation (ROSC) and a statistically insignificant survival increase from 23.8% to 32.2% (p=0.210). There were linear relationships between Clinical Frailty Scale (CFS) or Charlson Comorbidity Index (CCI) and diminished survival in the pooled data (both p<0.001). Both increasing frailty (measured by CFS) and comorbidity (measured by CCI) were associated with reduced survival from CPR. However, survival and ROSC during COVID-19 were no worse than before the pandemic.

77 Introducing a Do Not Attempt Cardio-Pulmonary Resuscitation (DNACPR) and treatment escalation plan (TEP) guideline in an acute hospital.

Abstract Background In hospital cardiopulmonary resuscitation has a 13–20% survival rate to hospital discharge. Do Not Attempt Cardiopulmonary Resuscitation (DNACPR) orders reduce resuscitation activity in people who are unlikely to survive due to underlying co-morbidities. The project aimed to identify current DNACPR practice and cardiac arrest calls in an acute care hospital and to improve communicating and documentation of DNACPR decisions. Methods A baseline audit of 20 medical charts identified variable practice on documenting DNACPR, inappropriate cardiac arrest calls and poor communication between the multidisciplinary team and patients on DNACPR decisions. A sub-committee of Medical Clinical Governance developed a hospital wide Guideline and Proforma for DNACPR that included a treatment escalation plan (TEP). Using quality improvement methodology and plan-do-study-act cycles the DNACPR proforma and decision support framework was refined and tested. Implementation strategies included stakeholder engagement, education, re-formatting documentation and recruitment of ward champions over a six month period. Post implementation audit measured compliance with completing the proforma and resuscitation calls. Results Clinical audit in February 2023 of 33 patient medical records identified 100% (n = 33) of patients with a DNACPR decision had the correct form in use. 96% documented communication with patients. All proformas were signed by a senior decision maker, included a TEP and were filed correctly in the patients notes. There was a 25% reduction in cardiac arrest calls. Only 56% of proformas were signed by nursing staff. Conclusion Implementing a standardised DNACPR proforma was feasible and improved documentation, visibility and ease of locating DNACPR decisions. It promoted timely communication with patients and families setting realistic goals of care in the TEP. In turn there has been a reduction in cardiac arrest calls. Nurses role in the process remains sub-optimal and there is a need to understand the experience of patients and families.

Swedish emergency medical dispatch centres’ ability to answer emergency medical calls and dispatch an ambulance in response to out-of-hospital cardiac arrest calls in accordance with the American Heart Association performance goals: An observational study

AimTo investigate the ability of Swedish Emergency Medical Dispatch Centres (EMDCs) to answer medical emergency calls and dispatch an ambulance for out-of-hospital cardiac arrest (OHCA) in accordance with the American Heart Association (AHA) performance goals in a 1-step (call connected directly to the EMDC) and a 2-step (call transferred to regional EMDC) procedure over 10 years, and to assess whether delays may be associated with 30-day survival. MethodObservational data from the Swedish Registry for Cardiopulmonary Resuscitation and EMDC. ResultsA total of 9,174,940 medical calls were answered (1-step). The median answer delay was 7.3 s (interquartile range [IQR], 3.6–14.5 s). Furthermore, 594,008 calls (6.1%) were transferred in a 2-step procedure, with a median answer delay of 39 s (IQR, 30–53 s). A total of 45,367 cases (0.5%, 1-step) were registered as OHCA, with a median answer delay of 7.2 s (IQR, 3.6–14.1 s) (AHA high-performance goal, 10 s). For 1-step procedure, no difference in 30-day survival was found regarding answer delay. For OHCA (1-step), an ambulance was dispatched after a median of 111.9 s (IQR, 81.7–159.9 s). Thirty-day survival was 10.8% (n = 664) when an ambulance was dispatched within 70 s (AHA high-performance) versus 9.3% (n = 2174) > 100 s (AHA acceptable) (p = 0.0013). Outcome data in the 2-step procedure was unobtainable. ConclusionThe majority of calls were answered within the AHA performance goals. When an ambulance was dispatched within the AHA high-performance standard in response to OHCA calls, survival was higher compared with calls when dispatch was delayed.

"If you miss that first step in the chain of survival, there is no second step"-Emergency ambulance call-takers' experiences in managing out-of-hospital cardiac arrest calls.

When a person has an out-of-hospital cardiac arrest (OHCA), calling the ambulance for help is the first link in the chain of survival. Ambulance call-takers guide the caller to perform life-saving interventions on the patient before the paramedics arrive at the scene, therefore, their actions, decisions and communication are integral to saving the patient's life. In 2021, we conducted open-ended interviews with 10 ambulance call-takers with the aim of understanding their experiences of managing these phone calls; and to explore their views on using a standardised call protocol and triage system for OHCA calls. We took a realist/essentialist methodological approach and applied an inductive, semantic and reflexive thematic analysis to the interview data to yield four main themes expressed by the call-takers: 1) time-critical nature of OHCA calls; 2) the call-taking process; 3) caller management; 4) protecting the self. The study found that call-takers demonstrated deep reflection on their roles in, not only helping the patient, but also the callers and bystanders to manage a potentially distressing event. Call-takers expressed their confidence in using a structured call-taking process and noted the importance of skills and traits such as active listening, probing, empathy and intuition, based on experience, in order to supplement the use of a standardised system in managing the emergency. This study highlights the often under-acknowledged yet critical role of the ambulance call-taker in being the first member of an emergency medical service that is contacted in the event of an OHCA.

Barriers to the Initiation of Telecommunicator-CPR during 9-1-1 Out-of-Hospital Cardiac Arrest Calls: A Qualitative Study

Introduction Fewer than 10% of individuals who suffer out-of-hospital cardiac arrest (OHCA) survive with good neurologic function. Bystander CPR more than doubles the chance of survival, and telecommunicator-CPR (T-CPR) during a 9-1-1 call substantially improves the frequency of bystander CPR. Objective We examined the barriers to initiation of T-CPR. Methods We analyzed the 9-1-1 call audio from 65 EMS-treated OHCAs from a single US 9-1-1 dispatch center. We initially conducted a thematic analysis aimed at identifying barriers to the initiation of T-CPR. We then conducted a conversation analysis that examined the interactions between telecommunicators and bystanders during the recognition phase (i.e., consciousness and normal breathing). Results We identified six process themes related to barriers, including incomplete or delayed recognition assessment, delayed repositioning, communication gaps, caller emotional distress, nonessential questions and assessments, and caller refusal, hesitation, or inability to act. We identified three suboptimal outcomes related to arrest recognition and delivery of chest compressions, which are missed OHCA identification, delayed OHCA identification and treatment, and compression instructions not provided following OHCA identification. A primary theme observed during missed OHCA calls was incomplete or delayed recognition assessment and included failure to recognize descriptors indicative of agonal breathing (e.g., “snoring”, “slow”) or to confirm that breathing was effective in an unconscious victim. Conclusions We observed that modifiable barriers identified during 9-1-1 calls where OHCA was missed, or treatment was delayed, were often related to incomplete or delayed recognition assessment. Repositioning delays were a common barrier to the initiation of chest compressions.

Abstract 293: Swedish Emergency Medical Dispatch Centres Ability To Answer And Manage Out Of Hospital Cardiac Arrest Calls In Accordance With The American Heart Association Performance Goals

Background: In witnessed OHCA each minute of delay to CPR reduce the probability of survival. Thus, Emergency Medical Dispatch Centres (EMDC) ability to rapidly answer the emergency call and dispatch EMS is crucial. The American Heart Association (AHA) states that the goal of public safety answering points in a high performing system is to answer OHCA calls within 10 seconds (s) and dispatch the first EMS unit within 60 s. Aim: The aim of the study was to 1) describe the EMDC call answer delay and 2) time to dispatch of the first EMS unit in OHCA calls in the region of Stockholm, during 10 years in accordance with the AHA goals for OHCA call handling. We also aimed to 3) investigate the effect of “late responders” on 30-day survival. Method: This is an observational study based on OHCA data from the region of Stockholm, Sweden, between year 2012 to 2021. Data was collected from the Swedish Registry for Cardiopulmonary Resuscitation (SRCR) and SOS Alarm AB database. Results: During the 10-year period, the EMDC answered a total of 12 298 calls with a corresponding OHCA-record in the SRCR. Complete data was available in 11 484 calls (93%) which were included in the analysis. Of these, 62% (n=7127) of OHCA calls were answered within 10 s, 20% (n=2306) in 10-20 s and 18% (n=2051) above 20 s. The median delay was 7.3 s [IQR 3.3, 15.1]. Dispatch of the first EMS unit was performed within 60 s in 9% (1053 of 11470), 23% (2640 of 11470) between 60-90 s and 68% (7777 of 11470) above 90 s, with a median delay of 114.5 s [IQR 81.4, 168.6]. Differences in call answer delay and time to dispatch of the first EMS unit over the 10-years are presented in Figure 1. Survival at 30-days was 11% (109 of 990) when EMS was dispatched within 60 s, 8.7% (215 of 2459) in 60-90 s and 10.3% (746 of 7210) above 90 s. Conclusion: In the region Stockholm the majority of emergency calls concerning OHCA were answered within the high performance system goal of 10 seconds. Additional effort is needed to achieve the 60 second goal for dispatch of the first EMS unit.

Curing the pandemic of misinformation on COVID-19 mRNA vaccines through real evidence-based medicine - Part 1

BackgroundIn response to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), several new pharmaceutical agents have been administered to billions of people worldwide, including the young and healthy at little risk from the virus. Considerable leeway has been afforded in terms of the pre-clinical and clinical testing of these agents, despite an entirely novel mechanism of action and concerning biodistribution characteristics.AimTo gain a better understanding of the true benefits and potential harms of the messenger ribonucleic acid (mRNA) coronavirus disease (COVID) vaccines.MethodsA narrative review of the evidence from randomised trials and real world data of the COVID mRNA products with special emphasis on BionTech/Pfizer vaccine.ResultsIn the non-elderly population the “number needed to treat” to prevent a single death runs into the thousands. Re-analysis of randomised controlled trials using the messenger ribonucleic acid (mRNA) technology suggests a greater risk of serious adverse events from the vaccines than being hospitalised from COVID-19. Pharmacovigilance systems and real-world safety data, coupled with plausible mechanisms of harm, are deeply concerning, especially in relation to cardiovascular safety. Mirroring a potential signal from the Pfizer Phase 3 trial, a significant rise in cardiac arrest calls to ambulances in England was seen in 2021, with similar data emerging from Israel in the 16–39-year-old age group.ConclusionIt cannot be said that the consent to receive these agents was fully informed, as is required ethically and legally. A pause and reappraisal of global vaccination policies for COVID-19 is long overdue.ContributionThis article highlights the importance of addressing metabolic health to reduce chronic disease and that insulin resistance is also a major risk factor for poor outcomes from COVID-19.

Increased emergency cardiovascular events among under-40 population in Israel during vaccine rollout and third COVID-19 wave

Cardiovascular adverse conditions are caused by coronavirus disease 2019 (COVID-19) infections and reported as side-effects of the COVID-19 vaccines. Enriching current vaccine safety surveillance systems with additional data sources may improve the understanding of COVID-19 vaccine safety. Using a unique dataset from Israel National Emergency Medical Services (EMS) from 2019 to 2021, the study aims to evaluate the association between the volume of cardiac arrest and acute coronary syndrome EMS calls in the 16–39-year-old population with potential factors including COVID-19 infection and vaccination rates. An increase of 25% was detected in both call types during January–May 2021, compared with the years 2019–2020. Using Negative Binomial regression models, the weekly emergency call counts were significantly associated with the rates of 1st and 2nd vaccine doses administered to this age group but were not with COVID-19 infection rates. While not establishing causal relationships, the findings raise concerns regarding vaccine-induced undetected severe cardiovascular side-effects and underscore the already established causal relationship between vaccines and myocarditis, a frequent cause of unexpected cardiac arrest in young individuals. Surveillance of potential vaccine side-effects and COVID-19 outcomes should incorporate EMS and other health data to identify public health trends (e.g., increased in EMS calls), and promptly investigate potential underlying causes.

The trajectory of repairs in the defibrillator sequence during emergency cardiac arrest calls – Balancing progressivity and intersubjectivity

Due to the urgent, time-sensitive nature of interactions in emergency ambulance phone calls, dealing with repairs (communication trouble) can be challenging. We investigate a critical medical emergency known as out-of-hospital cardiac arrest (OHCA) and focus on how ambulance call-takers handle repairs during an interactive sequence concerning the retrieval of automatic external defibrillators (AED). Clear communication about AEDs is vital, because the device can deliver a life-saving shock to an OHCA patient’s heart. We examined repair initiations, and their subsequent trajectories, during the defibrillator sequences in 58 OHCA emergency calls. We found evidence of competing influences in resolving such repairs: (1) providing a repair solution (including ensuring caller comprehension of what a defibrillator is) to achieve intersubjectivity that could resolve the question of defibrillator availability; or (2) progressing the call as swiftly as possible to an immediately applicable life-saving intervention such as cardiopulmonary resuscitation. The findings suggest that in certain institutional contexts, such as emergency medical service dispatch, the resolution of repairs in communication can take varying trajectories in order to achieve the most feasible goal in immediate time. We suggest that emergency medical services consider these trajectories in helping ambulance call-takers anticipate repairs in OHCA calls.

A 6-Year Thematic Review of Reported Incidents Associated With Cardiopulmonary Resuscitation Calls in a United Kingdom Hospital.

Critical incident reporting can be applied to cardiopulmonary resuscitation (CPR) events as a means of reducing further occurrences. We hypothesized that local CPR-related events might follow patterns only seen after a long period of analysis. We reviewed 6 years of local incidents associated with cardiac arrest calls. The following search terms were used to identify actual or potential resuscitation events: "resuscitation," "cardio-pulmonary," "CPR," "arrest," "heart attack," "DNR," "DNAR," "DNACPR," "Crash," "2222." All identified incidents were independently reviewed and categorized, looking for identifiable patterns. Nottingham University Hospitals is a large UK tertiary referral teaching hospital. A total of 1017 reports were identified, relating to 1069 categorizable incidents. During the same time, there were approximately 1350 cardiac arrest calls, although it should be noted that many arrest-related incidents were not associated with cardiac arrest call (e.g., failure to have the correct equipment available in the event of a cardiac arrest). Incidents could be broadly classified into 10 thematic areas: no identifiable incident (n = 189; 18%), failure to rescue (n = 133; 12%), staffing concerns (n = 134; 13%), equipment/drug concerns (n = 133; 12%), communication issues (n = 122; 10%), do-not-attempt-CPR decisions (n = 101; 9%), appropriateness of patient location or transfer (n = 96; 9%), concerns that the arrest may have been iatrogenic (n = 76; 7%), patient or staff injury (n = 43; 4%), and miscellaneous (n = 52; 5%). Specific patterns of events were seen within each category. By reviewing incidents, we were able to identify patterns only noticeable over a long time frame, which may be amenable to intervention. Our findings may be generalizable to other centers or encourage others to undertake this exercise themselves.

Early Changes in Patient Access Interval during the COVID-19 Pandemic

Background: COVID-19 was first reported in the United States in January 2020. Its spread throughout the country required EMS systems to rapidly adapt to patient needs while protecting EMS personnel. EMS agencies developed protocols requiring personnel to don enhanced personal protective equipment prior to patient contact. We hypothesized that the Patient Access Interval (PAI), defined as the time from wheels stopped on scene to initial patient contact, had increased during the COVID pandemic. This had the potential to affect patient outcomes, particularly in time-sensitive emergencies such as cardiac arrest or respiratory distress. Methods: This retrospective cohort study used commercial ambulance data from the four largest cities in Connecticut at two different time points: (Pre-COVID) March-May 2019, and (COVID) March-May 2020. PAI was calculated from contemporaneously reported scene times. Total cases were analyzed, and sub-analyses performed for calls located at extended care facilities (ECFs), for all emergent (Echo/Delta) calls, and for medical cardiac arrest calls. Results: 92,846 total cases were evaluated: 50,083 from 2019, and 42,763 from 2020. Cases that did not include necessary time data for PAI were removed, yielding 75,796 total cases (41,852 from 2019, 33,944 from 2020). The average PAI increased from 1 minute 55 seconds (1 m:55s) Pre-COVID to 2 m:18s COVID. ECF PAI increased from 2 m:39s to 3 m:42s. Echo/Delta PAI increased from 1 m:42s to 2 m:07s. Medical cardiac arrest PAI increased from 1 m:27s to 2 m:04s, and ECF cardiac arrest PAI increased from 2 m:18s to 4 m:35s (all comparisons p < 0.01). Conclusions: There were statistically significant increases in all studied PAIs during COVID. The 23 second increase in PAI for all calls may not have been clinically significant in most cases; however, for life-threatening patient presentations, the increase may have been particularly relevant. The increased PAI was compounded in the ECF environment, possibly due to state-mandated screening and temperature checks of EMS personnel before entering facilities. This was highlighted in the ECF cardiac arrest data, which demonstrated a clinically significant increase in PAI of 2m:17s. While this study was limited by the accuracy of contemporaneous time reports by EMS, the results support our hypothesis that PAI had increased during the COVID pandemic.

701 Resuscitation and Escalation Status Documentation in Acute Surgical Admissions: A Re-Audit Feb 2020

Abstract Aim Resuscitation and escalation of care are not always in the best interests of all patients. Planning ahead for patient’s escalation and CPR status, along with clear documentation can prevent inappropriate cardiac arrest calls. Previous audits on CPR and escalation status were conducted in 2017, 18 and 19 and the completion of CPR and escalation documentation was 0, 2 and 12% prospectively, with an improvement to 42% after an intervention in the 2019 audit. A re-audit for 2020 is conducted. Method A snapshot audit of patients who were admitted or referred to General Surgery between 10-17 Feb 20 were identified, admission record and medical notes reviewed. CPR and escalation status, presence of a DNACPR form, record of discussion with the patient or their family and the medical notes entry were reviewed. Results 46.6% (34/73) of admissions had CPR and escalation status completed. Out of the admissions, all patients who were deemed not for resuscitation had DNACPR forms in their admission notes. All of these were discussed with either the patient or with their family, and the decisions appropriately documented in their medical notes. Conclusions This re-audit had shown improvement in the documentation on escalation and CPR status. This area can be improved further. Our recommendations include: 1. To present this finding in the education meeting and 2. Junior doctors to highlight the documentation to the oncall consultant/ or registrar during the first senior review and 3. To conduct a questionnaire to explore the surgeons’ views on this subject.

Early recognition of a caller’s emotion in out-of-hospital cardiac arrest dispatching: An artificial intelligence approach

AimThis study aimed to develop an AI model for detecting a caller’s emotional state during out-of-hospital cardiac arrest calls by processing audio recordings of dispatch communications. MethodsAudio recordings of 337 out-of-hospital cardiac arrest calls from March-April 2011 were retrieved. The callers’ emotional state was classified based on the emotional content and cooperative scores. Mel-frequency cepstral coefficients extracted essential information from the voice signals. A support vector machine was utilised for the automatic judgement, and repeated random sub-sampling cross validation (RRS-CV) was applied to evaluate robustness. The results from the artificial intelligence classifier were compared with the consensus of expert reviewers. ResultsThe audio recordings were classified into five emotional content and cooperative score levels. The proposed model had an average positive predictive value of 72.97%, a negative predictive value of 93.47%, sensitivity of 38.76%, and specificity of 98.29%. If only the first 10 seconds of the recordings were considered, it had an average positive predictive value of 84.62%, a negative predictive value of 93.57%, sensitivity of 52.38%, and specificity of 98.64%. The artificial intelligence model’s performance maintained preferable results for emotionally stable cases. ConclusionArtificial intelligence models can possibly facilitate the judgement of callers’ emotional states during dispatch conversations. This model has the potential to be utilised in practice, by pre-screening emotionally stable callers, thus allowing dispatchers to focus on cases that are judged to be emotionally unstable. Further research and validation are required to improve the model's performance and make it suitable for the general population.

Utilizing timestamps of longitudinal electronic health record data to classify clinical deterioration events.

To propose an algorithm that utilizes only timestamps of longitudinal electronic health record data to classify clinical deterioration events. This retrospective study explores the efficacy of machine learning algorithms in classifying clinical deterioration events among patients in intensive care units using sequences of timestamps of vital sign measurements, flowsheets comments, order entries, and nursing notes. We design a data pipeline to partition events into discrete, regular time bins that we refer to as timesteps. Logistic regressions, random forest classifiers, and recurrent neural networks are trained on datasets of different length of timesteps, respectively, against a composite outcome of death, cardiac arrest, and Rapid Response Team calls. Then these models are validated on a holdout dataset. A total of 6720 intensive care unit encounters meet the criteria and the final dataset includes 830 578 timestamps. The gated recurrent unit model utilizes timestamps of vital signs, order entries, flowsheet comments, and nursing notes to achieve the best performance on the time-to-outcome dataset, with an area under the precision-recall curve of 0.101 (0.06, 0.137), a sensitivity of 0.443, and a positive predictive value of 0. 092 at the threshold of 0.6. This study demonstrates that our recurrent neural network models using only timestamps of longitudinal electronic health record data that reflect healthcare processes achieve well-performing discriminative power.

Comparison of Emergency Medical Dispatch Systems for Performance of Telecommunicator-Assisted Cardiopulmonary Resuscitation Among 9-1-1 Callers With Limited English Proficiency

Increasing bystander cardiopulmonary resuscitation (CPR) among racial/ethnic minority groups and culturally underserved populations is a key strategy in improving health care disparities in out-of-hospital cardiac arrest. To ascertain whether implementation of the Los Angeles Tiered Dispatch System (LA-TDS) was associated with improved performance of telecommunicator-assisted CPR (T-CPR) among 9-1-1 callers with limited English proficiency in the City of Los Angeles. This cohort study compared emergency medical services-treated, nontraumatic out-of-hospital cardiac arrest calls using the Medical Priority Dispatch System (MPDS) from January 1 to March 31, 2014, with calls using LA-TDS from January 1 to March 31, 2015. Trained data abstractors evaluated all 9-1-1 audio recordings for the initiation of T-CPR and the elapsed time to predefined events. Data were analyzed between January and December 2017. The primary outcome was the prevalence of T-CPR among 9-1-1 callers with limited English proficiency for field-confirmed nontraumatic cardiac arrests. Additional outcomes included T-CPR among callers with English proficiency and the elapsed time until key events in the call. Of the 1027 emergency medical services calls during the study periods, 597 met the inclusion criteria. A total of 289 calls (48%) were made using MPDS (263 callers with English proficiency, and 26 callers with limited English proficiency), and 308 calls (52%) were made using LA-TDS (273 callers with English proficiency, and 35 callers with limited English proficiency). No differences between MPDS and LA-TDS cohorts were found in age, sex, known comorbidities, arrest location (private vs public), or witnessed status. The prevalence of T-CPR among callers with limited English proficiency was significantly greater using LA-TDS (69%) vs MPDS (28%) (odds ratio [OR], 5.66; 95% CI, 1.79-17.85; P = .003). For callers with English proficiency, the prevalence of T-CPR improved from 55% using MPDS to 67% using LA-TDS (OR, 1.66; 95% CI, 1.15-2.41; P = .007). With LA-TDS, callers with limited English proficiency had a significant decrease in time to recognition of cardiac arrest (OR, 0.59; 95% CI, 0.41-0.84; P = .005) and dispatch of resources (OR, 0.71; 95% CI, 0.54-0.94; P = .02). The LA-TDS compared with MPDS was associated with increased performance of T-CPR for out-of-hospital cardiac arrests involving 9-1-1 callers with limited English proficiency. Further studies are needed in communities with a predominance of people with limited English proficiency to characterize bystander response, promote activation of the chain of survival, and clarify the precise elements of LA-TDS that can improve T-CPR performance.

Laboratory-derived early warning score for the prediction of in-hospital mortality, intensive care unit admission, medical emergency team activation and cardiac arrest in general medical wards.

General medical wards admit a varied cohort of patients from the emergency department, some of whom deteriorate during their hospital stay. Currently, we use vital signs based warning scores to predict patients at risk of imminent deterioration, but there is now a growing body of literature that commonly available laboratory results may also help to identify those at risk. To assess whether a laboratory-based admission score can predict in hospital mortality, intensive care unit (ICU) admission, medical emergency team (MET) activation or cardiac arrest in a cohort of Australian general medical patients admitted through the emergency department (ED). We performed a retrospective observational study of all general medical admissions to hospital through the ED in 2015. Admission pathology was used to calculate a risk score. In-patient outcomes of death, ICU transfer, MET call activation or cardiac arrest were collected from hospital records. We studied 2942 admissions derived from 2521 patients, with a median age of 81 years. There were 143 in-patient deaths, 82 ICU admissions, 277 MET calls and 14 cardiac arrest calls. The laboratory-based admission score had an area under the receiver operating characteristic curve (AUC-ROC) of 0.76 (95% confidence interval (CI): 0.72-0.80) for inpatient death, an AUC-ROC of 0.79 (95% CI: 0.66-0.93) for inpatient cardiac arrest, an AUC-ROC of 0.64 (95% CI: 0.58-0.70) for ICU transfer and an AUC-ROC of 0.59 (95% CI: 0.55-0.62) for MET call activation. When patients aged over 75 were analysed separately, the AUC-ROC for prediction of in-patient death was 0.74 (95% CI: 0.70-0.78) and increased to 0.86 (95% CI: 0.73-0.98) for the prediction of in-patient cardiac arrest. A simple laboratory-derived score obtained at patient admission is a fair to good predictor of subsequent in-patient death or cardiac arrest in general medical patients and in the older patient cohort. Prospective interventional studies are required to ascertain the clinical utility of this admission score.