Comparing outcomes of out-of-hospital cardiac arrest patients with initial shockable rhythm in Singapore and Osaka using population-based databases.

BACKGROUND: Previous research indicated outcomes among refractory out-of-hospital cardiac arrest (OHCA) patients with initial shockable rhythm were different in Singapore and Osaka, Japan, possibly due to the differences in access to extracorporeal cardiopulmonary resuscitation. However, this previous study had a risk of selection bias. To address this concern, this study aimed to evaluate the outcomes between Singapore and Osaka for OHCA patients with initial shockable rhythm using only population-based databases. METHODS: This was a secondary analysis of two OHCA population-based databases in Osaka and Singapore, including adult OHCA patients with initial shockable rhythm. A machine-learning-based prediction model was derived from the Osaka data (n = 3088) and applied to the PAROS-SG data (n = 2905). We calculated the observed-expected ratio (OE ratio) for good neurological outcomes observed in Singapore and the expected derived from the data in Osaka by dividing subgroups with or without prehospital ROSC. RESULTS: The one-month good neurological outcomes in Osaka and Singapore among patients with prehospital ROSC were 70% (791/1,125) and 57% (440/773), and among patients without prehospital ROSC were 10% (196/1963) and 2.8% (60/2,132). After adjusting patient characteristics, the outcome in Singapore was slightly better than expected from Osaka in patients with ROSC (OE ratio, 1.067 [95%CI 1.012 to 1.125]), conversely, it was worse than expected in patients without prehospital ROSC (OE ratio, 0.238 [95%CI 0.173 to 0.294]). CONCLUSION: This study showed the outcomes of OHCA patients without prehospital ROSC in Singapore were worse than expected derived from Osaka data even using population-based databases. (249/250 words).

Outcome assessment for out-of-hospital cardiac arrest patients in Singapore and Japan with initial shockable rhythm.

BACKGROUND: Singapore and Osaka in Japan have comparable population sizes and prehospital management; however, the frequency of ECPR differs greatly for out-of-hospital cardiac arrest (OHCA) patients with initial shockable rhythm. Given this disparity, we hypothesized that the outcomes among the OHCA patients with initial shockable rhythm in Singapore were different from those in Osaka. The aim of this study was to evaluate the outcomes of OHCA patients with initial shockable rhythm in Singapore compared to the expected outcomes derived from Osaka data using machine learning-based prediction models. METHODS: This was a secondary analysis of two OHCA databases: the Singapore PAROS database (SG-PAROS) and the Osaka-CRITICAL database from Osaka, Japan. This study included adult (18-74 years) OHCA patients with initial shockable rhythm. A machine learning-based prediction model was derived and validated using data from the Osaka-CRITICAL database (derivation data 2012-2017, validation data 2018-2019), and applied to the SG-PAROS database (2010-2016 data), to predict the risk-adjusted probability of favorable neurological outcomes. The observed and expected outcomes were compared using the observed-expected ratio (OE ratio) with 95% confidence intervals (CI). RESULTS: From the SG-PAROS database, 1,789 patients were included in the analysis. For OHCA patients who achieved return of spontaneous circulation (ROSC) on hospital arrival, the observed favorable neurological outcome was at the same level as expected (OE ratio: 0.905 [95%CI: 0.784-1.036]). On the other hand, for those who had continued cardiac arrest on hospital arrival, the outcomes were lower than expected (shockable rhythm on hospital arrival, OE ratio: 0.369 [95%CI: 0.258-0.499], and nonshockable rhythm, OE ratio: 0.137 [95%CI: 0.065-0.235]). CONCLUSION: This observational study found that the outcomes for patients with initial shockable rhythm but who did not obtain ROSC on hospital arrival in Singapore were lower than expected from Osaka. We hypothesize this is mainly due to differences in the use of ECPR.

myResponder Smartphone Application to Crowdsource Basic Life Support for Out-of-Hospital Cardiac Arrest: The Singapore Experience.

OBJECTIVE: The Singapore myResponder is a novel smartphone application developed by the Singapore Civil Defence Force (SCDF) that notifies volunteer first responders of a suspected out-of-hospital cardiac arrest (OHCA) case and locations of Automated External Defibrillators (AED) in the vicinity so that they can assist with resuscitation. We aimed to examine the performance of this application, challenges encountered, and future directions. Methods: We analyzed data from the myResponder app since its launch from April 2015 to July 2019. The number of installations, registered community first responders, suspected OHCA cases, notifications sent by the app, percentage of responders who accepted activation and arrived at scene were reviewed. A subgroup of taxi driving responders (within a 1.5-kilometer response radius) carrying an AED under a subsequent pilot program was also analyzed. Results: By July 2019, 46,689 responders were registered in the myResponder app. There were a total of 19,189 cases created for suspected OHCA, with a median of 358 cases per month (IQR 330-430), in which 10,073 responders accepted activation from myResponder and 4,955 arrived on-scene. A total of 135,599 notifications were sent for these cases, with a median of 7.1 notifications per case (IQR 4.3-8.7). In 2019, the percentages of responders who accepted notification and arrived on scene were 45.8% and 24.1%, respectively. 43% (1110/2581) of responders arrived before EMS crew. Conclusion: The myResponder mobile application is a feasible smart technology solution to improve community response to OHCA, and to increase bystander CPR and AED use. Future directions include increasing the number of active responders, improving response rates, app performance, and better data capture for quality improvement.

Improved Out-of-Hospital Cardiac Arrest Survival with a Comprehensive Dispatcher-Assisted CPR Program in a Developing Emergency Care System.

OBJECTIVE: Out-of-hospital cardiac arrest (OHCA) is associated with poor survival. Studies have demonstrated improved survival with early bystander cardiopulmonary resuscitation (BCPR). This study evaluated the impact of a dispatcher-assisted CPR (DA-CPR) program on BCPR rate and outcomes of OHCA in a developing emergency medical services (EMS) system setting. METHODS: Data were extracted from the national cardiac arrest registry. A before-after analysis was performed between OHCA cases with cardiac etiology conveyed by EMS from April 2010-June 2012 (pre-intervention) and July 2012-December 2015 (post-intervention). Primary outcomes were survival-to-discharge/30 days post-arrest and favorable cerebral performance (Glasgow-Pittsburgh cerebral performance categories 1 and 2). RESULTS: 6365 OHCA cases were analyzed with 2129 in the pre-intervention and 4236 in the post-intervention group. In the post-intervention group, there was an increase in BCPR rates from 24.8% to 53.8% (p < 0.001), adjusted OR 3.67 (aOR; 95%CI: 3.26-4.13). OHCA outcomes also improved with survival-to-discharge rates increasing from 3.0%-4.5% (p < 0.01), aOR 2.10 (95%CI: 1.40-3.17) and favorable cerebral performance increasing from 1.6% to 2.7% (p < 0.05), aOR 2.82 (95%CI: 1.65-4.82). In patients with initial shockable rhythm, BCPR without dispatcher assistance was associated with significantly higher odds of survival-to-discharge (aOR 1.67, 95%CI: 1.06-2.64) and favorable cerebral performance (aOR 2.32, 95%CI: 1.26-4.27) compared to no BCPR. CONCLUSION: Our study showed that a simplified DA-CPR program can be successfully implemented in a developing EMS system and can contribute to higher BCPR rate and in turn, improve OHCA survival. Future studies can examine bystanders' characteristics and quality of the CPR performed to understand their impact on survival.

Emergency medical dispatch services across Pan-Asian countries: a web-based survey.

BACKGROUND: Dispatch services (DS's) form an integral part of emergency medical service (EMS) systems. The role of a dispatcher has also evolved into a crucial link in patient care delivery, particularly in dispatcher assisted cardio-pulmonary resuscitation (DACPR) during out-of-hospital cardiac arrest (OHCA). Yet, there has been a paucity of research into the emerging area of dispatch science in Asia. This paper compares the characteristics of DS's, and state of implementation of DACPR within the Pan-Asian Resuscitation Outcomes (PAROS) network. METHODS: A cross-sectional descriptive survey addressing population characteristics, DS structures and levels of service, state of DACPR implementation (including protocols and quality improvement programs) among PAROS DS's. RESULTS: 9 DS's responded, representing a total of 23 dispatch centres from 9 countries that serve over 80 million people. Most PAROS DS's operate a tiered dispatch response, have implemented medical oversight, and tend to be staffed by dispatchers with a predominantly medical background. Almost all PAROS DS's have begun tracking key EMS indicators. 77.8% (n = 7) of PAROS DS's have introduced DACPR. Of the DS's that have rolled out DACPR, 71.4% (n = 5) provided instructions in over one language. All DS's that implemented DACPR and provided feedback to dispatchers offered feedback on missed OHCA recognition. The majority of DS's (83.3%; n = 5) that offered DACPR and provided feedback to dispatchers also implemented corrective feedback, while 66.7% (n = 4) offered positive feedback. Compression-only CPR was the standard instruction for PAROS DS's. OHCA recognition sensitivity varied widely in PAROS DS's, ranging from 32.6% (95% CI: 29.9-35.5%) to 79.2% (95% CI: 72.9-84.4%). Median time to first compression ranged from 120 s to 220 s. CONCLUSIONS: We found notable variations in characteristics and state of DACPR implementation between PAROS DS's. These findings will lay the groundwork for future DS and DACPR studies in the PAROS network.

Characteristics of Frequent Users of Emergency Medical Services in Singapore.

OBJECTIVES: This study aims to describe frequent users of Emergency Medical Services (EMS) conveyed to a Singapore tertiary hospital, focusing on a comparison between younger users (age <65) and older users in diagnoses and admission rates. METHODS: All patients conveyed by EMS to a tertiary hospital 4 times or more over a 1-year period in 2015 had their EMS ambulance charts and Emergency Department (ED) electronic records retrospectively analyzed (n = 243), with admission the primary outcome. RESULTS: The 243 frequent users were analyzed with a combined total of 1,705 visits, out of a total of 10,183 patients with 12,839 visits conveyed by EMS to Singapore General Hospital (SGH) in 2015. Younger frequent users (<65 years age) were found to be predominantly male (79.6%, p = 0.001) and were on average responsible for more visits than elderly frequent users (8.6 vs. 5.7, p = 0.004). Medical co-morbidities were significantly more prevalent in older users. Younger frequent users were more likely to be smokers (60.2% vs. 22.3%), heavy drinkers (51.3% vs. 8.5%), substance abusers (12.4% vs. 0.8%), and bad debtors (49.6% vs. 20.0%, p < 0.001). A larger proportion presented with altered mental states (11.7% vs. 5.4%, p < 0.001) and alcohol related diagnoses (34.7% vs. 5.3%, p < 0.001). Many were picked up from public areas (45.5% vs. 19.6%, p < 0.001), and had lower acuity triage scores at both EMS (p < 0.001) and ED (p = 0.001). They had lower admission rates (40.5% vs. 78.7%, p < 0.001) and shorter length of stay (4.3 vs. 5.9 days, p < 0.001). Univariable and multivariable analysis showed alcohol related diagnoses, history of alcohol abuse and lower triage scores were less likely to require admissions. CONCLUSION: Frequent EMS users consume a disproportionate amount of healthcare resources. Two broad subgroups of patients were identified: younger patients with social issues and older patients with multiple medical conditions. EMS usage by older patients was significantly associated with higher rates of admission.

Comparison of Outcomes and Characteristics of Emergency Medical Services (EMS)-Witnessed, Bystander-Witnessed, and Unwitnessed Out-of-Hospital Cardiac Arrests in Singapore.

Objectives: The objective was to compare the survival outcomes of emergency medical services (EMS)-witnessed to bystander-witnessed, and unwitnessed out-of-hospital cardiac arrests (OHCA) in Singapore. Secondary aims are to describe the 5-year trend in survival rates of EMS-witnessed arrests. Methods: This was a retrospective analysis of the Singapore's OHCA registry data from 2011 to 2015. Excluded from the analysis were patients younger than 18 years old, arrests of traumatic etiology, resuscitation not attempted, and cases not conveyed by EMS. The primary outcome was survival to hospital discharge or 30 days post-arrest. Secondary outcomes were return of spontaneous circulation (ROSC) and survival to hospital admission. Results: 8,394 cases were analyzed, with 650 (7.7%) EMS-witnessed arrests, 4480 (53.4%) bystander-witnessed arrests, and 3264 (38.9%) unwitnessed arrests. Among EMS-witnessed arrests, the majority were presumed to be of cardiac etiology (62.8%) and the most common presenting rhythm was pulseless electrical activity (PEA; 57.2%). Survival to discharge or 30th day post-arrest was higher in EMS-witnessed arrests compared to bystander-witnessed and unwitnessed arrests (11.2% vs. 5.3% and 1.3%, p < 0.001). Survival to discharge for EMS-witnessed cases increased from 2011 (13.2%) to 2015 (18.9%). Conclusions: EMS-witnessed OHCAs were more likely to have favorable outcomes compared to bystander-witnessed and unwitnessed OHCAs. High PEA rates in EMS-witnessed arrests were associated with older patients with underlying preexisting medical conditions. Increasing public awareness on recognition of prodromal symptoms and early activation of EMS could improve post-arrest survival and neurological outcomes of OHCA.

Ethnic and Neighborhood Socioeconomic Differences In Incidence and Survival From Out-Of-Hospital Cardiac Arrest In Singapore.

Objective: We aimed to examine the association of ethnicity and socioeconomic status (SES) with Out-of-Hospital Cardiac Arrest (OHCA) incidence and 30-day survival in Singapore. Methods: We analyzed the Singapore cohort of Pan-Asia Resuscitation Outcome Study (PAROS), a multi-center, prospective OHCA registry between 2010 and 2015. The Singapore Socioeconomic Disadvantage Index (SEDI) score, obtained according to zip code, was used as surrogate for neighborhood SES. Age-adjusted OHCA incidence and Utstein survival were calculated by ethnicity and SES. Utstein survival was defined as the number of cardiac OHCA cases with initial rhythm of ventricular fibrillation witnessed by a bystander who survived 30-days or until hospital discharge. Logistic regression was used to investigate association of ethnicity with 30-day and Utstein survivals. Results: Our study population comprised 8,900 patients: 6,453 Chinese, 1,472 Malays, and 975 Indians. The overall age-adjusted incidence ratios (95% CI) for Malay/Chinese and Indian/Chinese were 1.93 (1.83-2.04) and 1.95 (1.83-2.08), respectively. The overall age-adjusted incidence ratios (95% CI) for average/low and high/low SEDI group were 1.12 (0.95-1.33) and 1.29 (1.08-1.53), respectively. Malay showed lesser Utstein survival of 8.1% compared to Chinese (14.6%) and Indian (20.4%) [p = 0.018]. Ethnicity did not reach statistical significance (p = 0.072) in forward selection model of Utstein survival, while SEDI score and category were not significant (p > 0.2 and p = 0.349). Conclusions: We found Malay and Indian communities to be at higher risks of OHCA compared to Chinese, and additionally, the Malay community is at higher risk of subsequent mortality than the Chinese and Indian communities. These disparities were not explained by neighborhood SES.

Psychological Comfort of Paramedics with Field Death Pronouncement: A National Asian Study to Prepare Paramedics for Field Termination of Resuscitation.

OBJECTIVES: Paramedics' decision to terminate field resuscitation without a physician present may depend on personal and external factors. This study investigates factors associated with paramedic psychological comfort with termination of resuscitation (TOR) to inform future training. METHODS: We administered an anonymous survey to all practicing paramedics in a large urban Asian Emergency Medical Services system where formal TOR training had not yet been conducted and field TOR was not routinely applied. The survey assessed psychological comfort using the validated Psychological Comfort Total (PCT) scale (summed score of 28 items, with higher scores representing greater comfort). We examined scores associated with four personal (prior resolution of personal loss, knowledge of survival probability, religious affiliation and experience with death pronouncements) and two external (location of patient and perceived trust of family) factors. Data were entered into Excel and analyzed by t-tests and ANOVA. RESULTS: Response rate was 73.6% (254/345). Respondents were 30.3 years (mean, SD 7.1) with 7.2 years (mean, SD 5.54) of practice experience. Over 60% had been involved in 6 or more field death pronouncements in the prior 12 months. Higher PCT scores were associated with prior resolution of personal loss and knowledge of survival probability. Lower PCT scores were associated with patient location in a public place and perceived family lack of trust. PCT scores were not associated with paramedic religious affiliation or number of prior death pronouncements. CONCLUSIONS: Paramedic psychological comfort with field death pronouncement is associated with personal and external factors. Since paramedic comfort is important for protocol adoption, TOR education should target not only knowledge, but also public arena management, communication skills for engaging with families, and help paramedics resolve prior personal loss.

Association of the Emergency Medical Services-Related Time Interval with Survival Outcomes of Out-of-Hospital Cardiac Arrest Cases in Four Asian Metropolitan Cities Using the Scoop-and-Run Emergency Medical Services Model.

BACKGROUND: Response time interval (RTI) and scene time interval (STI) are key time variables in the out-of-hospital cardiac arrest (OHCA) cases treated and transported via emergency medical services (EMS). OBJECTIVE: We evaluated distribution and interactive association of RTI and STI with survival outcomes of OHCA in four Asian metropolitan cities. METHODS: An OHCA cohort from Pan-Asian Resuscitation Outcome Study (PAROS) conducted between January 2009 and December 2011 was analyzed. Adult EMS-treated cardiac arrests with presumed cardiac origin were included. A multivariable logistic regression model with an interaction term was used to evaluate the effect of STI according to different RTI categories on survival outcomes. Risk-adjusted predicted rates of survival outcomes were calculated and compared with observed rate. RESULTS: A total of 16,974 OHCA cases were analyzed after serial exclusion. Median RTI was 6.0 min (interquartile range [IQR] 5.0-8.0 min) and median STI was 12.0 min (IQR 8.0-16.1). The prolonged STI in the longest RTI group was associated with a lower rate of survival to discharge or of survival 30 days after arrest (adjusted odds ratio [aOR] 0.59; 95% confidence interval [CI] 0.42-0.81), as well as a poorer neurologic outcome (aOR 0.63; 95% CI 0.41-0.97) without an increasing chance of prehospital return of spontaneous circulation (aOR 1.12; 95% CI 0.88-1.45). CONCLUSIONS: Prolonged STI in OHCA with a delayed response time had a negative association with survival outcomes in four Asian metropolitan cities using the scoop-and-run EMS model. Establishing an optimal STI based on the response time could be considered.

Termination of Resuscitation Rules to Predict Neurological Outcomes in Out-of-Hospital Cardiac Arrest for an Intermediate Life Support Prehospital System.

AIM: Futile resuscitation can lead to unnecessary transports for out-of-hospital cardiac arrest (OHCA). The Basic Life Support (BLS) and Advanced Life Support (ALS) termination of resuscitation (TOR) guidelines have been validated with good results in North America. This study aims to evaluate the performance of these two rules in predicting neurological outcomes of OHCA patients in Singapore, which has an intermediate life support Emergency Medical Services (EMS) system. METHODS: A retrospective cohort study was carried out on Singapore OHCA data collected from April 2010 to May 2012 for the Pan-Asian Resuscitation Outcomes Study (PAROS). The outcomes of each rule were compared to the actual neurological outcomes of the patients. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and predicted transport rates of each test were evaluated. RESULTS: A total of 2,193 patients had cardiac arrest of presumed cardiac etiology. TOR was recommended for 1,411 patients with the BLS-TOR rule, with a specificity of 100% (91.9, 100.0) for predicting poor neurological outcomes, PPV 100% (99.7, 100.0), sensitivity 65.7% (63.6, 67.7), NPV 5.6% (4.1, 7.5), and transportation rate 35.6%. Using the ALS-TOR rule, TOR was recommended for 587 patients, specificity 100% (91.9, 100.0) for predicting poor neurological outcomes, PPV 100% (99.4, 100.0), sensitivity 27.3% (25.4, 29.3), NPV 2.7% (2.0, 3.7), and transportation rate 73.2%. BLS-TOR predicted survival (any neurological outcome) with specificity 93.4% (95% CI 85.3, 97.8) versus ALS-TOR 98.7% (95% CI 92.9, 99.8). CONCLUSION: Both the BLS and ALS-TOR rules had high specificities and PPV values in predicting neurological outcomes, the BLS-TOR rule had a lower predicted transport rate while the ALS-TOR rule was more accurate in predicting futility of resuscitation. Further research into unique local cultural issues would be useful to evaluate the feasibility of any system-wide implementation of TOR.

Variation of current protocols for managing out-of-hospital cardiac arrest in prehospital settings among Asian countries.

BACKGROUND/PURPOSE: Protocols for managing patients with out-of-hospital cardiac arrest (OHCA) may vary due to legal, cultural, or socioeconomic concerns. We sought to assess international variation in policies and protocols related to OHCA. METHODS: A brief survey was developed by consensus. Elicited information included protocols for managing patients with nontraumatic OHCA or traumatic OHCA, policies for using automated external defibrillators (AEDs) during transportation of patients with ongoing resuscitation, and application of terminations of resuscitation (TOR) rules in prehospital settings in the respondent's city or country. The populations of interest were emergency physicians, medical directors of emergency medical services (EMS), and policy makers. RESULTS: Responses were obtained from eight cities in six Asian countries. Only one (12.5%) city applied TOR rules for OHCAs. Do-not-resuscitate (DNR) orders were valid in prehospital settings in five (62.5%) cities. All cities used AEDs for nontraumatic OHCAs; seven (87.5%) cities did not routinely use AEDs for traumatic OHCAs. For nontraumatic OHCAs, four (50%) cities performed 2 minutes of on-scene cardiopulmonary resuscitation (CPR) and then transported the patients with ongoing resuscitation to hospitals; three (37.5%) cities performed 4 minutes of on-scene CPR; one (12.5%) city allowed variation in the duration of on-scene CPR. CONCLUSION: International variation in practices and polices related to OHCAs do exist. Concerns regarding prehospital TOR rules include medical evidence, legal considerations, EMS manpower, public perception, medical oversight, education, EMS characteristics, and cost-effectiveness analysis. Further research is needed to achieve consensus regarding management protocols, especially for EMS that perform resuscitation during transportation of OHCA patients.

Prehospital Trauma Care in Singapore.

Prehospital emergency care in Singapore has taken shape over almost a century. What began as a hospital-based ambulance service intended to ferry medical cases was later complemented by an ambulance service under the Singapore Fire Brigade to transport trauma cases. The two ambulance services would later combine and come under the Singapore Civil Defence Force. The development of prehospital care systems in island city-state Singapore faces unique challenges as a result of its land area and population density. This article defines aspects of prehospital trauma care in Singapore. It outlines key historical milestones and current initiatives in service, training, and research. It makes propositions for the future direction of trauma care in Singapore. The progress Singapore has made given her circumstances may serve as lessons for the future development of prehospital trauma systems in similar environments. Key words: Singapore; trauma; prehospital emergency care; emergency medical services.

Dynamic ambulance reallocation for the reduction of ambulance response times using system status management.

OBJECTIVES: Dynamically reassigning ambulance deployment locations throughout a day to balance ambulance availability and demands can be effective in reducing response times. The objectives of this study were to model dynamic ambulance allocation plans in Singapore based on the system status management (SSM) strategy and to evaluate the dynamic deployment plans using a discrete event simulation (DES) model. METHODS: The geographical information system-based analysis and mathematical programming were used to develop the dynamic ambulance deployment plans for SSM based on ambulance calls data from January 1, 2011, to June 30, 2011. A DES model that incorporated these plans was used to compare the performance of the dynamic SSM strategy against static reallocation policies under various demands and travel time uncertainties. RESULTS: When the deployment plans based on the SSM strategy were followed strictly, the DES model showed that the geographical information system-based plans resulted in approximately 13-second reduction in the median response times compared to the static reallocation policy, whereas the mathematical programming-based plans resulted in approximately a 44-second reduction. The response times and coverage performances were still better than the static policy when reallocations happened for only 60% of all the recommended moves. CONCLUSIONS: Dynamically reassigning ambulance deployment locations based on the SSM strategy can result in superior response times and coverage performance compared to static reallocation policies even when the dynamic plans were not followed strictly.

Machine learning prediction of refractory ventricular fibrillation in out-of-hospital cardiac arrest using features available to EMS.

Background: Shock-refractory ventricular fibrillation (VF) or ventricular tachycardia (VT) is a treatment challenge in out-of-hospital cardiac arrest (OHCA). This study aimed to develop and validate machine learning models that could be implemented by emergency medical services (EMS) to predict refractory VF/VT in OHCA patients. Methods: This was a retrospective study examining adult non-traumatic OHCA patients brought into the emergency department by Singapore EMS from the Pan-Asian Resuscitation Outcomes Study (PAROS) registry. Data from April 2010 to March 2020 were extracted for this study. Refractory VF/VT was defined as VF/VT persisting or recurring after at least one shock. Features were selected based on expert clinical opinion and availability to dispatch prior to arrival at scene. Multivariable logistic regression (MVR), LASSO and random forest (RF) models were investigated. Model performance was evaluated using receiver operator characteristic (ROC) area under curve (AUC) analysis and calibration plots. Results: 20,713 patients were included in this study, of which 860 (4.1%) fulfilled the criteria for refractory VF/VT. All models performed comparably and were moderately well-calibrated. ROC-AUC were 0.732 (95% CI, 0.695 - 0.769) for MVR, 0.738 (95% CI, 0.701 - 0.774) for LASSO, and 0.731 (95% CI, 0.690 - 0.773) for RF. The shared important predictors across all models included male gender and public location. Conclusion: The machine learning models developed have potential clinical utility to improve outcomes in cases of refractory VF/VT OHCA. Prediction of refractory VF/VT prior to arrival at patient's side may allow for increased options for intervention both by EMS and tertiary care centres.

Corrigendum to "Development and Asian-wide validation of the Grade for Interpretable Field Triage (GIFT) for predicting mortality in pre-hospital patients using the Pan-Asian Trauma Outcomes Study (PATOS)" [The Lancet Regional Health - Western Pacific 34 (2023) 100733].

[This corrects the article DOI: 10.1016/j.lanwpc.2023.100733.].

Increasing neurologically intact survival after out-of-hospital cardiac arrest among elderly: Singapore Experience.

Objectives: With more elderly presenting with Out-of-Hospital Cardiac Arrests (OHCAs) globally, neurologically intact survival (NIS) should be the aim of resuscitation. We aimed to study the trend of OHCA amongst elderly in a large Asian registry to identify if age is independently associated with NIS and factors associated with NIS. Methods: All adult OHCAs aged ≥18 years attended by emergency medical services (EMS) from April 2010 to December 2019 in Singapore was extracted from the Pan-Asian Resuscitation Outcomes Study (PAROS) registry. Cases pronounced dead at scene, non-EMS transported, traumatic OHCAs and OHCAs in ambulances were excluded. Patient characteristics and outcomes were compared across four age categories (18-64, 65-79, 80-89, ≥90). Multivariable logistic regression analysis determined the factors associated with NIS. Results: 19,519 eligible cases were analyzed. OHCA incidence increased with age almost doubling in octogenarians (from 312/100,000 in 2011 to 652/100,000 in 2019) and tripling in those ≥90 years (from 458/100,000 in 2011 to 1271/100,000 in 2019). The proportion of patients with NIS improved over time for the 18-64, 65-79- and 80-89-years age groups, with the greatest improvement in the youngest group. NIS decreased with each increasing year of age and minute of response time. NIS increased in the arrests of presumed cardiac etiology, witnessed and bystander CPR. Conclusions: Survival with good outcomes has increased even amongst the elderly. Regardless of age, NIS is possible with good-quality CPR, highlighting its importance. End-of-life planning is a complex yet necessary decision that requires qualitative exploration with elderly, their families and care providers.

Variable importance analysis with interpretable machine learning for fair risk prediction.

Machine learning (ML) methods are increasingly used to assess variable importance, but such black box models lack stability when limited in sample sizes, and do not formally indicate non-important factors. The Shapley variable importance cloud (ShapleyVIC) addresses these limitations by assessing variable importance from an ensemble of regression models, which enhances robustness while maintaining interpretability, and estimates uncertainty of overall importance to formally test its significance. In a clinical study, ShapleyVIC reasonably identified important variables when the random forest and XGBoost failed to, and generally reproduced the findings from smaller subsamples (n = 2500 and 500) when statistical power of the logistic regression became attenuated. Moreover, ShapleyVIC reasonably estimated non-significant importance of race to justify its exclusion from the final prediction model, as opposed to the race-dependent model from the conventional stepwise model building. Hence, ShapleyVIC is robust and interpretable for variable importance assessment, with potential contribution to fairer clinical risk prediction.

Inter hospital external validation of interpretable machine learning based triage score for the emergency department using common data model.

Emergency departments (ED) are complex, triage is a main task in the ED to prioritize patient with limited medical resources who need them most. Machine learning (ML) based ED triage tool, Score for Emergency Risk Prediction (SERP), was previously developed using an interpretable ML framework with single center. We aimed to develop SERP with 3 Korean multicenter cohorts based on common data model (CDM) without data sharing and compare performance with inter-hospital validation design. This retrospective cohort study included all adult emergency visit patients of 3 hospitals in Korea from 2016 to 2017. We adopted CDM for the standardized multicenter research. The outcome of interest was 2-day mortality after the patients' ED visit. We developed each hospital SERP using interpretable ML framework and validated inter-hospital wisely. We accessed the performance of each hospital's score based on some metrics considering data imbalance strategy. The study population for each hospital included 87,670, 83,363 and 54,423 ED visits from 2016 to 2017. The 2-day mortality rate were 0.51%, 0.56% and 0.65%. Validation results showed accurate for inter hospital validation which has at least AUROC of 0.899 (0.858-0.940). We developed multicenter based Interpretable ML model using CDM for 2-day mortality prediction and executed Inter-hospital external validation which showed enough high accuracy.

A quantile regression analysis of ambulance response time.

BACKGROUND: Shorter ambulance response time (ART) contributes to improved clinical outcomes. Various methods have been used to analyze ART. OBJECTIVES: We aimed to compare the use of quantile regression with the standard ordinary least squares (OLS) model for identifying factors associated with ART in Singapore. A secondary aim was to determine the relative importance of patient-level (e.g., gender and ethnicity) versus system-level (e.g., call volumes within the last one hour) factors contributing to longer ART. METHODS: We conducted a retrospective review of data electronically captured from ambulance dispatch records and patient case notes of emergency calls to the national ambulance service from January to May 2006 (n = 30,687). The primary outcome was ART, defined as the time taken for an ambulance to arrive at the scene upon receiving an emergency call, and modeled as a function of patient- and system-level factors. We used a quantile regression model to account for potential heterogeneous effects of explanatory variables on ART across different quantiles of the ART distribution, and compared estimates derived with the corresponding OLS estimates. RESULTS: Quantile regression estimates suggested that the call volume in the previous one hour predicted increased ART, with the effect being more pronounced in higher ART quantiles. At the 90th and 50th percentiles of ART, each additional call in the last one hour was predicted to increase ART to the next call from the same area by 93 and 57 seconds, respectively. The corresponding OLS estimate was 58 seconds. Patient factors had little effect on ART. CONCLUSION: The quantile regression model is more useful than the OLS model for estimating ART, revealing that in Singapore, ART is influenced heterogeneously by the volume of emergency calls in the past one hour.