RESUMEN
Objective: The Score for Emergency Risk Prediction (SERP) is a novel mortality risk prediction score which leverages machine learning in supporting triage decisions. In its derivation study, SERP-2d, SERP-7d and SERP-30d demonstrated good predictive performance for 2-day, 7-day and 30-day mortality. However, the dataset used had significant class imbalance. This study aimed to determine if addressing class imbalance can improve SERP's performance, ultimately improving triage accuracy. Methods: The Singapore General Hospital (SGH) emergency department (ED) dataset was used, which contains 1,833,908 ED records between 2008 and 2020. Records between 2008 and 2017 were randomly split into a training set (80%) and validation set (20%). The 2019 and 2020 records were used as test sets. To address class imbalance, we used random oversampling and random undersampling in the AutoScore-Imbalance framework to develop SERP+-2d, SERP+-7d, and SERP+-30d scores. The performance of SERP+, SERP, and the commonly used triage risk scores was compared. Results: The developed SERP+ scores had five to six variables. The AUC of SERP+ scores (0.874 to 0.905) was higher than that of the corresponding SERP scores (0.859 to 0.894) on both test sets. This superior performance was statistically significant for SERP+-7d (2019: Z = -5.843, p < 0.001, 2020: Z = -4.548, p < 0.001) and SERP+-30d (2019: Z = -3.063, p = 0.002, 2020: Z = -3.256, p = 0.001). SERP+ outperformed SERP marginally on sensitivity, specificity, balanced accuracy, and positive predictive value measures. Negative predictive value was the same for SERP+ and SERP. Additionally, SERP+ showed better performance compared to the commonly used triage risk scores. Conclusions: Accounting for class imbalance during training improved score performance for SERP+. Better stratification of even a small number of patients can be meaningful in the context of the ED triage. Our findings reiterate the potential of machine learning-based scores like SERP+ in supporting accurate, data-driven triage decisions at the ED.
RESUMEN
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.