CJEM最新文献

Study objective: This study investigates the potential to improve emergency department (ED) triage using machine learning models by comparing their predictive performance with the Canadian Triage Acuity Scale (CTAS) in identifying the need for critical care within 12 h of ED arrival.

Methods: Three machine learning models (LASSO regression, gradient-boosted trees, and a deep learning model with embeddings) were developed using retrospective data from 670,841 ED visits to the Jewish General Hospital from June 2012 to Jan 2021. The model outcome was the need for critical care within the first 12 h of ED arrival. Metrics, including the areas under the receiver-operator characteristic curve (ROC) and precision-recall curve (PRC) were used for performance evaluation. Shapley additive explanation scores were used to compare predictor importance.

Results: The three machine learning models (deep learning, gradient-boosted trees and LASSO regression) had areas under the ROC of 0.926 ± 0.003, 0.912 ± 0.003 and 0.892 ± 0.004 respectively, and areas under the PRC of 0.27 ± 0.01, 0.24 ± 0.01 and 0.23 ± 0.01 respectively. In comparison, the CTAS score had an area under the ROC of 0.804 ± 0.006 and under the PRC of 0.11 ± 0.01. The predictors of most importance were similar between the models.

Conclusions: Machine learning models outperformed CTAS in identifying, at the point of ED triage, patients likely to need early critical care. If validated in future studies, machine learning models such as the ones developed here may be considered for incorporation in future revisions of the CTAS triage algorithm, potentially improving discrimination and reliability.

Objectives: Abdominal ultrasound is used for diagnosing appendicitis in patients with right lower quadrant abdominal pain. Between 45 and 82% of radiology performed ultrasounds are indeterminate for appendicitis and computed tomography is required for diagnostic confirmation. Our study aims to determine predictors to rule out appendicitis when ultrasound is indeterminate.

Methods: We performed a health records review of adult emergency department (ED) patients presenting with symptoms suspicious for appendicitis and indeterminate ultrasound to two academic EDs between June 2019 and July 2020. The outcome was appendicitis diagnosis within 30 days of the index ED visit. We used multivariable logistic regression, identifying a cut-off threshold for continuous variables with cubic spline, and chose the parsimonious model to develop a binary decision rule. We report Odds ratios (OR) and diagnostic performance with 95% confidence intervals (CI).

Results: Overall, 463 patients (mean age 30.3 years (SD 10.5 years), 74.9% female) were included. Appendicitis was diagnosed in 45 patients (9.7% [95% CI 7.2-12.8%]). After ultrasound, computed tomography was performed in 227 patients (49.0%) and 39 patients (17.2%) were diagnosed with appendicitis. Among the 236 patients who did not have a subsequent computed tomography, 6 (2.6%) patients had appendicitis. Neutrophil count > 5.5 × 10⁹/L (OR 1.21 [95% CI 1.12-1.30]) and secondary signs of inflammation on ultrasound (OR 2.16 [1.07-4.37]) were associated with a higher likelihood of appendicitis (C-statistic 0.77 [95% CI 0.70-0.84]). The absence of both predictors had a sensitivity of 88.9% (95% CI 76.0-96.3%), specificity of 45.7% (95% CI 40.8-50.6%) and a negative predictive value of 0.97 (95% CI 0.94-0.99) to rule out appendicitis.

Conclusion: For patients suspected of appendicitis and indeterminate ultrasound, the absence of an elevated neutrophil count and secondary signs of inflammation are associated with a low probability of appendicitis.

Purpose: Intersex people make up 1.7-4% of the population of North America. A recent scoping review of emergency department (ED) relevant literature for the care of sexual and gender minorities found almost no representation of this population. Intersex people have unique equity, diversity, and inclusion needs in the ED, so we undertook a review of international guidelines to identify ED-relevant recommendations.

Methods: Using the PRISMA criteria, we systematically searched OVID Medline, EMBASE, CINAHL, and the gray literature for any clinical practice guideline (CPG) or best practice statement (BPS) published until Oct 21, 2022. We included articles in English, which included care of intersex people of any age, in any setting, region, or nation, and were of national or international in scope. We excluded primary research, systematic or narrative reviews, non-CPS or BPS statements, editorials, articles of regional or individual hospital scope, or if a more recent version had been published. Recommendations relevant to the ED were identified and the guideline and recommendations scored for quality using the AGREE-II and AGREE-REX tools respectively.

Results: Of 1599 studies identified, 1400 studies were excluded, 199 full-text reviews completed, and 95 studies included for evaluation. There were no ED-relevant recommendations found among these guideline documents.

Conclusion: A systematic review of the literature for ED-relevant guidelines for the care of Intersex populations returned no results. Given the risk of increasing barriers to care for intersex people, and the increasing use of the ED for primary care, the requirements of Intersex people need to be investigated and integrated into future development of a CPG for care of sexual and gender minority populations in the ED.

Introduction: There are many limitations to utilizing on-duty emergency department (ED) physicians as Base Hospital Physicians for paramedic telephone consultations. We aimed to examine the impact of a specialized and centralized Online Medical Consultation program for paramedic consultations on system-relevant performance.

Methods: This is a before-after study with concurrent control using health record review of audio recordings over a 6-month period before and after implementation of the Online Medical Consultation program. The primary outcome was the duration of paramedic consultation calls. The secondary outcomes included number of calls with orders that contradicted existing medical directives, number of calls with orders outside of paramedic scope of practice, number of calls with Base Hospital Physician requiring clarification on medical directives, and number of calls with Base Hospital Physician interrupting the paramedic during the call.

Results: We included 220 consultation calls. The patients' mean age was 54.5 years. Most consultation calls (70.5%) were for mandatory consultations and 22.7% were voluntary. Most consultations were related to cardiac arrest (43.6%), combative patients (15.0%), and analgesia (13.6%). Before-after comparisons for total call duration showed that mean call duration decreased in Ottawa from 4:28 to 4:05 min (p = 0.77) and decreased in Kingston from 4:50 to 4:13 min (p = 0.49). There were no significant differences in our secondary outcomes.

Conclusions: The Online Medical Consultation program was implemented and removed the responsibility of responding to online medical consultations for on-duty emergency physicians in Ottawa. The total call duration was not significantly different between groups. Additional time intervals and adherence to protocol benefits were also not statistically significant due to low baseline incidence.

Introduction: The primary objective was to evaluate the effectiveness of telemedicine for improving clinical follow-up for survivors of sexual assault and intimate partner violence after an emergency department (ED) visit. The Sexual Assault and Partner Abuse Care Program (SAPACP) is an ED-based clinic for survivors of sexual assault/intimate partner violence. Virtual Visit, a telemedicine platform, was introduced at SAPACP in January 2020, allowing patients to attend follow-up virtually.

Methods: This is a seven-year analysis of sexual assault/intimate partner violence cases presenting to an ED from 1 Jan 2015 to 31 Mar 2022. Interrupted time series and segmented linear regression models with a first lag autoregressive covariance structure were used to analyze follow-up rates post-Virtual Visit and to predict a counterfactual trend.

Results: Between 2015 and 2022, there were 3317 sexual assault/intimate partner violence case visits, with 2406 visits being pre-Virtual Visit (1 Jan 2015-31 Dec 2019) and 911 visits post-Virtual Visit (1 Jun 2020-31 Mar 2022). Within the 911 post-Virtual Visit visits, 315 (35%) had virtual follow-ups, 291 (32%) had in-person follow-ups, and 305 (33%) did not return for follow-up. Post-Virtual Visit, there was an immediate 10% increase in the proportion of overall follow-ups while accounting for pre-Virtual Visit trends, which was sustained over two years. The proportion of overall follow-up pre-Virtual Visit was 48% (95% CI: 46.0-50.0%), and was 70.0% (95.0% CI: 67.0-73.0%) post-Virtual Visit. Looking at sub-groups, follow-up for sexual assault cases reached 75.0% (95% CI: 71.0-78.0%) and physical/verbal cases reached 64.0% (95% CI: 60.0-69.0%) post-Virtual Visit.

Conclusion: The implementation of Virtual Visit led to an immediate 10% increase in the proportion of overall follow-ups, which was sustained over a two-year period, while accounting for pre-Virtual Visit trends. These findings indicate that telemedicine can help improve clinical follow-up among survivors of sexual assault/intimate partner violence.