IMPORTANCEIncreased wait times and long lengths of stay in emergency departments (EDs) are associated with poor patient outcomes. Systems to improve ED efficiency would be useful. Specifically, minimizing the time to diagnosis by developing novel workflows that expedite test ordering can help accelerate clinical decision-making. OBJECTIVE To explore the use of machine learning-based medical directives (MLMDs) to automate diagnostic testing at triage for patients with common pediatric ED diagnoses. DESIGN, SETTING, AND PARTICIPANTS Machine learning models trained on retrospective electronic health record data were evaluated in a decision analytical model study conducted at the ED of the Hospital for Sick Children Toronto, Canada. Data were collected on all patients aged 0 to 18 years presenting to the ED from July 1, 2018, to June 30, 2019 (77 219 total patient visits). EXPOSURE Machine learning models were trained to predict the need for urinary dipstick testing, electrocardiogram, abdominal ultrasonography, testicular ultrasonography, bilirubin level testing, and forearm radiographs. MAIN OUTCOMES AND MEASURES Models were evaluated using area under the receiver operator curve, true-positive rate, false-positive rate, and positive predictive values. Model decision thresholds were determined to limit the total number of false-positive results and achieve high positive predictive values. The time difference between patient triage completion and test ordering was assessed for each use of MLMD. Error rates were analyzed to assess model bias. In addition, model explainability was determined using Shapley Additive Explanations values. RESULTS There was a total of 42 238 boys (54.7%) included in model development; mean (SD) age of the children was 5.4 (4.8) years. Models obtained high area under the receiver operator curve (0.89-0.99) and positive predictive values (0.77-0.94) across each of the use cases. The proposed implementation of MLMDs would streamline care for 22.3% of all patient visits and make test results available earlier by 165 minutes (weighted mean) per affected patient. Model explainability for each MLMD demonstrated clinically relevant features having the most influence on model predictions.Models also performed with minimal to no sex bias. CONCLUSIONS AND RELEVANCEThe findings of this study suggest the potential for clinical automation using MLMDs. When integrated into clinical workflows, MLMDs may have the potential to autonomously order common ED tests early in a patient's visit with explainability provided to patients and clinicians.

Health care workers (HCWs) are at higher risk for SARS-CoV-2 infection and may play a role in transmitting the infection to vulnerable patients and members of the community. This is particularly worrisome in the context of asymptomatic infection. We performed a cross-sectional study looking at asymptomatic SARS-CoV-2 infection in HCWs. We screened asymptomatic HCWs for SARS-CoV-2 via PCR. Complementary viral genome sequencing was performed on positive swab specimens. A seroprevalence analysis was also performed using multiple assays. Asymptomatic health care worker cohorts had a combined swab positivity rate of 29/5776 (0.50%, 95%CI 0.32–0.75) relative to a comparative cohort of symptomatic HCWs, where 54/1597 (3.4%) tested positive for SARS-CoV-2 (ratio of symptomatic to asymptomatic 6.8:1). SARS-CoV-2 seroprevalence among 996 asymptomatic HCWs with no prior known exposure to SARS-CoV-2 was 1.4–3.4%, depending on assay. A novel in-house Coronavirus protein microarray showed differing SARS-CoV-2 protein reactivities and helped define likely true positives vs. suspected false positives. Our study demonstrates the utility of routine screening of asymptomatic HCWs, which may help to identify a significant proportion of infections.

We screened three separate cohorts of healthcare workers for SARS-CoV-2 via nasopharyngeal swab PCR. A seroprevalence analysis using multiple assays was performed in a subgroup. The asymptomatic health care worker cohorts had a combined positivity rate of 29/5776 (0.50%, 95%CI 0.32-0.75) compared to the symptomatic cohort rate of 54/1597 (3.4%) (ratio of symptomatic to asymptomatic 6.8:1). Sequencing demonstrated several variants. The seroprevalence (n=996) was 1.4-3.4% depending on assay. Protein microarray analysis showed differing SARS-CoV-2 protein reactivities and helped define likely true positives vs. suspected false positives. Routine screening of asymptomatic health care workers helps identify a significant proportion of infections.

No abstract