Martini, Wayne A

Tackling hospital emergency department (ED) overcrowding is a paramount challenge for healthcare systems. To combat this issue, an innovative approach is to identify patients who can be served vertically (i.e., in a seated position) and route them to a dedicated area termed the vertical processing pathway (VPP). Successfully implementing this design requires understanding which patients should be routed to the VPP and when. Currently, the decision to leverage the VPP is made in an ad hoc fashion. To assist our partner hospital and other EDs in capturing the value of the VPP, we develop a machine learning model that provides personalized risk scores predicting whether each arriving patient needs an ED bed. We use these scores as input to a stochastic patient flow model and analytically characterize the optimal VPP policy that minimizes length of stay. We then combine our results to derive an interpretable VPP patient streaming protocol suitable for implementation in practice and conduct a before-and-after experiment in which we leverage empirical analyses to evaluate the impact of integrating it into practice. The implemented protocol led to an 11-minute (4.2%) reduction in ED length of stay in our partner hospital without any adverse effect on quality-of-care outcomes. This effect was statistically significant and remained robust after controlling for various confounding factors and potential sources of endogeneity. Our VPP protocol can also be utilized in other EDs, offering operational improvements without requiring additional resources. This paper was accepted by Stefan Scholtes, healthcare management. Funding: S. Saghafian received financial support from the NSF [Grant CMMI-15 62645], “Data-Driven Management of Post-Transplant Medications,” which partially enabled this work, as well as funding and support from the Harvard Data Science Initiative and the Harvard Mossavar-Rahmani Center for Business and Government. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2024.07517 .

Background/Objectives: Machine learning (ML) models can predict hospital admission from emergency department (ED) triage data with areas under the receiver operating characteristic curve (AUC) exceeding 0.85. Whether incorporating the assigned provider’s identity—as a proxy for unmeasured practice variation—improves prediction has not been systematically studied. We aimed to compare 10 supervised ML classifiers for predicting hospital admission at ED triage, with and without provider identity, and to characterize model reasoning using SHapley Additive exPlanations (SHAP). Methods: We conducted a retrospective cohort study of 186,094 ED visits (2020–2023, training) and 58,151 visits (2024, temporal holdout test) at one academic tertiary-care ED. Ten classifiers spanning linear, distance-based, tree-based, ensemble, probabilistic, and neural network families were each trained in two conditions: baseline (23 triage features) and with provider identity appended. SHAP TreeExplainer was applied to the top-performing models (CatBoost and XGBoost). Results: The admission rate was 31.3% (training) and 31.7% (test). CatBoost achieved the highest baseline AUC of 0.8906 (0.8878–0.8933). Adding provider identity produced negligible AUC changes across all models (ΔAUC range: −0.0029 to +0.0015; all DeLong p > 0.05). SHAP analysis identified ESI level, respiratory rate, temperature, complaint category, and age as the dominant predictors, with clinically intuitive directionality. Conclusions: Provider identity does not meaningfully improve ML prediction of hospital admission beyond standard triage variables. The observed 28-percentage-point variation in provider admission rates is explained by patient case-mix differences than with independent practice pattern effects on prediction. SHAP provides transparent, clinically interpretable explanations suitable for bedside decision support.

Background/Objectives: Recent advances in artificial intelligence (AI) and machine learning (ML) enable targeted optimization of emergency department (ED) operations. We examine how reworking an ED’s vertical processing pathway (VPP) using AI- and ML-driven recommendations affected patient throughput. Methods: We trained a non-linear ML model using triage data from 49,350 ED encounters to generate a personalized risk score that predicted whether an incoming patient is suitable for vertical processing. This model was integrated into a stochastic patient flow framework using queueing theory to derive an optimized VPP design. The resulting protocol prioritized a vertical assessment for patients with Emergency Severity Index (ESI) scores of 4 and 5, as well as 3 when the chief complaints involved skin, urinary, or eye issues. In periods of ED saturation, our data-driven protocol suggested that any waiting room patient should become VPP eligible. We implemented this protocol during a 13-week prospective trial and evaluated its effect on ED performance using before-and-after data. Results: Implementation of the optimized VPP protocol reduced the average ED length of stay (LOS) by 10.75 min (4.15%). Adjusted analyses controlling for potential confounders during the study period estimated a LOS reduction between 7.5 and 11.9 min (2.89% and 4.60%, respectively). No adverse effects were observed in the quality metrics, including 72 h ED revisit or hospitalization rates. Conclusions: A personalized, data-driven VPP protocol, enabled by ML predictions, significantly improved the ED throughput while preserving care quality. Unlike standard fast-track systems, this approach adapts to ED saturation and patient acuity. The methodology is customizable to patient populations and ED operational characteristics, supporting personalized patient flow optimization across diverse emergency care settings.

Purpose This study investigates medical students’ opinions on the optimal timing for taking the emergency medicine (EM) clerkship and its effects on preparedness for subsequent clerkships and the USMLE (United States Medical Licensing Examination) Step 2 examination. Methods A cross-sectional survey was conducted among third- and fourth-year students at the Mayo Clinic Alix School of Medicine. The survey collected opinions on the optimal timing for the EM clerkship and its impact on self-efficacy as well as perceived preparedness for future clerkships and the USMLE Step 2 examination. Results Only 5% of third-year EM clerkship students felt they would have been more prepared in their fourth year. Conversely, 70% of fourth-year students believed they were more prepared than if they had taken it in their third year. Students were evenly split on if an earlier EM clerkship would have been helpful for subsequent clerkships. 88% of pre-USMLE Step 2 students found it helpful for Step 2, while 37% of post-USMLE Step 2 students felt it would have helped their preparedness. Additionally, 96% of students stated the EM clerkship increased their confidence in seeing patients. 43% of those who took the clerkship in their fourth year reported they would have considered EM as a specialty had they taken it earlier. Most students recommended taking the EM clerkship during the second 6 months of clerkships. Conclusion There are mixed opinions among medical students regarding the helpfulness of taking the EM clerkship earlier for subsequent clerkships and USMLE Step 2 preparedness. However, the majority recommend taking the EM clerkship earlier. A significant proportion of participants would have considered EM as a specialty if they had taken the clerkship earlier. Furthermore, 96% reported increased self-efficacy. Therefore, encouraging students to take the EM clerkship earlier may increase self-confidence and allow for earlier career exploration.

Abstract Context Point-of-care ultrasound (POCUS) has diverse applications across various clinical specialties, serving as an adjunct to clinical findings and as a tool for increasing the quality of patient care. Owing to its multifunctionality, a growing number of medical schools are increasingly incorporating POCUS training into their curriculum, some offering hands-on training during the first 2 years of didactics and others utilizing a longitudinal exposure model integrated into all 4 years of medical school education. Midwestern University Arizona College of Osteopathic Medicine (MWU-AZCOM) adopted a 4-year longitudinal approach to include POCUS education in 2017. There is a small body of published research supporting this educational model, but there is not much data regarding how this approach with ultrasound curriculum translates to real-world changes in POCUS use by graduate student clinicians having received this model of education. Objectives The objectives of this study are to determine the frequency of POCUS use by MWU-AZCOM graduates and to assess how a 4-year longitudinal ultrasound curriculum may enhance the abilities of MWU-AZCOM graduates to perform and interpret ultrasound imaging in specific residency programs. Methods The study was approved by the MWU Institutional Review Board (#IRBAZ-5169, approval date October 3, 2022). An anonymous novel 12-question survey was conducted utilizing Research Electronic Data Capture (REDCap), a secure online platform, and distributed to MWU-AZCOM 2021 and 2022 graduates via email. Survey questions were aimed at assessing frequency of use, utilization of different imaging modalities, reasons for utilizing POCUS, barriers/enablers to utilizing POCUS, ultrasound training, and confidence in performing scans and interpreting POCUS imaging. All of the 104 surveys returned were included in the study. Statistical software R version 4.3 was utilized to conduct statistical analyses. Results Of the 484 surveys distributed, 104 were completed (21.5 % response rate). Responses came from residents working in 14 different specialties, 50 in primary care and 54 in nonprimary care. Of all respondents, 85.6 % currently utilize POCUS in their practice on at least a monthly basis and 77.0 % believe that their POCUS training in medical school enriches their current practice in residency. The top five modalities utilized by residents were procedures (89.9 %), cardiac (88.8 %), pulmonary (82.0 %), Focused Assessment with Sonography for Trauma (FAST, 73.0 %), and vascular (71.9 %). Respondents recognized POCUS as a beneficial diagnostic tool (97.8 %) and reported enhancements in physical examination skills (58.4 %) and professional growth (61.8 %). Facilitators for POCUS adoption included cost-effectiveness (82.0 %), diagnostic differentiation (78.7 %), and safety (79.8 %). Barriers included a lack of trained faculty (27.9 %), absence of necessary equipment (26.9 %), and cost of equipment (22.1 %). Participants demonstrated high confidence levels in performing (74.0 %) and interpreting (76.0 %) POCUS, with 43.3 % believing that their POCUS training enhanced their attractiveness as residency candidates. Conclusions This study supports the positive impact of a 4-year longitudinal POCUS curriculum on graduates’ practice. It emphasizes the link between MWU-AZCOM’s curriculum and real-world clinical needs. Addressing identified barriers and advancing hands-on training can further enhance POCUS understanding, ensuring that future physicians are well-prepared to leverage its diagnostic potential across medical specialties.