Problem definition: Translate data from electronic health records (EHR) into accurate predictions on patient flows and inform daily decision making at a major hospital. Academic/practical relevance: In a constrained hospital environment, forecasts on patient demand patterns could help match capacity and demand and improve hospital operations. Methodology: We use data from 63,432 admissions at a large academic hospital (50% female, median age 64 years old, median length of stay 3.12 days). We construct an expertise-driven patient representation on top of their EHR data and apply a broad class of machine learning methods to predict several aspects of patient flows. Results: With a unique patient representation, we estimate short-term discharges, identify long-stay patients, predict discharge destination, and anticipate flows in and out of intensive care units with accuracy in the 80%+ range. More importantly, we implement this machine learning pipeline into the EHR system of the hospital and construct prediction-informed dashboards to support daily bed placement decisions. Managerial implications: Our study demonstrates that interpretable machine learning techniques combined with EHR data can be used to provide visibility on patient flows. Our approach provides an alternative to deep learning techniques that is equally accurate, interpretable, frugal in data and computational power, and production ready.

Problem definition: Turn raw data from Electronic Health Records into accurate predictions on patient flows and inform daily decision-making at a major hospital. Practical Relevance: In a hospital environment under increasing financial and operational stress, forecasts on patient demand patterns could help match capacity and demand and improve hospital operations. Methodology: We use data from 63,432 admissions at a large academic hospital (50.0% female, median age 64 years old, median length-of-stay 3.12 days). We construct an expertise-driven patient representation on top of their EHR data and apply a broad class of machine learning methods to predict several aspects of patient flows. Results: With a unique patient representation, we estimate short-term discharges, identify long-stay patients, predict discharge destination and anticipate flows in and out of intensive care units with accuracy in the 80%+ range. More importantly, we implement this machine learning pipeline into the EHR system of the hospital and construct prediction-informed dashboards to support daily bed placement decisions. Managerial Implications: Our study demonstrates that interpretable machine learning techniques combined with EHR data can be used to provide visibility on patient flows. Our approach provides an alternative to deep learning techniques which is equally accurate, interpretable, frugal in data and computational power, and production-ready.

IMPORTANCE Alternative methods for hospital occupancy forecasting, essential information in hospital crisis planning, are necessary in a novel pandemic when traditional data sources such as disease testing are limited.OBJECTIVE To determine whether mandatory daily employee symptom attestation data can be used as syndromic surveillance to estimate COVID-19 hospitalizations in the communities where employees live. DESIGN, SETTING, AND PARTICIPANTSThis cohort study was conducted from April 2, 2020, to November 4, 2020, at a large academic hospital network of 10 hospitals accounting for a total of 2384 beds and 136 000 discharges in New England. The participants included 6841 employees who worked on-site at hospital 1 and lived in the 10 hospitals' service areas.EXPOSURE Daily employee self-reported symptoms were collected using an automated text messaging system from a single hospital. MAIN OUTCOMES AND MEASURESMean absolute error (MAE) and weighted mean absolute percentage error (MAPE) of 7-day forecasts of daily COVID-19 hospital census at each hospital. RESULTS Among 6841 employees living within the 10 hospitals' service areas, 5120 (74.8%) were female individuals and 3884 (56.8%) were White individuals; the mean (SD) age was 40.8 (13.6) years, and the mean (SD) time of service was 8.8 (10.4) years. The study model had a MAE of 6.9 patients with COVID-19 and a weighted MAPE of 1.5% for hospitalizations for the entire hospital network. The individual hospitals had an MAE that ranged from 0.9 to 4.5 patients (weighted MAPE ranged from 2.1% to 16.1%). For context, the mean network all-cause occupancy was 1286 during this period, so an error of 6.9 is only 0.5% of the network mean occupancy. Operationally, this level of error was negligible to the incident command center. At hospital 1, a doubling of the number of employees reporting symptoms (which corresponded to 4 additional employees reporting symptoms at the mean for hospital 1) was associated with a 5% increase in COVID-19 hospitalizations at hospital 1 in 7 days (regression coefficient, 0.05; 95% CI, 0.02-0.07; P < .001).CONCLUSIONS AND RELEVANCE This cohort study found that a real-time employee health attestation tool used at a single hospital could be used to estimate subsequent hospitalizations in 7 days at hospitals throughout a larger hospital network in New England.

Anomalous biliary anatomy is frequently encountered by surgeons during cholecystectomy. Importance of its recognition lies in avoiding serious biliary injuries. One such anomaly is cholecystohepatic duct. We describe rare clinical situation wherein agenesis of CHD along with cholecystohepatic duct was mistaken for hilar stricture.

No abstract