Predicting Next Day Discharge via Electronic Health Record Audit Logs (DSI-SRP)
This DSI-SRP fellowship funded Xinmeng Zhang to work in the laboratory of Professor You Chen in the Department of Biomedical Informatics during the summer of 2019. Xinmeng anticipates graduating in 2022 and is majoring in Computer Science.
The project funded by this fellowship aimed to understand the intersection between electronic health records (EHR) and clinicians’ decision-making process when discharging patients. Innovations in health information technology have allowed care providers and patients to generate healthcare data via a variety of sources such as electronic health record (EHR) systems. Data in EHRs contain longitudinal patient data, healthcare service, and a high volume of operational actions. Most of the existing studies focus on medical data in the EHRs. Audit logs which is another type of data in the EHRs have seldom been investigated. EHR audit logs document the clinical activities and information exchanges between healthcare providers and are useful for inferring clinical workflows and clinical treatment decision paths. Building upon the research project funded by DSI in the summer of 2019, we conducted a long-term project to predict whether a patient would be discharged within 24 hours. The project was conducted on the cohort of all adults admitted to Vanderbilt University Medical Center in 2019, which includes 26,283 inpatient stays, 133,398 patient-day observations, and 819 types of user-EHR interactions. Our findings indicate that EHR audit log data has the potential to reflect clinicians’ insights into a patients’ clinical status in the discharge decision-making process.
In addition to receiving support through a DSI-SRP fellowship, this project was supported and facilitated by the DSI Data Science Team through their regular summer workshops and demo sessions.
Image cited as:
“Mobile Medical Unit” by BC Gov Photos is licensed under CC BY-NC-ND 2.0”