87 Automated data processes in clinical operations – a pilot study for the cardiac M&M meetings. (15th December 2021)

• Record Type: Journal Article
• Authors: Briggs, Lydia; Serrato, Maria; Issitt, Richard W; Ng, Cho; Booth, John; Key, Daniel; Shah, Mohsin; Spiridou, Anastassia; Bryant, William A
• Is Part Of: Archives of disease in childhood. Volume 106(2021)Supplement 3
• Page Start: A32
• Page End: A33
• Publication Date: 2021-12-15
• Subjects: Children -- Diseases -- Periodicals; Infants -- Diseases -- Periodicals; 618.920005
• DOI: 10.1136/archdischild-2021-gosh.87
• Languages: English
• ISSNs: 0003-9888

Abstract

Abstract : The EPR system enables digital recording of patient health information and has enabled a huge amount of data to be available at a clinician's fingertips. A large amount of data is stored which can make sourcing uncommon or niche diagnoses and events difficult. Some patient events such as post-surgery complications are required for reporting purposes as well as patient care and department management. However, such information can take a long time for analysts to manually read and capture from the various areas in EPIC such as doctors' notes, scan summaries, procedures and laboratory results. By utilising the data extraction processes used in the Digital Research Environment (DRE), information of interest can be extracted from the back-end databases of the EPR system and stored in a structured and queryable format. These data can then be wrangled, filtered and joined together easily using a generalised tool developed in R, Python and SQL programming languages. Informative and interactive analytics in the form of graphs and tables can then be presented to display the information of interest in clinical team meetings to track and manage operational insights and patient care within the hospital over time. Here we present a tool which displays patient and hospital level information and visualisations on defined features of interest for the Cardiology M&M meetings as a pilot study, which is presented as an interactive Shiny app that the Cardiac department can explore. By utilising generalisable code developed in R and integrating the DRE extraction processes, the analytics displayed can be refreshed automatically to any date by the input of only a few commands. This app therefore has the potential to save department analyst time from repeated and manual data sourcing for rare complications post-surgery as well as improving overall data quality.