Evaluating potential adverse patient outcomes associated with delay of procedures due to COVID-19

Overview

During the coronavirus disease of 2019 (COVID-19), hospital systems postponed non-essential medical procedures to accommodate a surge of critically ill patients, and patients avoided hospital visits for routine healthcare. We developed two complementary approaches, one for inpatient surgical procedures (surgical_delay.r) and the other for outpatient screening tests (screening_delay.r), to help hospital systems understand and predict the impact of delaying procedures on patient health outcomes using historical electronic health record (EHR) data.

Reference:

Please direct questions to neil.zheng@vumc.org and wei-qi.wei@vumc.org

Requirements

• OMOP Common Data Model Version 5
• R Packages:
  • dplyr
  • SqlRender
  • DatabaseConnector
  • survival
  • lubridate
  • MASS

How to use

• Clone or download repository in desired location.
• Update surgical_delay.r or screening_delay.r for your analysis:
  • Input database connection settings
  • Update procedure and diagnosis filters
  • Adjust SQL code as needed if using different table names
  • Adjust R code as needed for analysis if using different covariates or outcomes
    (Note: we included insurance information as a covariate and cancer stage at diagnosis as an outcome, both of which are not OMOP CDM)
• Additional details included in surgical_delay.r or screening_delay.r