E-News Exclusive

Data Analytics Could Prevent Testing Bottlenecks During Future Pandemics

Breaking research demonstrates the efficacy of two data analytics-based strategies that clinical labs employed to meet COVID-19 testing demands during the height of the pandemic. These findings, published in the data science issue of American Association for Clinical Chemistry’s (AACC) The Journal of Applied Laboratory Medicine, give labs a blueprint for using data analytics to ensure patient access to testing during future infectious disease outbreaks.

A major challenge labs faced during the first two years of the pandemic was keeping up with testing demands in the midst of chronic supply shortages. While this isn’t currently a serious issue with US COVID-19 cases being relatively low, it’s only a matter of time before labs could be grappling with this problem again—either because of a surge in COVID-19 cases caused by a new variant, or because of a new infectious disease outbreak entirely.

One team of researchers led by Rohit B. Sangal, MD, MBA, FACEP, of the Yale University School of Medicine, has shown that labs can use EHR systems to ensure scarce testing resources are optimally allocated during a pandemic.

View the full study at https://academic.oup.com/jalm/article/8/1/98/6848829.

During the omicron variant surge of December 2021 to January 2022, Sangal’s health care system developed guidelines to ensure that limited SARS-CoV-2 tests and combined tests for SARS-CoV-2, flu, and respiratory syncytial virus (RSV) were used on the appropriate patients. To help clinicians adhere to these guidelines when ordering tests, Sangal’s team implemented a redesign of the EHR’s test ordering interface on December 22, 2021.

Following this, Sangal’s team analyzed test ordering data from the three weeks before the redesign and the three weeks after and found that the EHR redesign successfully changed testing patterns to align with guidelines. For symptomatic patients who were discharged from the emergency department, COVID-19 + flu/RSV testing decreased 49% while testing for COVID-19 + flu-only increased 160%. Meanwhile, for symptomatic patients who were admitted to the hospital, COVID-19 + flu/RSV testing increased 128%. Not only does this mean that the right patients were getting the right tests but also these changes saved approximately 437 test cartridges per week, thereby preserving the limited supply of testing resources.

“A simple EHR order redesign was associated with increased adherence to institutional guidelines for SARS-CoV-2 and influenza testing amidst supply chain limitations necessitating optimal allocation of scarce testing resources,” Sangal says. “With continually shifting resource availability, clinician education is not sufficient. Rather, system-based interventions embedded within existing workflows can better align resources and serve the testing needs of the community.”

These findings build on data analytics work from the start of the pandemic led by Daniel T. Holmes, MD, of St. Paul’s Hospital in Vancouver, Canada, which demonstrated how labs can use data automation to handle outbreak-related testing surges.

View the full study at https://academic.oup.com/jalm/article/8/1/41/6840464.

During the first year of the pandemic, supply shortages for SARS-CoV-2 testing instruments began to affect the ability of St. Paul’s Hospital to keep up with testing demand. In response, Holmes’s team used open-source software tools (Linux, bash, R, RShiny, ShinyProxy, and Docker) to develop an automated workflow that manages and optimizes all steps of the SARS-CoV-2 testing process. From September to December 2020, this automated workflow decreased the lab’s consumption of reagents for SARS-CoV-2 testing by approximately 58%.

In his group’s paper in The Journal of Applied Laboratory Medicine, Holmes says, “We describe our strategy for data automation for singleton and pooled sample [testing] for SARS-CoV-2 with extension to other viral PCR [polymerase chain reaction] assays. The open-source software tools used and the software development and operational deployment strategy are explained. This work will give the reader direction on how to develop and deploy similar tools should the need arise.”

— Source: AACC