COVID-19 sent health care leaders worldwide on a quest for timely, accurate reporting. Nearly overnight, metrics related to testing, confirmed cases, capacity and ventilator use became critical to organizations’ pandemic response and recovery efforts. With time in short supply, organizations began searching for resources to automate data collection. Leaders at Universal Health Services, Inc. (UHS) turned to the Cerner enterprise data warehouse (EDW) to consolidate and summarize key COVID-19 metrics across its 26 hospitals spanning seven states.
"Initially, our sites each pulled certain data points and provided them to corporate reporting teams by a given time each day," said Ehab Hanna, MD, vice president, chief medical information officer. "But that process was manual and had room for variance, depending upon the person reporting and the time of day they submitted. It was time-consuming and didn't allow for reliable data trending."
Hanna engaged the UHS reporting and IT teams to automate enterprisewide COVID-19 data collection, improve tracking and facilitate analysis using the Cerner EDW, HealtheEDWSM.
"We looked at which figures the sites reported and determined how to capture the information while removing that manual onus from the sites," said Sheree Whitley, BSN, RN, senior clinical informaticist.
The UHS reporting team pulls information from the organization’s Cerner Millennium® electronic health record (EHR) into HealtheEDW to create a reporting dataset, and then uses an interactive dashboard to present the insights.
The dashboard displays COVID-19 outcome metrics — including tests, confirmed cases, deaths and discharges — from the previous day and in aggregate. At the hospital level, the metrics are adjusted to reflect the hospital's size, helping users do a comparative analysis across the UHS network.
"By normalizing our data based on the number of beds at each site, we can see if a specific hospital or region is getting hit harder than another," said Melissa Mulvey, software engineer.
Leaders and infection prevention staff rely on the tool to track the virus’s progression across UHS facilities and geographic regions.
"Our infection preventionists use the dashboard when they meet to review where we are at various hospitals — looking at each site’s curve and how long it's taking sites to double their COVID-19 rate," said Whitley. "And regional leaders can group their hospitals within the dashboard and view the group to quickly see how we're doing in a particular area."
The dashboard also tracks daily trending census metrics — including number of people awaiting test results, patients in the ICU and patients on ventilators. The data helps leaders understand each hospital's current status and spot broader trends.
As clinicians' needs evolve in response to the virus, UHS IT teams adapt their Cerner system and workflows accordingly. With each change, reporting teams quickly update the dashboard to continue capturing key information.
"We keep iterating because it's not like you have the requirements in advance," said Agnes Canlas, manager, business intelligence and analytics. "We're all responding to a crisis. So as one team adjusts Millennium workflows and creates new catalog codes or event codes in the system, we're building the datasets for reporting. HealtheEDW is flexible enough to do what we want to do."
Canlas and her team expect that iterative process to continue for the foreseeable future. They view the dashboard as a vital tool for informing UHS' new normal.
"The dashboard will assist as we resume elective procedures by tracking screening tests, but also by watching COVID-19 metrics as things reopen and restrictions are lessened," said Whitley. "Are our COVID-19 numbers declining? Are we seeing spikes in certain locations? We'll be able to monitor as things return to a new normal."
As the world responds to the COVID-19 pandemic, Cerner’s work continues to support health care providers and communities across the globe. Learn about pre-built COVID-19 dashboards and reports available in Cerner HealtheAnayticsSM.
