Validation Methodology of Healthcare-Associated Infection Device Day Denominators When Switching Electronic Medical Records. (October 2020)

• Record Type: Journal Article
• Authors: Luong, Lan; Simkins, Michelle; Snyders, Rachael; Gase, Kathleen Anne; Leone, Carole; Sykora, Carol; Hoehner, Christine; Babcock, Hilary
• Is Part Of: Infection control and hospital epidemiology. Volume 41(2020)Supplement 1
• Page Start: s428
• Page End: s429
• Publication Date: 2020-10
• Subjects: Nosocomial infections -- Epidemiology -- Periodicals; Health facilities -- Sanitation -- Periodicals; Hospital buildings -- Sanitation -- Periodicals; Cross Infection -- Periodicals; Epidemiology -- Periodicals; Hospitals -- Periodicals; Infection Control -- Periodicals; 614.44
• DOI: 10.1017/ice.2020.1090
• Languages: English
• ISSNs: 0899-823X

Abstract

Abstract : Background: From August 2017 to June 2018, 11 hospitals within a large healthcare system switched from multiple different electronic medical records (EMRs) to 1 EMR. At the time of this transition, the NHSN provided guidelines to validate healthcare-associated infection (HAI) denominators when switching from manual denominator collection to electronic denominator collection, but the NHSN did not give guidelines for validation when switching from 1 EMR to another. We aimed to build a validation process to ensure the accuracy of central-line and urinary catheter days reported to the NHSN after switching EMRs. Methods: Our validation process began with a statistical phase followed by a targeted manual validation phase. The statistical phase used 3 prediction methods (linear regression, time series analysis, and statistical process control [SPC] charts) to forecast device days after the EMR switch for units within hospitals. Models were developed using baseline data from the old EMR (January 2015 through the new EMR implementation). Using prespecified criteria for each method to determine discrepancies, we built a decision tree to identify units needing manual validation. Any unit that failed the statistical phase would need to participate in the manual validation phase, using a midnight census and direct visualization of devices. The manual validation process was composed of 14-day blocks. At the end of each block, if manual device days were within ±5% of EMR device days, they were considered validated. Manual validation would be repeated in 14-day blocks until 2 consecutive blocks passed within ±5%. Results: Overall, 157 units were evaluated for urinary catheter days and central-line days. Among them, 143 units passed the statistical validation test for urinary catheter days and 151 passed for central-line days. There was no specific pattern when comparing forecasted versus actual device days. The manual validation process for the 20 failing units (14 urinary catheter and 6 central-line units) is ongoing; preliminary results identified issues with missing nursing documentation in the EMR and with inaccurate manual counting of device days. There were no systematic discrepancies associated with the new EMR. Conclusions: We developed a novel validation process using statistical prediction methods supplemented with a targeted manual process. This process saved resources by identifying the units that need manual validation. Discrepancies were largely related to nursing documentation, which the infection prevention team addressed with additional training. Funding: None Disclosures: None