Garbage In, Garbage Out? Negative Impact of Physiological Waveform Artifacts in a Hospital Clinical Data Warehouse

Frederick H Kuo; Mohamed A Rehman; Luis M Ahumada

doi:10.1007/s10916-024-02128-1

Garbage In, Garbage Out? Negative Impact of Physiological Waveform Artifacts in a Hospital Clinical Data Warehouse

J Med Syst. 2024 Nov 25;48(1):109. doi: 10.1007/s10916-024-02128-1.

Authors

Frederick H Kuo^{1

2}, Mohamed A Rehman^{3

4}, Luis M Ahumada⁵

Affiliations

¹ Department of Anesthesia and Pain Medicine, Johns Hopkins All Children's Hospital, St Petersburg, FL, USA. frederick.kuo@jhmi.edu.
² Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA. frederick.kuo@jhmi.edu.
³ Department of Anesthesia and Pain Medicine, Johns Hopkins All Children's Hospital, St Petersburg, FL, USA.
⁴ Department of Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
⁵ Center for Pediatric Data Science and Analytics Methodology, Johns Hopkins All Children's Hospital, St Petersburg, Florida, USA.

PMID: 39586893
DOI: 10.1007/s10916-024-02128-1

Abstract

Hospitals around the world are deploying increasingly advanced systems to collect and store high-resolution physiological patient data for quality improvement and research. However, data accuracy, completeness, consistency, and contextual validity remain issues. This report highlights a data artifact known as waveform clipping in our hospital's physiological data capture system that went unnoticed for years, limiting data analysis and delaying several research projects. We aim to raise awareness in the medical informatics community about the importance of careful system setup, ongoing data validation, and close cooperation between clinicians and data scientists.

Keywords: Artifacts; Big data; Information systems; Medical informatics.

MeSH terms

Artifacts*
Data Accuracy
Data Warehousing
Humans