Salience of Medical Concepts of Inside Clinical Texts and Outside Medical Records for Referred Cardiovascular Patients

Sungrim Moon; Sijia Liu; David Chen; Yanshan Wang; Douglas L Wood; Rajeev Chaudhry; Hongfang Liu; Paul Kingsbury

doi:10.1007/s41666-019-00044-5

Salience of Medical Concepts of Inside Clinical Texts and Outside Medical Records for Referred Cardiovascular Patients

J Healthc Inform Res. 2019 Jan 28;3(2):200-219. doi: 10.1007/s41666-019-00044-5. eCollection 2019 Jun.

Authors

Sungrim Moon¹, Sijia Liu^{1

2}, David Chen¹, Yanshan Wang¹, Douglas L Wood³, Rajeev Chaudhry⁴, Hongfang Liu¹, Paul Kingsbury¹

Affiliations

¹ Department of Health Sciences Research, Mayo Clinic, Rochester, MN USA.
² Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY USA.
³ Center For Innovation, Mayo Clinic, Rochester, MN USA.
⁴ Department of Medicine and Center for Translational Informatics, Mayo Clinic, Rochester, MN USA.

Abstract

Outside medical records (OMRs) accompanying referred patients are frequently sent as faxes from external healthcare providers. Accessing useful and relevant information from these OMRs in a timely manner is a challenging task due to a combination of the presence of machine-illegible information and the limited system interoperability inherent in healthcare. Little research has been done on investigating information in OMRs. This paper evaluated overlapping and non-overlapping medical concepts captured from digitally faxed OMRs for patients transferring to the Department of Cardiovascular Medicine and from clinical consultant notes generated at the Mayo Clinic. We used optical character recognition (OCR) techniques to make faxed OMRs machine-readable and used natural language processing (NLP) techniques to capture clinical concepts from both machine-readable OMRs and Mayo clinical notes. We measured the level of overlap in medical concepts between OMRs and Mayo clinical narratives in the quantitative approaches and assessed the salience of concepts specific to Cardiovascular Medicine by calculating the ratio of those mentioned concepts relative to an independent clinical corpus. Among the concepts collected from the OMRs, 11.19% of those were also present in the Mayo clinical narratives that were generated within the 3 months after their initial encounter at the Mayo Clinic. For those common concepts, 73.97% were identified in initial consultant notes (ICNs) and 26.03% were captured over subsequent follow-up consultant notes (FCNs). These findings implied that information collected from the OMRs is potentially informative for patient care, but some valuable information (additionally identified in FCNs) collected from the OMRs is not fully used in an earlier stage of the care process. The concepts collected from the ICNs have the highest salience to Cardiovascular Medicine (0.112) compared to concepts in OMRs and concepts in FCNs. Additionally, unique concepts captured in ICNs (unseen in OMRs or FCNs) carried the most salient information (0.094), which demonstrated that ICNs provided the most informative concepts for the care of transferred patients.

Keywords: Electronic health record; Medical concept evaluation; Medical concept matching; Natural language processing; Optical character recognition; Outside medical records.

Abstract

Grants and funding