Strategies for handling missing clinical data for automated surgical site infection detection from the electronic health record

Zhen Hu; Genevieve B. Melton; Elliot G. Arsoniadis; Yan Wang; Mary R. Kwaan; Gyorgy J. Simon

doi:10.1016/j.jbi.2017.03.009

Strategies for handling missing clinical data for automated surgical site infection detection from the electronic health record

Zhen Hu, Genevieve B. Melton, Elliot G. Arsoniadis, Yan Wang, Mary R. Kwaan, Gyorgy J. Simon

Research output: Contribution to journal › Article › peer-review

68 Scopus citations

Abstract

Proper handling of missing data is important for many secondary uses of electronic health record (EHR) data. Data imputation methods can be used to handle missing data, but their use for analyzing EHR data is limited and specific efficacy for postoperative complication detection is unclear. Several data imputation methods were used to develop data models for automated detection of three types (i.e., superficial, deep, and organ space) of surgical site infection (SSI) and overall SSI using American College of Surgeons National Surgical Quality Improvement Project (NSQIP) Registry 30-day SSI occurrence data as a reference standard. Overall, models with missing data imputation almost always outperformed reference models without imputation that included only cases with complete data for detection of SSI overall achieving very good average area under the curve values. Missing data imputation appears to be an effective means for improving postoperative SSI detection using EHR clinical data.

Original language	English (US)
Pages (from-to)	112-120
Number of pages	9
Journal	Journal of Biomedical Informatics
Volume	68
DOIs	https://doi.org/10.1016/j.jbi.2017.03.009
State	Published - Apr 1 2017

Bibliographical note

Funding Information:
This research was supported by the University of Minnesota Academic Health Center Faculty Development Award (GS, GM), the American Surgical Association Foundation (GM), the Agency for Healthcare Research and Quality (R01HS24532-01A1), and the National Institutes of Health (NIH) Clinical and Translational Science Award (CTSA) program (8UL1TR000114-02). The authors also thank Fairview Health Services for support of this research.

Publisher Copyright:
© 2017

Keywords

Electronic health records
Missing data
Surgical site infections

Access

10.1016/j.jbi.2017.03.009

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5474942

OpenUrl availability

Full text

Cite this

@article{e60d20b431c74f28ab1880164b4daaee,

title = "Strategies for handling missing clinical data for automated surgical site infection detection from the electronic health record",

abstract = "Proper handling of missing data is important for many secondary uses of electronic health record (EHR) data. Data imputation methods can be used to handle missing data, but their use for analyzing EHR data is limited and specific efficacy for postoperative complication detection is unclear. Several data imputation methods were used to develop data models for automated detection of three types (i.e., superficial, deep, and organ space) of surgical site infection (SSI) and overall SSI using American College of Surgeons National Surgical Quality Improvement Project (NSQIP) Registry 30-day SSI occurrence data as a reference standard. Overall, models with missing data imputation almost always outperformed reference models without imputation that included only cases with complete data for detection of SSI overall achieving very good average area under the curve values. Missing data imputation appears to be an effective means for improving postoperative SSI detection using EHR clinical data.",

keywords = "Electronic health records, Missing data, Surgical site infections",

author = "Zhen Hu and Melton, {Genevieve B.} and Arsoniadis, {Elliot G.} and Yan Wang and Kwaan, {Mary R.} and Simon, {Gyorgy J.}",

note = "Funding Information: This research was supported by the University of Minnesota Academic Health Center Faculty Development Award (GS, GM), the American Surgical Association Foundation (GM), the Agency for Healthcare Research and Quality (R01HS24532-01A1), and the National Institutes of Health (NIH) Clinical and Translational Science Award (CTSA) program (8UL1TR000114-02). The authors also thank Fairview Health Services for support of this research. Publisher Copyright: {\textcopyright} 2017",

year = "2017",

month = apr,

day = "1",

doi = "10.1016/j.jbi.2017.03.009",

language = "English (US)",

volume = "68",

pages = "112--120",

journal = "Journal of Biomedical Informatics",

issn = "1532-0464",

publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Strategies for handling missing clinical data for automated surgical site infection detection from the electronic health record

AU - Hu, Zhen

AU - Melton, Genevieve B.

AU - Arsoniadis, Elliot G.

AU - Wang, Yan

AU - Kwaan, Mary R.

AU - Simon, Gyorgy J.

N1 - Funding Information: This research was supported by the University of Minnesota Academic Health Center Faculty Development Award (GS, GM), the American Surgical Association Foundation (GM), the Agency for Healthcare Research and Quality (R01HS24532-01A1), and the National Institutes of Health (NIH) Clinical and Translational Science Award (CTSA) program (8UL1TR000114-02). The authors also thank Fairview Health Services for support of this research. Publisher Copyright: © 2017

PY - 2017/4/1

Y1 - 2017/4/1

N2 - Proper handling of missing data is important for many secondary uses of electronic health record (EHR) data. Data imputation methods can be used to handle missing data, but their use for analyzing EHR data is limited and specific efficacy for postoperative complication detection is unclear. Several data imputation methods were used to develop data models for automated detection of three types (i.e., superficial, deep, and organ space) of surgical site infection (SSI) and overall SSI using American College of Surgeons National Surgical Quality Improvement Project (NSQIP) Registry 30-day SSI occurrence data as a reference standard. Overall, models with missing data imputation almost always outperformed reference models without imputation that included only cases with complete data for detection of SSI overall achieving very good average area under the curve values. Missing data imputation appears to be an effective means for improving postoperative SSI detection using EHR clinical data.

AB - Proper handling of missing data is important for many secondary uses of electronic health record (EHR) data. Data imputation methods can be used to handle missing data, but their use for analyzing EHR data is limited and specific efficacy for postoperative complication detection is unclear. Several data imputation methods were used to develop data models for automated detection of three types (i.e., superficial, deep, and organ space) of surgical site infection (SSI) and overall SSI using American College of Surgeons National Surgical Quality Improvement Project (NSQIP) Registry 30-day SSI occurrence data as a reference standard. Overall, models with missing data imputation almost always outperformed reference models without imputation that included only cases with complete data for detection of SSI overall achieving very good average area under the curve values. Missing data imputation appears to be an effective means for improving postoperative SSI detection using EHR clinical data.

KW - Electronic health records

KW - Missing data

KW - Surgical site infections

UR - http://www.scopus.com/inward/record.url?scp=85016029838&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016029838&partnerID=8YFLogxK

U2 - 10.1016/j.jbi.2017.03.009

DO - 10.1016/j.jbi.2017.03.009

M3 - Article

C2 - 28323112

AN - SCOPUS:85016029838

SN - 1532-0464

VL - 68

SP - 112

EP - 120

JO - Journal of Biomedical Informatics

JF - Journal of Biomedical Informatics

ER -

Strategies for handling missing clinical data for automated surgical site infection detection from the electronic health record

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this