help button home button JAMIA Bigger figures
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH
First published October 12, 2005 as JAMIA PrePrint; doi:10.1197/jamia.M1911
Journal of the American Medical Informatics Association 2006;13(1):52-60
© 2006 American Medical Informatics Association

A more recent version of this article appeared on January 1, 2006
This Article
Right arrow Full Text (PDF)
Right arrow Appendix
Right arrow All Versions of this Article:
M1911v1
13/1/52    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Demner-Fushman, D.
Right arrow Articles by Thoma, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Demner-Fushman, D.
Right arrow Articles by Thoma, G.

Submitted on July 15, 2005
Accepted on September 15, 2005

Automatically Identifying Health Outcome Information in MEDLINE records

Dina Demner-Fushman MD, PhD1*, Barbara Few RN, MSN, MSI1, Susan E. Hauser PhD1, and George Thoma PhD1

Affiliation of the authors: 1 Lister Hill National Center for Biomedical Communications, Communications Engineering Branch, National Library of Medicine, National Institutes of Health, Bethesda, MD

* To whom correspondence should be addressed.

Objective Understanding the effect of a given intervention on the patient's health outcome is one of the key elements in providing optimal patient care. This study presents a methodology for automatic identification of outcomes-related information in medical text and evaluates its potential in satisfying clinical information needs related to healthcare outcomes.

An annotation scheme based on evidence based medicine (EBM) model for critical appraisal of evidence was developed and used to annotate 633 MEDLINEr citations. Textual, structural and meta-information features essential to outcome identification were learned from the created collection and used to develop an automatic system. Accuracy of automatic outcome identification was assessed in an intrinsic evaluation, and in an extrinsic evaluation, in which ranking of MEDLINE search results obtained using PubMedr Clinical Queries relied upon identified outcome statements.

The accuracy and positive predictive value of outcome identification were calculated. Effectiveness of the outcome-based ranking was measured using mean average precision and precision at rank ten.

Results Automatic outcome identification achieved 88 to 93 percent accuracy. The positive predictive value of individual sentences identified as outcomes ranged from 30 to 37 percent. Outcome-based ranking improved retrieval accuracy tripling mean average precision and achieving 389 percent improvement in precision at rank ten.

Conclusion Preliminary results in outcome-based document ranking show potential validity of the EBM model-based approach in timely delivery of information critical to clinical decision support at the point of service.






HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH
Copyright © 1994 by the American Medical Informatics Association.