Generating and evaluating a propensity model using textual features from electronic medical records

Propensity score (PS) methods are commonly used to control for confounding in comparative effectiveness studies. Electronic health records (EHRs) contain much unstructured data that could be used as proxies for potential confounding factors. The goal of this study was to assess whether the unstructu...

Full description

Saved in:

Bibliographic Details
Published in:	PLOS ONE Vol. 14; no. 3; p. e0212999
Main Authors:	Afzal, Zubair, Masclee, Gwen M. C., Sturkenboom, Miriam C. J. M., Kors, Jan A., Schuemie, Martijn J.
Format:	Journal Article
Language:	English
Published:	United States Public Library of Science (PLoS) 04.03.2019 Public Library of Science
Subjects:	Adult Age Aged Aged, 80 and over Algorithms Anti-inflammatory agents Automation Bleeding Channeling Chi-square test Classification Clinical trials Cohort Studies Comparative studies Confidence intervals Confounding Factors, Epidemiologic COX-2 inhibitors Cyclooxygenase 2 Inhibitors Cyclooxygenase 2 Inhibitors - adverse effects Cyclooxygenase-2 Data Interpretation, Statistical Dosage and administration Drug therapy Electronic Health Records Electronic Health Records - statistics & numerical data Electronic medical records Electronic records EMC NIHES-03-77-01 Family medical history Female Gastrointestinal Hemorrhage Gastrointestinal Hemorrhage - chemically induced Gastrointestinal Hemorrhage - epidemiology Health Health care Health informatics Health services Humans Inflammation Inhibitors Intestine Male Matching Medical records Medicine Medicine and health sciences Methods Middle Aged Netherlands Netherlands - epidemiology Nonsteroidal anti-inflammatory drugs Oxygenase Patients Physical Sciences Primary care Product safety Propensity Score Proportional Hazards Models Prostaglandin endoperoxide synthase Proxy Q R R&D Regression analysis Research & development Research and Analysis Methods Research Article Risk Risk Assessment Risk Assessment - methods Risk factors Science Sex Statistical analysis Studies Surveillance Technology application Unstructured data Netherlands United States > US
ISSN:	1932-6203, 1932-6203
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Propensity score (PS) methods are commonly used to control for confounding in comparative effectiveness studies. Electronic health records (EHRs) contain much unstructured data that could be used as proxies for potential confounding factors. The goal of this study was to assess whether the unstructured information can also be used to construct PS models that would allow to properly deal with confounding. We used an example of coxibs (Cox-2 inhibitors) vs. traditional NSAIDs and the risk of upper gastro-intestinal bleeding as example, since this association is often confounded due to channeling of coxibs to patients at higher risk of upper gastro-intestinal bleeding. In a cohort study of new users of nonsteroidal anti-inflammatory drugs (NSAIDs) from the Dutch Integrated Primary Care Information (IPCI) database, we identified all patients who experienced an upper gastrointestinal bleeding (UGIB). We used a large-scale regularized regression to fit two PS models using all structured and unstructured information in the EHR. We calculated hazard ratios (HRs) to estimate the risk of UGIB among selective cyclo-oxygenase-2 (COX-2) inhibitor users compared to nonselective NSAID (nsNSAID) users. The crude hazard ratio of UGIB for COX-2 inhibitors compared to nsNSAIDs was 0.50 (95% confidence interval 0.18-1.36). Matching only on age resulted in an HR of 0.36 (0.11-1.16), and of 0.35 (0.11-1.11) when further adjusted for sex. Matching on PS only, the first model yielded an HR of 0.42 (0.13-1.38), which reduced to 0.35 (0.96-1.25) when adjusted for age and sex. The second model resulted in an HR of 0.42 (0.13-1.39), which dropped to 0.31 (0.09-1.08) after adjustment for age and sex. PS models can be created using unstructured information in EHRs. An incremental benefit was observed by matching on PS over traditional matching and adjustment for covariates.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 ObjectType-Article-2 ObjectType-Undefined-1 ObjectType-Feature-3 content type line 23 Competing Interests: This study was supported by the VICI project 91896632 of the Netherlands Organization for Health Research and Development ZonMw. Janssen Research and Development provided support in the form of salaries for author M.J.S. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
ISSN:	1932-6203 1932-6203
DOI:	10.1371/journal.pone.0212999