EUVIMEDEUVIMED entdecken
Open Access

Scalable Feature Engineering from Electronic Free Text Notes to Supplement Confounding Adjustment of Claims‐Based Pharmacoepidemiologic Studies

Richard Wyss, Joseph M Plasek iD, Li Zhou, Lily G Bessette iD, Sebastian Schneeweiss iD, Jeremy A Rassen iD, Theodore Tsacogianis, Kueiyu Joshua Lin

Clinical Pharmacology & Therapeutics · 2023

DOI: 10.1002/cpt.2826 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Clinical Pharmacology & Therapeutics, 2023. Verfasst von Richard Wyss und 7 weiteren Autor:innen. Beteiligte Einrichtung: Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Mass…. Hauptthemen laut MEDLINE: Medicare, Electronic Health Records. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NLM NIH HHS, National Institutes of Health.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Natural language processing (NLP) tools turn free‐text notes (FTNs) from electronic health records (EHRs) into data features that can supplement confounding adjustment in pharmacoepidemiologic studies. However, current applications are difficult to scale. We used unsupervised NLP to generate high‐dimensional feature spaces from FTNs to improve prediction of drug exposure and outcomes compared with claims‐based analyses. We linked Medicare claims with EHR data to generate three cohort studies comparing different classes of medications on the risk of various clinical outcomes. We used “bag‐of‐words” to generate features for the top 20,000 most prevalent terms from FTNs. We compared machine learning (ML) prediction algorithms using different sets of candidate predictors: Set1 (39 researcher‐specified variables), Set2 (Set1 + ML‐selected claims codes), and Set3 (Set1 + ML‐selected NLP‐generated features), vs. Set4 (Set1 + 2 + 3). When modeling treatment choice, we observed a consistent pattern across the examples: ML models utilizing Set4 performed best followed by Set2, Set3, then Set1. When modeling the outcome risk, there was little to no improvement beyond models based on Set1. Supplementing claims data with NLP‐generated features from free text notes improved prediction of prescribing choices but had little or no improvement on clinical risk prediction. These findings have implications for strategies to improve confounding using EHR data in pharmacoepidemiologic studies.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MedicareElectronic Health RecordsAgedUnited StatesHumansCohort StudiesNatural Language ProcessingAlgorithms

Autor:innen und Institutionen

  • Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA.
  • Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA.
  • Aetion, Inc, New York, New York, USA.
  • Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA.

ORCID-Profile: Joseph M Plasek, Lily G Bessette, Sebastian Schneeweiss, Jeremy A Rassen

Förderung

  • NLM NIH HHS
  • National Institutes of Health

Zitieren

APA

Wyss, R., Plasek, J. M., Zhou, L., Bessette, L. G., Schneeweiss, S., Rassen, J. A., Tsacogianis, T., & Lin, K. J. (2023). Scalable Feature Engineering from Electronic Free Text Notes to Supplement Confounding Adjustment of Claims‐Based Pharmacoepidemiologic Studies. Clinical Pharmacology & Therapeutics. https://doi.org/10.1002/cpt.2826

Vancouver

Wyss R, Plasek JM, Zhou L, Bessette LG, Schneeweiss S, Rassen JA, et al. Scalable Feature Engineering from Electronic Free Text Notes to Supplement Confounding Adjustment of Claims‐Based Pharmacoepidemiologic Studies. Clinical Pharmacology & Therapeutics. 2023. doi:10.1002/cpt.2826

BibTeX
@article{Wyss2023,
  author = {Wyss, Richard and Plasek, Joseph M and Zhou, Li and Bessette, Lily G and Schneeweiss, Sebastian and Rassen, Jeremy A and Tsacogianis, Theodore and Lin, Kueiyu Joshua},
  title = {Scalable Feature Engineering from Electronic Free Text Notes to Supplement Confounding Adjustment of Claims‐Based Pharmacoepidemiologic Studies},
  journal = {Clinical Pharmacology & Therapeutics},
  year = {2023},
  doi = {10.1002/cpt.2826}
}
Kostenlos starten →