Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023;16(4):505-515.
doi: 10.4310/22-sii739. Epub 2023 Apr 14.

Estimating individualized treatment rules for multicategory type 2 diabetes treatments using electronic health records

Jitong Lou  1 , Yuanjia Wang  2 , Lang Li  3 , Donglin Zeng  4
Affiliations

Estimating individualized treatment rules for multicategory type 2 diabetes treatments using electronic health records

Jitong Lou et al. Stat Interface. 2023.

Abstract

In this article, we propose a general framework to learn optimal treatment rules for type 2 diabetes (T2D) patients using electronic health records (EHRs). We first propose a joint modeling approach to characterize patient's pretreatment conditions using longitudinal markers from EHRs. The estimation accounts for informative measurement times using inverse-intensity weighting methods. The predicted latent processes in the joint model are used to divide patients into a finite of subgroups and, within each group, patients share similar health profiles in EHRs. Within each patient group, we estimate optimal individualized treatment rules by extending a matched learning method to handle multicategory treatments using a one-versus-one approach. Each matched learning for two treatments is implemented by a weighted support vector machine with matched pairs of patients. We apply our method to estimate optimal treatment rules for T2D patients in a large sample of EHRs from the Ohio State University Wexner Medical Center. We demonstrate the utility of our method to select the optimal treatments from four classes of drugs and achieve a better control of glycated hemoglobin than any one-size-fits-all rules.

Keywords: Electronic health records; Individualized treatment rules; Latent process; Machine learning; Multicategory treatments; Type 2 diabetes.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Flow chart of learning optimal ITRs using OSU-WMC EHRs.
Figure 2.
Figure 2.
Averages of normalized measurements by health markers and patient subgroups using 24-month-data before baseline treatment dates. Red: more severe status than the overall sample average in terms of a health marker; Blue: healthier status than the overall sample average in terms of a health marker; White: overall sample average status in terms of a health marker.
Figure 3.
Figure 3.
The empirical value function for the expected HbA1c level (%) using 2-fold cross-validation with 100 repeats (a lower value is more beneficial).
Figure 4.
Figure 4.
The distribution of observed treatments versus treatments recommended by estimated ITRs within each subgroup.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Abramowitz M and Stegun IA (1965). Handbook of Mathematical Functions: With Formulas, Graphs, and Mathematical Tables. Applied mathematics series. Dover Publications, New York, NY, USA. MR0415956
    1. Andersen PK and Gill RD (1982). Cox’s regression model for counting processes: a large sample study. The Annals of Statistics 10 1100–1120. MR0673646
    1. Antonelli J, Cefalu M, Palmer N and Agniel D (2018). Doubly robust matching estimators for high dimensional confounding adjustment. Biometrics 74 1171–1179. MR3908135 - PMC - PubMed
    1. American Diabetes Association (2018). Pharmacologic approaches to glycemic treatment: standards of medical care in diabetes–2018. Diabetes Care 41 73–85. - PubMed
    1. American Diabetes Association (2021). Standards of Medical Care in Diabetes—2021. Diabetes Care 44 S73–S150. - PubMed

LinkOut - more resources