Learning latent heterogeneity for type 2 diabetes patients using longitudinal health markers in electronic health records.

Publisher: Wiley Country of Publication: England NLM ID: 8215016 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1097-0258 (Electronic) Linking ISSN: 02776715 NLM ISO Abbreviation: Stat Med Subsets: MEDLINE

Original Publication: Chichester ; New York : Wiley, c1982-

Diabetes Mellitus, Type 2* ; Electronic Health Records*; Cohort Studies ; Glycated Hemoglobin/analysis ; Humans ; Retrospective Studies

Electronic health records (EHRs) from type 2 diabetes (T2D) patients consist of longitudinally and sparsely measured health markers at clinical encounters. Our goal is to use such data to learn latent patterns that can inform patient's health status related to T2D while accounting for challenges in retrospectively collected EHRs. To handle challenges such as correlated longitudinal measurements, irregular and informative encounter times, and mixed marker types, we propose multivariate generalized linear models to learn latent patient subgroups. In our model, covariate effects were time-dependent and latent Gaussian processes were introduced to model between-marker correlations over time. Using inferred latent processes, we integrated the irregularly measured health markers of mixed types into composite scores and applied hierarchical clustering to learn latent subgroup structures among T2D patients. Application to an EHR dataset of T2D patients showed different trends of age, sex, and race effects on hypertension/high blood pressure, total cholesterol, glycated hemoglobin, high-density lipoprotein, and medications. The associations among these markers varied over time during the study window. Clustering results revealed four subgroups, each with distinct health status. The same patterns were further confirmed using new EHR records of the same cohort. We developed a novel latent model to integrate longitudinal health markers in EHRs and characterize patient latent heterogeneities. Analysis indicated that there were distinct subgroups of T2D patients, suggesting that effective healthcare managements for these patients should be performed separately for each subgroup.
(© 2021 John Wiley & Sons, Ltd.)

Stat Med. 2006 Apr 30;25(8):1307-22. (PMID: 16217846)
J Med Internet Res. 2005 Mar 14;7(1):e3. (PMID: 15829475)
Stat Interface. 2008;1(1):179-195. (PMID: 18978950)
J Biomed Inform. 2014 Dec;52:199-211. (PMID: 25038555)
J R Stat Soc Series B Stat Methodol. 2015 Sep;77(4):755-776. (PMID: 26568699)
Stat Med. 2002 Nov 15;21(21):3197-217. (PMID: 12375299)
J Am Coll Cardiol. 2014 Jul 1;63(25 Pt B):2889-934. (PMID: 24239923)
J Am Coll Cardiol. 2018 May 15;71(19):e127-e248. (PMID: 29146535)
Health Serv Res. 2012 Aug;47(4):1522-40. (PMID: 22250953)
Stat Methods Med Res. 2014 Feb;23(1):42-59. (PMID: 22523185)
N Engl J Med. 2011 Sep 1;365(9):825-33. (PMID: 21879900)
Diabetes Care. 2018 Jan;41(Suppl 1):S73-S85. (PMID: 29222379)

R21 MH117458 United States MH NIMH NIH HHS; R01 NS073671 United States NS NINDS NIH HHS; R01 GM124104 United States GM NIGMS NIH HHS

Keywords: electronic health records; generalized linear models; kernel smoothing; latent process; type 2 diabetes

0 (Glycated Hemoglobin A)

Date Created: 20210215 Date Completed: 20210629 Latest Revision: 20221207

20231215

PMC8033418

10.1002/sim.8880

33586187