Identification of Clinically Meaningful Plasma Transfusion Subgroups Using Unsupervised Random Forest Clustering

Research output: Contribution to journalArticle

Abstract

Statistical techniques such as propensity score matching and instrumental variable are commonly employed to "simulate" randomization and adjust for measured confounders in comparative effectiveness research. Despite such adjustments, the results of these methods apply essentially to an "average" patient. However, as patients show significant heterogeneity in their responses to treatments, this average effect is of limited value. It does not account for individual level variabilities, which can deviate substantially from the population average. To address this critical problem, we present a framework that allows the discovery of clinically meaningful homogeneous subgroups with differential effects of plasma transfusion using unsupervised random forest clustering. Subgroup analysis using two blood transfusion datasets show that considerable variablilities exist between the subgroups and population in both the treatment effect of plasma transfusion on bleeding and mortality and risk factors for these outcomes. These results support the customization of blood transfusion therapy for the individual patient.

Original languageEnglish (US)
Pages (from-to)1332-1341
Number of pages10
JournalAMIA ... Annual Symposium proceedings. AMIA Symposium
Volume2017
StatePublished - Jan 1 2017

Fingerprint

Cluster Analysis
Blood Transfusion
Comparative Effectiveness Research
Propensity Score
Random Allocation
Population
Therapeutics
Hemorrhage
Mortality
Datasets

Keywords

  • bleeding
  • Plasma transfusion
  • subgroup analysis.
  • unsupervised learning

ASJC Scopus subject areas

  • Medicine(all)

Cite this

@article{e9b7ba45bdf84b50b110c0cf29c1136c,
title = "Identification of Clinically Meaningful Plasma Transfusion Subgroups Using Unsupervised Random Forest Clustering",
abstract = "Statistical techniques such as propensity score matching and instrumental variable are commonly employed to {"}simulate{"} randomization and adjust for measured confounders in comparative effectiveness research. Despite such adjustments, the results of these methods apply essentially to an {"}average{"} patient. However, as patients show significant heterogeneity in their responses to treatments, this average effect is of limited value. It does not account for individual level variabilities, which can deviate substantially from the population average. To address this critical problem, we present a framework that allows the discovery of clinically meaningful homogeneous subgroups with differential effects of plasma transfusion using unsupervised random forest clustering. Subgroup analysis using two blood transfusion datasets show that considerable variablilities exist between the subgroups and population in both the treatment effect of plasma transfusion on bleeding and mortality and risk factors for these outcomes. These results support the customization of blood transfusion therapy for the individual patient.",
keywords = "bleeding, Plasma transfusion, subgroup analysis., unsupervised learning",
author = "Che Ngufor and Matthew Warner and Murphree, {Dennis H.} and Liu, {Hongfang D} and Carter, {Rickey E.} and Curtis Storlie and Kor, {Daryl J}",
year = "2017",
month = "1",
day = "1",
language = "English (US)",
volume = "2017",
pages = "1332--1341",
journal = "AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium",
issn = "1559-4076",
publisher = "American Medical Informatics Association",

}

TY - JOUR

T1 - Identification of Clinically Meaningful Plasma Transfusion Subgroups Using Unsupervised Random Forest Clustering

AU - Ngufor, Che

AU - Warner, Matthew

AU - Murphree, Dennis H.

AU - Liu, Hongfang D

AU - Carter, Rickey E.

AU - Storlie, Curtis

AU - Kor, Daryl J

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Statistical techniques such as propensity score matching and instrumental variable are commonly employed to "simulate" randomization and adjust for measured confounders in comparative effectiveness research. Despite such adjustments, the results of these methods apply essentially to an "average" patient. However, as patients show significant heterogeneity in their responses to treatments, this average effect is of limited value. It does not account for individual level variabilities, which can deviate substantially from the population average. To address this critical problem, we present a framework that allows the discovery of clinically meaningful homogeneous subgroups with differential effects of plasma transfusion using unsupervised random forest clustering. Subgroup analysis using two blood transfusion datasets show that considerable variablilities exist between the subgroups and population in both the treatment effect of plasma transfusion on bleeding and mortality and risk factors for these outcomes. These results support the customization of blood transfusion therapy for the individual patient.

AB - Statistical techniques such as propensity score matching and instrumental variable are commonly employed to "simulate" randomization and adjust for measured confounders in comparative effectiveness research. Despite such adjustments, the results of these methods apply essentially to an "average" patient. However, as patients show significant heterogeneity in their responses to treatments, this average effect is of limited value. It does not account for individual level variabilities, which can deviate substantially from the population average. To address this critical problem, we present a framework that allows the discovery of clinically meaningful homogeneous subgroups with differential effects of plasma transfusion using unsupervised random forest clustering. Subgroup analysis using two blood transfusion datasets show that considerable variablilities exist between the subgroups and population in both the treatment effect of plasma transfusion on bleeding and mortality and risk factors for these outcomes. These results support the customization of blood transfusion therapy for the individual patient.

KW - bleeding

KW - Plasma transfusion

KW - subgroup analysis.

KW - unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85058739104&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058739104&partnerID=8YFLogxK

M3 - Article

C2 - 29854202

AN - SCOPUS:85058739104

VL - 2017

SP - 1332

EP - 1341

JO - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

JF - AMIA ... Annual Symposium proceedings / AMIA Symposium. AMIA Symposium

SN - 1559-4076

ER -