Phase-Type Survival Trees and Mixed Distribution Survival Trees for Clustering Patients' Hospital Length of Stay

Lalit Garg, Sally McClean, Brian Meenan, Peter Millard

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Clinical investigators, health professionals and managers are often interested in developing criteria for clustering patients into clinically meaningful groups according to their expected length of stay. In this paper, we propose two novel types of survival trees; phase-type survival trees and mixed distribution survival trees, which extend previous work on exponential survival trees. The trees are used to cluster the patients with respect to length of stay where partitioning is based on covariates such as gender, age at the time of admission and primary diagnosis code. Likelihood ratio tests are used to determine optimal partitions. The approach is illustrated using nationwide data available from the English Hospital Episode Statistics (HES) database on stroke-related patients, aged 65 years and over, who were discharged from English hospitals over a 1-year period.
LanguageEnglish
Pages57-72
JournalInformatica
Volume22
Issue number1
Publication statusPublished - 11 Apr 2011

Fingerprint

Managers
Health
Statistics

Keywords

  • decision support
  • clinical databases
  • phases of care
  • estimating group
  • service time

Cite this

@article{e8d02093aaf240ea984047662ef3de5e,
title = "Phase-Type Survival Trees and Mixed Distribution Survival Trees for Clustering Patients' Hospital Length of Stay",
abstract = "Clinical investigators, health professionals and managers are often interested in developing criteria for clustering patients into clinically meaningful groups according to their expected length of stay. In this paper, we propose two novel types of survival trees; phase-type survival trees and mixed distribution survival trees, which extend previous work on exponential survival trees. The trees are used to cluster the patients with respect to length of stay where partitioning is based on covariates such as gender, age at the time of admission and primary diagnosis code. Likelihood ratio tests are used to determine optimal partitions. The approach is illustrated using nationwide data available from the English Hospital Episode Statistics (HES) database on stroke-related patients, aged 65 years and over, who were discharged from English hospitals over a 1-year period.",
keywords = "decision support, clinical databases, phases of care, estimating group, service time",
author = "Lalit Garg and Sally McClean and Brian Meenan and Peter Millard",
year = "2011",
month = "4",
day = "11",
language = "English",
volume = "22",
pages = "57--72",
journal = "Informatica",
issn = "0868-4952",
publisher = "IOS Press",
number = "1",

}

Phase-Type Survival Trees and Mixed Distribution Survival Trees for Clustering Patients' Hospital Length of Stay. / Garg, Lalit; McClean, Sally; Meenan, Brian; Millard, Peter.

In: Informatica, Vol. 22, No. 1, 11.04.2011, p. 57-72.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Phase-Type Survival Trees and Mixed Distribution Survival Trees for Clustering Patients' Hospital Length of Stay

AU - Garg, Lalit

AU - McClean, Sally

AU - Meenan, Brian

AU - Millard, Peter

PY - 2011/4/11

Y1 - 2011/4/11

N2 - Clinical investigators, health professionals and managers are often interested in developing criteria for clustering patients into clinically meaningful groups according to their expected length of stay. In this paper, we propose two novel types of survival trees; phase-type survival trees and mixed distribution survival trees, which extend previous work on exponential survival trees. The trees are used to cluster the patients with respect to length of stay where partitioning is based on covariates such as gender, age at the time of admission and primary diagnosis code. Likelihood ratio tests are used to determine optimal partitions. The approach is illustrated using nationwide data available from the English Hospital Episode Statistics (HES) database on stroke-related patients, aged 65 years and over, who were discharged from English hospitals over a 1-year period.

AB - Clinical investigators, health professionals and managers are often interested in developing criteria for clustering patients into clinically meaningful groups according to their expected length of stay. In this paper, we propose two novel types of survival trees; phase-type survival trees and mixed distribution survival trees, which extend previous work on exponential survival trees. The trees are used to cluster the patients with respect to length of stay where partitioning is based on covariates such as gender, age at the time of admission and primary diagnosis code. Likelihood ratio tests are used to determine optimal partitions. The approach is illustrated using nationwide data available from the English Hospital Episode Statistics (HES) database on stroke-related patients, aged 65 years and over, who were discharged from English hospitals over a 1-year period.

KW - decision support

KW - clinical databases

KW - phases of care

KW - estimating group

KW - service time

M3 - Article

VL - 22

SP - 57

EP - 72

JO - Informatica

T2 - Informatica

JF - Informatica

SN - 0868-4952

IS - 1

ER -