Predicting High Cost Patients at Point of Admission using Network Science

Karthik Srinivasan, Faiz Currim, Sudha Ram

Research output: Contribution to journalArticle

Abstract

Data mining models for high-cost patient encounter prediction at the point-of-admission (HPEPP) in inpatient wards are scarce in literature. This is due to lack of availability of relevant features at such an early stage of treatment. In this study, we create a disease co-occurrence network (DCN) using a subset of the State Inpatient database of Arizona. We explore this network for community formation and structural properties to create new input features for HPEPP models. Tree-based data mining models are trained using input feature sets including these new network features and distinct disease communities in the DCN are identified. We propose community membership and high-cost propensity scores as two network based features for HPEPP modeling. We compare the performance of models with different input feature sets and find that the new features significantly improve the accuracy sensitivity and specificity of prediction models. This model has the potential to improve targeted care management and reduce health care expenditure.

Original languageEnglish (US)
JournalIEEE Journal of Biomedical and Health Informatics
DOIs
StateAccepted/In press - Dec 12 2017

Fingerprint

Costs and Cost Analysis
Data Mining
Costs
Inpatients
Propensity Score
Data mining
Health Expenditures
Bioelectric potentials
Databases
Delivery of Health Care
Health care
Sensitivity and Specificity
Structural properties
Availability
Therapeutics

Keywords

  • Data models
  • Diabetes
  • disease co-occurrence networks
  • Diseases
  • healthcare data analysis
  • high cost patients
  • Hypertension
  • Informatics
  • point of admission predictive modeling
  • Predictive models

ASJC Scopus subject areas

  • Biotechnology
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Health Information Management

Cite this

Predicting High Cost Patients at Point of Admission using Network Science. / Srinivasan, Karthik; Currim, Faiz; Ram, Sudha.

In: IEEE Journal of Biomedical and Health Informatics, 12.12.2017.

Research output: Contribution to journalArticle

@article{fb07c6a321b34fefbc4dbbbe05810aaf,
title = "Predicting High Cost Patients at Point of Admission using Network Science",
abstract = "Data mining models for high-cost patient encounter prediction at the point-of-admission (HPEPP) in inpatient wards are scarce in literature. This is due to lack of availability of relevant features at such an early stage of treatment. In this study, we create a disease co-occurrence network (DCN) using a subset of the State Inpatient database of Arizona. We explore this network for community formation and structural properties to create new input features for HPEPP models. Tree-based data mining models are trained using input feature sets including these new network features and distinct disease communities in the DCN are identified. We propose community membership and high-cost propensity scores as two network based features for HPEPP modeling. We compare the performance of models with different input feature sets and find that the new features significantly improve the accuracy sensitivity and specificity of prediction models. This model has the potential to improve targeted care management and reduce health care expenditure.",
keywords = "Data models, Diabetes, disease co-occurrence networks, Diseases, healthcare data analysis, high cost patients, Hypertension, Informatics, point of admission predictive modeling, Predictive models",
author = "Karthik Srinivasan and Faiz Currim and Sudha Ram",
year = "2017",
month = "12",
day = "12",
doi = "10.1109/JBHI.2017.2783049",
language = "English (US)",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Predicting High Cost Patients at Point of Admission using Network Science

AU - Srinivasan, Karthik

AU - Currim, Faiz

AU - Ram, Sudha

PY - 2017/12/12

Y1 - 2017/12/12

N2 - Data mining models for high-cost patient encounter prediction at the point-of-admission (HPEPP) in inpatient wards are scarce in literature. This is due to lack of availability of relevant features at such an early stage of treatment. In this study, we create a disease co-occurrence network (DCN) using a subset of the State Inpatient database of Arizona. We explore this network for community formation and structural properties to create new input features for HPEPP models. Tree-based data mining models are trained using input feature sets including these new network features and distinct disease communities in the DCN are identified. We propose community membership and high-cost propensity scores as two network based features for HPEPP modeling. We compare the performance of models with different input feature sets and find that the new features significantly improve the accuracy sensitivity and specificity of prediction models. This model has the potential to improve targeted care management and reduce health care expenditure.

AB - Data mining models for high-cost patient encounter prediction at the point-of-admission (HPEPP) in inpatient wards are scarce in literature. This is due to lack of availability of relevant features at such an early stage of treatment. In this study, we create a disease co-occurrence network (DCN) using a subset of the State Inpatient database of Arizona. We explore this network for community formation and structural properties to create new input features for HPEPP models. Tree-based data mining models are trained using input feature sets including these new network features and distinct disease communities in the DCN are identified. We propose community membership and high-cost propensity scores as two network based features for HPEPP modeling. We compare the performance of models with different input feature sets and find that the new features significantly improve the accuracy sensitivity and specificity of prediction models. This model has the potential to improve targeted care management and reduce health care expenditure.

KW - Data models

KW - Diabetes

KW - disease co-occurrence networks

KW - Diseases

KW - healthcare data analysis

KW - high cost patients

KW - Hypertension

KW - Informatics

KW - point of admission predictive modeling

KW - Predictive models

UR - http://www.scopus.com/inward/record.url?scp=85038874422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038874422&partnerID=8YFLogxK

U2 - 10.1109/JBHI.2017.2783049

DO - 10.1109/JBHI.2017.2783049

M3 - Article

C2 - 29990022

AN - SCOPUS:85038874422

JO - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

ER -