An automated model to identify heart failure patients at risk for 30-day readmission or death using electronic medical record data

Ruben Amarasingham, Billy J. Moore, Ying P. Tabak, Mark H. Drazner, Christopher A. Clark, Song Zhang, W. Gary Reed, Timothy S. Swanson, Ying Ma, Ethan A. Halm

Research output: Contribution to journalArticle

276 Citations (Scopus)

Abstract

Background: A real-time electronic predictive model that identifies hospitalized heart failure (HF) patients at high risk for readmission or death may be valuable to clinicians and hospitals who care for these patients. Methods: An automated predictive model for 30-day readmission and death was derived and validated from clinical and nonclinical risk factors present on admission in 1372 HF hospitalizations to a major urban hospital between January 2007 and August 2008. Data were extracted from an electronic medical record. The performance of the electronic model was compared with mortality and readmission models developed by the Center for Medicaid and Medicare Services (CMS models) and a HF mortality model derived from the Acute Decompensated Heart Failure Registry (ADHERE model). Results: The 30-day mortality and readmission rates were 3.1% and 24.1% respectively. The electronic model demonstrated good discrimination for 30 day mortality (C statistic 0.86) and readmission (C statistic 0.72) and performed as well, or better than, the ADHERE model and CMS models for both outcomes (C statistic ranges: 0.72-0.73 and 0.56-0.66 for mortality and readmissions respectively; P < 0.05 in all comparisons). Markers of social instability and lower socioeconomic status improved readmission prediction in the electronic model (C statistic 0.72 vs. 0.61, P < 0.05). Conclusions: Clinical and social factors available within hours of hospital presentation and extractable from an EMR predicted mortality and readmission at 30 days. Incorporating complex social factors increased the model's accuracy, suggesting that such factors could enhance risk adjustment models designed to compare hospital readmission rates.

Original languageEnglish (US)
Pages (from-to)981-988
Number of pages8
JournalMedical Care
Volume48
Issue number11
DOIs
StatePublished - Nov 2010

Fingerprint

Electronic Health Records
Heart Failure
Mortality
Risk Adjustment
Centers for Medicare and Medicaid Services (U.S.)
Patient Readmission
Urban Hospitals
Social Class
Registries
Patient Care
Hospitalization

Keywords

  • health policy
  • informatics
  • quality improvement

ASJC Scopus subject areas

  • Public Health, Environmental and Occupational Health

Cite this

An automated model to identify heart failure patients at risk for 30-day readmission or death using electronic medical record data. / Amarasingham, Ruben; Moore, Billy J.; Tabak, Ying P.; Drazner, Mark H.; Clark, Christopher A.; Zhang, Song; Reed, W. Gary; Swanson, Timothy S.; Ma, Ying; Halm, Ethan A.

In: Medical Care, Vol. 48, No. 11, 11.2010, p. 981-988.

Research output: Contribution to journalArticle

Amarasingham, Ruben ; Moore, Billy J. ; Tabak, Ying P. ; Drazner, Mark H. ; Clark, Christopher A. ; Zhang, Song ; Reed, W. Gary ; Swanson, Timothy S. ; Ma, Ying ; Halm, Ethan A. / An automated model to identify heart failure patients at risk for 30-day readmission or death using electronic medical record data. In: Medical Care. 2010 ; Vol. 48, No. 11. pp. 981-988.
@article{9e1fbc64d0f04954b31f928c69ae4b4e,
title = "An automated model to identify heart failure patients at risk for 30-day readmission or death using electronic medical record data",
abstract = "Background: A real-time electronic predictive model that identifies hospitalized heart failure (HF) patients at high risk for readmission or death may be valuable to clinicians and hospitals who care for these patients. Methods: An automated predictive model for 30-day readmission and death was derived and validated from clinical and nonclinical risk factors present on admission in 1372 HF hospitalizations to a major urban hospital between January 2007 and August 2008. Data were extracted from an electronic medical record. The performance of the electronic model was compared with mortality and readmission models developed by the Center for Medicaid and Medicare Services (CMS models) and a HF mortality model derived from the Acute Decompensated Heart Failure Registry (ADHERE model). Results: The 30-day mortality and readmission rates were 3.1{\%} and 24.1{\%} respectively. The electronic model demonstrated good discrimination for 30 day mortality (C statistic 0.86) and readmission (C statistic 0.72) and performed as well, or better than, the ADHERE model and CMS models for both outcomes (C statistic ranges: 0.72-0.73 and 0.56-0.66 for mortality and readmissions respectively; P < 0.05 in all comparisons). Markers of social instability and lower socioeconomic status improved readmission prediction in the electronic model (C statistic 0.72 vs. 0.61, P < 0.05). Conclusions: Clinical and social factors available within hours of hospital presentation and extractable from an EMR predicted mortality and readmission at 30 days. Incorporating complex social factors increased the model's accuracy, suggesting that such factors could enhance risk adjustment models designed to compare hospital readmission rates.",
keywords = "health policy, informatics, quality improvement",
author = "Ruben Amarasingham and Moore, {Billy J.} and Tabak, {Ying P.} and Drazner, {Mark H.} and Clark, {Christopher A.} and Song Zhang and Reed, {W. Gary} and Swanson, {Timothy S.} and Ying Ma and Halm, {Ethan A.}",
year = "2010",
month = "11",
doi = "10.1097/MLR.0b013e3181ef60d9",
language = "English (US)",
volume = "48",
pages = "981--988",
journal = "Medical Care",
issn = "0025-7079",
publisher = "Lippincott Williams and Wilkins",
number = "11",

}

TY - JOUR

T1 - An automated model to identify heart failure patients at risk for 30-day readmission or death using electronic medical record data

AU - Amarasingham, Ruben

AU - Moore, Billy J.

AU - Tabak, Ying P.

AU - Drazner, Mark H.

AU - Clark, Christopher A.

AU - Zhang, Song

AU - Reed, W. Gary

AU - Swanson, Timothy S.

AU - Ma, Ying

AU - Halm, Ethan A.

PY - 2010/11

Y1 - 2010/11

N2 - Background: A real-time electronic predictive model that identifies hospitalized heart failure (HF) patients at high risk for readmission or death may be valuable to clinicians and hospitals who care for these patients. Methods: An automated predictive model for 30-day readmission and death was derived and validated from clinical and nonclinical risk factors present on admission in 1372 HF hospitalizations to a major urban hospital between January 2007 and August 2008. Data were extracted from an electronic medical record. The performance of the electronic model was compared with mortality and readmission models developed by the Center for Medicaid and Medicare Services (CMS models) and a HF mortality model derived from the Acute Decompensated Heart Failure Registry (ADHERE model). Results: The 30-day mortality and readmission rates were 3.1% and 24.1% respectively. The electronic model demonstrated good discrimination for 30 day mortality (C statistic 0.86) and readmission (C statistic 0.72) and performed as well, or better than, the ADHERE model and CMS models for both outcomes (C statistic ranges: 0.72-0.73 and 0.56-0.66 for mortality and readmissions respectively; P < 0.05 in all comparisons). Markers of social instability and lower socioeconomic status improved readmission prediction in the electronic model (C statistic 0.72 vs. 0.61, P < 0.05). Conclusions: Clinical and social factors available within hours of hospital presentation and extractable from an EMR predicted mortality and readmission at 30 days. Incorporating complex social factors increased the model's accuracy, suggesting that such factors could enhance risk adjustment models designed to compare hospital readmission rates.

AB - Background: A real-time electronic predictive model that identifies hospitalized heart failure (HF) patients at high risk for readmission or death may be valuable to clinicians and hospitals who care for these patients. Methods: An automated predictive model for 30-day readmission and death was derived and validated from clinical and nonclinical risk factors present on admission in 1372 HF hospitalizations to a major urban hospital between January 2007 and August 2008. Data were extracted from an electronic medical record. The performance of the electronic model was compared with mortality and readmission models developed by the Center for Medicaid and Medicare Services (CMS models) and a HF mortality model derived from the Acute Decompensated Heart Failure Registry (ADHERE model). Results: The 30-day mortality and readmission rates were 3.1% and 24.1% respectively. The electronic model demonstrated good discrimination for 30 day mortality (C statistic 0.86) and readmission (C statistic 0.72) and performed as well, or better than, the ADHERE model and CMS models for both outcomes (C statistic ranges: 0.72-0.73 and 0.56-0.66 for mortality and readmissions respectively; P < 0.05 in all comparisons). Markers of social instability and lower socioeconomic status improved readmission prediction in the electronic model (C statistic 0.72 vs. 0.61, P < 0.05). Conclusions: Clinical and social factors available within hours of hospital presentation and extractable from an EMR predicted mortality and readmission at 30 days. Incorporating complex social factors increased the model's accuracy, suggesting that such factors could enhance risk adjustment models designed to compare hospital readmission rates.

KW - health policy

KW - informatics

KW - quality improvement

UR - http://www.scopus.com/inward/record.url?scp=78049334037&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049334037&partnerID=8YFLogxK

U2 - 10.1097/MLR.0b013e3181ef60d9

DO - 10.1097/MLR.0b013e3181ef60d9

M3 - Article

C2 - 20940649

AN - SCOPUS:78049334037

VL - 48

SP - 981

EP - 988

JO - Medical Care

JF - Medical Care

SN - 0025-7079

IS - 11

ER -