Predicting 30-Day Hospital Readmissions in Acute Myocardial Infarction: The AMI "READMITS" (Renal Function, Elevated Brain Natriuretic Peptide, Age, Diabetes Mellitus, Nonmale Sex, Intervention with Timely Percutaneous Coronary Intervention, and Low Systolic Blood Pressure) Score

Oanh K Nguyen, Anil N Makam, Christopher Clark, Song Zhang, Sandeep R Das, Ethan A Halm

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

BACKGROUND: Readmissions after hospitalization for acute myocardial infarction (AMI) are common. However, the few currently available AMI readmission risk prediction models have poor-to-modest predictive ability and are not readily actionable in real time. We sought to develop an actionable and accurate AMI readmission risk prediction model to identify high-risk patients as early as possible during hospitalization. METHODS AND RESULTS: We used electronic health record data from consecutive AMI hospitalizations from 6 hospitals in north Texas from 2009 to 2010 to derive and validate models predicting all-cause nonelective 30-day readmissions, using stepwise backward selection and 5-fold cross-validation. Of 826 patients hospitalized with AMI, 13% had a 30-day readmission. The first-day AMI model (the AMI "READMITS" score) included 7 predictors: renal function, elevated brain natriuretic peptide, age, diabetes mellitus, nonmale sex, intervention with timely percutaneous coronary intervention, and low systolic blood pressure, had an optimism-corrected C-statistic of 0.73 (95% confidence interval, 0.71-0.74) and was well calibrated. The full-stay AMI model, which included 3 additional predictors (use of intravenous diuretics, anemia on discharge, and discharge to postacute care), had an optimism-corrected C-statistic of 0.75 (95% confidence interval, 0.74-0.76) with minimally improved net reclassification and calibration. Both AMI models outperformed corresponding multicondition readmission models. CONCLUSIONS: The parsimonious AMI READMITS score enables early prospective identification of high-risk AMI patients for targeted readmissions reduction interventions within the first 24 hours of hospitalization. A full-stay AMI readmission model only modestly outperformed the AMI READMITS score in terms of discrimination, but surprisingly did not meaningfully improve reclassification.

Original languageEnglish (US)
JournalJournal of the American Heart Association
Volume7
Issue number8
DOIs
StatePublished - Apr 17 2018

Fingerprint

Patient Readmission
Brain Natriuretic Peptide
Percutaneous Coronary Intervention
Hypotension
Diabetes Mellitus
Myocardial Infarction
Blood Pressure
Kidney
Hospitalization
Subacute Care
Confidence Intervals
Aptitude
Electronic Health Records
Diuretics
Calibration
Anemia

Keywords

  • acute myocardial infarction
  • health services research
  • hospital performance
  • prediction
  • readmission

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine

Cite this

@article{ff0e2b4e87cc46f1996c04d4d92ad398,
title = "Predicting 30-Day Hospital Readmissions in Acute Myocardial Infarction: The AMI {"}READMITS{"} (Renal Function, Elevated Brain Natriuretic Peptide, Age, Diabetes Mellitus, Nonmale Sex, Intervention with Timely Percutaneous Coronary Intervention, and Low Systolic Blood Pressure) Score",
abstract = "BACKGROUND: Readmissions after hospitalization for acute myocardial infarction (AMI) are common. However, the few currently available AMI readmission risk prediction models have poor-to-modest predictive ability and are not readily actionable in real time. We sought to develop an actionable and accurate AMI readmission risk prediction model to identify high-risk patients as early as possible during hospitalization. METHODS AND RESULTS: We used electronic health record data from consecutive AMI hospitalizations from 6 hospitals in north Texas from 2009 to 2010 to derive and validate models predicting all-cause nonelective 30-day readmissions, using stepwise backward selection and 5-fold cross-validation. Of 826 patients hospitalized with AMI, 13{\%} had a 30-day readmission. The first-day AMI model (the AMI {"}READMITS{"} score) included 7 predictors: renal function, elevated brain natriuretic peptide, age, diabetes mellitus, nonmale sex, intervention with timely percutaneous coronary intervention, and low systolic blood pressure, had an optimism-corrected C-statistic of 0.73 (95{\%} confidence interval, 0.71-0.74) and was well calibrated. The full-stay AMI model, which included 3 additional predictors (use of intravenous diuretics, anemia on discharge, and discharge to postacute care), had an optimism-corrected C-statistic of 0.75 (95{\%} confidence interval, 0.74-0.76) with minimally improved net reclassification and calibration. Both AMI models outperformed corresponding multicondition readmission models. CONCLUSIONS: The parsimonious AMI READMITS score enables early prospective identification of high-risk AMI patients for targeted readmissions reduction interventions within the first 24 hours of hospitalization. A full-stay AMI readmission model only modestly outperformed the AMI READMITS score in terms of discrimination, but surprisingly did not meaningfully improve reclassification.",
keywords = "acute myocardial infarction, health services research, hospital performance, prediction, readmission",
author = "Nguyen, {Oanh K} and Makam, {Anil N} and Christopher Clark and Song Zhang and Das, {Sandeep R} and Halm, {Ethan A}",
year = "2018",
month = "4",
day = "17",
doi = "10.1161/JAHA.118.008882",
language = "English (US)",
volume = "7",
journal = "Journal of the American Heart Association",
issn = "2047-9980",
publisher = "Wiley-Blackwell",
number = "8",

}

TY - JOUR

T1 - Predicting 30-Day Hospital Readmissions in Acute Myocardial Infarction

T2 - The AMI "READMITS" (Renal Function, Elevated Brain Natriuretic Peptide, Age, Diabetes Mellitus, Nonmale Sex, Intervention with Timely Percutaneous Coronary Intervention, and Low Systolic Blood Pressure) Score

AU - Nguyen, Oanh K

AU - Makam, Anil N

AU - Clark, Christopher

AU - Zhang, Song

AU - Das, Sandeep R

AU - Halm, Ethan A

PY - 2018/4/17

Y1 - 2018/4/17

N2 - BACKGROUND: Readmissions after hospitalization for acute myocardial infarction (AMI) are common. However, the few currently available AMI readmission risk prediction models have poor-to-modest predictive ability and are not readily actionable in real time. We sought to develop an actionable and accurate AMI readmission risk prediction model to identify high-risk patients as early as possible during hospitalization. METHODS AND RESULTS: We used electronic health record data from consecutive AMI hospitalizations from 6 hospitals in north Texas from 2009 to 2010 to derive and validate models predicting all-cause nonelective 30-day readmissions, using stepwise backward selection and 5-fold cross-validation. Of 826 patients hospitalized with AMI, 13% had a 30-day readmission. The first-day AMI model (the AMI "READMITS" score) included 7 predictors: renal function, elevated brain natriuretic peptide, age, diabetes mellitus, nonmale sex, intervention with timely percutaneous coronary intervention, and low systolic blood pressure, had an optimism-corrected C-statistic of 0.73 (95% confidence interval, 0.71-0.74) and was well calibrated. The full-stay AMI model, which included 3 additional predictors (use of intravenous diuretics, anemia on discharge, and discharge to postacute care), had an optimism-corrected C-statistic of 0.75 (95% confidence interval, 0.74-0.76) with minimally improved net reclassification and calibration. Both AMI models outperformed corresponding multicondition readmission models. CONCLUSIONS: The parsimonious AMI READMITS score enables early prospective identification of high-risk AMI patients for targeted readmissions reduction interventions within the first 24 hours of hospitalization. A full-stay AMI readmission model only modestly outperformed the AMI READMITS score in terms of discrimination, but surprisingly did not meaningfully improve reclassification.

AB - BACKGROUND: Readmissions after hospitalization for acute myocardial infarction (AMI) are common. However, the few currently available AMI readmission risk prediction models have poor-to-modest predictive ability and are not readily actionable in real time. We sought to develop an actionable and accurate AMI readmission risk prediction model to identify high-risk patients as early as possible during hospitalization. METHODS AND RESULTS: We used electronic health record data from consecutive AMI hospitalizations from 6 hospitals in north Texas from 2009 to 2010 to derive and validate models predicting all-cause nonelective 30-day readmissions, using stepwise backward selection and 5-fold cross-validation. Of 826 patients hospitalized with AMI, 13% had a 30-day readmission. The first-day AMI model (the AMI "READMITS" score) included 7 predictors: renal function, elevated brain natriuretic peptide, age, diabetes mellitus, nonmale sex, intervention with timely percutaneous coronary intervention, and low systolic blood pressure, had an optimism-corrected C-statistic of 0.73 (95% confidence interval, 0.71-0.74) and was well calibrated. The full-stay AMI model, which included 3 additional predictors (use of intravenous diuretics, anemia on discharge, and discharge to postacute care), had an optimism-corrected C-statistic of 0.75 (95% confidence interval, 0.74-0.76) with minimally improved net reclassification and calibration. Both AMI models outperformed corresponding multicondition readmission models. CONCLUSIONS: The parsimonious AMI READMITS score enables early prospective identification of high-risk AMI patients for targeted readmissions reduction interventions within the first 24 hours of hospitalization. A full-stay AMI readmission model only modestly outperformed the AMI READMITS score in terms of discrimination, but surprisingly did not meaningfully improve reclassification.

KW - acute myocardial infarction

KW - health services research

KW - hospital performance

KW - prediction

KW - readmission

UR - http://www.scopus.com/inward/record.url?scp=85061502670&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061502670&partnerID=8YFLogxK

U2 - 10.1161/JAHA.118.008882

DO - 10.1161/JAHA.118.008882

M3 - Article

C2 - 29666065

AN - SCOPUS:85061502670

VL - 7

JO - Journal of the American Heart Association

JF - Journal of the American Heart Association

SN - 2047-9980

IS - 8

ER -