Predicting the risk of acute care readmissions among rehabilitation inpatients: A machine learning approach

被引:18
作者
Xue, Yajiong [1 ]
Liang, Huigang [1 ,2 ,3 ]
Norbury, John [4 ]
Gillis, Rita [5 ]
Killingworth, Brenda [1 ]
机构
[1] East Carolina Univ, Coll Business, Dept Management Informat Syst, Greenville, NC USA
[2] East Carolina Univ, Coll Business, Ctr Healthcare Management Syst, Greenville, NC USA
[3] East Carolina Univ, Big Data & Analyt Res Cluster, Greenville, NC USA
[4] East Carolina Univ, Brody Sch Med, Dept Phys Med & Rehabil, Greenville, NC USA
[5] Vidant Hlth, Greenville, NC USA
关键词
Readmission; Rehabilitation; Machine learning; Functional independence measure; STATUS OUTPERFORMS COMORBIDITIES; FUNCTIONAL STATUS; HOSPITAL READMISSION; IMPAIRMENT; DISCHARGE;
D O I
10.1016/j.jbi.2018.09.009
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Introduction: Readmission from inpatient rehabilitation facilities to acute care hospitals is a serious problem. This study aims to develop a predictive model based on machine learning algorithms to identify patients at high risk of readmission. Methods: A retrospective dataset (2001-2017) including 16,902 patients admitted into a large inpatient rehabilitation facility in North Carolina was collected in 2017. Three types of machine learning models with different predictors were compared in 2018. The model with the highest c-statistic was selected as the best model and further tested by using five sets of training and validation data with different split time. The optimum threshold for classification was identified. Results: The logistic regression model with only functional independence measures has the highest validation c-statistic at 0.852. Using this model to predict the recent 5 years acute care readmissions yielded high discriminative ability (c-statistics: 0.841-0.869). Larger training data yielded better performance on the test data. The default cutoff (0.5) resulted in high specificity (> 0.997) but low sensitivity (< 0.07). The optimum threshold helped to achieve a balance between sensitivity (0.754-0.867) and specificity (0.747-0.780). Conclusions: This study demonstrates that functional independence measures can be analyzed by using machine learning algorithms to predict acute care readmissions, thus improving the effectiveness of preventive medicine.
引用
收藏
页码:143 / 148
页数:6
相关论文
共 25 条
[1]  
Alekseyev K., 2017, Patient Safety Quality Improvement Journal, V5, P488
[2]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[3]   Predictors of Discharge to Acute Care after Inpatient Rehabilitation in Severely Affected Stroke Patients [J].
Chung, Duc M. ;
Niewczyk, Paulette ;
DiVita, Margaret ;
Markello, Sam ;
Granger, Carl .
AMERICAN JOURNAL OF PHYSICAL MEDICINE & REHABILITATION, 2012, 91 (05) :387-392
[4]   Diagnoses and Timing of 30-Day Readmissions After Hospitalization for Heart Failure, Acute Myocardial Infarction, or Pneumonia [J].
Dharmarajan, Kumar ;
Hsieh, Angela F. ;
Lin, Zhenqiu ;
Bueno, Hector ;
Ross, Joseph S. ;
Horwitz, Leora I. ;
Barreto-Filho, Jose Augusto ;
Kim, Nancy ;
Bernheim, Susannah M. ;
Suter, Lisa G. ;
Drye, Elizabeth E. ;
Krumholz, Harlan M. .
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, 2013, 309 (04) :355-363
[5]   A VALIDATION OF THE FUNCTIONAL INDEPENDENCE MEASUREMENT AND ITS PERFORMANCE AMONG REHABILITATION INPATIENTS [J].
DODDS, TA ;
MARTIN, DP ;
STOLOV, WC ;
DEYO, RA .
ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION, 1993, 74 (05) :531-536
[6]  
Fingar K., 2015, TRENDS HOSP READMISS
[7]   Return to the Primary Acute Care Service Among Patients With Multiple Myeloma on an Acute Inpatient Rehabilitation Unit [J].
Fu, Jack B. ;
Lee, Jay ;
Shin, Ben C. ;
Silver, Julie K. ;
Smith, Dennis W. ;
Shah, Jatin J. ;
Bruera, Eduardo .
PM&R, 2017, 9 (06) :571-578
[8]   Functional Impairment and Hospital Readmission in Medicare Seniors [J].
Greysen, S. Ryan ;
Cenzer, Irena Stijacic ;
Auerbach, Andrew D. ;
Covinsky, Kenneth E. .
JAMA INTERNAL MEDICINE, 2015, 175 (04) :559-565
[9]   Interventions to Reduce 30-Day Rehospitalization: A Systematic Review [J].
Hansen, Luke O. ;
Young, Robert S. ;
Hinami, Keiki ;
Leung, Alicia ;
Williams, Mark V. .
ANNALS OF INTERNAL MEDICINE, 2011, 155 (08) :520-U94
[10]   Functional Status Impairment Is Associated With Unplanned Readmissions [J].
Hoyer, Erik H. ;
Needham, Dale M. ;
Miller, Jason ;
Deutschendorf, Amy ;
Friedman, Michael ;
Brotman, Daniel J. .
ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION, 2013, 94 (10) :1951-1958