Appl Clin Inform 2019; 10(02): 316-325
DOI: 10.1055/s-0039-1688553
Research Article
Georg Thieme Verlag KG Stuttgart · New York

Development and Prospective Validation of a Machine Learning-Based Risk of Readmission Model in a Large Military Hospital

Carly Eckert
1   KenSci Inc., Seattle, Washington, United States
,
Neris Nieves-Robbins
2   Office of the U.S. Army Surgeon General, Defense Health Headquarters (Health Information Technology/CMIO Office), Falls Church, Virginia, United States
,
Elena Spieker
3   Clinical Informatics Division, Madigan Army Medical Center, Joint Base Lewis-McChord, Tacoma, Washington, United States
,
Tom Louwers
1   KenSci Inc., Seattle, Washington, United States
,
David Hazel
1   KenSci Inc., Seattle, Washington, United States
,
James Marquardt
1   KenSci Inc., Seattle, Washington, United States
,
Keith Solveson
3   Clinical Informatics Division, Madigan Army Medical Center, Joint Base Lewis-McChord, Tacoma, Washington, United States
,
Anam Zahid
1   KenSci Inc., Seattle, Washington, United States
,
Muhammad Ahmad
1   KenSci Inc., Seattle, Washington, United States
,
Richard Barnhill
3   Clinical Informatics Division, Madigan Army Medical Center, Joint Base Lewis-McChord, Tacoma, Washington, United States
,
T. Greg McKelvey
1   KenSci Inc., Seattle, Washington, United States
,
Robert Marshall
3   Clinical Informatics Division, Madigan Army Medical Center, Joint Base Lewis-McChord, Tacoma, Washington, United States
,
Eric Shry
3   Clinical Informatics Division, Madigan Army Medical Center, Joint Base Lewis-McChord, Tacoma, Washington, United States
,
Ankur Teredesai
1   KenSci Inc., Seattle, Washington, United States
› Author Affiliations
Funding This work was funded by a research grant from the Army Medical Department (AMEDD) Advanced Medical Technology Initiative (AAMTI) provided by the Telemedicine and Advanced Technical Research Center (TATRC).
Further Information

Publication History

24 April 2018

22 March 2019

Publication Date:
08 May 2019 (online)

Abstract

Background Thirty-day hospital readmissions are a quality metric for health care systems. Predictive models aim to identify patients likely to readmit to more effectively target preventive strategies. Many risk of readmission models have been developed on retrospective data, but prospective validation of readmission models is rare. To the best of our knowledge, none of these developed models have been evaluated or prospectively validated in a military hospital.

Objectives The objectives of this study are to demonstrate the development and prospective validation of machine learning (ML) risk of readmission models to be utilized by clinical staff at a military medical facility and demonstrate the collaboration between the U.S. Department of Defense's integrated health care system and a private company.

Methods We evaluated multiple ML algorithms to develop a predictive model for 30-day readmissions using data from a retrospective cohort of all-cause inpatient readmissions at Madigan Army Medical Center (MAMC). This predictive model was then validated on prospective MAMC patient data. Precision, recall, accuracy, and the area under the receiver operating characteristic curve (AUC) were used to evaluate model performance. The model was revised, retrained, and rescored on additional retrospective MAMC data after the prospective model's initial performance was evaluated.

Results Within the initial retrospective cohort, which included 32,659 patient encounters, the model achieved an AUC of 0.68. During prospective scoring, 1,574 patients were scored, of whom 152 were readmitted within 30 days of discharge, with an all-cause readmission rate of 9.7%. The AUC of the prospective predictive model was 0.64. The model achieved an AUC of 0.76 after revision and addition of further retrospective data.

Conclusion This work reflects significant collaborative efforts required to operationalize ML models in a complex clinical environment such as that seen in an integrated health care system and the importance of prospective model validation.

Protection of Human and Animal Subjects

This study was performed in compliance with the World Medical Association Declaration of Helsinki on Ethical Principles for Medical Research Involving Human Subjects and was reviewed by MAMC Institutional Review Board.


 
  • References

  • 1 Boccuti C, Casillas G. Aiming for fewer hospital U-turns: the Medicare hospital readmission reduction program. Policy Brief 2015
  • 2 Kripalani S, Theobald CN, Anctil B, Vasilevskis EE. Reducing hospital readmission rates: current strategies and future directions. Annu Rev Med 2014; 65 (01) 471-485
  • 3 Frizzell JD, Liang L, Schulte PJ. , et al. Prediction of 30-day all-cause readmissions in patients hospitalized for heart failure: comparison of machine learning and other statistical approaches. JAMA Cardiol 2017; 2 (02) 204-209
  • 4 Allaudeen N, Schnipper JL, Orav EJ, Wachter RM, Vidyarthi AR. Inability of providers to predict unplanned readmissions. J Gen Intern Med 2011; 26 (07) 771-776
  • 5 Billings J, Blunt I, Steventon A, Georghiou T, Lewis G, Bardsley M. Development of a predictive model to identify inpatients at risk of re-admission within 30 days of discharge (PARR-30). BMJ Open 2012; 2 (04) e001667
  • 6 Donzé JD, Williams MV, Robinson EJ. , et al. International validity of the hospital score to predict 30-day potentially avoidable hospital readmissions. JAMA Intern Med 2016; 176 (04) 496-502
  • 7 Kansagara D, Englander H, Salanitro A. , et al. Risk prediction models for hospital readmission: a systematic review. JAMA 2011; 306 (15) 1688-1698
  • 8 Zhou H, Della PR, Roberts P, Goh L, Dhaliwal SS. Utility of models to predict 28-day or 30-day unplanned hospital readmissions: an updated systematic review. BMJ Open 2016; 6 (06) e011060
  • 9 Amarasingham R, Moore BJ, Tabak YP. , et al. An automated model to identify heart failure patients at risk for 30-day readmission or death using electronic medical record data. Med Care 2010; 48 (11) 981-988
  • 10 Goldstein BA, Navar AM, Pencina MJ, Ioannidis JP. Opportunities and challenges in developing risk prediction models with electronic health records data: a systematic review. J Am Med Inform Assoc 2017; 24 (01) 198-208
  • 11 Krumholz HM. Big data and new knowledge in medicine: the thinking, training, and tools needed for a learning health system. Health Aff (Millwood) 2014; 33 (07) 1163-1170
  • 12 Freund Y, Schapire RE. A decision-theoretic generalization of on-line learning and an application to boosting. In: European Conference on Computational Learning Theory. 1995: 23-37
  • 13 Futoma J, Morris J, Lucas J. A comparison of models for predicting early hospital readmissions. J Biomed Inform 2015; 56: 229-238
  • 14 Shameer K, Johnson KW, Yahi A. , et al. Predictive modeling of hospital readmission rates using electronic medical record-wide machine learning: a case-study using Mount Sinai Heart Failure Cohort. In: Pacific Symposium on Biocomputing. 2017: 276-287
  • 15 Yang C, Delcher C, Shenkman E, Ranka S. Predicting 30-day all-cause readmissions from hospital inpatient discharge data. In: e-Health Networking, Applications and Services (Healthcom), 2016 IEEE 18th International Conference on IEEE. 2016: 1-6
  • 16 Basu Roy S, Teredesai A, Zolfaghar K. , et al. Dynamic hierarchical classification for patient risk-of-readmission. In: Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2015: 1691-1700
  • 17 Caruana R, Lou Y, Gehrke J, Koch P, Sturm M, Elhadad N. Intelligible models for healthcare: predicting pneumonia risk and hospital 30-day readmission. In: Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2015: 1721-1730
  • 18 Hao S, Wang Y, Jin B. , et al. Development, validation and deployment of a real time 30-day hospital readmission risk assessment tool in the Maine Healthcare Information Exchange. PLoS One 2015; 10 (10) e0140271
  • 19 Final Report to the Secretary of Defense: Military Health System Review; 2014
  • 20 Barrett M, Wier L, Jiang J, Steiner C. . All-Case Readmissions by Payer and Age, 2009–2013. Healthcare Cost and Utilization Project. Statistical Brief #199. Agency for Healthcare Research and Quality; 2015
  • 21 Evaluation of the TRICARE Program. Fiscal Year. 2017 Report to Congress. Available at: https://health.mil/Military-Health-Topics/Access-Cost-Quality-and-Safety/Health-Care-Program-Evaluation/Annual-Evaluation-of-the-TRICARE-Program . Accessed March 30, 2018
  • 22 World Health Organization. International Statistical Classification of Diseases and Related Health Problems. Geneva, Switzerland: World Health Organization; 2004
  • 23 Wang W. Some fundamental issues in ensemble methods. In: Proceedings of IEEE World Congress on Computational Intelligence. 2008: 2244-2251
  • 24 Zolfaghar K, Meadem N, Teredesai A, Roy SB, Chin SC, Muckian B. Big data solutions for predicting risk-of-readmission for congestive heart failure patients. In: Big Data, 2013 IEEE International Conference. 2013: 64-71
  • 25 Hon CP, Pereira M, Sushmita S, Teredesai A, De Cock M. Risk stratification for hospital readmission of heart failure patients: A machine learning approach. In: Proceedings of the 7th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics. 2016: 491-492
  • 26 Liu R, Zolfaghar K, Chin SC, Roy SB, Teredesai A. A framework to recommend interventions for 30-day heart failure readmission risk. In: Data Mining (ICDM), 2014 IEEE International Conference. 2014: 911-916
  • 27 Rao VR, Zolfaghar K, Hazel DK, Mandava V, Roy SB, Teredesai A. . Readmissions Score as a Service (RaaS)
  • 28 García-Laencina PJ, Sancho-Gómez JL, Figueiras-Vidal AR, Verleysen M. K nearest neighbours with mutual information for simultaneous classification and missing data imputation. Neurocomputing 2009; 72 (7–9): 1483-1493
  • 29 Abdi H, Williams LJ. Principal component analysis. Wiley Interdiscip Rev Comput Stat 2010; 2 (04) 433-459
  • 30 Verbiest N, Vermeulen K, Teredesai A. Evaluation of Classification Methods. CRC Press; 2014
  • 31 Hastie TJ, Tibshirani RJ, Friedman JH. The Elements of Statistical Learning: Data Mining Inference and Prediction. 2nd ed. Springer; 2009
  • 32 Breiman L. Random forests. Mach Learn 2001; 45 (01) 5-32
  • 33 Cox DR. The regression analysis of binary sequences. J R Stat Soc [Ser A] 1958; 215-242
  • 34 Friedman JH. Stochastic gradient boosting. Comput Stat Data Anal 2002; 38 (04) 367-378
  • 35 Fluss R, Faraggi D, Reiser B. Estimation of the Youden index and its associated cutoff point. Biom J 2005; 47 (04) 458-472
  • 36 Kohavi R. A study of cross-validation and bootstrap for accuracy estimation and model selection. IJCAI (U S) 1995; 14 (02) 1137-1145
  • 37 Hawkins DM, Basak SC, Mills D. Assessing model fit by cross-validation. J Chem Inf Comput Sci 2003; 43 (02) 579-586
  • 38 van Walraven C, Dhalla IA, Bell C. , et al. Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community. CMAJ 2010; 182 (06) 551-557
  • 39 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40 (05) 373-383
  • 40 Escobar GJ, Turk BJ, Ragins A. , et al. Piloting electronic medical record-based early detection of inpatient deterioration in community hospitals. J Hosp Med 2016; 11 (Suppl. 01) S18-S24
  • 41 Choudhry SA, Li J, Davis D, Erdmann C, Sikka R, Sutariya B. A public-private partnership develops and externally validates a 30-day hospital readmission risk prediction model. Online J Public Health Inform 2013; 5 (02) 219
  • 42 Blumenthal D. Launching HITECH. N Engl J Med 2010; 362 (05) 382-385
  • 43 Pan F, Converse T, Ahn D, Salvetti F, Donato G. Feature selection for ranking using boosted trees. In: Proceedings of the 18th ACM Conference on Information and Knowledge Management. 2009: 2025-2028
  • 44 Amarasingham R, Patel PC, Toto K. , et al. Allocating scarce resources in real-time to reduce heart failure readmissions: a prospective, controlled study. BMJ Qual Saf 2013; 22 (12) 998-1005
  • 45 Moons KGM, Altman DG, Reitsma JB. , et al. Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration. Ann Intern Med 2015; 162 (01) W1-73
  • 46 Siontis GC, Tzoulaki I, Castaldi PJ, Ioannidis JP. External validation of new risk prediction models is infrequent and reveals worse prognostic discrimination. J Clin Epidemiol 2015; 68 (01) 25-34
  • 47 Hsu KW, Pathak N, Srivastava J, Tschida G, Bjorklund E. Data mining-based tax audit selection: a case study of a pilot project at the Minnesota Department of Revenue. In: Real World Data Mining Applications; 2015: 221-245
  • 48 Borbora Z, Srivastava J, Hsu KW, Williams D. Churn prediction in MMORPGs using player motivation theories and an ensemble approach. In: 2011 IEEE Third International Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third International Conference on Social Computing. 2011: 157-164
  • 49 Altman DG, Royston P. What do we mean by validating a prognostic model?. Stat Med 2000; 19 (04) 453-473
  • 50 Widmer G, Kubat M. Learning in the presence of concept drift and hidden contexts. Mach Learn 1996; 23 (01) 69-101
  • 51 Davis SE, Lasko TA, Chen G, Siew ED, Matheny ME. Calibration drift in regression and machine learning models for acute kidney injury. J Am Med Inform Assoc 2017; 24 (06) 1052-1061
  • 52 Krumholz HM, Chaudhry SI, Spertus JA, Mattera JA, Hodshon B, Herrin J. Do non-clinical factors improve prediction of readmission risk?. JACC Heart Fail 2016; 4 (01) 12-20
  • 53 Calvillo-King L, Arnold D, Eubank KJ. , et al. Impact of social factors on risk of readmission or mortality in pneumonia and heart failure: systematic review. J Gen Intern Med 2013; 28 (02) 269-282
  • 54 Taylor AJ, Meyer GS, Morse RW, Pearson CE. Can characteristics of a health care system mitigate ethnic bias in access to cardiovascular procedures? Experience from the Military Health Services System. J Am Coll Cardiol 1997; 30 (04) 901-907
  • 55 Clark JY, Thompson IM. Military rank as a measure of socioeconomic status and survival from prostate cancer. South Med J 1994; 87 (11) 1141-1144
  • 56 Chaudhary MA, Sharma M, Scully RE. , et al. Universal insurance and an equal access healthcare system eliminate disparities for Black patients after traumatic injury. Surgery 2018; 163 (04) 651-656
  • 57 Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates DW. Adverse drug events occurring following hospital discharge. J Gen Intern Med 2005; 20 (04) 317-323
  • 58 Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 2009; 360 (14) 1418-1428
  • 59 Lipton ZC. . The Mythos of Model Interpretability. arXiv; 2016