Predictive Modeling for End-of-Life Pain Outcome Using Electronic Health Records

Electronic health record (EHR) systems are being widely used in the healthcare industry nowadays, mostly for monitoring the progress of the patients. EHR data analysis has become a big data problem as data is growing rapidly. Using a nursing EHR system, we built predictive models for determining what factors influence pain in end-of-life (EOL) patients. Utilizing different modeling techniques, we developed coarse-grained and fine-grained models to predict patient pain outcomes. The coarse-grained models help predict the outcome at the end of each hospitalization, whereas fine-grained models help predict the outcome at the end of each shift, thus providing a trajectory of predicted outcomes over the entire hospitalization. These models can help in determining effective treatments for individuals and groups of patients and support standardization of care where appropriate. Using these models may also lower the cost and increase the quality of end-of-life care. Results from these techniques show significantly accurate predictions.

[1]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[2]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[3]  Haiden A Huskamp,et al.  Health care costs in the last week of life: associations with end-of-life conversations. , 2009, Archives of internal medicine.

[4]  Elizabeth Yakel,et al.  Maintaining a consistent big picture: meaningful use of a Web-based POC EHR system. , 2012, International journal of nursing knowledge.

[5]  Simon Lin,et al.  Data mining issues and opportunities for building nursing knowledge , 2003, J. Biomed. Informatics.

[6]  David W. Aha,et al.  Instance-Based Learning Algorithms , 1991, Machine Learning.

[7]  Michael A. Arbib,et al.  The handbook of brain theory and neural networks , 1995, A Bradford book.

[8]  William A. Knaus,et al.  A controlled trial to improve care for seriously ill hospitalized patients. The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT). The SUPPORT Principal Investigators. , 1995, JAMA.

[9]  N. Macdonald,et al.  Montreal prognostic score: estimating survival of patients with non-small cell lung cancer using clinical biomarkers , 2013, British Journal of Cancer.

[10]  V. Tilden,et al.  Sampling Challenges in End-of-Life Research: Case-Finding for Family Informants , 2002, Nursing research.

[11]  G. Bulechek,et al.  Nursing Interventions Classification , 2000 .

[12]  Kenneth Gronbach.  The Age Curve: How to Profit from the Coming Demographic Storm , 2008 .

[13]  Karl Pearson F.R.S..  X. On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling , 2009 .

[14]  P. Benner.  From novice to expert. , 2004, Nursing standard (Royal College of Nursing (Great Britain) : 1987).

[15]  Thomas J. Smith,et al.  A high-volume specialist palliative care unit and team may reduce in-hospital end-of-life care costs. , 2003, Journal of palliative medicine.

[16]  Enrique Espinosa,et al.  Development and validation of a prognostic nomogram for terminally ill cancer patients. , 2011, Journal of the National Cancer Institute.

[17]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[18]  Andrew J Vickers,et al.  Prediction models in cancer care , 2011, CA: a cancer journal for clinicians.

[19]  Bonnie J. Wakefield,et al.  Iowa Outcomes Project: Nursing Outcomes Classification (NOC) , 2000 .

[20]  V. Tilden,et al.  Sampling challenges in nursing home research. , 2013, Journal of the American Medical Directors Association.

[21]  David D. Lewis,et al.  Naive (Bayes) at Forty: The Independence Assumption in Information Retrieval , 1998, ECML.

[22]  S. Brunak,et al.  Mining electronic health records: towards better research applications and clinical care , 2012, Nature Reviews Genetics.

[23]  K. Pearson.  On the Criterion that a Given System of Deviations from the Probable in the Case of a Correlated System of Variables is Such that it Can be Reasonably Supposed to have Arisen from Random Sampling , 1900 .

[24]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[25]  Judea Pearl,et al.  Bayesian Networks , 1998, Encyclopedia of Social Network Analysis and Mining. 2nd Ed..

[26]  W. N. Street,et al.  A NURSING CARE PLAN RECOMMENDER SYSTEM USING A DATA MINING APPROACH , 2008 .

[27]  Rashid Ansari,et al.  Current State of Pain Care for Hospitalized Patients at End of Life , 2013, The American journal of hospice & palliative care.

[28]  Blaz Zupan,et al.  Predictive data mining in clinical medicine: Current issues and guidelines , 2008, Int. J. Medical Informatics.

[29]  T. Hsia,et al.  A FRAMEWORK FOR DESIGNING NURSING KNOWLEDGE MANAGEMENT SYSTEMS , 2006 .