Opinion versus practice regarding the use of rehabilitation services in home care: an investigation using machine learning algorithms

BackgroundResources for home care rehabilitation are limited, and many home care clients who could benefit do not receive rehabilitation therapy. The interRAI Contact Assessment (CA) is a new screening instrument comprised of a subset of interRAI Home Care (HC) items, designed to be used as a preliminary assessment to identify which potential home care clients should be referred for a full assessment, or for services such as rehabilitation. We investigated which client characteristics are most relevant in predicting rehabilitation use in the full interRAI HC assessment.MethodsWe applied two algorithms from machine learning and data mining ― the LASSO and the random forest ― to frequency matched interRAI HC and service utilization data for home care clients in Ontario, Canada.ResultsAnalyses confirmed the importance of functional decline and mobility variables in targeting rehabilitation services, but suggested that other items in use as potential predictors may be less relevant. Six of the most highly ranked items related to ambulation. Diagnosis of cancer was highly associated with decreased rehabilitation use; however, cognitive status was not.ConclusionsInconsistencies between variables considered important for classifying clients who need rehabilitation and those identified in this study based on use may indicate a discrepancy in the client characteristics considered relevant in theory versus actual practice.

[1]  Mu Zhu,et al.  Kernels and Ensembles : Perspectives on Statistical Learning , 2008 .

[2]  S. Rigatti.  Random Forest. , 2017, Journal of insurance medicine.

[3]  A. Giusti,et al.  An analysis of the feasibility of home rehabilitation among elderly people with proximal femoral fractures. , 2006, Archives of physical medicine and rehabilitation.

[4]  J. Hirdes,et al.  The Method for Assigning Priority Levels (MAPLe): A new decision-support system for allocating home care resources , 2008, BMC medicine.

[5]  J. Hirdes,et al.  Health Outcomes among the Frail Elderly in Communities and Institutions: Use of the Minimum Data Set (MDS) to Create Effective Linkages between Research and Policy , 1997 .

[6]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[7]  Mu Zhu,et al.  Using Machine Learning to Plan Rehabilitation for Home Care Clients: Beyond "Black-Box" Predictions , 2014, Machine Learning in Healthcare Informatics.

[8]  Roy Romanow,et al.  Building on Values: Report of the Commission on the Future of Health Care in Canada [Reports] , 2002 .

[9]  J. Hirdes,et al.  Sharing clinical information across care settings: the birth of an integrated assessment system , 2009, BMC health services research.

[10]  Dursun Delen,et al.  A machine learning-based approach to prognostic analysis of thoracic transplantations , 2010, Artif. Intell. Medicine.

[11]  Frank Knoefel,et al.  State of the art in geriatric rehabilitation. Part I: review of frailty and comprehensive geriatric assessment. , 2003, Archives of physical medicine and rehabilitation.

[12]  Peter C Austin,et al.  Automated variable selection methods for logistic regression produced unstable models for predicting acute myocardial infarction mortality. , 2004, Journal of clinical epidemiology.

[13]  Leo Breiman,et al.  Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author) , 2001 .

[14]  Mu Zhu,et al.  Using machine learning algorithms to guide rehabilitation planning for home care clients , 2007, BMC Medical Informatics Decis. Mak..

[15]  Kerry Kuluski,et al.  Aging at home: integrating community-based care for older persons. , 2009, HealthcarePapers.

[16]  N. Ikegami,et al.  Comprehensive Clinical Assessment in Community Setting: Applicability of the MDS‐HC , 1997, Journal of the American Geriatrics Society.

[17]  Naoki Ikegami,et al.  Home care quality indicators (HCQIs) based on the MDS-HC. , 2004, The Gerontologist.

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

[19]  D. Forbes,et al.  Information sharing with rural family caregivers during care transitions of hip fracture patients , 2014, International journal of integrated care.

[20]  D. Madigan,et al.  [Least Angle Regression]: Discussion , 2004 .

[21]  Frank Knoefel,et al.  State of the art in geriatric rehabilitation. Part II: clinical challenges. , 2003, Archives of physical medicine and rehabilitation.

[22]  A. Heinemann,et al.  Functional outcome following rehabilitation of the cancer patient. , 1996, Archives of physical medicine and rehabilitation.

[23]  Mu Zhu,et al.  Stochastic Stepwise Ensembles for Variable Selection , 2010, 1003.5930.

[24]  Gary Naglie,et al.  Interdisciplinary inpatient care for elderly people with hip fracture: a randomized controlled trial. , 2002, CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne.

[25]  Joshua J. Armstrong,et al.  Rehabilitation therapies for older clients of the Ontario home care system: regional variation and client-level predictors of service provision , 2015, Disability and rehabilitation.

[26]  H. Allore,et al.  A program to prevent functional decline in physically frail, elderly persons who live at home. , 2002, The New England journal of medicine.

[27]  William R. Hazzard advisor.  Hazzard's geriatric medicine and gerontology / , 2017 .

[28]  M. Creditor.  Hazards of Hospitalization of the Elderly , 1993, Annals of Internal Medicine.

[29]  J. Hirdes,et al.  interRAI Contact Assessment (CA) Form and User’s Manual: A Screening Level Assessment for Emergency Department and Intake from Community/Hospital. Version 9.2 , 2010 .

[30]  J. Hirdes.  Addressing the health needs of frail elderly people: Ontario's experience with an integrated health information system. , 2006, Age and ageing.

[31]  N. Nusbaum.  Rehabilitation and the older cancer patient. , 2004, The American journal of the medical sciences.

[32]  T. Scialla,et al.  Rehabilitation for elderly patients with cancer asthenia: making a transition to palliative care , 2000, Palliative medicine.

[33]  Alessandra Marengoni,et al.  Predictors of successful rehabilitation in geriatric patients: subgroup analysis of patients with cognitive impairment , 2007, Aging clinical and experimental research.

[34]  R. Kuisma,et al.  A randomized, controlled comparison of home versus institutional rehabilitation of patients with hip fracture , 2002, Clinical rehabilitation.

[35]  Wendy Armstrong,et al.  Building on Values: The Future of Health Care in Canada , 2005 .

[36]  L. Gitlin,et al.  A Randomized Trial of a Multicomponent Home Intervention to Reduce Functional Difficulties in Older Adults , 2006, Journal of the American Geriatrics Society.

[37]  B. Turlach.  Discussion of "Least Angle Regression" by Efron, Hastie, Johnstone and Tibshirani , 2004 .

[38]  B E Fries,et al.  Integrated Health Information Systems Based on the RAI/MDS Series of Instruments , 1999, Healthcare management forum.

[39]  Matthias Egger,et al.  Inpatient rehabilitation specifically designed for geriatric patients: systematic review and meta-analysis of randomised controlled trials , 2010, BMJ : British Medical Journal.

[40]  N. Macdonald,et al.  Rehabilitation for patients with advanced cancer , 2014, Canadian Medical Association Journal.

[41]  R. Tibshirani,et al.  Regression shrinkage and selection via the lasso: a retrospective , 2011 .

[42]  Maria Crotty,et al.  Patient and caregiver outcomes 12 months after home-based therapy for hip fracture: a randomized controlled trial. , 2003, Archives of physical medicine and rehabilitation.

[43]  Sue Jones,et al.  Rehabilitation at home following early discharge after hip surgery , 1998 .

[44]  Nicolai Meinshausen,et al.  Relaxed Lasso , 2007, Comput. Stat. Data Anal..

[45]  Ashutosh Kumar Singh,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2010 .

[46]  Richard Schulz,et al.  Effect of an In‐Home Occupational and Physical Therapy Intervention on Reducing Mortality in Functionally Vulnerable Older People: Preliminary Findings , 2006, Journal of the American Geriatrics Society.

[47]  M Colombo,et al.  The impact of cognitive impairment on the rehabilitation process in geriatrics. , 2004, Archives of gerontology and geriatrics. Supplement.

[48]  P. Bühlmann,et al.  The group lasso for logistic regression , 2008 .

[49]  G. Carrière,et al.  Seniors' use of home care. , 2006, Health reports.

[50]  A. Hershkovitz,et al.  Factors affecting short-term rehabilitation outcomes of disabled elderly patients with proximal hip fracture. , 2007, Archives of physical medicine and rehabilitation.

[51]  Dursun Delen,et al.  Predicting the graft survival for heart-lung transplantation patients: An integrated data mining methodology , 2009, Int. J. Medical Informatics.

[52]  H. Zou.  The Adaptive Lasso and Its Oracle Properties , 2006 .

[53]  J. Hirdes,et al.  Reliability of the interRAI suite of assessment instruments: a 12-country study of an integrated health information system , 2008, BMC health services research.

[54]  Mu Zhu,et al.  The K-nearest neighbor algorithm predicted rehabilitation potential better than current Clinical Assessment Protocol. , 2007, Journal of clinical epidemiology.

[55]  Sanjay Asthana,et al.  Hazzard's Geriatric Medicine and Gerontology , 2016 .

[56]  P. Stolee,et al.  Inpatient versus home-based rehabilitation for older adults with musculoskeletal disorders: a systematic review , 2012, Clinical rehabilitation.

[57]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.