Getting a Better Return on our Investment in Health Information Systems

Lessons from Home-Based Rehabilitation in Ontario

At present, we have a very limited understanding of the allocation and outcomes of home care in Ontario, particularly the rehabilitation services that could benefit older clients with MSK disorders. Importantly, there is considerable evidence of the feasibility and effectiveness of rehabilitation for older persons in home-based settings (Gitlin, 2006a; Gitlin, 2006b). Despite knowledge that MSK disorders such as hip fracture are very costly to the system and can result in a cascade of adverse events, and that home-based rehabilitation can improve functional outcomes, older MSK patients do not always receive the rehabilitation services they need:

• 71.2% of older home care clients assessed as having rehabilitation potential did not receive any type of rehabilitation therapy (Hirdes et al., 2004).
• In a recent study of resource utilization in eight Ontario-based home care programs, only 26% of home care clients with hip fracture received rehabilitation (Poss et al., 2005).
• Many fracture patients are not evaluated and do not receive osteoporosis treatment (Jaglal, 2003).
• 41% of home care clients with osteoporosis did not receive pharmacotherapy (Vik et al., 2007).

High quality, accessible information is essential for clinical, administrative and health policy decision-making, and to monitor and improve system-wide quality, efficiency and outcomes. Health information systems are fundamental to the optimal performance and effectiveness of health systems (Bickenbach, 2003; Hollander & Prince, 2002; Leggatt & Leatt, 1997) and to ensuring health system accountability (Ontario MOHLTC, 2005).This is especially true as we move toward more integrated health systems (Beland et al., 2006a; 2006b; Shortell et al., 1993) with corresponding needs for greater communication of client-centred information to ensure smooth transitions between services, when quality of care and quality of life are often most at risk (Coleman, 2003; Naylor, 2003).

Recognition of the benefits of standardized information systems has led to these systems being mandated or strongly encouraged in many sectors of the health system. This is requiring major investments in the development and testing of assessment systems and reporting mechanisms, in implementation and training, and in the development and acquisition of supporting software. Despite these investments, and the valuable informational resources that have been developed, the potential of these systems to support clinical, consumer, administrative and policy decisions is far from being realized (Marshall et al., 2000; Teare & Weiler, 2003; Zelmer, 2004; Egan et al., submitted).

Inadequate provision of rehabilitation is in part a reflection of resource constraints, but also reflects shortcomings in the way the health system manages and uses information on home care clients. Better use and management of health information systems would close many of these information gaps and improve the provision of rehabilitation services to older MSK home care clients, thereby increasing their quality of care and quality of life. Improved use and management of information would result in better identification of rehabilitation clients and their needs, and more effective communication of rehabilitation goals and treatment plans as clients are discharged from one setting to another (Stolee et al., 2006).

The appropriate management and use of health information is particularly critical in MSK rehabilitation. Better use of information systems could lead to identification of factors that would predict successful rehabilitation and better target limited resources. More appropriate targeting of rehabilitation therapy could be achieved through more informed care planning, but rehabilitation decisions are particularly challenging. For acute care patients, diagnoses are often clearly defined. By contrast, rehabilitation patients have considerable variability even within specific diagnostic categories. Assessment of rehabilitation potential for older patients is not always straightforward and is often complicated by medical complexity and multiple co-morbidities (Knoefel et al., 2003; Wells et al., 2003a; 2003b), and requires management by multiple health professionals across multiple health settings (Borrie et al., 2005; Coleman, 2003).

Home care clients requiring rehabilitation are at a critical turning point in terms of their future functioning and quality of life, and their potential to live independently. Use of information systems to ensure appropriate and equitable access to rehabilitation would achieve major benefits for the health, quality of life, and independence of older persons with MSK disorders. There will also be system benefits through decreased costs, more appropriate resource use, and avoided institutional placements.

The interRAI/Minimum Data Set (MDS) instruments are a comprehensive assessment and problem identification system developed by an international consortium of researchers (interRAI). The interRAI Home Care assessment instrument (RAI-HC, or MDS-HC, Morris, et al., 1997) has been developed for home care settings. Since 2002 in Ontario, the RAI-HC assessment instrument has been mandated for use with all Community Care Access Centre (CCAC)ⁱ clients who are anticipated to be long stay (> 60 days). Repeat assessments of the RAI-HC are completed at intervals of approximately 180 days. Assessment items include: personal items, referral information, cognition, communication and hearing, vision, mood and behaviour, physical functioning, continence, disease diagnoses, preventive health measures, nutrition status, oral health, skin condition, environmental assessment, and formal and informal service use.

Our program of research – which we call “InfoRehab”ⁱⁱ – is aimed at understanding how we can make better use of the rich resource of health information, provided through assessment systems such as the RAI-HC, in the rehabilitation of older persons with MSK disorders. Our work to date has focused on home-based rehabilitation through the home care programs of Ontario’s CCACs. This research has had two major directions:

• We aim to understand the barriers and facilitators of effective use of health information by home care administrators, case managers and service providers, and to find ways to improve the use of this information in the rehabilitation of home care clients; and
• We aim to explore the potential of advanced statistical techniques to answer pressing questions relevant to MSK health and quality of life of older persons – with the large amount of data currently available in databases such as those provided by the RAI-HC, we believe there is potential for use of new data mining and machine learning techniques to answer questions such as: Which clients have the most potential to benefit from rehabilitation? and: What are the most important predictors of functional decline or improvement?

So far, we have found that home care case managers have needs for health information about their clients that are currently not being met (Egan et al., submitted). In addition, they often have little or no access to the RAI-HC data despite the electronic nature of the assessment, and thus rarely have these data to inform their own care planning. Nor do they access aggregated RAI-HC data that could be used to compare across case managers or across sites. There are as yet no mechanisms to transmit RAI-HC assessments or summary information electronically from the case manager assessment to rehabilitation provider agencies or to permit access to the health information on the internet. This leads to potential missed information, duplication in assessment processes, and increased burden for clients. Our current research is aimed at uncovering a comprehensive set of barriers and facilitators to the use of data in home care settings, and creating recommendations around solutions and facilitation strategies to improve the use of health information.

On the statistical analysis side, we have applied machine learning tools (the K-nearest neighbors algorithm, and the support vector machine) to predict the rehabilitation potential of older home care clients. Our results showed that these data-driven machine learning algorithms produced significantly better predictions than existing clinical assessment protocols. This pilot work also demonstrated how machine learning techniques can “set the bar” for clinical predictions (Zhu, et al., 2007a; 2007b), and how machine learning can be used to refine clinical protocols to achieve comparable performance (Zhu, et al., 2007b). The machine learning algorithms achieved a level of prediction accuracy that provides a target for clinical protocols to achieve – in this way serving as a potential “quality control” mechanism - and also provided insights into which predictors were most important and how these variables should be coded. We believe our work to date supports continued investigation of the potential for advanced machine learning algorithms and other statistical techniques to support care planning for home care rehabilitation.

Health information systems offer tremendous potential benefits for health care planning and decision-making, but continued research will be essential if we wish to take full advantage of these investments.

references

• Beland, F., Bergman, H., Lebel, P., Clarfield, A. M., Tousignant, P., Contandriopoulos, A. P., et al. (2006a). A system of integrated care for older persons with disabilities in Canada: Results from a randomized controlled trial. The Journals of Gerontology.Series A, Biological Sciences and Medical sciences, 61(4), 367-373.
• Beland, F., Bergman, H., Lebel, P., Dallaire, L., Fletcher, J., Contandriopoulos, A. P., et al. (2006b). Integrated services for frail elders (SIPA): A trial of a model for Canada. Canadian Journal on Aging, 25(1), 5-42.
• Bickenbach, J.E. (2003). Functional status and health information in Canada: Proposals and prospects. Health Care Finance Review, 24, 89-102.
• Borrie MJ, Stolee P, Knoefel FD, Wells JL, & Seabrook JA. (2005). Current best practices in geriatric rehabilitation in Canada. Geriatrics Today: Canadian Journal of Geriatric Medicine and Geriatric Psychiatry, 8, 148-153.
• Canadian Institutes for Health Research Institute of Musculoskeletal Health and Arthritis (IMHA). Facts and Figures. Available online: http://www.cihr-irsc.gc.ca/e/27028.html. Retrieved March 2005.
• Coleman, E.A. (2003). Falling through the cracks: Challenges and opportunities for improving transitional care for persons with continuous complex care needs. Journal of the American Geriatrics Society, 51, 549-555.
• Commission on the Future of Health Care in Canada (Romanow commission). (2002). Building on Values: The Future of Health Care in Canada. Saskatoon: Commission on the Future of Health Care in Canada.
• Egan, M., Wells, J., Byrne, K., Jaglal, S., Stolee, P., Chesworth, B. & Hillier, L. (submitted). Maximizing usefulness of universal assessment data in home care case management: A qualitative study..
• Gitlin LN, Hauck WW, Winter L, Dennis MP, & Schulz R. (2006a). Effect of an in-home occupational and physical therapy intervention on reducing mortality in functionally vulnerable older people: Preliminary findings. Journal of the American Geriatrics Society, 54, 950-955.
• Gitlin LN, Winter L, Dennis MP, Corcoran M, Schinfeld S, & Hauck WW (2006b). A randomized trial of a multicomponent home intervention to reduce functional difficulties in older adults. Journal of the American Geriatrics Society, 54(5), 809-16.
• Hirdes JP, Fries BE, Morris JN, Ikegami N, Zimmerman D, Dalby DM, Aliaga P, Hammer S, & Jones R. (2004). Home care quality indicators (HCQIs) based on the MDS-HC. Gerontologist 44, 665-679.
• Jaglal, S.B. (2003). Development of a Model for Integrated Post-Fracture Care in Ontario. Ontario Women’s Health Council, 1-92.
• Knoefel, F., Helliwell, B., Seabrook, J.A., Borrie, M.J., Stolee, P., & Wells, J.L. (2003). A comparison of functional independence and medical complexity in geriatric and physical medicine rehabilitation inpatients. Geriatrics Today: Journal of the Canadian Geriatrics Society, 6, 90-94.
• Leggatt, S.G., & Leatt, P. (1997). A framework for assessing the performance of integrated health delivery systems. Healthcare Management Forum, 10(1), 11-26.
• Marshall, M.N., Shekelle, P.G., Leatherman, S., & Brook, R.H. (2000). The public release of performance data. What do we expect to gain? A review of the evidence. Journal of the American Medical Association, 283, 1866-1874.
• Morris, J.N., Fries, B.E., Steel, K., Ikegami, N., Bernabei, R., Carpenter, G.I., Gilgen, R., Hirdes, J.P., & Topinkova, E. (1997). Comprehensive clinical assessment in community setting: Applicability of the MDS-HC. Journal of the American Geriatrics Society, 45, 1017-1024.
• Naylor, M. (2003). Transitional care of older adults. In J.J. Fitzpatrick, P.G. Archbold, & B.J. Stewart (Eds), Annual Review of Nursing Research, Vol. 20. New York: Springer.
• Ontario Ministry of Health and Long-Term Care (2005). Transforming healthcare. URL: http://www.health.gov.on.ca/transformation/ Accessed January 2, 2005.
• Poss, J., Hirdes., J.P., Reidel, K., Chase, M., Fries, B.E., & McKillop, I. (2005). An Evaluation of the Use of the RUG-III/HC Case Mix Classification System to Estimate Resource Intensity of Long-Stay CCAC Clients. Technical Report Submitted to Central South and Central West CCACs, Ontario.
• Public Health Agency of Canada. (1998). Economic Burden of Illness in Canada as cited In: Institute of Musculoskeletal Health and Arthritis (2005). Fact Sheet. URL: http://www.cihr-irsc.gc.ca/e/27442.html
• Randall, G.E. & Williams, A.P. (2006). Exploring limits to market-based reform: Managed competition and rehabilitation home care services in Ontario. Social Science and Medicine, 62, 1594-1604.
• Shortell, S.N., Gillies, R.R., Anderson, D.A., Mitchell, J.B., & Morgan, K.L. (1993). Creating organized delivery systems: The barriers and facilitators. Hospital and Health Services Administration, 38(4), 447-466.
• Stolee P, Hillier LM, Webster F, O’Callaghan C (2006). Stroke care in long-term care facilities in Southwestern Ontario. Topics in Stroke Rehabilitation. 13, 97-108.
• Teare, G.F. & Weiler, L. (2003). Hospital Report 2003: Complex Continuing Care System Integration Change Technical Report. URL: http://www.oha.com/Client/OHA/OHA_LP4W_LND_WebStation.nsf/resources/CCCSICTechReport/$file/2003CCCSICTechReport.pdf
• Vik SA, Jantzi M, Poss J, Hirdes J, Hanley DA, Hogan DB, & Maxwell CJ. (2007). Factors associated with pharmacologic treatment of osteoporosis in an older home care population. Journal of Gerontology: Medical Sciences, 62A, 872-878.
• Wells JL, Seabrook JA, Stolee P, Borrie MJ, & Knoefel F. (2003a). State of the art in geriatric rehabilitation, Part I: Review of frailty and comprehensive geriatric assessment. Archives of Physical Medicine and Rehabilitation 84, 890-7.
• Wells JL, Seabrook JA, Stolee P, Borrie MJ, & Knoefel F. (2003b). State of the art in geriatric rehabilitation, Part II: Clinical challenges. Archives of Physical Medicine and Rehabilitation, 84, 898-903.
• Zelmer, J. (2004). From access to quality to outcomes - Using health information to improve health care. Canadian Institute for Health Information Data Users’ Conference, September, Ottawa, ON.
• Zhu M, Chen W, Hirdes JP, & Stolee P (2007a). The K-nearest neighbors algorithm predicted rehabilitation potential better than current clinical assessment protocol. Journal of Clinical Epidemiology, 60, 1015-1021.
• Zhu M, Zhang Z, Hirdes JP, & Stolee P (2007b). Using machine learning algorithms to guide rehabilitation planning for home care clients. BMC Medical Informatics and Decision-Making, 7(1), 41.