http://informahealthcare.com/dre ISSN 0963-8288 print/ISSN 1464-5165 online Disabil Rehabil, 2015; 37(7): 625–631 ! 2014 Informa UK Ltd. DOI: 10.3109/09638288.2014.935494
REHABILITATION AND PRACTICE
Rehabilitation therapies for older clients of the Ontario home care system: regional variation and client-level predictors of service provision Joshua J. Armstrong1, Mu Zhu2, John P. Hirdes3, and Paul Stolee3,4 1
Geriatric Medicine Research, Dalhousie University, Halifax, Nova Scotia, Canada, 2Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, Ontario, Canada, 3School of Public Health and Health Systems, University of Waterloo, Waterloo, Ontario, Canada, and 4 Schlegel – University of Waterloo Research Institute for Aging, Waterloo, Ontario, Canada Abstract
Keywords
Purpose: To examine regional variation in service provision and identify the client characteristics associated with occupational therapy (OT) and physiotherapy (PT) services for older adults in the Ontario Home Care System. Methods: Secondary analyses of a provincial database containing comprehensive assessments (RAI-HC) linked with service utilization data from every older long-stay home care client in the system between 2005 and 2010 (n ¼ 299 262). Hierarchical logistic regression models were used to model the dependent variables of OT and PT service use within 90 d of the initial assessment. Results: Regional differences accounted for 9% of the variation in PT service provision and 20% of OT service provision. After controlling for the differences across regions, the most powerful predictors of service provision were identified for both OT and PT. The most highly associated client characteristics related to PT service provision were hip fracture, impairments in activities of daily living/instrumental activities of daily living, cerebrovascular accidents, and cognitive impairment. For OT, hazards in the home environment was the most powerful predictor of future service provision. Conclusions: Where a client lived was an important determinant of service provision in Ontario, raising the possibility of inequities in access to rehabilitation services. Health care planners and policy makers should review current practices and make adjustments to meet the increasing and changing needs for rehabilitation therapies of the aging population.
Geriatrics, multi-level modeling, occupational therapy, physiotherapy History Received 15 November 2013 Revised 10 June 2014 Accepted 12 June 2014 Published online 1 July 2014
ä Implications for Rehabilitation
For older adults in home care, the goal of rehabilitation therapy services is to allow individuals to maintain or improve physical functioning, quality of life and overall independence while living within their community. Previous research has demonstrated that a large proportion of home care clients specifically identified as having rehabilitation potential do not receive it. This article used clinical assessment data to identify the predictors of and barriers to rehabilitation services for older adults in the Ontario Home Care System. Barriers of PT included dementia diagnosis and French as a first language. Barriers to OT included dementia diagnosis. Policies and practices related to service provision for older adults should be reconsidered if we are going to meet the demands of aging populations and increasing rates of functional and cognitive impairments.
Introduction Many studies have found rehabilitation therapies in home-based settings to be effective in improving outcomes for older adults [1–7]. A 2012 systematic literature review found that home-based
Address for correspondence: Paul Stolee, PhD, School of Public Health and Health Systems, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, Canada N2L 1P3. Fax: (519) 888-4362. E-mail: [email protected]
rehabilitation produced outcomes either equal or better than rehabilitation based within hospitals, for older persons with musculoskeletal disorders [8]. Despite this, in the Canadian province of Ontario, research has demonstrated that a large proportion of home care clients specifically identified as having rehabilitation potential do not receive it [9–11]. The Ontario provincial home care system has been geographically organized into 14 regions referred to as Local Health Integration Networks (LHINs). This regional division of the system was designed to allow each region to tailor its services to
626
J. J. Armstrong et al.
their populations. Eligibility for the receipt of home care services is determined by case managers who then arrange for health care providers – nurses, physiotherapists, social workers, registered dieticians, occupational therapists, speech therapists, and personal support workers – to provide a range of care and support services to individuals living within the community. Although the province aims to ensure equal access to publicly funded services for all residents of the province, research studies have found substantial disparities in the provision of home care services and rehabilitation therapies across the geographic regions of Ontario [12–15]. This indicates that where an older client lives could dictate whether or not they have access to rehabilitation therapy services, despite their actual or perceived needs. Due to an aging population and the associated increase in a number of chronic diseases, disabilities, and complex conditions, the demand of rehabilitation services across the continuum of care is expected to rise in the years to come [16]. In a publicly funded home care system where there are individuals with unmet needs and reported regional differences, there is a need for research studies to develop understanding of current practices in order to optimize the system for the growing older population. The first objective of this study is to quantify the variation in the provision of physiotherapy (PT) and occupational therapy (OT) services for older adults across the 14 health regions of the Ontario Home Care System. After accounting for regional variation, a secondary objective is to identify client characteristics that are associated with users and non-users of PT and OT services. These objectives will be met using hierarchical logistic regression modeling of assessment data from a large provincial database.
Methods This study is a population-level, observational study based on the initial inter-RAI Home Care (RAI-HC) assessment and service utilization information from every long-stay home care client in Ontario, Canada, from 2005 to 2010 (n ¼ 299 262). Analyses excluded all home care clients under the age of 65 and all assessments performed on older adults in hospitals for placement purposes. The province of Ontario is organized by geography into 14 health regions known as LHINs, each of which includes a Community Care Access Centre (CCAC) which acts as a single entry point for persons who require home care services. The Office of Research Ethics at the University of Waterloo provided ethics clearance for this study. Data sources The inter-RAI home care assessment instrument, the RAI-HC (or MDS-HC) [17] is used to collect health information on home care clients in the Canadian province of Ontario as well as many other jurisdictions [18,19] . The RAI-HC is mandated in Ontario for assessment of all individuals who are enter the home care system and are expected to be long stay (i.e. individuals requiring service for 60 d or more) [20]. As a comprehensive assessment instrument, the RAI-HC collects information using over 300 items from a wide range of domains. Once this information is collected by CCAC case managers, it is used to inform and guide comprehensive care and service planning in community-based settings. RAI-HC items have been shown to have good inter-rater and test–retest reliability [17,21]. Measures Independent variables Due to the large number of variables collected in the RAI-HC assessment, the Andersen and Newman Framework of Health
Disabil Rehabil, 2015; 37(7): 625–631
Services Utilization [22] was used to organize the predictor variables from the RAI-HC into three categories: Predisposing Factors, Enabling Factors, and Needs Factors. Predisposing factors included variables that are thought to influence an individual’s propensity to use services before the need for those services is present (e.g. sex, age, first language). Enabling factors are client characteristics that facilitate or inhibit the ease at which services can be obtained (e.g. home environment, education, income). Since the RAI-HC collects information on clinical characteristics for use in care planning across the multiple health care disciplines (e.g. nursing, rehabilitation, social work), many domains contained in the instrument are related to need. Because of this, need factors were separated into three separate categories: (1) inter-RAI scales, (2) activities of daily living (ADL) and Instrumental ADL (IADL) impairments, and (3) disease diagnoses. Within the electronic RAI-HC information system, a variety of scales are generated automatically for each individual client. These scales are designed to act as status and outcome measures that can be used to summarize client characteristics in a number of different domains. The following inter-RAI scales are utilized in this study: Changes in Health, End-stage disease and Signs and Symptoms (CHESS) score (a measure of health instability) [23,24]; Activities of Daily Living Self-performance Hierarchy Scale [25,26]; Instrumental Activities of Daily Living Performance Scale [26,27]; Cognitive Performance Scale [26,28]; and a Pain Scale [29]. The RAI-HC contains 18 specific items related to ADLs and IADLs. All of these individual items were included in the analyses. These items capture information about the client’s performance in the various domains in the last 7 d for IADLs, and for the last 3 d for ADLs. The ADL and IADL items were dichotomized (independent versus requiring assistance). The final subcategory within the Need Factors group is Disease Diagnoses. In total, 18 diagnoses were included in the analyses, each of the variables was dichotomized (not present versus present). Dependent variables The dependent variables in this study are the provision of OT and PT services after entering the home care system for the first time. Using unique client identifiers, RAI-HC assessment data were linked to health care service use records from the Ontario Association of Community Care Access Centres (OACCAC). OT and PT service data were dichotomized so that analyses could examine the differences between rehabilitation service users and non-users. The window of time for service utilization that was examined was 90 d post RAI-HC assessment. Statistical analyses Due to the hierarchical nature of the provincial data (home care clients nested within health regions (CCACs)), the modeling approach that was chosen to evaluate rehabilitation service provision was hierarchical logistic regression. This approach allows for the estimation of the degree to which health regions are associated with service provision and the creation of statistical models that control for any regional variation. Consistent with other multi-level modeling studies [30–32], a multistep procedure was utilized: (1) examination of the variability in provision of OT and PT services across geographic regions, (2) estimating the association of client characteristics with PT/OT services while accounting for region effects, and (3) the development of multivariate models that incorporate the most powerful clientlevel predictors from step (2). All hierarchical logistic models
Rehabilitation therapy services for older clients
DOI: 10.3109/09638288.2014.935494
were created using a random intercept approach that accounted for the differences in service provision between CCACs [33]. All client-level predictors were treated as fixed effects. In the first step of these analyses, null models were created to identify the amount of variance due to region. These null models indicate the amount of variance in OT and PT service provision that can be attributed to the client’s location within a CCAC as indicated by the intra-class correlation coefficient [34]. In step 2 of the multi-level logistic analyses, separate multivariate models were created for each of the predictor variable groupings (enabling factors, predisposing factors, disease diagnoses, ADL/ IADLs, and scales). The results of these models were used to create final multivariate models for both PT and OT that incorporated the strongest predictive variables from step 2 (odds ratio 41.2 and odds ratio 50.83). These odds ratio cut-off values are used to identify stronger predictor variables and were based upon similar research using a large database of RAI-HC data [35]. Area under ROC curve (AUC) results were examined across all models to identify which sets of variables were best at predicting OT and PT service utilization. AUCs are a commonly used measure of accuracy in biomedical informatics [36]. If a model can perfectly predict a target variable, their AUC would be equal to 1.0; the minimum value of AUC is 0.5 which would represent no improvement over random assignment of predictions. Analyses were conducted in SAS version 9.1.3 using PROC GLIMMIX [37].
Results Of the 299 262 older adults long-stay home care clients, the majority were female (62.5%) with an average age of 81.1 years. Of these clients, 26.5% received OT services within 3 months of their initial RAI-HC assessment and 21.4% received PT services within 3 months of their initial assessment.
627
Multi-level models – step 1 (null models) For OT, the ICC was equal to 0.21. This indicates that just over 20% of the variance in OT service provision was due to where a client lived. For PT, the ICC was smaller (ICC ¼ 0.09) and indicated that 9% of the variance in PT service provision was due to a client’s home region. Multi-level models – step 2 (predictor group models) Table 1 highlights the results from the 10 models developed in Step 2 (5 variable groups 2 outcomes). The results of these models were utilized to reduce the predictor variables to create final models. The strongest predictor (OR41.2 or OR50.83) varied slightly between the two rehabilitation services. Multi-level models – step 3 (final models) Odd ratio plots for the OT and PT models are found in Figures 1 and 2. Of all of the most predictive clinical variables, the factors that were most predictive of OT were Environmental Hazards in Home (OR ¼ 1.91, 95% CI ¼ 1.87–1.96), Parkinson’s disease (OR ¼ 1.48, 95% CI ¼ 1.42–1.55), needing assistance transferring surfaces (OR ¼ 1.41, 95% CI ¼ 1.36–1.47), and needing assistance with shopping (OR ¼ 1.39, 95% CI ¼ 1.34–1.43). Dementia was highly associated with not receiving OT services: (OR ¼ 0.77, 95% CI ¼ 0.75–0.80). The top predictor variables for PT provision included having a hip fracture (OR ¼ 2.08, 95% CI ¼ 1.99–2.16), needing assistance with outside locomotion (OR ¼ 1.55, 95% CI ¼ 1.51–1.58), shopping (OR¼1.51, 95% CI ¼ 1.46–1.58), and non-hip related fracture (OR¼1.47, 95% CI ¼ 1.43–1.52). Three variables were highly predictive of older clients not receiving PT services: having a diagnosis of dementia (OR ¼ 0.60, 95% CI ¼ 0.57–0.63), French as the primary spoken language (OR ¼ 0.77, 95% CI ¼ 0.71–0.85), and cancer (OR ¼ 0.76, 95% CI ¼ 0.74–0.79).
Table 1. Top predictor variables from Step 2 multivariate analyses. Occupational therapy Predictor group Predisposing
Enabling Need – Scales Need – ADL/IADL Variables
Need – Disease Diagnoses
a
Not a strong predictor.
Variable French as primary language Sex Married Age 65–74 Age 75–84 Age 85+ Environmental Hazard in home Caregiver in distress CHESS Pain Scale CPS Transfer Surfaces Shopping Outside Mobility Bed Mobility Dressing Lower Body Managing Finances Managing Medications Phone Use Housework Transportation Bathing Parkinson’s Cerebrovascular disease Fracture (non-hip) Dementia Cancer Hip Fracture
OR a a a a a a 2.25 1.28 1.25 a a 1.51 1.44 1.29 1.23 1.24 a a a a a a 1.71 1.33 1.23 0.80 a a
Physiotherapy
95% CI
OR
a a a a a a 2.21–2.30 1.25–1.325 1.24–1.26 a a 1.45–1.58 1.39–1.49 1.27–1.33 1.18–1.33 1.20–1.28 a a a a a a 1.64–1.78 1.30–1.36 1.20–1.27 0.78–0.82 a a
0.75 a 1.21
95% CI
0.71–0.80 a 1.19–1.23 Reference Group 0.99 0.97–1.01 0.83 0.81–0.86 1.89 1.85–1.93 a a a a 1.25 1.24–1.26 0.71 0.70–0.72 1.39 1.33–1.45 1.50 1.44–1.56 1.65 1.61–1.69 a a 1.61 1.56–1.67 0.81 0.79–0.83 0.77 0.75–0.79 0.64 0.62–0.65 1.35 1.30–1.42 1.32 1.29–1.36 1.30 1.27–1.33 1.66 1.59–1.74 a a 1.79 1.74–1.85 0.43 0.42–0.44 0.82 0.80–0.84 2.41 2.32–2.51
628
J. J. Armstrong et al.
Disabil Rehabil, 2015; 37(7): 625–631
Figure 1. Odds ratios plot for final PT multivariate model.
Figure 2. Odds ratios plot for final OT multivariate model.
Area under the curve Table 2 contains the AUC values for all models in this study. Of the three Andersen and Newman variable groupings, need Factors were the most predictive sets (note that these models also had the largest number of predictors). The highest AUCs came from creating the final models that used the strongest predictors from across all three categories of the Andersen-Newman framework (OT AUC ¼ 0.6833; PT AUC ¼ 0.7225).
Discussion This study has investigated the regional variation and client-level predictors of the utilization of OT and PT in a population of older long-stay clients in the Ontario Home Care System. The multilevel modeling approach indicated that for the years 2005–2010,
Table 2. AUC for all multi-level multivariate models.
Model Predisposing factors Enabling factors Need factors 1. Scales 2. ADL/IADL 3. Disease Diagnoses Final model
Occupational therapy AUC
Physiotherapy AUC
0.6183 0.6460
0.6240 0.6350
0.6639 0.6542 0.6291 0.6833
0.6919 0.6876 0.6716 0.7229
20% of the variation in the provision of OT can be accounted for by region. For PT, this amount was less but still substantial at 9%. Findings of sizable regional variations in the Ontario Home Care System have been found previously [12–15] and this study
DOI: 10.3109/09638288.2014.935494
illustrates that variation across health regions continues to be an issue. While some amount of variation is expected, access to rehabilitation therapy services for older adults should be equal for everyone within the province, irrespective of where they reside. Whether this variation is due to funding differences or policy and practices of the individual regions or CCACs, these findings indicate that there is a need to examine and modify the home care system to ensure that all older residents of Ontario have equal access to rehabilitation services. Using a provincial database, the results of this study highlight the characteristics of older adults that have recently driven the provision and non-provision of rehabilitation services. The most highly associated client-level factors related to PT were hip fracture, impairments in ADL/IADL, cerebrovascular accidents, and Parkinson’s disease. PT has a clear and direct role to play for older adults with these issues. For example, past research has illustrated that up to one fifth of hip fracture patients become functionally dependent [38]. Home care PT services play an important role in maximizing the status of patients with a hip fracture after leaving the hospital and returning to the community. As all of these factors related to PT utilization are associated with aging, the results provide further evidence that the need for PT services in home care will increase as the proportion of older adults in the population increases. Governments, policy makers, and health care systems planners should be planning for increases in demand for PT services and modifying the system to meet the changing needs of the population. The results also clearly illustrated that dementia significantly decreased the likelihood that an older home care client receives PT services. This indicates that for older long-stay clients, dementia may be a barrier to receiving PT services in the Ontario Home Care System. Research has shown that cognitive impairment can impact the effectiveness and efficiency of PT in older adults [39]. However, research has also shown that PT can lead to positive outcomes for older adults with cognitive impairments [40–44]. With the number of older adults experiencing dementia expected to increase dramatically [45] and a growing understanding of the links between physical frailty and cognitive impairment [46], decision makers and health care managers need to re-examine the role that home care PT can play for older adults with dementia. French as the primary language was also related to a reduced rate PT service provision. However, Theriault and Stones found that receipt of home care services in Ontario did not vary significantly between francophones and anglophones across the province [47]. The potentially controversial finding of this study that French older adults may be marginalized in regards to home care rehabilitation services requires further research. The results could possibly be attributed to the fact that many FrancoOntarians live in the North Eastern CCAC and Champlain regions which contains more rural areas and access to OT and PT in home care may be limited for all home care clients in the region (not just for French-speaking clients). For OT, hazards in the home environment were highly predictive of service provision. This coincides with research findings that indicate that OT is primarily used for home assessment purposes in the Ontario Home Care System [48]. This finding also corresponds with a recent review of American home care services that found adaptation and environment as the most frequently referenced intervention [49]. Within the Ontario context, this finding illustrates the increasing emphasis of OT on adaptation and environment while decreasing its focus on remediation. Increased complexity as indicated by the CHESS scale was also associated with higher rates of OT provision. The CHESS scale was designed to identify individuals who have unstable health and may be at risk of serious decline. This association with OT may be related to the need to adapt the
Rehabilitation therapy services for older clients
629
client’s home to address functional decline and home safety issues. Parkinson’s disease was also associated with higher rates of OT provision. Due to the impact of Parkinson’s disease on both motor and non-motor systems in the brain, OTs can work with clients to assist with executing meaningful daily occupations and addressing issues that may interfere with these activities. Similarly to PT, having a diagnosis of dementia was related to a reduced rate of OT service use. This result is surprising given the evidence of effectiveness [50,51] and cost-effectiveness [52] of rehabilitation for persons with cognitive impairment. More research is warranted to explore these findings. Of the predictor variables organized by the Andersen Newman framework, need factors were most predictive of service utilization. This replicates findings of previous studies that found that characteristics related to functional and health problems are most predictive of home care service utilization [53]. When comparing the AUCs from across the models, the highest AUCs (the most accurate models) were from the models that incorporated factors from all three domains of the Andersen Newman framework. This demonstrates the strength of the Andersen Newman framework as it encompasses the wide spectrum of factors that influence health care service utilization. Compared to the general population, examining a wide spectrum of factors may be even more important for geriatric populations due to the increasing complexity that has been found in the health profiles of the oldest-old [54]. Limitations This study is not without limitations. First, the sample that was used for the analysis contained only long-stay clients and did not include the home care clients who used services for less than 60 d. Therefore, the results of this study should not be generalized to all rehabilitation services for older adults in home care. However, with the expanded use of a new assessment instrument (inter-RAI Contact Assessment), future studies using standardized provincial home care data would provide a more complete picture of rehabilitation services for older adults by examining both shortstay and long-stay clients. A second limitation of this study is that the analyses examined only the first assessment upon entering the home care system and service use within a short time period post assessment (window for the target variables was 3 months post RAI-HC assessment). Future research would benefit from examining the use of home care services over the entire time that clients spend in the system. This study also did not examine factors that related to changes in the home care system that occurred from 2005 to 2010. As the home care system (and any health care system for that matter) is dynamic, changes in how the system provides services would have changed to some degree during the study period. A final limitation of this study is that the outcome of interest was dichotomized and the level of service was not considered. While the goal of this study was to determine the differences between users and non-users, the amount of service provided may be of interest to specific decision makers at the regional and provincial levels. Future research can address this question of amount of service while controlling for regional differences by using hierarchical generalized linear models.
Conclusions This study found regional differences between the health regions of Ontario for OT and PT services for older adults. In a publiclyfunded home care system, where persons live should not impact their ability to access services. The wide variations in the rates of home care rehabilitation services highlights the importance of adapting funding to ensure that all older residents of Ontario have
630
J. J. Armstrong et al.
equal access. The analyses of the provincial database also revealed the main drivers and barriers to receiving rehabilitative services. At a time when demand for rehabilitation is increasing and health care resources are constrained, it will be important to utilize available health information to optimize the services for older adults.
Declaration of interest This research was supported by a grant from the Canadian Institutes of Health Research. The lead author was supported by an Ontario Graduate Scholarship. The authors report no conflicts of interest.
References 1. Gill TM, Baker DI, Gottschalk M, et al. A program to prevent functional decline in physically frail, elderly persons who live at home. N Engl J Med 2002;347:1068–74. 2. Kuisma R. A randomized, controlled comparison of home versus institutional rehabilitation of patients with hip fracture. Clin Rehabil 2002;16:553–61. 3. Crotty M, Whitehead C, Miller M, Gray S. Patient and caregiver outcomes 12 months after home-based therapy for hip fracture: a randomized controlled trial. Arch Phys Med Rehabil 2003;84: 1237–9. 4. Gitlin LN, Hauck WW, Winter L, et al. Effect of an in-home occupational and physical therapy intervention on reducing mortality in functionally vulnerable older people: preliminary findings. J Am Geriatr Soc 2006;54:950–5. 5. Gitlin LN, Winter L, Dennis MP, et al. A randomized trial of a multicomponent home intervention to reduce functional difficulties in older adults. J Am Geriatr Soc 2006;54:809–16. 6. Giusti A, Barone A, Oliveri M, et al. An analysis of the feasibility of home rehabilitation among elderly people with proximal femoral fractures. Arch Phys Med Rehabil 2006;87:826–31. 7. Cook RJ, Berg K, Lee K, et al. Rehabilitation in home care is associated with functional improvement and preferred discharge. Arch Phys Med Rehabil 2013;94:1038–47. 8. Stolee P, Lim SN, Wilson L, Glenny C. Inpatient versus home-based rehabilitation for older adults with musculoskeletal disorders: a systematic review. Clin Rehabil 2012;26:387–402. 9. Hirdes JP, Fries BE, Morris JN, et al. Home care quality indicators (HCQIs) based on the MDS-HC. Gerontologist 2004;44:665–79. 10. Borrie MJ, Stolee P, Knoefel FD, et al. Current best practices in geriatric rehabilitation in canada. Geriatrics Today: J Can Geriatr Soc 2005;8:148–53. 11. Jaglal S, McIsaac W, Hawker G, et al. Information needs in the management of osteoporosis in family practice: an illustration of the failure of the current guideline implementation process. Osteoporosis Int 2003;14:672–6. 12. Coyte PC, Young W. Regional variations in the use of home care services in Ontario, 1993/95. CMAJ 1999;161:376–80. 13. Mahomed, Nizar Lau JTC, Lin MKS, et al. Significant variation exists in home care services following total joint arthroplasty. J Rheumatol 2004;31:973–5. 14. Shanmugasegaram S, Oh P, Reid RD, et al. Cardiac rehabilitation barriers by rurality and socioeconomic status: a cross-sectional study. Int J Equity Health 2013;12:72. 15. Kitchen PW, Allison P, Donna RW. Socio-spatial patterns of home care use in Ontario, Canada: a case study. Health Place 2011;17: 195–206. 16. Landry MD, Jaglal S, Wodchis WP, et al. Analysis of factors affecting demand for rehabilitation services in Ontario, Canada: a health-policy perspective. Disabil Rehabil 2008;30:1837–47. 17. Morris JN, Fries BE, Steel K, et al. Comprehensive clinical assessment in community setting: applicability of the MDS-HC. J Am Geriatr Soc 1997;45:1017–24. 18. Hirdes JP, Fries BE, Morris JN, et al. Integrated health information systems based on the RAI/MDS series of instruments. Health Manage Forum 1999;12:30–40. 19. Canadian Association on Home Care. Portraits of Home Care in Canada. 2013. Available from: http://www.cdnhomecare.ca/content.php?sec=4 [last accessed 4 Jul 2013].
Disabil Rehabil, 2015; 37(7): 625–631
20. Hirdes JP. Addressing the health needs of frail elderly people: Ontario’s experience with an integrated health information system. Age Ageing 2006;35:329–31. 21. Hirdes JP, Ljunggren G, Morris JN, et al. Reliability of the interRAI suite of assessment instruments: a 12-country study of an integrated health information system. BMC Health Serv Res 2008;8:277. doi: 10.1186/1472-6963-8-277. 22. Andersen R, Newman JF. Societal and individual determinants of medical care utilization in the United States. Milbank Mem Fd Quart 1973;51:95–124. 23. Hirdes J. The MDS-CHESS scale: a new measure to predict mortality in institutionalized older people. J Am Geriatr Soc 2003; 51:96–100. 24. Armstrong JJ, Stolee P, Hirdes JP, Poss JW. Examining three frailty conceptualizations in their ability to predict negative outcomes for home-care clients. Age Ageing 2010;39:755–8. 25. Morris JN, Fries BE, Morris SA. Scaling ADLs within the MDS. J Gerontol – Series A Biol Sci Med Sci 1999;54:M546–53. 26. Landi F, Tua E, Onder G, et al. Minimum data set for home care: a valid instrument to assess frail older people living in the community. Med Care 2000;38:1184–90. 27. Morris JN, Carpenter I, Berg K, Jones RN. Outcome measures for use with home care clients. Can J Aging 2000;19:87–105. 28. Morris JN, Fries BE, Mehr DR, et al. MDS cognitive performance scale. J Gerontol 1994;49:M174–82. 29. Fries BE, Simon SE, Morris JN, et al. Pain in U.S. nursing homes: validating a pain scale for the minimum data set. Gerontologist 2001;41:173–9. 30. Murnaghan DA, Leatherdale ST, Sihvonen M, Kekki P. A multilevel analysis examining the association between school-based smoking policies, prevention programs and youth smoking behavior: evaluating a provincial tobacco control strategy. Health Educ Res 2008;23:1016–28. 31. Leatherdale S. The association between overweight and school policies on physical activity: a multilevel analysis among elementary school youth in the PLAY-on study. Health Educ Res 2010;25: 1061–73. 32. Hobin E, Leatherdale S, Manske S, et al. A multilevel examination of gender differences in the association between features of the school environment and physical activity among a sample of grades 9 to 12 students in Ontario, Canada. BMC Publ Health 2012;12:74. 33. Luke, DA. Multilevel modeling. Thousand Oaks (CA): Sage Publications; 2004. 34. Snijders T, Bosker R. Multilevel analysis: an introduction to basic and advanced multilevel modeling. Thousand Oaks (CA): Sage Publications; 1999. 35. The´riault E´, Guthrie DM. Conference presentation at Canadian Association on Gerontology Conference – Vancouver 2012. Abstract. Available from: http://cagacg.ca/conference/cag2012/ [last accessed 9 Jun 2014]. 36. Lasko T, Bhagwat J, Zou K, Ohno L. The use of receiver operating characteristic curves in biomedical informatics. J Biomed Inform 2005;38:404–15. 37. SAS Corporation. 2008 SAS 9.1 Documentation. SAS Institute Inc., Cary, NC, USA; 2006. 38. Jaglal SB, McIsaac WJ, Hawker G, et al. Information needs in the management of osteoporosis in family practice: an illustration of the failure of the current guideline implementation process. Osteoporosis Int 2003;14:672–6. 39. Colombo M, Guaita A, Cottino M, et al. The impact of cognitive impairment on the rehabilitation process in geriatrics. Arch Gerontol Geriatr 2004;38:85–92. 40. Ghisla M. Predictors of successful rehabilitation in geriatric patients: subgroup analysis of patients with cognitive impairment. Aging Clin Exp Res 2007;19:417–23. 41. Hershkovitz AB, Shai B. The association between patients’ cognitive status and rehabilitation outcome in a geriatric day hospital. Disabil Rehabil 2007;29:333–7. 42. Naglie G, Tansey C, Kirkland JL, et al. Interdisciplinary inpatient care for elderly people with hip fracture: a randomized controlled trial. CMAJ 2002;167:25–32. 43. Colombo M, Guaita A, Cottino M, et al. The impact of cognitive impairment on the rehabilitation process in geriatrics. Arch Gerontol Geriatr 2004;38:85–92.
DOI: 10.3109/09638288.2014.935494
44. Ghisla M. Predictors of successful rehabilitation in geriatric patients: subgroup analysis of patients with cognitive impairment. Aging Clin Exp Res 2007;19:417–23. 45. Alzheimer Society of Canada. Rising Tide report 2010. Available from: http://www.alzheimer.ca/en/Get-involved/Raise-your-voice/ ~/media/Files/national/Advocacy/ASC_Rising%20Tide_Full%20 Report_Eng.ashx [last accessed 8 Jul 2013]. 46. Robertson D, Savva G, Kenny R. Frailty and cognitive impairment – a review of the evidence and causal mechanisms. Age Res Rev 2013; 12:840–51. 47. The´riault E´, Stones M. Differences between the francophone and anglophone ethno-linguistic groups in the Ontario home-care setting. Senior Care Canada 2009;3:10–14. 48. Armstrong J. Rehabilitation therapy services for older long–stay clients in the ontario home care system – chapter 3. [dissertation]. Waterloo, Ontario: University of Waterloo; 2013. Available from the Electronic thesis and dissertation database: http://uwspace.uwaterloo.ca/handle/10012/6 [last accessed 4 Jul 2013].
Rehabilitation therapy services for older clients
631
49. Craig D. Current occupational therapy publications in home health: a scoping review. Am J Occup Ther 2012;66:338–47. 50. Graff M, Vernooij-Dassen M, Thijssen M, et al. Community based occupational therapy for patients with dementia and their care givers: Randomised controlled trial. Brit Med J 2006;333: 1196–9. 51. Graff M, Vernooij-Dassen M, Thijssen M, et al. Effects of community occupational therapy on quality of life, mood, and health status in dementia patients and their caregivers: a randomized controlled trial. J Gerontol A-Biol 2007;62:1002–9. 52. Graff M, Vernooij-Dassen M, Thijssen M, et al. Community occupational therapy for older patients with dementia and their care givers: cost effectiveness study. Brit Med J 2008;336: 134–8. 53. Forbes DA, Stewart N, Morgan D, et al. Individual determinants of home-care nursing and housework assistance. CJNR 2003;35:14–36. 54. Murtaugh C, Peng T, Totten A, et al. Complexity in geriatric home healthcare. J Healthcare Qual 2009;31:34–43.
Copyright of Disability & Rehabilitation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
REHABILITATION AND PRACTICE
Rehabilitation therapies for older clients of the Ontario home care system: regional variation and client-level predictors of service provision Joshua J. Armstrong1, Mu Zhu2, John P. Hirdes3, and Paul Stolee3,4 1
Geriatric Medicine Research, Dalhousie University, Halifax, Nova Scotia, Canada, 2Department of Statistics and Actuarial Science, University of Waterloo, Waterloo, Ontario, Canada, 3School of Public Health and Health Systems, University of Waterloo, Waterloo, Ontario, Canada, and 4 Schlegel – University of Waterloo Research Institute for Aging, Waterloo, Ontario, Canada Abstract
Keywords
Purpose: To examine regional variation in service provision and identify the client characteristics associated with occupational therapy (OT) and physiotherapy (PT) services for older adults in the Ontario Home Care System. Methods: Secondary analyses of a provincial database containing comprehensive assessments (RAI-HC) linked with service utilization data from every older long-stay home care client in the system between 2005 and 2010 (n ¼ 299 262). Hierarchical logistic regression models were used to model the dependent variables of OT and PT service use within 90 d of the initial assessment. Results: Regional differences accounted for 9% of the variation in PT service provision and 20% of OT service provision. After controlling for the differences across regions, the most powerful predictors of service provision were identified for both OT and PT. The most highly associated client characteristics related to PT service provision were hip fracture, impairments in activities of daily living/instrumental activities of daily living, cerebrovascular accidents, and cognitive impairment. For OT, hazards in the home environment was the most powerful predictor of future service provision. Conclusions: Where a client lived was an important determinant of service provision in Ontario, raising the possibility of inequities in access to rehabilitation services. Health care planners and policy makers should review current practices and make adjustments to meet the increasing and changing needs for rehabilitation therapies of the aging population.
Geriatrics, multi-level modeling, occupational therapy, physiotherapy History Received 15 November 2013 Revised 10 June 2014 Accepted 12 June 2014 Published online 1 July 2014
ä Implications for Rehabilitation
For older adults in home care, the goal of rehabilitation therapy services is to allow individuals to maintain or improve physical functioning, quality of life and overall independence while living within their community. Previous research has demonstrated that a large proportion of home care clients specifically identified as having rehabilitation potential do not receive it. This article used clinical assessment data to identify the predictors of and barriers to rehabilitation services for older adults in the Ontario Home Care System. Barriers of PT included dementia diagnosis and French as a first language. Barriers to OT included dementia diagnosis. Policies and practices related to service provision for older adults should be reconsidered if we are going to meet the demands of aging populations and increasing rates of functional and cognitive impairments.
Introduction Many studies have found rehabilitation therapies in home-based settings to be effective in improving outcomes for older adults [1–7]. A 2012 systematic literature review found that home-based
Address for correspondence: Paul Stolee, PhD, School of Public Health and Health Systems, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, Canada N2L 1P3. Fax: (519) 888-4362. E-mail: [email protected]
rehabilitation produced outcomes either equal or better than rehabilitation based within hospitals, for older persons with musculoskeletal disorders [8]. Despite this, in the Canadian province of Ontario, research has demonstrated that a large proportion of home care clients specifically identified as having rehabilitation potential do not receive it [9–11]. The Ontario provincial home care system has been geographically organized into 14 regions referred to as Local Health Integration Networks (LHINs). This regional division of the system was designed to allow each region to tailor its services to
626
J. J. Armstrong et al.
their populations. Eligibility for the receipt of home care services is determined by case managers who then arrange for health care providers – nurses, physiotherapists, social workers, registered dieticians, occupational therapists, speech therapists, and personal support workers – to provide a range of care and support services to individuals living within the community. Although the province aims to ensure equal access to publicly funded services for all residents of the province, research studies have found substantial disparities in the provision of home care services and rehabilitation therapies across the geographic regions of Ontario [12–15]. This indicates that where an older client lives could dictate whether or not they have access to rehabilitation therapy services, despite their actual or perceived needs. Due to an aging population and the associated increase in a number of chronic diseases, disabilities, and complex conditions, the demand of rehabilitation services across the continuum of care is expected to rise in the years to come [16]. In a publicly funded home care system where there are individuals with unmet needs and reported regional differences, there is a need for research studies to develop understanding of current practices in order to optimize the system for the growing older population. The first objective of this study is to quantify the variation in the provision of physiotherapy (PT) and occupational therapy (OT) services for older adults across the 14 health regions of the Ontario Home Care System. After accounting for regional variation, a secondary objective is to identify client characteristics that are associated with users and non-users of PT and OT services. These objectives will be met using hierarchical logistic regression modeling of assessment data from a large provincial database.
Methods This study is a population-level, observational study based on the initial inter-RAI Home Care (RAI-HC) assessment and service utilization information from every long-stay home care client in Ontario, Canada, from 2005 to 2010 (n ¼ 299 262). Analyses excluded all home care clients under the age of 65 and all assessments performed on older adults in hospitals for placement purposes. The province of Ontario is organized by geography into 14 health regions known as LHINs, each of which includes a Community Care Access Centre (CCAC) which acts as a single entry point for persons who require home care services. The Office of Research Ethics at the University of Waterloo provided ethics clearance for this study. Data sources The inter-RAI home care assessment instrument, the RAI-HC (or MDS-HC) [17] is used to collect health information on home care clients in the Canadian province of Ontario as well as many other jurisdictions [18,19] . The RAI-HC is mandated in Ontario for assessment of all individuals who are enter the home care system and are expected to be long stay (i.e. individuals requiring service for 60 d or more) [20]. As a comprehensive assessment instrument, the RAI-HC collects information using over 300 items from a wide range of domains. Once this information is collected by CCAC case managers, it is used to inform and guide comprehensive care and service planning in community-based settings. RAI-HC items have been shown to have good inter-rater and test–retest reliability [17,21]. Measures Independent variables Due to the large number of variables collected in the RAI-HC assessment, the Andersen and Newman Framework of Health
Disabil Rehabil, 2015; 37(7): 625–631
Services Utilization [22] was used to organize the predictor variables from the RAI-HC into three categories: Predisposing Factors, Enabling Factors, and Needs Factors. Predisposing factors included variables that are thought to influence an individual’s propensity to use services before the need for those services is present (e.g. sex, age, first language). Enabling factors are client characteristics that facilitate or inhibit the ease at which services can be obtained (e.g. home environment, education, income). Since the RAI-HC collects information on clinical characteristics for use in care planning across the multiple health care disciplines (e.g. nursing, rehabilitation, social work), many domains contained in the instrument are related to need. Because of this, need factors were separated into three separate categories: (1) inter-RAI scales, (2) activities of daily living (ADL) and Instrumental ADL (IADL) impairments, and (3) disease diagnoses. Within the electronic RAI-HC information system, a variety of scales are generated automatically for each individual client. These scales are designed to act as status and outcome measures that can be used to summarize client characteristics in a number of different domains. The following inter-RAI scales are utilized in this study: Changes in Health, End-stage disease and Signs and Symptoms (CHESS) score (a measure of health instability) [23,24]; Activities of Daily Living Self-performance Hierarchy Scale [25,26]; Instrumental Activities of Daily Living Performance Scale [26,27]; Cognitive Performance Scale [26,28]; and a Pain Scale [29]. The RAI-HC contains 18 specific items related to ADLs and IADLs. All of these individual items were included in the analyses. These items capture information about the client’s performance in the various domains in the last 7 d for IADLs, and for the last 3 d for ADLs. The ADL and IADL items were dichotomized (independent versus requiring assistance). The final subcategory within the Need Factors group is Disease Diagnoses. In total, 18 diagnoses were included in the analyses, each of the variables was dichotomized (not present versus present). Dependent variables The dependent variables in this study are the provision of OT and PT services after entering the home care system for the first time. Using unique client identifiers, RAI-HC assessment data were linked to health care service use records from the Ontario Association of Community Care Access Centres (OACCAC). OT and PT service data were dichotomized so that analyses could examine the differences between rehabilitation service users and non-users. The window of time for service utilization that was examined was 90 d post RAI-HC assessment. Statistical analyses Due to the hierarchical nature of the provincial data (home care clients nested within health regions (CCACs)), the modeling approach that was chosen to evaluate rehabilitation service provision was hierarchical logistic regression. This approach allows for the estimation of the degree to which health regions are associated with service provision and the creation of statistical models that control for any regional variation. Consistent with other multi-level modeling studies [30–32], a multistep procedure was utilized: (1) examination of the variability in provision of OT and PT services across geographic regions, (2) estimating the association of client characteristics with PT/OT services while accounting for region effects, and (3) the development of multivariate models that incorporate the most powerful clientlevel predictors from step (2). All hierarchical logistic models
Rehabilitation therapy services for older clients
DOI: 10.3109/09638288.2014.935494
were created using a random intercept approach that accounted for the differences in service provision between CCACs [33]. All client-level predictors were treated as fixed effects. In the first step of these analyses, null models were created to identify the amount of variance due to region. These null models indicate the amount of variance in OT and PT service provision that can be attributed to the client’s location within a CCAC as indicated by the intra-class correlation coefficient [34]. In step 2 of the multi-level logistic analyses, separate multivariate models were created for each of the predictor variable groupings (enabling factors, predisposing factors, disease diagnoses, ADL/ IADLs, and scales). The results of these models were used to create final multivariate models for both PT and OT that incorporated the strongest predictive variables from step 2 (odds ratio 41.2 and odds ratio 50.83). These odds ratio cut-off values are used to identify stronger predictor variables and were based upon similar research using a large database of RAI-HC data [35]. Area under ROC curve (AUC) results were examined across all models to identify which sets of variables were best at predicting OT and PT service utilization. AUCs are a commonly used measure of accuracy in biomedical informatics [36]. If a model can perfectly predict a target variable, their AUC would be equal to 1.0; the minimum value of AUC is 0.5 which would represent no improvement over random assignment of predictions. Analyses were conducted in SAS version 9.1.3 using PROC GLIMMIX [37].
Results Of the 299 262 older adults long-stay home care clients, the majority were female (62.5%) with an average age of 81.1 years. Of these clients, 26.5% received OT services within 3 months of their initial RAI-HC assessment and 21.4% received PT services within 3 months of their initial assessment.
627
Multi-level models – step 1 (null models) For OT, the ICC was equal to 0.21. This indicates that just over 20% of the variance in OT service provision was due to where a client lived. For PT, the ICC was smaller (ICC ¼ 0.09) and indicated that 9% of the variance in PT service provision was due to a client’s home region. Multi-level models – step 2 (predictor group models) Table 1 highlights the results from the 10 models developed in Step 2 (5 variable groups 2 outcomes). The results of these models were utilized to reduce the predictor variables to create final models. The strongest predictor (OR41.2 or OR50.83) varied slightly between the two rehabilitation services. Multi-level models – step 3 (final models) Odd ratio plots for the OT and PT models are found in Figures 1 and 2. Of all of the most predictive clinical variables, the factors that were most predictive of OT were Environmental Hazards in Home (OR ¼ 1.91, 95% CI ¼ 1.87–1.96), Parkinson’s disease (OR ¼ 1.48, 95% CI ¼ 1.42–1.55), needing assistance transferring surfaces (OR ¼ 1.41, 95% CI ¼ 1.36–1.47), and needing assistance with shopping (OR ¼ 1.39, 95% CI ¼ 1.34–1.43). Dementia was highly associated with not receiving OT services: (OR ¼ 0.77, 95% CI ¼ 0.75–0.80). The top predictor variables for PT provision included having a hip fracture (OR ¼ 2.08, 95% CI ¼ 1.99–2.16), needing assistance with outside locomotion (OR ¼ 1.55, 95% CI ¼ 1.51–1.58), shopping (OR¼1.51, 95% CI ¼ 1.46–1.58), and non-hip related fracture (OR¼1.47, 95% CI ¼ 1.43–1.52). Three variables were highly predictive of older clients not receiving PT services: having a diagnosis of dementia (OR ¼ 0.60, 95% CI ¼ 0.57–0.63), French as the primary spoken language (OR ¼ 0.77, 95% CI ¼ 0.71–0.85), and cancer (OR ¼ 0.76, 95% CI ¼ 0.74–0.79).
Table 1. Top predictor variables from Step 2 multivariate analyses. Occupational therapy Predictor group Predisposing
Enabling Need – Scales Need – ADL/IADL Variables
Need – Disease Diagnoses
a
Not a strong predictor.
Variable French as primary language Sex Married Age 65–74 Age 75–84 Age 85+ Environmental Hazard in home Caregiver in distress CHESS Pain Scale CPS Transfer Surfaces Shopping Outside Mobility Bed Mobility Dressing Lower Body Managing Finances Managing Medications Phone Use Housework Transportation Bathing Parkinson’s Cerebrovascular disease Fracture (non-hip) Dementia Cancer Hip Fracture
OR a a a a a a 2.25 1.28 1.25 a a 1.51 1.44 1.29 1.23 1.24 a a a a a a 1.71 1.33 1.23 0.80 a a
Physiotherapy
95% CI
OR
a a a a a a 2.21–2.30 1.25–1.325 1.24–1.26 a a 1.45–1.58 1.39–1.49 1.27–1.33 1.18–1.33 1.20–1.28 a a a a a a 1.64–1.78 1.30–1.36 1.20–1.27 0.78–0.82 a a
0.75 a 1.21
95% CI
0.71–0.80 a 1.19–1.23 Reference Group 0.99 0.97–1.01 0.83 0.81–0.86 1.89 1.85–1.93 a a a a 1.25 1.24–1.26 0.71 0.70–0.72 1.39 1.33–1.45 1.50 1.44–1.56 1.65 1.61–1.69 a a 1.61 1.56–1.67 0.81 0.79–0.83 0.77 0.75–0.79 0.64 0.62–0.65 1.35 1.30–1.42 1.32 1.29–1.36 1.30 1.27–1.33 1.66 1.59–1.74 a a 1.79 1.74–1.85 0.43 0.42–0.44 0.82 0.80–0.84 2.41 2.32–2.51
628
J. J. Armstrong et al.
Disabil Rehabil, 2015; 37(7): 625–631
Figure 1. Odds ratios plot for final PT multivariate model.
Figure 2. Odds ratios plot for final OT multivariate model.
Area under the curve Table 2 contains the AUC values for all models in this study. Of the three Andersen and Newman variable groupings, need Factors were the most predictive sets (note that these models also had the largest number of predictors). The highest AUCs came from creating the final models that used the strongest predictors from across all three categories of the Andersen-Newman framework (OT AUC ¼ 0.6833; PT AUC ¼ 0.7225).
Discussion This study has investigated the regional variation and client-level predictors of the utilization of OT and PT in a population of older long-stay clients in the Ontario Home Care System. The multilevel modeling approach indicated that for the years 2005–2010,
Table 2. AUC for all multi-level multivariate models.
Model Predisposing factors Enabling factors Need factors 1. Scales 2. ADL/IADL 3. Disease Diagnoses Final model
Occupational therapy AUC
Physiotherapy AUC
0.6183 0.6460
0.6240 0.6350
0.6639 0.6542 0.6291 0.6833
0.6919 0.6876 0.6716 0.7229
20% of the variation in the provision of OT can be accounted for by region. For PT, this amount was less but still substantial at 9%. Findings of sizable regional variations in the Ontario Home Care System have been found previously [12–15] and this study
DOI: 10.3109/09638288.2014.935494
illustrates that variation across health regions continues to be an issue. While some amount of variation is expected, access to rehabilitation therapy services for older adults should be equal for everyone within the province, irrespective of where they reside. Whether this variation is due to funding differences or policy and practices of the individual regions or CCACs, these findings indicate that there is a need to examine and modify the home care system to ensure that all older residents of Ontario have equal access to rehabilitation services. Using a provincial database, the results of this study highlight the characteristics of older adults that have recently driven the provision and non-provision of rehabilitation services. The most highly associated client-level factors related to PT were hip fracture, impairments in ADL/IADL, cerebrovascular accidents, and Parkinson’s disease. PT has a clear and direct role to play for older adults with these issues. For example, past research has illustrated that up to one fifth of hip fracture patients become functionally dependent [38]. Home care PT services play an important role in maximizing the status of patients with a hip fracture after leaving the hospital and returning to the community. As all of these factors related to PT utilization are associated with aging, the results provide further evidence that the need for PT services in home care will increase as the proportion of older adults in the population increases. Governments, policy makers, and health care systems planners should be planning for increases in demand for PT services and modifying the system to meet the changing needs of the population. The results also clearly illustrated that dementia significantly decreased the likelihood that an older home care client receives PT services. This indicates that for older long-stay clients, dementia may be a barrier to receiving PT services in the Ontario Home Care System. Research has shown that cognitive impairment can impact the effectiveness and efficiency of PT in older adults [39]. However, research has also shown that PT can lead to positive outcomes for older adults with cognitive impairments [40–44]. With the number of older adults experiencing dementia expected to increase dramatically [45] and a growing understanding of the links between physical frailty and cognitive impairment [46], decision makers and health care managers need to re-examine the role that home care PT can play for older adults with dementia. French as the primary language was also related to a reduced rate PT service provision. However, Theriault and Stones found that receipt of home care services in Ontario did not vary significantly between francophones and anglophones across the province [47]. The potentially controversial finding of this study that French older adults may be marginalized in regards to home care rehabilitation services requires further research. The results could possibly be attributed to the fact that many FrancoOntarians live in the North Eastern CCAC and Champlain regions which contains more rural areas and access to OT and PT in home care may be limited for all home care clients in the region (not just for French-speaking clients). For OT, hazards in the home environment were highly predictive of service provision. This coincides with research findings that indicate that OT is primarily used for home assessment purposes in the Ontario Home Care System [48]. This finding also corresponds with a recent review of American home care services that found adaptation and environment as the most frequently referenced intervention [49]. Within the Ontario context, this finding illustrates the increasing emphasis of OT on adaptation and environment while decreasing its focus on remediation. Increased complexity as indicated by the CHESS scale was also associated with higher rates of OT provision. The CHESS scale was designed to identify individuals who have unstable health and may be at risk of serious decline. This association with OT may be related to the need to adapt the
Rehabilitation therapy services for older clients
629
client’s home to address functional decline and home safety issues. Parkinson’s disease was also associated with higher rates of OT provision. Due to the impact of Parkinson’s disease on both motor and non-motor systems in the brain, OTs can work with clients to assist with executing meaningful daily occupations and addressing issues that may interfere with these activities. Similarly to PT, having a diagnosis of dementia was related to a reduced rate of OT service use. This result is surprising given the evidence of effectiveness [50,51] and cost-effectiveness [52] of rehabilitation for persons with cognitive impairment. More research is warranted to explore these findings. Of the predictor variables organized by the Andersen Newman framework, need factors were most predictive of service utilization. This replicates findings of previous studies that found that characteristics related to functional and health problems are most predictive of home care service utilization [53]. When comparing the AUCs from across the models, the highest AUCs (the most accurate models) were from the models that incorporated factors from all three domains of the Andersen Newman framework. This demonstrates the strength of the Andersen Newman framework as it encompasses the wide spectrum of factors that influence health care service utilization. Compared to the general population, examining a wide spectrum of factors may be even more important for geriatric populations due to the increasing complexity that has been found in the health profiles of the oldest-old [54]. Limitations This study is not without limitations. First, the sample that was used for the analysis contained only long-stay clients and did not include the home care clients who used services for less than 60 d. Therefore, the results of this study should not be generalized to all rehabilitation services for older adults in home care. However, with the expanded use of a new assessment instrument (inter-RAI Contact Assessment), future studies using standardized provincial home care data would provide a more complete picture of rehabilitation services for older adults by examining both shortstay and long-stay clients. A second limitation of this study is that the analyses examined only the first assessment upon entering the home care system and service use within a short time period post assessment (window for the target variables was 3 months post RAI-HC assessment). Future research would benefit from examining the use of home care services over the entire time that clients spend in the system. This study also did not examine factors that related to changes in the home care system that occurred from 2005 to 2010. As the home care system (and any health care system for that matter) is dynamic, changes in how the system provides services would have changed to some degree during the study period. A final limitation of this study is that the outcome of interest was dichotomized and the level of service was not considered. While the goal of this study was to determine the differences between users and non-users, the amount of service provided may be of interest to specific decision makers at the regional and provincial levels. Future research can address this question of amount of service while controlling for regional differences by using hierarchical generalized linear models.
Conclusions This study found regional differences between the health regions of Ontario for OT and PT services for older adults. In a publiclyfunded home care system, where persons live should not impact their ability to access services. The wide variations in the rates of home care rehabilitation services highlights the importance of adapting funding to ensure that all older residents of Ontario have
630
J. J. Armstrong et al.
equal access. The analyses of the provincial database also revealed the main drivers and barriers to receiving rehabilitative services. At a time when demand for rehabilitation is increasing and health care resources are constrained, it will be important to utilize available health information to optimize the services for older adults.
Declaration of interest This research was supported by a grant from the Canadian Institutes of Health Research. The lead author was supported by an Ontario Graduate Scholarship. The authors report no conflicts of interest.
References 1. Gill TM, Baker DI, Gottschalk M, et al. A program to prevent functional decline in physically frail, elderly persons who live at home. N Engl J Med 2002;347:1068–74. 2. Kuisma R. A randomized, controlled comparison of home versus institutional rehabilitation of patients with hip fracture. Clin Rehabil 2002;16:553–61. 3. Crotty M, Whitehead C, Miller M, Gray S. Patient and caregiver outcomes 12 months after home-based therapy for hip fracture: a randomized controlled trial. Arch Phys Med Rehabil 2003;84: 1237–9. 4. Gitlin LN, Hauck WW, Winter L, et al. Effect of an in-home occupational and physical therapy intervention on reducing mortality in functionally vulnerable older people: preliminary findings. J Am Geriatr Soc 2006;54:950–5. 5. Gitlin LN, Winter L, Dennis MP, et al. A randomized trial of a multicomponent home intervention to reduce functional difficulties in older adults. J Am Geriatr Soc 2006;54:809–16. 6. Giusti A, Barone A, Oliveri M, et al. An analysis of the feasibility of home rehabilitation among elderly people with proximal femoral fractures. Arch Phys Med Rehabil 2006;87:826–31. 7. Cook RJ, Berg K, Lee K, et al. Rehabilitation in home care is associated with functional improvement and preferred discharge. Arch Phys Med Rehabil 2013;94:1038–47. 8. Stolee P, Lim SN, Wilson L, Glenny C. Inpatient versus home-based rehabilitation for older adults with musculoskeletal disorders: a systematic review. Clin Rehabil 2012;26:387–402. 9. Hirdes JP, Fries BE, Morris JN, et al. Home care quality indicators (HCQIs) based on the MDS-HC. Gerontologist 2004;44:665–79. 10. Borrie MJ, Stolee P, Knoefel FD, et al. Current best practices in geriatric rehabilitation in canada. Geriatrics Today: J Can Geriatr Soc 2005;8:148–53. 11. Jaglal S, McIsaac W, Hawker G, et al. Information needs in the management of osteoporosis in family practice: an illustration of the failure of the current guideline implementation process. Osteoporosis Int 2003;14:672–6. 12. Coyte PC, Young W. Regional variations in the use of home care services in Ontario, 1993/95. CMAJ 1999;161:376–80. 13. Mahomed, Nizar Lau JTC, Lin MKS, et al. Significant variation exists in home care services following total joint arthroplasty. J Rheumatol 2004;31:973–5. 14. Shanmugasegaram S, Oh P, Reid RD, et al. Cardiac rehabilitation barriers by rurality and socioeconomic status: a cross-sectional study. Int J Equity Health 2013;12:72. 15. Kitchen PW, Allison P, Donna RW. Socio-spatial patterns of home care use in Ontario, Canada: a case study. Health Place 2011;17: 195–206. 16. Landry MD, Jaglal S, Wodchis WP, et al. Analysis of factors affecting demand for rehabilitation services in Ontario, Canada: a health-policy perspective. Disabil Rehabil 2008;30:1837–47. 17. Morris JN, Fries BE, Steel K, et al. Comprehensive clinical assessment in community setting: applicability of the MDS-HC. J Am Geriatr Soc 1997;45:1017–24. 18. Hirdes JP, Fries BE, Morris JN, et al. Integrated health information systems based on the RAI/MDS series of instruments. Health Manage Forum 1999;12:30–40. 19. Canadian Association on Home Care. Portraits of Home Care in Canada. 2013. Available from: http://www.cdnhomecare.ca/content.php?sec=4 [last accessed 4 Jul 2013].
Disabil Rehabil, 2015; 37(7): 625–631
20. Hirdes JP. Addressing the health needs of frail elderly people: Ontario’s experience with an integrated health information system. Age Ageing 2006;35:329–31. 21. Hirdes JP, Ljunggren G, Morris JN, et al. Reliability of the interRAI suite of assessment instruments: a 12-country study of an integrated health information system. BMC Health Serv Res 2008;8:277. doi: 10.1186/1472-6963-8-277. 22. Andersen R, Newman JF. Societal and individual determinants of medical care utilization in the United States. Milbank Mem Fd Quart 1973;51:95–124. 23. Hirdes J. The MDS-CHESS scale: a new measure to predict mortality in institutionalized older people. J Am Geriatr Soc 2003; 51:96–100. 24. Armstrong JJ, Stolee P, Hirdes JP, Poss JW. Examining three frailty conceptualizations in their ability to predict negative outcomes for home-care clients. Age Ageing 2010;39:755–8. 25. Morris JN, Fries BE, Morris SA. Scaling ADLs within the MDS. J Gerontol – Series A Biol Sci Med Sci 1999;54:M546–53. 26. Landi F, Tua E, Onder G, et al. Minimum data set for home care: a valid instrument to assess frail older people living in the community. Med Care 2000;38:1184–90. 27. Morris JN, Carpenter I, Berg K, Jones RN. Outcome measures for use with home care clients. Can J Aging 2000;19:87–105. 28. Morris JN, Fries BE, Mehr DR, et al. MDS cognitive performance scale. J Gerontol 1994;49:M174–82. 29. Fries BE, Simon SE, Morris JN, et al. Pain in U.S. nursing homes: validating a pain scale for the minimum data set. Gerontologist 2001;41:173–9. 30. Murnaghan DA, Leatherdale ST, Sihvonen M, Kekki P. A multilevel analysis examining the association between school-based smoking policies, prevention programs and youth smoking behavior: evaluating a provincial tobacco control strategy. Health Educ Res 2008;23:1016–28. 31. Leatherdale S. The association between overweight and school policies on physical activity: a multilevel analysis among elementary school youth in the PLAY-on study. Health Educ Res 2010;25: 1061–73. 32. Hobin E, Leatherdale S, Manske S, et al. A multilevel examination of gender differences in the association between features of the school environment and physical activity among a sample of grades 9 to 12 students in Ontario, Canada. BMC Publ Health 2012;12:74. 33. Luke, DA. Multilevel modeling. Thousand Oaks (CA): Sage Publications; 2004. 34. Snijders T, Bosker R. Multilevel analysis: an introduction to basic and advanced multilevel modeling. Thousand Oaks (CA): Sage Publications; 1999. 35. The´riault E´, Guthrie DM. Conference presentation at Canadian Association on Gerontology Conference – Vancouver 2012. Abstract. Available from: http://cagacg.ca/conference/cag2012/ [last accessed 9 Jun 2014]. 36. Lasko T, Bhagwat J, Zou K, Ohno L. The use of receiver operating characteristic curves in biomedical informatics. J Biomed Inform 2005;38:404–15. 37. SAS Corporation. 2008 SAS 9.1 Documentation. SAS Institute Inc., Cary, NC, USA; 2006. 38. Jaglal SB, McIsaac WJ, Hawker G, et al. Information needs in the management of osteoporosis in family practice: an illustration of the failure of the current guideline implementation process. Osteoporosis Int 2003;14:672–6. 39. Colombo M, Guaita A, Cottino M, et al. The impact of cognitive impairment on the rehabilitation process in geriatrics. Arch Gerontol Geriatr 2004;38:85–92. 40. Ghisla M. Predictors of successful rehabilitation in geriatric patients: subgroup analysis of patients with cognitive impairment. Aging Clin Exp Res 2007;19:417–23. 41. Hershkovitz AB, Shai B. The association between patients’ cognitive status and rehabilitation outcome in a geriatric day hospital. Disabil Rehabil 2007;29:333–7. 42. Naglie G, Tansey C, Kirkland JL, et al. Interdisciplinary inpatient care for elderly people with hip fracture: a randomized controlled trial. CMAJ 2002;167:25–32. 43. Colombo M, Guaita A, Cottino M, et al. The impact of cognitive impairment on the rehabilitation process in geriatrics. Arch Gerontol Geriatr 2004;38:85–92.
DOI: 10.3109/09638288.2014.935494
44. Ghisla M. Predictors of successful rehabilitation in geriatric patients: subgroup analysis of patients with cognitive impairment. Aging Clin Exp Res 2007;19:417–23. 45. Alzheimer Society of Canada. Rising Tide report 2010. Available from: http://www.alzheimer.ca/en/Get-involved/Raise-your-voice/ ~/media/Files/national/Advocacy/ASC_Rising%20Tide_Full%20 Report_Eng.ashx [last accessed 8 Jul 2013]. 46. Robertson D, Savva G, Kenny R. Frailty and cognitive impairment – a review of the evidence and causal mechanisms. Age Res Rev 2013; 12:840–51. 47. The´riault E´, Stones M. Differences between the francophone and anglophone ethno-linguistic groups in the Ontario home-care setting. Senior Care Canada 2009;3:10–14. 48. Armstrong J. Rehabilitation therapy services for older long–stay clients in the ontario home care system – chapter 3. [dissertation]. Waterloo, Ontario: University of Waterloo; 2013. Available from the Electronic thesis and dissertation database: http://uwspace.uwaterloo.ca/handle/10012/6 [last accessed 4 Jul 2013].
Rehabilitation therapy services for older clients
631
49. Craig D. Current occupational therapy publications in home health: a scoping review. Am J Occup Ther 2012;66:338–47. 50. Graff M, Vernooij-Dassen M, Thijssen M, et al. Community based occupational therapy for patients with dementia and their care givers: Randomised controlled trial. Brit Med J 2006;333: 1196–9. 51. Graff M, Vernooij-Dassen M, Thijssen M, et al. Effects of community occupational therapy on quality of life, mood, and health status in dementia patients and their caregivers: a randomized controlled trial. J Gerontol A-Biol 2007;62:1002–9. 52. Graff M, Vernooij-Dassen M, Thijssen M, et al. Community occupational therapy for older patients with dementia and their care givers: cost effectiveness study. Brit Med J 2008;336: 134–8. 53. Forbes DA, Stewart N, Morgan D, et al. Individual determinants of home-care nursing and housework assistance. CJNR 2003;35:14–36. 54. Murtaugh C, Peng T, Totten A, et al. Complexity in geriatric home healthcare. J Healthcare Qual 2009;31:34–43.
Copyright of Disability & Rehabilitation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
