Examining Variation in Access to Long-term Home Care Services for Ontario Seniors

Transcription

1 Examining Variation in Access to Long-term Home Care Services for Ontario Seniors By Erin Elizabeth Patterson A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Lawrence S. Bloomberg Faculty of Nursing University of Toronto Copyright by Erin Patterson 2016

2 Examining Variation in Access to Long-term Home care Services for Ontario Seniors Erin Elizabeth Patterson Doctor of Philosophy Lawrence S. Bloomberg Faculty of Nursing University of Toronto 2016 Abstract Background: As Canada s population ages, a greater number of seniors will require formal and informal care to remain in their homes. To effectively maintain seniors in their homes, equitable access to home care services is required. However, variations in access to these services have persisted in Ontario for decades. Purpose: The purpose of this study was to examine factors affecting access to long-term home care for seniors identified as at moderate risk of adverse outcomes and to determine whether variation in access to these services exists across Ontario. Methods: Administrative data were used to conduct a secondary analysis. A non-experimental design was employed. Factors associated with three dimensions of access to home care (propensity, intensity, wait time) were examined across four outcome categories: no home-based care, home-based nursing services alone, home-based supportive care services alone, and home-based nursing and supportive care services together. Results: Personal characteristics, caregiver characteristics and regional characteristics were found to affect all dimensions of access. Factors associated with propensity to receive services varied across outcome categories and included having a dementia diagnosis, dependence in performing activities of daily living, having a previous fall, previous health care utilization, and the region in which the patient resides. Similarly, factors associated with intensity of services were dependent on type of care received and included education, continence and living with a caregiver for nursing services and cognitive status, functional status, health status, and living with a caregiver for supportive care services. While, intensity of nursing services did not vary across regions, intensity of supportive care services was significantly ii

3 different across the province. Factors affecting wait times included sex, functional status, health status, and receiving nursing care. Conclusions: Inequity in access to home care services persists across Ontario. Variation in propensity to receive services, intensity of supportive care services received, and wait times was evident across regions. Inequitable access to services could result in negative outcomes for those unable to access services. Equitable access to home-based care is vital to ensuring Ontario seniors are able to live and receive care in their preferred location, which for most seniors is at home. iii

4 Acknowledgements I would like to express my sincere gratitude and appreciation to all of those who helped and inspired me over the last five years. With out the help and support of many, I would not have been able to complete my dissertation. To my supervisor Dr. Ann Tourangeau, I cannot thank you enough for your ongoing support and guidance. Throughout the program you provided me countless opportunities to expand my knowledge and skills related to research and for that I cannot thank you enough. Thank you for always challenging me. Because of your mentorship I have truly grown as an individual and researcher during my time in the program, thank you! I must also acknowledge the other members of my committee, Drs. Audrey Laporte and Kathy McGilton. To Dr. Laporte, thank you for helping me to view health and healthcare from a different perspective. You have been influential in broadening the way I think about health research and encouraged me to expand my knowledge and skills related to statistical methods, thank you! To Dr. McGilton, thank you for your insight and encouragement throughout the research process. Your thoughtful comments and feedback were invaluable and instrumental in helping me to better understand my research findings, thank you! I must also thank my friends in the program, Margaret Saari and Heather Thompson. Heather, I would not have made it though first year without you. Thank you for showing me the ropes and encouraging me to keep going! Margaret, thank you for your ongoing support and encouragement. Thank you for pushing me to work harder, to learn more, and to think outside of the box. I cannot thank you enough for your support over these last four years, thank you! Finally, I would like to thank my family for their support over the last five years. In particular, I must express my most sincere and heartfelt gratitude to my Husband, Andrew, who supported me in so many ways throughout the program. I truly could not have not it without you Andrew, thank you! I must also express profound gratitude to my Mom, Nancy who has been influential in so many ways. Mom, I cannot express how thankful I am for your love and support, I could not have done this without you, thank you! iv

5 Table of Contents Chapter 1: Background... 1 Seniors and Ontario s Health Care System... 1 Aging-in-Place... 2 Accessing Home Care in Ontario... 3 Determining Eligibility... 3 Variation in Access to Home Care... 7 Study Purpose... 7 Chapter 2: Literature Review... 8 Defining Access... 8 Service Availability... 8 Utilization... 8 Barriers to Access... 8 Relevance and Effectiveness... 9 Equity and Access... 9 Literature Review Methods Personal Characteristics Informal Caregiver Characteristics Region and Regional Characteristics Agency Characteristics Health System Characteristics Limitations of Current Evidence The Problem Statement and Research Questions Hypothesized Model Chapter 3: Methods Research Design Data Sources Data Quality Measures Access Personal Characteristics Caregiver Characteristics Regional Characteristics Agency Characteristics Health System Characteristics Study Population and Sample Inclusion Criteria Data Analysis v

6 Sample Description Propensity Models: Logistic / Probit Regression Intensity Models: Selection Modeling Wait Time Model: Selection Modeling Ethical Issues Risks and Benefits Chapter 4: Results Descriptive Findings and Tests of Difference Sample Demographic Characteristics Personal Characteristics Caregiver Characteristics Health System Characteristics Regression Model Results Propensity to Receive Home Care Services Intensity of Home Care Services Received Wait Time to First Home Care Service Visit: Selection Model Summary of Results Research Question One Research Question Two Research Question Three Research Question Four Research Question Five Research Question Six Research Question Seven Chapter 5: Discussion, Implications, and Conclusions Propensity to Receive Home care Services: Discussion and Interpretation of Results Intensity of Home care Services Received: Interpretation and Discussion of Results Wait Time to First Home Care Service Visit: Interpretation and Discussion of Results Variation in Access to Home Care Services: Discussion and Interpretation of Results Study Limitations Study Implications, Areas for Future Research, and Recommendations Conclusion References Appendices vi

7 List of Tables Table 1. Table 2a. Table 2b. Table 2c. Table 2d. Table 2e. Table 2f. Table 2g. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. Table 10. Table 11. Table 12. Table 13. Table 14. Table 15. Table 16. Study Concepts, Variables, and Data Sources Concept Measurement: Access Variables Concept Measurement: Demographic Variables Concept Measurement: Cognitive Status and Functional Status Variables Concept Measurement: Health Status and Continence Variables Concept Measurement: Informal Help and Previous Health Care Utilization Variables Concept Measurement: Caregiver Characteristic Variables Concept Measurement: Regional Characteristic Variables Descriptive Statistics and Tests of Difference: Sample Demographics Descriptive Statistics and Tests of Difference: Access Variables Descriptive Statistics and Tests of Difference: Cognitive Status and Functional Status Descriptive Statistics and Tests of Difference: Health Status and Previous Health Care Utilization Descriptive Statistics and Tests of Difference: Informal Help and Caregiver Characteristics Descriptive Statistics and Tests of Difference: Regional Characteristics Propensity to Receive Home Care Services: The Results of the Multinomial Probit Model Intensity of Nursing Services Received (Nursing Services Alone): Results of a Selection Model Intensity of Nursing Services Received (Nursing and Supportive Care Services): Results of a Selection Model Intensity of Supportive Care Services Received (Supportive Care Services Alone): Results of a Selection Model Intensity of Supportive Care Services Received (Nursing and Supportive Care Services): Results of a Selection Model Wait Time to First Home Care Service Visit: Results of a Selection Model Propensity to Receive Home Care: Model Summary Intensity of Home Care Services Received: Model Comparison vii

8 List of Figures Figure 1: Figure 2: CCAC Population Based Model of Care Hypothesized Model of Factors Affecting Access to Long-term Home Care viii

9 List of Appendices Appendix A: Appendix B: Propensity to Receive Home Care Services: Full Regression Model Results Propensity to Receive Home Care Services: Marginal Effects and 95% Confidence Intervals ix

10 Chapter 1: Background The purpose of this research study was to examine variation in access to long-term home care services for Ontario seniors. As Canada s population ages, a greater number of seniors will require assistance from both formal and informal care providers to remain in their homes and out of long-term care facilities. Understanding which factors predict access to home care, the amount of home care services received, and the wait time for home care services can highlight sources of variation in access to formal home care. Understanding how access varies can inform strategies to address variation, resulting in more equitable access to long-term home care for seniors in need across the Province of Ontario. In this first chapter, information related to seniors and health care in Ontario is presented along with background to the importance of the study. Next, the processes for gaining access to home care are discussed, and the issue of variation in access is explored. Finally the study purpose is outlined. Seniors and Ontario s Health Care System Canada s population is aging. Canadians age 65 and over represent the fastest growing age group, a trend that is likely to continue as a result of lower birth rates, longer life expectancy, and the aging baby boomer generation (Human Resources and Skill Development Canada, 2012). Utilization of health care services increases dramatically with advanced age. Moreover, older Canadians tend to use more health care resources across health care sectors than their younger counterparts (Nie, Wang, Tracey, Moineddin, & Upshur, 2007). In addition, seniors are the most frequent users of long-term care services provided in both community and institutional settings. The percentages of home care recipients and long-term care (LTC) residents age 65 and over are 82 and 95 percent respectively (Canadian Institute for Health Information [CIHI], 2011a). In the Province of Ontario, Canada, institutional LTC accounts for 7.5 percent of the provincial health care budget. Ontario spends approximately 3.4 billion dollars annually on LTC homes (LTC Innovation Expert Panel, 2012), an amount that will continue to grow as the population ages. It is estimated that Ontario will require an additional 9,465 LTC beds by 2016 at a cost of approximately 1.15 billion dollars. Subsequently, an additional 454 million dollars annually will be required in operating costs (LTC Innovation Expert Panel, 2012). The 19,000 Ontarians awaiting placement in institutional LTC facilities (nursing homes) demonstrates a need for greater LTC capacity across the province, with many of these individuals waiting for placement in acute care hospital beds (LTC Innovation Expert Panel, 2012). Patients who could be more appropriately cared for in other settings occupy 14 percent of acute care hospital beds across the province. This situation negatively affects Ontario s capacity to provide timely acute care hospital services (LTC Innovation Expert Panel, 2012). Community based LTC offers an alternative to facility based LTC for seniors who are not able to live independently. For a portion of seniors facility based LTC is the most appropriate location of care. However, evidence suggests that many seniors waiting for LTC placement could be adequately cared for in their homes at a lower cost to the health care system given the availability of appropriate informal and formal supports (Kuluski, Williams, Berta, & Laporte, 2012; Williams et al., 2009). Canadian researchers 1

11 concluded that between 30 and 50 percent of seniors waiting for LTC placement in the Toronto area could be adequately and cost-effectively cared for in the community (Williams et al., 2009). In a similar study conducted in Thunder Bay and surrounding areas, it was found that while eight percent of individuals in urban Thunder Bay could be diverted from LTC to community-based care, the potential diversion rate for seniors across more rural areas of the region was 50 percent (Kuluski et al., 2012). These findings indicate that availability of resources in the community to care for seniors requiring support to remain in their homes is an important factor driving admissions to LTC (Kuluski et al., 2012). This finding supports the notion that, given availability of adequate formal and informal supports in the community, care provided in individuals homes could substitute for placement in LTC for a portion of seniors. The fact that this does not occur to a greater degree raises questions about what barriers might be at work against more seniors receiving care in their homes. Aging-in-Place Aging-in-place, a principle adopted by many provincial health ministries, supports enabling seniors to remain safely in their homes and communities despite diminished health and functioning. This principle also recognizes the desire of most Canadian seniors to remain living in their homes for as long as possible (Health Council of Canada, 2012). Ontario s Provincial Government has recognized the potential benefits of supporting seniors to age in the community through the Aging at Home Strategy and the Home First Philosophy. The Aging at Home Strategy, announced in 2007, was focused on maintaining seniors in their homes by providing innovative programs geared toward enhancing healthy aging and independence. Various programs funded through the Aging at Home Strategy were implemented at the level of the Local Health Integration Networks (LHIN) and target needs of seniors locally (Hamilton Niagara Haldimand Brant Local Health Integration Network, 2008). A total of 1.1 billion dollars over four years was invested in seniors care through aging at home programs. The Home First Philosophy, a transition management philosophy devised by the LHINs in 2011, aims to support seniors to return home following discharge from acute care hospitals (Health Council of Canada, 2012). The goal of the Home First Philosophy is to more effectively plan for hospital discharge by engaging patients and family members in decision making related to post-acute care provision. The ultimate goal of this transition management philosophy is to transition elderly patients back to community settings and to avoid new admissions to residential LTC (Health Council of Canada, 2012). For many seniors, being able to age-in-place requires access to adequate informal and formal support. Informal support is integral to keeping seniors in their homes (Williams et al., 2009) and is most often provided by a spouse, child, or other family member. Formal support is sometimes required when informal caregivers are unable to provide certain aspects of care or when caregivers are not able to fully cope with care provision. Personal support workers (PSWs) and nurses most often provide in-home formal care services. In 2008, Ontario s Ministry of Health and Long-Term Care (MOHLTC) recognized the importance of adequate formal support for Ontario seniors and their caregivers by increasing the number of home support and home nursing hours available to eligible community residents (O. Reg. 2

12 164/08). Approximately 160,000 Ontarians were assessed for long-term home care services in fiscal year (CIHI, 2013). Accessing Home Care in Ontario In Ontario, publicly funded home care is accessed through 14 Community Care Access Centers (CCACs). The CCACs are the single point of access for publicly funded home care and residential LTC in the province. Care Coordinators, employed by CCACs, assess Ontario seniors identified as potentially needing formal home care or LTC placement. The role of the Care Coordinator is to determine eligibility for and frequency of home care services based on health need within a fixed budget and funding parameters of the CCAC. Care Coordinators do not provide direct patient care; rather, PSWs, nurses, and allied health professionals employed by private for-profit and not-for-profit home care service provider organizations deliver care to patients in their homes. These organizations hold contracts with the CCACs to provide specific amounts and types of care in specific geographic areas. Ontario residents requiring home care services are identified as either short-stay patients or long-stay patients. Short-stay patients have an expected length of home care service less than 60 days, while long-stay or long-term patients are expected to require home care services for greater than 60 days. The focus of this study was the population of Ontario seniors identified as requiring long-term home care services. Determining Eligibility Eligibility for home care services is determined by CCACs based on criteria outlined in provincial regulation, information provided in the home care referral, and an initial assessment usually conducted in hospital or over the telephone. According to the Health Insurance Act individuals are deemed eligible for nursing or other professional services if: 1. The services required are insured services and the individual in need of services is insured under the Ontario Health Insurance Plan (OHIP); and 2. The following conditions are met: a) the services are necessary to maintain the individual in his or her home; b) the individual s needs cannot be met through out-patient care; c) the individual requires at least one professional service (i.e., nursing, physiotherapy, occupational therapy, etc.); d) the services are provided in the individual s home which has been deemed a suitable environment for care provision; e) the services are available in the area in which the individual resides; and f) the services are expected to result in progress towards rehabilitation. According to the Home Care and Community Services Act, 1994, an individual is eligible for personal support services if: 1. the individual in need of services and is insured under OHIP; and 2. the location of care has been deemed a suitable environment for care provision. An individual is eligible for homemaking services if: 1. the individual requires personal support services in addition to homemaking; 3

13 2. the individual receives personal support services and homemaking services from an informal caregiver who requires assistance with care provision to continue providing required personal care; or 3. the individual requires constant supervision and the caregiver requires assistance with homemaking tasks. To assist in determining eligibility, the Care Coordinator assesses each patient using the Resident Assessment Instrument Contact Assessment (RAI-CA). The RAI-CA is a shortened version of the more lengthy RAI assessment tools and is used to support clinical decision-making prior to completion of the RAI-Home Care (RAI-HC). The RAI-CA is most often completed over the phone within 72 hours of referral. The RAI-CA helps Care Coordinators categorize patients into one of five categories ranging from well to complex. All five categories are included in Figure 1, which depicts the CCACs client care model. Figure 1. CCAC Population Based Model of Care. Note. Based on Care Coordinator assessment, individuals are categorized into a patient population based on their health condition, level of independence, risk of requiring acute care, and socio-economic factors. Individuals can move between populations when care needs change and an acute care episode can occur at any time across populations. Reprinted with the permission of the Office of the Auditor General of Ontario from CCACs Community Care Access Centres Home Care Program by the Office of the Auditor General of Ontario, 2015, Annual Report 2015, Copyright 2015 by the Queen s Printer for Ontario. If the initial assessment results in the individual being categorized as complex, chronic or community independent, the Care Coordinator must completed a RAI-HC assessment within 14 days of the RAI-CA. The RAI-HC must be completed in-person at the individual s home or place of residence and is used to develop the client care plan. 4

14 As of 2002, Care Coordinators have assessed Ontario seniors in need of long-term formal support or LTC placement using the RAI-HC. Outcome scales derivable from the RAI-HC include Activities of Daily Living (ADL) Hierarchy; Instrumental Activities of Daily Living (IADL) Scale; Cognitive Performance Scale (CPS); Depression Rating Scale; Pain Scale; Pressure Ulcer Risk Scale; and the Changes in Health, End-stage Disease, and Symptoms and Signs (CHESS) Score (CIHI, 2012a). Using RAI-HC data, assessed seniors are assigned a MAPLe or Method for Assigning Priority Level score (Hirdes, Poss, & Curtin-Telegdi, 2008). The MAPLe is based on a variety of clinical variables in the RAI- HC and reflects the individual s level of risk for admission to LTC. The MAPLe is comprised of five levels: low, mild, moderate, high, and very high (Hirdes et al., 2008). Seniors assessed as having a low MAPLe score are generally independent, have no physical limitations and only mild cognitive impairment without behavioural problems. For this group, home support may be needed to provide assistance to cope with mild loss of cognitive function (Health Council of Canada, 2012). Seniors assessed as having a mild MAPLe score experience some challenges with instrumental activities of daily living (IADLs) / activities of daily living (ADLs) such as housework, bathing, and transportation. These patients may require assistance to complete IADLs / ADLs (Health Council of Canada, 2012). Seniors assessed as having a moderate MAPLe score show increased physical and / or cognitive impairment that threatens their ability to live independently. This group of patients may have difficulty managing medications, housework, and meal preparation, and require an increased level of care (Health Council of Canada, 2012). Those assessed as having a high MAPLe score experience more complex health problems and often exhibit limited physical functioning and / or more severe cognitive impairment. These patients are at increased risk of LTC placement and their caregivers are more likely to be experiencing distress (Health Council of Canada, 2012). Finally, those assessed as having a very high MAPLe score experience severe physical and / or cognitive impairment and are no longer able to live independently in their homes. This group of clients often experience falls, behavioural issues, and wandering. These patients are at highest risk of LTC placement and caregiver distress (Health Council of Canada, 2012). In the fiscal year, 158,836 community dwelling Ontarians were assessed using the RAI-HC (CIHI 2013). This group of Ontario residents was, on average, 77 years of age and predominantly female. The majority of those assessed, 58.7 percent, were identified as having a low to moderate MAPLe score, while 41.3 percent had a high or very high MAPLe score. Approximately 70 percent of those assessed received home support services, while 43.8 percent received home health services such as nursing, physical therapy, and occupational therapy (CIHI, 2013). Care Coordinators use the RAI-HC to help determine the intensity of supportive care services required for long-stay home care patients (CIHI, 2012a; Hirdes et al., 2008). In addition to the RAI-HC, decisions around intensity of home care services are based on Care Coordinators clinical judgment and the availability of resources within the region. As such, criteria used to determine eligibility for publicly funded home care services may vary across the province. Generally, seniors at higher levels of risk are given priority for home care services (Hirdes et al., 2008). 5

15 To help determine the intensity or amount of personal support services required by individual patients, the CCACs have also developed a scoring method, the RAI score (The Office of the Ontario Auditor General, 2015). The RAI score is the sum of the MAPLe, ADL Hierarchy, CPS, CHESS scale, and the IADL capacity / difficulty scales (C-L., Sinn, personal communication, September 18, 2015). The score ranges from 0 to 28 with higher scores indicating greater need. This scoring method is not endorsed by interrai (The Office of the Ontario Auditor General, 2015). The use of the RAI score is problematic for several reasons. First, the ADL Hierarchy and CPS are included in both the MAPLe and RAI scores; as such, components of these scales are over-represented as the RAI score also contains the MAPLe. Second, components of the RAI score are conceptually different bringing into question the construct validity of this scoring method. While the MAPLe predicts placement in institutional LTC, the CHESS is negatively correlated with institutionalization. Finally, the meaning of the RAI score is not defined. For example what is common among individuals with a RAI score of 10 and how does a score of 10 differ from a score of 15? Despite these identified deficiencies, CCACs are currently using the RAI score to determine priority for and intensity of personal support services (C-L., Sinn, personal communication, September 18, 2015). However, the cutoff score used to determine who gains access to home care services is not consistent across the province. While an individual with a RAI score of 7 may receive services in one CCAC, they may not receive services in another (The Office of the Ontario Auditor General, 2015). Although research examining care coordinator decision-making is limited, research evidence supports the notion that care coordinator home care service allocation decisions vary across patients (Fraser, & Estabrooks, 2008). Through a review of the literature, Fraser and Estabrooks (2008) identified four categories of factors affecting care coordinator allocation decisions including: Care Coordinatorrelated factors, patient-related factors, information-related factors, and system / program-related factors. Patient characteristics such as cognitive status, presence of an informal care provider, and patient preference were identified as more influential than other types of factors in Care Coordinator allocation decisions (Fraser & Estabrooks, 2008). Care Coordinator-related factors included education, gender, being a licensed social worker, and intake specialization. Discretionary practices among and between Care Coordinators were also evident as Care Coordinators applied rules and policies differently across patients when making decisions around the allocation of home care resources (Fraser & Estabrooks, 2008). A recent Ontario based study also found that Care Coordinator-related factors affected allocation decisions (Kohli, 2009). Kohli identified Care Coordinator values of safety, independence, and patient focused care as influential in care coordinator priority setting. In addition, the location of the Care Coordinator caseload (rural or urban), their years of experience in home care, and having recently experienced providing informal care were factors identified as affecting CCAC Care Coordinator allocation decisions (Kohli, 2009). While understanding Care Coordinator decision-making related to the allocation of home care services is important, it is complex and influenced by numerous factors (Fraser & Estabrooks, 2008). As such, examining allocation of home care at the Care Coordinator level was beyond the scope of this study. Rather, this study focused on variation in access to home care services at the CCAC or regional level. 6

16 Variation in Access to Home Care Historically, variation in access to home care services has existed across the Province of Ontario. Using data collected between , before the restructuring of Ontario s home care system and the creation of the CCACs, Coyte and Young (1999) noted regional differences in the rates of home care services received across the province. Issues related to inequity in access to home care were factors leading to the creation of CCACs and the implementation of market-modeled home care in Ontario (Denton, Zeytinoglu, Kusch, & Davies, 2007). However, following the introduction of CCACs inequity in access persisted. Laporte et al. (2007) found that independent of health care need, the region or CCAC in which an individual lives was a significant predictor of the probability an individual received home care services. Further, again controlling for health need, the region or CCAC also predicted the intensity of home care services received among long-term home care patients. Unfortunately, no research was located examining whether or not the introduction of CCACs improved equity in access across Ontario. More recently, variations in access to home care services for seniors across CCACs have been observed. A 2010 report by Ontario s Auditor General found this variation to be especially visible among seniors assessed as having moderate-risk for institutionalization (The Office of the Ontario Auditor General, 2010). Understanding factors affecting access to home care services can uncover sources of variation in access for Ontario seniors. Further, understanding how access varies can inform strategies to address variation, resulting in more equitable access to long-term home care for seniors in need across the province. Study Purpose Access to publicly funded homecare is determined by CCAC Care Coordinators and is based on four components: provincial regulation, the outcomes of the standardized RAI-HC (including the MAPLe), the Care Coordinators clinical judgment, and the availability of resources within the region. As such, criteria used to determine eligibility for publicly funded home care services may vary across the province. The purpose of this study was to examine factors affecting access to long-term home care services for Ontario seniors (age 65 and over) identified as at moderate risk of adverse outcomes and to determine whether variation in access to long-term home care services exists across the province. 7

17 Chapter 2: Literature Review In this chapter, a review of relevant literature is presented. First, the concept of access is discussed and defined. Next, methods used to search for relevant literature are described and factors determining access to home care services are identified. Finally, the research problem is described, research questions are proposed, and a hypothesized model of factors affecting access to long-term home care for Ontario seniors is presented. Defining Access Access is a multidimensional concept. Simply stated, access refers to an individual s ability to obtain needed health care services (Khan & Bhardwaj, 1994). Aday and Andersen (1981) define access as those dimensions, which describe the potential and actual entry of a given population group to the health care delivery system (p.6). While Gulzar (1999) defines access to health care as the fit among personal, socio-cultural, economic, and system-related factors that enable individuals, families, and communities to have timely, needed, necessary, continuous, and satisfactory health care services (p. 17). This second definition more accurately reflects the multiple dimensions of access discussed in the literature. Gulliford et al. (2002) identified four dimensions of access including 1) service availability; 2) utilization of services and barriers to access; 3) relevance, effectiveness and access; and 4) equity and access. These dimensions will be used to guide the following discussion around access to health care. Service Availability The dimension, service availability, reflects the notion that access requires the availability of an adequate supply of health care services (Gulliford et al., 2002). The availability of services has also been referred to as potential access (Khan & Bhardwaj, 1994). Penchansky and Thomas (1981) identified a similar dimension availability, which reflects the relationship between the existing supply of and demand for health care services. Utilization Utilization of services, an indicator of access, reflects the notion that while services may be available, individuals often experience difficulties utilizing needed services (Gulliford et al., 2002). Donabedian (1972) described the importance of actual utilization when evaluating access, noting, the proof of access is use of services, not simply the presence of a facility (p. 111). Utilization is also termed realized access, which occurs when barriers to access are overcome and entry into the health care system is gained (Khan & Bhardwaj, 1994). Barriers to Access Barriers to access are frequently described as a means of conceptualizing access to health care. Several dimensions of access identified by Penchansky and Thomas (1981) reflect barriers to access including accessibility, accommodation, affordability, and acceptability. Accessibility refers to the location of health care services and the ability of the individual to access those services (Penchansky & Thomas, 1981). Similar to accessibility, Khan and Bhardwaj (1994) identified geographic location, or spatial access, as a potential barrier to service utilization for some populations. Accommodation refers to the relationship between the way in which health care services are organized and delivered and the service users ability to utilize services, given these factors (Penchansky & Thomas, 1981). For example, 8

18 if a home care patient requires treatment in the evening, but the home care provider only operates between the hours of 9:00 a.m. and 5:00 p.m. the organization of care is a barrier to access for this patient. Affordability refers to the relationship between the cost of services and the service users ability to pay for those services (Penchansky & Thomas, 1981). Acceptability refers to the suitability of the service providers personal and professional characteristics to the service user, as well as the suitability of the service users personal characteristics to the service provider (Penchansky & Thomas, 1981). Similarly, Gulliford et al. (2002) identified three categories of barriers to access: personal barriers, financial barriers and organizational barriers. Personal barriers reflect the individuals ability or willingness to recognize their need for health care services, as well as their attitudes and beliefs about health care resulting from previous experiences with the health care system. These personal barriers are not unlike the dimension of acceptability identified by Penchansky and Thomas (1981). Financial barriers, similar to the dimension of affordability (Penchansky & Thomas, 1981), relate to an individuals ability to cope with the costs of health care (Gulliford et al., 2002). Regardless of whether the system user incurs the cost of care, financial barriers my still exist (e.g., travel costs, opportunity costs from time lost working, etc.). Organizational barriers reflect systemic factors that impact access to care, for example wait lists for services and variations in referral practices (Gulliford et al., 2002). These organizational barriers are not unlike those of accessibility and accommodation described by Penchansky and Thomas (1981). Relevance and Effectiveness Gulliford et al., (2002) also identified Relevance and Effectiveness as an important dimension of access. This dimension relates to the need for the right health care service, provided at the right time, and in the right location. Relevance and effectiveness also encompasses the notion of best possible health outcomes as an indicator of access (Gulliford et al., 2002). Gulzar (1999) also identified health outcomes as relevant to the concept of access to health care. Effective access is achieved when use of health care services is demonstrated to improve patients health status (Gulzar, 1999). Equity and Access Equity is an important dimension of access reflecting fairness in the distribution of and access to health care services (Gulliford et al., 2002). According to Aday and Andersen (1981) equity exists when health care services are distributed based on the needs of the population. In contrast, inequity exists when the distribution of health care services is based on factors other than need such as race, socioeconomic status or geography (Aday & Andersen, 1981). Equity can be further described as horizontal equity or fairness in access to health care services for populations with equivalent need (Gulliford et al., 2002), and vertical equity. Vertical equity refers to the idea that populations with different health care needs (e.g., acute care patients as compared to patients requiring home care services) receive the same access to appropriate services to address their needs (Gulliford et al., 2002). In this research study, three dimensions of access were examined including propensity to receive services, intensity of services received, and wait-time to first home care provider visit. Compared to those definitions presented above, the conceptualization of access used in this study was simplistic. Propensity reflects whether or not an individual is granted access to publicly funded home care services, 9

19 intensity reflects the amount of services an individual receives, and wait time to first home care service visit reflects the number of days spent waiting to receive home care services after gaining access to the home care system. Propensity and intensity relate to utilization, a dimension of access descried by Donabedian (1972), Gulliford et al., (2002), and Khan and Bhardwaj, (1994). Wait time to first home care service visit is an organizational barrier to accessing home care services described by Gulliford et al., (2002), reflecting the concept of accessibility described by Penchansky and Thomas (1981). Many of the other dimensions of access described above are examined in this study as factors influencing access to home care. These factors are described further in the literature review and relate to the following dimensions of access: personal and organizational barriers to access as well as equity. Literature Review Methods A thorough understanding of the concept of access helped to inform this literature review. The dimensions availability and utilization informed the selection of key search terms. To identify relevant literature, several databases were accessed including the Cumulative Index for Nursing and Allied Health Literature (CIHAHL), Scopus, Web of Science, Pubmed, Medline, and Econlit. Search terms included home health care, professional home nursing, home rehabilitation, home visits, home maker services, home nursing, long-term care, home healthcare, home visiting, and home support alone and in combination with access*, health services accessibility, availability, eligibility, and utilization. The search was limited to English language articles published after For this literature search, a specific population of home care patients was not selected as it was anticipated that research related to access / utilization of home care services would be limited. In addition, the reference lists of key publications were reviewed, resulting in the identification of additional publications. Across databases, a total of 1,681 articles were identified, a further 23 articles were identified through hand searching reference lists. After reviewing the titles and abstracts of each article, 132 articles were selected for full-text review. Articles were eliminated if (a) the findings of research were not reported (e.g. policy discussions, etc.), (b) the outcome of interest was not access to or utilization of formal home care services, or (c) the data used in the analysis were collected prior to Ultimately, 39 articles were selected for inclusion in the review. The majority of studies were conducted in the United States (n = 19). Seven studies were conducted in Canada and four in Sweden. One study was conducted in each of Australia, England, Japan, and South Korea. Four studies were conducted using international data from multiple countries. The majority of studies used secondary data (n = 24). Two studies were literature reviews. In 23 studies, participants were seniors 65 years of age and older, the other studies included home care recipients younger than 65, Case Mangers / Care Coordinators, and informal caregivers. The results of the literature review are organized according to categories of predictors identified in the literature as influencing access to or utilization of home care services. These categories include personal characteristics, informal caregiver characteristics, regional characteristics, home care provider agency characteristics, and health system characteristics. 10

20 Personal Characteristics Personal characteristics include factors related to the individual receiving home care services, such as age, sex, socioeconomic status, marital status, and education as well as factors reflecting the individual s health and need for care. The factors described below were identified in the literature as predictors of either the propensity to receive home care services, the intensity of services received, or both propensity and intensity. Age Age has consistently been found to predict access to / utilization of formal home care services. Many studies have reported a positive relationship between age and home care service use controlling for health status, with increasing age predicting greater use of services (Allan, Funk, Reid, & Colutier- Fisher, 2011; Balia & Brau, 2014; Blomgren, Martikainen, Martelin, & Koskinen, 2008; Crocker Houde, 1998; Hawranik & Strain, 2001; Lagergren, 1994; Laporte et al., 2007; Larsson, Thorslund, & Forsell, 2004; Lehning, Kin, & Dunkle, 2013; McAuley, Spector & Nostrand, 2009; Meinow, Kareholt, & Lagergren, 2005; Mery, Wodchis, & Laporte, 2015; Rodriguez, 2013; Stoddart, Whitely, Harvey, & Sharp, 2002; Torrez, Estes, & Linkens, 1998). Laporte et al., (2007) found increasing age to predict both propensity to receive services and intensity of services received among Ontario residents covered under the Ontario Health Insurance Plan (OHIP). Further, women over the age of 65 and men over the age of 75 were more likely to receive long-term home care services than their younger counterparts (Laporte et al., 2007). In a synthesis of the literature (Kadushin, 2004), the majority of identified studies reported increasing age to predict greater use of a variety of home care services including nursing, home support, mobile meals, physical therapy, occupational therapy, and social work (Kadushin, 2004). In contrast, one Canadian study reported a negative relationship between nursing intensity and age. Masucci, Guerriere, Sogurski & Coyte (2013) found those aged less than 60 years received significantly more nursing visits than those aged 62 to 82. One possible explanation for these contrasting findings is the study sample. Masucci et al. (2013) included only palliative patients. No other study included in this review focused exclusively on the palliative patient population. Overall, these findings suggest increasing age predicts access to home care services for non-palliative home care patients. Sex There was little research that found sex to be a predictor of access to home care services. Previous research has found sex to predict both propensity (Balia & Brau, 2014; Laporte et al., 2007; Rodriguez, 2013) and intensity of home care service use (Laporte et al., 2007). Being female predicted propensity to receive services, while being male predicted greater intensity of services received. Larsson and Thorslund (2002) also examined sex as a predictor of access to home care. Although sex initially predicted the probability of receiving services, the addition of living arrangement (lives alone) resulted in sex no longer predicting home care service use. Laporte et al. (2007) did not include living arrangement in their model. The findings of Larsson and Thorslund (2002) suggest living arrangement may act as a mediator in the relationship between sex and propensity to receive home care services. Indeed, Mery et al. (2015) found the inclusion of an interaction between sex (male) and living with a partner resulted in a significant relationship between sex (being male) and propensity to receive home support services. 11

21 Males residing without their partner were significantly more likely to receive home support services while males residing with their partner were significantly less likely to receive home support services. Although evidence supporting sex as a predictor of propensity to receive home care services is inconclusive (Kadushin, 2004; Larsson & Thorslund, 2002), evidence does support including sex as a predictor of intensity of services received (Laporte et al., 2007). Race Race as a predictor of access to home care services was primarily evaluated in studies based in the United States (Goins & Hobbs, 2001; Kirby & Lau, 2010; Li, 2006; McAuley et al., 2009; Torrez et al., 1998). Being non-white was associated with experiencing barriers to accessing home care services (Li, 2006). Being white was associated with greater use of home care services (McAuley et al., 2009) and more agency certified visits (Torrez et al., 1998). Being African American was associated with fewer home care visits (Crocker Houde, 1998). In contrast, Goins and Hobbs (2001) found that being non-white was associated with greater use of home care services among elderly home care clients. Evidence related to race as a predictor of access to home care services is contradictory. A review of the literature (Kadushin, 2004) revealed that only 31 percent of studies included in the review found race to be a significant predictor of propensity to receive services and only 20 percent found race to be a significant predictor of intensity of services received. Race was not included as a factor affecting access to home care in Canadian studies as data on race are generally not collected in Canada. For this reason, race is not included in the models tested in this study. Marital Status Marital status has been found to predict access to home care services. Those who never married, were widowed or divorced, or who were married but whose spouse is unable to provide care were more likely to receive home care services than those who were married (Corazzini-Gomez, 2002; Lagergren, 1994; McAuley, et al., 2009; Stoddart et al., 2002). However, in a review of the literature Kadushin (2004) reported that support for marital status as a predictor of access to home care services, was uncertain. In Kadushin s (2004) review of the literature, approximately half of all studies included were published prior to 1993, and as such were not included in this review. The inclusion of different studies may explain this contradictory finding. Socioeconomic Status Socioeconomic status (SES) or income level has been examined as a predictor of access to home care services. Several studies found low SES to be associated with greater propensity to receive services and greater intensity of services received (Laporte et al., 2007; Masucci et al., 2013; Mery et al., 2015; Rodriguea 2013). This finding is supported by a recent review of the literature that found persons with low SES were more likely to use home care services even after health status was adjusted for (Goodridge, Hawranik, Duncan, & Turner, 2012). However, a second, earlier review of the literature found the evidence around SES as a predictor of utilization of home care services to be less certain (Kadushin, 2004). In addition, in one American study, lower income (less than $25,000/year) was associated with experiencing barriers to accessing home care services (Li, 2006). Similarly, Balia and Brau (2014) found higher income to be associated with greater intensity of nursing services. The authors 12

22 postulate the positive relationship between income and nursing intensity that resulted from their analysis may be due to the possible inclusion of privately purchased nursing services in the nursing intensity variable used (Balia & Brau, 2014). When considering only those studies examining publicly funded home care services, the literature seems to support a positive relationship between low income and access to home care services Education Blomgren et al. (2008), Larsson et al. (2004), and Stoddart et al. (2002) examined education in relation to propensity to receive services and reported persons with lower levels or fewer years of education to be more likely to use publicly funded home care services. In contrast, Balia and Brau (2014) examined the relationship between education and intensity of services received and found higher levels of education to be related to increased intensity of nursing services. These findings suggest that it is possible that education affects the various dimensions of access (propensity and intensity) differently. Health Status Health status reflects the individuals overall physical and mental health. Overall health status was identified as a significant predictor of propensity to receive home care services and intensity of services received among Ontario residents. Poor health status was associated with greater use of home care services (Laporte, et al., 2007). Many factors comprising health status also predicted access to home care services when examined individually. The number of chronic conditions an individual suffered from was positively related to use of long-term home care services (Allan et al., 2011; Balia & Brau, 2014; Lehning, et al., 2013; Rodriguez, 2013). Decreasing emotional health was also found to be related to increased use of home care services (Stoddart et al., 2002). Additionally, several specific illnesses predicted utilization of home care services including arthritis, heart disease, diabetes (McAuley et al., 2009), stroke, cancer (Mery et al., 2015), respiratory disease (Noguchi & Shimizutani, 2009), and poor foot health (Stoddart et al., 2002). Individuals living with these conditions were more likely to receive formal home care services (McAuley et al., 2009; Noguchi & Shimizutani, 2009; Stoddart et al., 2002). Further, an individual s prognosis has also been found to predict eligibility for home care, such that the worse the prognosis the more likely the individual was to be eligible for home care services (Corazzini- Gomez, 2002). In addition, in two studies, the impact of visual impairment on utilization of home care services was examined. Worsening eyesight was found to predict greater use of publicly funded home care services (Stoddart et al., 2009; Tay et al., 2007). Balia and Brau, 2014 and Lehning et al. (2013) found the impact of health on moderate activities to predict utilization of community based services, suggesting that perhaps it is not health status alone, but the impact of health on functional status that necessitates utilization of long-term home care services. Functional status Functional status reflects an individual s ability to carry out daily tasks such as mobilization, bathing, toileting, meal preparation, shopping, and banking. Functional status consistently predicted utilization of / access to long-term home care services. Functional status has been defined and examined in various ways including functional disability (Larergren, 1994; Masucci et al., 2013; Mery et al., 2015), mobility limitations (Balia & Brau, 2014; Blomgren et al., 2008; Corazzini-Gomez, 2002; Goins & Hobbs, 13