Specialized Geriatric Services Use by Older Home Care Clients

Abstract

Background: Geriatric medicine specialists are experts in the care of vulnerable, older adults with complex medical and psychosocial needs. In Canada, their scope of practice and expertise is communicated through the 5M Framework, and includes the following domains: mind, mobility, multicomplexity, medications, and what matters most to patients. Specialized geriatric resources are limited in Ontario, with an estimated shortage of the full time equivalent of over 100 geriatricians. Beyond simply a lack of resources, there is poor integration, communication, and collaboration between specialized geriatric services (SGS), and other community-based health services, such as home care services, primary care, and outpatient specialist care. The Regional Geriatric Programs of Ontario (RGPO) and the Ontario Ministry of Health and Long-term Care (MOHLTC) expressed an interest in developing a mechanism to assist in the allocation of limited resources by targeting older adults who would most benefit and improving integration between these care sectors.

To date, however, there has been little literature published on the practice patterns of geriatric medicine and the determinants of contact with this specialist discipline by community-dwelling older adults. Geriatric organizations and associations in Canada and the United States have published statements on the role and target population of geriatric medicine but there is a lack of empirical evidence on this topic. Standardized assessments in home care provide a wealth of health information that may be linked to service use data to investigate this topic. As older home care clients represent a complex and high needs subset of the general population of community-dwelling older adults, this is an appropriate population to study and target for SGS. While there is interest in examining the broader use of community-based SGS, it is not captured in administrative services data. However, it is possible to examine contact with geriatric medicine as a component of specialized geriatric care.

Objectives: The objectives of this dissertation are: 1) to investigate patterns of health care services use by older, home care clients in Ontario, with a focus on contact with geriatric medicine; 2) to identify determinants of contact with geriatric medicine; and 3) to examine determinants of frequent use of community-based physician services as a proxy for need for specialized geriatric care.

Methods: The sample included long-stay, community-dwelling, home care clients, 60 years of age and older, in Ontario (N=196,444). For each unique client, their Resident Assessment Instrument – Home Care (RAI-HC) admission assessment was linked to Ontario Health Insurance Plan (OHIP) billing records (contact with physician services on an outpatient basis), National Ambulatory Care Reporting System records (NACRS; unplanned emergency department visits), and the Discharge Abstract Database (DAD; hospital admissions). Service use was counted in the 90 days pre-assessment, 90 days post-assessment, and six months post-assessment. Descriptive statistics were used to describe the frequency of contact with various services and to compare home care clients with and without geriatric medicine contact. Logistic regression was used to examine the associations between home care client characteristics and contact with geriatric medicine on an outpatient basis (one or more contacts in 90 days post-assessment), and frequent contact with all physician disciplines on an outpatient basis (nine or more contacts in 90 days post-assessment).

Results: While almost half of the sample (49.6%) had contact with physicians four or more times in the 90 days post-assessment, only 5.2% of older home care clients had any contact with geriatric medicine during that time period. Nonetheless, almost half of the sample had multiple needs within the domains of the expertise of geriatric medicine according to the 5M Framework. While family medicine plays a gatekeeping role in the Ontario health care system, increased frequency of contact with family medicine did not result in much of an increase in any subsequent contact with geriatric medicine. Home care clients who had contact with geriatric medicine had lower odds of subsequent acute care services use than those without contact. However, the benefit varied when stratified by client characteristics. There appeared to be less benefit for those who were acutely ill, complex and unstable, and more benefit for those who were cognitively and functionally impaired. These findings may indicate the need for a more upstream approach whereby geriatric medicine is involved in a client’s care before acute issues lead to functional decline and risk of caregiver distress and institutionalization. In simple logistic regression analyses, functional and cognitive impairment, mental health conditions, risk of caregiver distress and institutionalization were found to be significantly associated with higher odds of geriatric medicine contact while pain, medical instability and complexity, and risk of unplanned ED visits were associated with lower odds. However, provincial experts in the care of older adults have identified all of the above as important for referral to SGS. In the final multivariable model, adjusted for regional effects, female sex, difficulties accessing the home, impaired locomotion outside of the home, good prospects of recovery, diagnosis of hemiplegia/hemiparesis, and cancer were associated with lower odds of geriatric contact. Age, worsening of decision-making, dementia, hallucinations, Parkinsonism, osteoporosis, and risk of caregiver distress and institutionalization (MAPLe score) were associated with higher odds of geriatric contact.

Frequent contact with outpatient physicians in general was expected to be a proxy for need for SGS as these home care clients likely had multiple, unmet needs and would benefit from the holistic, patient-centred approach of geriatric medicine. Interestingly, many of the factors driving frequent attendance were the same factors associated with lower odds of geriatric medicine contact. In the final multivariable model, adjusted for regional effects and age, married status, functional improvement potential, congestive heart failure, irregularly irregular pulse, cancer, treatments changed in last 90 days, nine or more medications, medical complexity and instability (CHESS), and at risk for unplanned ED visits (DIVERT) had higher odds of frequent attendance. Female sex, impaired locomotion outside of the home, cognitive impairment, dementia, stroke, multiple sclerosis, Parkinsonism, hip fracture, unusually poor hygiene, older age, and need for urgent referral (AUA) were associated with lower odds of frequent attendance.

Conclusions: This dissertation provides empirically-based insight into the current practice patterns and determinants of community-based geriatric medicine use in Ontario and highlights the need for a decision support mechanism to rationally and equitably identify older home care clients who may benefit from referral to SGS in a timely manner. As a result of this research, a decision support tool is proposed which incorporates insights from historical practice patterns, in addition to provincial expertise in the care of older adults, and the 5M Framework, used nationally to describe the scope and expertise of geriatric medicine. According to the proposed tool, home care clients at risk for caregiver distress and institutionalization (based on current practice patterns), with medical instability and complexity (based on provincial expertise), and needs within multiple domains of the 5M Framework should be identified through regular home care assessment for consideration for referral to SGS. A tool that is compatible with standardized assessments within home care and other care sectors will allow for allocation of resources more rationally and equitably, and enhance communication and integration across care providers, sectors, and agencies. Future research should evaluate the proposed tool and explore implementation considerations.