Abstract
Background
Home care supports older adults living in the community by providing medical, rehabilitative, and personal care at home. Across Canada there is wide variability in public funding models for home care, which may also be paid for privately. Our objective was to compare individual characteristics across different home care payment groups and examine associations between sociodemographic factors, health status, and private payment for care.
Methods
We included formal home care users from the Canadian Longitudinal Study on Aging (CLSA) between 2015 and 2021 and classified them into three groups based on how much of their home care was paid for out-of-pocket: none, part, or all. We used descriptive statistics to compare the individual and home care characteristics of the three groups. We used unadjusted and adjusted multinomial logistic regression models to examine associations with the home care payment groups.
Results
Of 44,817 participants in the CLSA, 3,580 were formal home care users. Using weighted proportions, 6.8% of the CLSA were home care users, and of these 46.2% reported paying nothing out-of-pocket, 12.7% paid partially, and 41.0% paid all costs. Individuals who paid all costs reported the best health, whereas those who paid partially reported the worst. Meal preparation/homemaking and housework/maintenance services were more commonly paid for privately, while medical care was more likely to be publicly funded. Higher-income individuals were more likely to pay entirely out-of-pocket and large provincial variations were noted across payment groups.
Conclusions
Private home care is common in Canada, particularly for non-medical services. Income-related disparities may limit access for those unable to pay, contributing to inequities in aging. Policies ensuring equitable access to essential services will be critical as demand for home care grows.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12913-025-12975-4.
Keywords: Home care, Community care, Health equity, Out-of-pocket expenditures, CLSA
Background
Home care provides health and personal care services to older adults in their place of residence that can help them live independently in the community [1]. Home care may be delivered on an on-going or short-term basis to meet medical, rehabilitative, and personal needs. Home care includes services such as nursing, physiotherapy, and occupational therapy, and support services like personal support and homemaking [2]. Care delivered by professionals is commonly referred to as formal home care, while support services provided by family, friends, or volunteers may be referred to as informal care or family-provided care [3].
The organization and funding of home care systems vary widely across high-income countries. Northern European countries such as Sweden and Norway have universal public systems administered through municipalities [4], while Japan, South Korea, and Germany fund home care through public long-term care insurance [5]. In the United States, home care may be provided by public Medicare and Medicaid programs with varying eligibility requirements [6]. In Canada, publicly funded access to medically necessary health care is guaranteed by the federal Canada Health Act with most health care administered through provincial health systems [7]. However, unlike physician or hospital services, home care is not considered an insured health service but an extended service and thus is not subject to the provisions of the act. Consequently, while all provinces provide some publicly funded home care, there is wide variability in eligibility thresholds, available services, maximum coverage amounts, and funding and delivery models [8]. Most provinces employ financial testing for certain home care services, with a notable exception of Ontario, Canada’s largest province.
Research has shown that publicly funded home care alone does not sufficiently meet the needs of many Canadians. To address these needs, individuals and families may provide informal care or purchase professional home care privately [9]. Precise data on the relative size of public and private expenditures on home care in Canada is not available, but estimates on the proportion of care that is privately purchased have ranged from one-quarter to one-half [10–12]. Purchasing care privately can be an effective way for individuals and families to meet their home care needs, but it may not be possible for all families and individuals, raising concerns about equity in health and aging [13, 14].
Canadian home care research examining both publicly funded and privately purchased care has been limited, and there has not been a national analysis of home care users focused on describing differences between populations that are paying and not paying privately for care. The objective of this paper is to characterize formal home care use and users in Canada across categories defined by how much individuals and families pay for home care: none, part, or all of the costs. We will further examine associations between health and sociodemographic characteristics and the likelihood of paying for care. As the demand for home care in Canada continues to exceed demand, our findings can inform policy and strategy to ensure all Canadians have the opportunity to age well.
Methods
Study design and setting
We conducted a cross-sectional study of middle-aged and older Canadians.
Data source
All data used in this study came from the Canadian Longitudinal Study on Aging (CLSA), for which a detailed methodology has been published previously [15]. The CLSA is a nationally generalizable, longitudinal study of 51,388 community dwelling adults aged 45–85 years at baseline (2011–2015). The CLSA is composed of a Tracking and Comprehensive cohort. Tracking cohort participants were sampled from all ten Canadian provinces and had their data collected through phone interviews. Comprehensive cohort participants were sampled from populations within 25–50 km of eleven data collection sites in seven provinces and had their data collected through a combination of in-person assessments and phone interviews. Eligibility criteria at baseline required participants to be physically and cognitively capable of independent participation. Residents of the territories, those living on First Nations reserves, full-time Canadian Armed Forces members, and people living in long-term residential care settings were excluded in the baseline recruitment. Copies of all CLSA questionnaires and data support documentation can be located on the study website [16]. CLSA collects new waves of data approximately every three years. We included data from Follow-Up 1 (2015–2018) and Follow-Up 2 (2018–2021). While the CLSA is a longitudinal study, our analysis only considered a single wave at a time and is thus cross-sectional.
Eligibility criteria
We included all formal home care users in both the Comprehensive and Tracking cohorts at CLSA Follow-Up 1 (2015–2018) in our primary analysis.
Measures
Formal home care
In the CLSA, formal home care use is defined as short-term or long-term assistance from professionals at home because of a health condition or limitation that affects daily life [17]. We extracted data on service types received, duration of care, and weekly hours of care. A full description of the home care services included is available in Supplementary Material 1. Formal home care users reported whether they or someone in their family paid directly for some or all the care they received with options including 1) Yes, we paid all the cost, 2.) Yes, we paid part of the cost, 3.) No, there was no cost, and 4.) No, we didn’t pay any of the cost. By grouping the two responses that reported no direct payment, we created three categories of home care users, those who paid all, part, or none of the costs of care. We considered this variable to be nominal rather than ordered as the levels represent qualitatively different categories rather than progressive stages on a continuum. Partial payment, for example, could be due to cost-sharing of public home care services or supplementation with private care on top of publicly funded care. We also extracted weekly out-of-pocket costs for home care. All home care variables were self-reported with a 12-month recall window.
Individual characteristics
We extracted data on sociodemographic and health status of participants. Sociodemographic data were age, sex, province of residence, rurality, living alone, partner status, income (< 50k, 50-100k, 100k + CAD), cultural/racial background (White/non-White), immigration history, and education (Less than secondary, secondary, post-secondary). Health variables included measures of the activities of daily living (ADLs), instrumental activities of daily living (IADLs), a validated cognitive impairment indicator [18], depression (Center of Epidemiologic Studies Depression Scale, 10-item version [19]), self-reported anxiety, self-reported general health, and self-reported chronic conditions (heart disease, lung disease, chronic kidney disease, stroke, dementia, diabetes, hypertension, cancer, rheumatoid arthritis, and osteoarthritis).
Analysis
We reported the prevalence of home care use among CLSA participants with both weighted and unweighted samples. We created a descriptive profile of the individual characteristics of home care recipients by payment category (paid none, paid part, paid all). We used medians with first and third quartiles, or means with standard deviations, to describe continuous variables and counts and proportions for categorical variables. We compared differences in individual characteristics across the three groups with Kruskal-Wallis tests for continuous variables and chi-square tests for categorical variables. We also compared the formal home care characteristics of each payment group using the same summary measures and statistical tests.
We fit multinomial logistic regression models to examine associations with home care payment categories as we considered this to be a nominal variable [20]. The dependent variable for this analysis was payment category (paid none, paid part, paid all). Paid none was considered the reference category to enable interpretation of associations in terms of increased odds of paying for care vs. not paying for care. Independent variables for the models included sex, age, rural residence, province of residence, household income, savings and investments, functional impairment categories, assistive device use, cognitive impairment, depression, anxiety, self-reported general health, and count of chronic conditions. These variables were chosen from the literature as the most important variables likely influencing individual use of public and private home care.
First, we fit models individually for each independent variable and from these models we reported odds ratios (ORs) and 95% confidence intervals. Next, we fit models with all independent variables included and reported the adjusted odds ratios (aORs) and 95% confidence intervals. The savings and investments and household income variables were highly correlated, therefore we only included household income in the adjusted model, as income is more frequently reported in the literature. We performed Hosmer-Lemeshow goodness of fit tests and reported the area under the receiver operating curve for the adjusted models. We also report e-values, which are the minimum strength of an unmeasured confounder with both the outcomes and exposure needed to explain away an observed association [21]. All analyses were performed in R-4.4.1 [22]. We received ethics approval from the Hamilton Integrated Research Ethics Board (2023–16413-C) in May 2023. Data were accessed from February 2024.
Sensitivity analysis
We repeated our adjusted multinomial regression analysis replacing income categories with savings and investments categories (< 50k, 50k-100k, 100k-1 million, 1 million + CAD). We also repeated the descriptive analysis at Follow-Up 2 of CLSA (2018–2021) to investigate changes as the CLSA cohort ages.
Results
Of the 44,817 unique participants in the CLSA at Follow-Up 1, 3,580 were formal home care users. Of this number, 1,213 reported that they paid nothing for home care, 455 reported that they paid part of the costs, and 1880 reported that they paid all of the costs. Data were missing on home care payment for 32 individuals. When weighted to better reflect the Canadian population, 6.8% of CLSA participants were formal home care users, with 46.2% in the paid none group, 12.7% in the paid part group, and 41.0% in the paid all group.
The median age of home care users was 75 years and 60.1% were female. Across the payment groups there were notable differences in sociodemographic and health variables (Table 1). Compared to those who paid part or all of their home care costs, individuals who paid nothing out-of-pocket were slightly younger and more likely to be male. Compared to those who paid part or none of the costs, those who paid all were in higher income and savings and investments categories. Overall, individuals who paid part of the costs had the highest levels of functional impairment and worse self-reported health, followed by those who paid none of the costs, with those who paid all of the costs reporting the best health.
Table 1.
Characteristics of CLSA formal home care users by home care payment group, 2015–2018
| Characteristic | Home Care Payment Group | p | ||
|---|---|---|---|---|
| Participants that Did Not Pay for Home Care Costs (N = 1213) | Participants that Paid Part of Home Care Costs (N = 455) | Participants that Paid All of Home Care Costs (N = 1880) |
||
| Sociodemographic | ||||
| Sex, F | 622 (51.4%) | 288 (63.3%) | 1222 (65.0%) | < 0.001 |
| Age yrs, median (q1, q3) | 73 (63, 81) | 76 (66, 83) | 78 (68, 83) | < 0.001 |
| Living alone | 485 (41.2%) | 223 (50.6%) | 872 (47.6%) | < 0.001 |
| Married/partnered | 586 (48.4%) | 197 (43.3%) | 879 (46.8%) | 0.18 |
| Rural residence | 122 (10.1%) | 43 (9.5%) | 132 (7.0%) | 0.0081 |
| Province | ||||
| Alberta | 125 (10.3%) | 40 (08.8%) | 193 (10.3%) | < 0.001 |
| British Columbia | 146 (12.0%) | 64 (14.1%) | 414 (22.0%) | |
| Manitoba | 116 (9.6%) | 39 (8.6%) | 144 (7.7%) | |
| New Brunswick | 40 (3.3%) | 20 (4.4%) | 42 (2.2%) | |
| Newfoundland and Labrador | 58 (4.8%) | 31 (6.8%) | 64 (3.4%) | |
| Nova Scotia | 113 (09.3%) | 24 (05.3%) | 145 (7.7%) | |
| Ontario | 331 (27.3%) | 65 (14.3%) | 352 (18.7%) | |
| Prince Edward Island | 21 (1.7%) | 14 (3.1%) | 33 (1.8%) | |
| Quebec | 241 (19.9%) | 146 (32.1%) | 450 (23.9%) | |
| Saskatchewan | 22 (01.8%) | 12 (02.6%) | 43 (2.3%) | |
| Household Income (CAD) | ||||
| <50k | 609 (56.1%) | 239 (58.0%) | 693 (41.7%) | < 0.001 |
| 50k to 100k | 288 (26.5%) | 116 (28.2%) | 624 (37.6%) | |
| 100k+ | 189 (17.4%) | 57 (13.8%) | 344 (20.7%) | |
| Savings and Investments (CAD) | ||||
| <50k | 450 (42.3%) | 161 (41.7%) | 433 (26.8%) | < 0.001 |
| 50k to 100k | 192 (18.1%) | 75 (19.4%) | 310 (19.2%) | |
| 100k − 1 mil | 357 (33.6%) | 132 (34.2%) | 725 (44.8%) | |
| 1 million+ | 65 (6.1%) | 18 (4.7%) | 149 (9.2%) | |
| Cultural/Racial Background | ||||
| White | 1154 (95.2%) | 441 (96.9%) | 1826 (97.2%) | 0.0131 |
| Non-White | 58 (4.8%) | 14 (3.1%) | 53 (2.8%) | |
| Immigration history | ||||
| Immigrant | 171 (14.1%) | 55 (12.1%) | 330 (17.6%) | 0.0029 |
| Non-Immigrant | 1041 (85.9%) | 400 (87.9%) | 1550 (82.5%) | |
| Education | ||||
| Secondary School Not Completed | 167 (13.8%) | 51 (11.3%) | 151 (8.1%) | < 0.001 |
| Completed Secondary School | 270 (22.3%) | 80 (17.7%) | 356 (19.0%) | |
| Completed Post-Secondary School | 773 (63.9%) | 321 (71.0%) | 1364 (72.9%) | |
| Health Status | ||||
| Functional impairment1 | ||||
| None | 506 (43.0%) | 121 (27.7%) | 763 (41.5%) | < 0.001 |
| Mild | 415 (35.2%) | 188 (43.0%) | 826 (44.9%) | |
| Moderate or greater | 257 (21.8%) | 128 (29.3%) | 249 (13.6%) | |
| Count of ADLs (mean, sd)2 | 0.7 (1.2) | 1.0 (1.4) | 0.5 (0.9) | < 0.001 |
| Count of IADLs (mean, sd)3 | 1.0 (1.4) | 1.4 (1.6) | 0.8 (1.2) | < 0.001 |
| Cognitive Impairment | 43 (4.6%) | 15 (4.3%) | 58 (3.8%) | 0.65 |
| Assistive device use (any) | 773 (63.7%) | 280 (61.5%) | 928 (49.4%) | < 0.001 |
| CES-D-10 score (mean, sd)4 | 7.8 (5.7) | 8.1 (6.1) | 7.2 (5.2) | 0.001 |
| Self-Reported Anxiety | 182 (15.8%) | 64 (14.7%) | 200 (11.1%) | < 0.001 |
| Self-reported health | ||||
| Excellent/very good | 339 (28.1%) | 118 (26.0%) | 637 (34.1%) | < 0.001 |
| Good | 415 (34.4%) | 149 (32.8%) | 695 (37.2%) | |
| Fair/poor | 452 (37.5%) | 187 (41.2%) | 538 (28.8%) | |
| Count of chronic conditions (mean, SD)5 | 2.7 (1.8) | 2.8 (1.7) | 2.5 (1.7) | < 0.001 |
| Dementia | 21 (1.8%) | 5 (1.1%) | 14 (0.8%) | 0.039 |
| Heart Failure | 290 (25.3%) | 105 (24.3%) | 434 (24.2%) | 0.80 |
| COPD6 | 195 (17.0%) | 69 (15.9%) | 243 (13.6%) | 0.034 |
| Diabetes | 398 (34.6%) | 139 (32.0%) | 504 (28.0%) | < 0.001 |
| Hypertension | 665 (57.7%) | 245 (56.5%) | 969 (53.9%) | 0.12 |
1. Older Americans Resources and Services Multidimensional Functional Assessment
2. Activities of daily living
3. Instrumental activities of daily living
4. Center for Epidemiologic Studies Depression Scale, 10-item
5. Includes heart disease, lung disease, chronic kidney disease, stroke, dementia, diabetes, hypertension, cancer, rheumatoid arthritis, and osteoarthritis
6. Chronic obstructive pulmonary disease
Home care characteristics
Use of formal home care services varied notably by payment group (Table 2). Medical care was the most common service in the paid none group (50.5%), followed by meal preparation/homemaking (29.8%) and physical therapy (22.8%). In the paid part group, meal preparation/homemaking (59.6%) was the most common service, followed by housework/maintenance (38.5%) and personal care (35.8%). Meal preparation/homemaking (52.8%) was also the most common in the paid all group, followed by housework/maintenance (52.8%). The individuals who paid part used the highest number of services, with nearly a quarter reporting the use of four or more services, followed by the paid none and paid all groups. Similarly, the paid part group had the highest average weekly hours of care (7.1) followed by paid none (5.0) and paid all (4.8).
Table 2.
Formal home care services among CLSA participants, 2015–2018
| Home Care Service Type | Participants that Did Not Pay for Home Care Costs (N = 1213) | Participants that Paid Part of Home Care Costs (N = 455) | Participants that Paid All of Home Care Costs (N = 1880) |
|---|---|---|---|
| N (%) | N (%) | N (%) | |
| Personal Care | 304 (25.06%) | 163 (35.82%) | 214 (11.38%) |
| Medical Care | 613 (50.54%) | 148 (32.53%) | 156 (8.30%) |
| Coordinating Care | 72 (5.94%) | 40 (8.79%) | 48 (2.55%) |
| Housework/Maintenance | 235 (19.37%) | 175 (38.46%) | 993 (52.82%) |
| Transportation | 184 (15.17%) | 124 (27.25%) | 220 (11.70%) |
| Meal Preparation/Homemaking | 362 (29.84%) | 271 (59.56%) | 1052 (55.96%) |
| Physical Therapy | 277 (22.84%) | 110 (24.18%) | 138 (7.34%) |
| Training and Adaptation | 90 (7.42%) | 45 (9.89%) | 48 (2.55)% |
| Other Professional Services | 91 (7.50%) | 25 (5.49%) | 59 (3.14%) |
| Number of Services Used | |||
| 1 | 701 (57.79%) | 186 (40.88%) | 1225 (65.16%) |
| 2 | 251 (20.69%) | 99 (21.76%) | 443 (23.56%) |
| 3 | 131 (10.80%) | 64 (14.07%) | 111 (5.90%) |
| 4+ | 130 (10.72%) | 106 (23.30%) | 101 (5.37%) |
| Hours per week of care, mean (sd) | 5.0 (14.5) | 7.1 (18.3) | 4.8 (13.2) |
| Hours per week of care, median (q1, q3) | 2 (1, 4) | 3 (1, 5) | 2, (1, 4) |
| Weekly out of pocket costs, mean (sd) | $0 ($0) | $369 ($1328) | $316 ($810) |
| Weekly out of pocket costs, median (q1, q3) | $0 ($0, $0) | $80 ($40, $200) | $110 ($60, $220) |
The distribution of payment groups was strongly associated with the types of services individuals received (Supplementary Material 2). Among those who received medical care, over two-thirds (66.8%) were in the paid none group while only 17.0% were in the paid all group. Conversely, among those who received meal preparation/homemaking, only 21.5% in the paid none group while 62.4% were in the paid all group. Among those who received housework/maintenance, only 16.7% were in the paid none group while 70.8% were in the paid all group.
Associations of sociodemographic and health variables with payment group
The unadjusted associations reflect the differences noted in the descriptive analysis (Table 3). Compared to those who did not pay, those who paid all were in higher income and savings categories and were less likely to use assistive devices, while those who paid part of the cost had higher levels of functional impairment. Female sex was positively associated with both paying part or all of the costs of home care compared to paying none. Individuals residing in rural areas were less likely to pay for part or all of their home care. There were few differences between the unadjusted and fully adjusted models (Table 4). However, the associations between income and province with payment groups grew more pronounced. Compared to those in the lowest income category, individuals in the middle category (aOR 2.09, 95%CI (1.64, 2.66)) and highest category (aOR 2.44 (1.81, 3.30)) were more likely to be in the paid all rather than the paid none group. Compared to Ontario, individuals in Newfoundland and Labrador (aOR 3.75 (1.35, 3.68)) were the most likely to be in the paid part group while individuals in British Columbia (aOR 3.29 (2.40, 4.53)) were the most likely to be in the paid all group. There were no provinces in which home care users were less likely to be in the paid part or all groups than Ontario.
Table 3.
Unadjusted multinomial logistic regression models on home care payment group
| Variable | Paid Part vs. Paid None | Paid All vs. Paid None |
|---|---|---|
| OR (95% CI) | OR (95% CI) | |
| Sociodemographics | ||
| Sex, F | 1.63 (1.31, 2.04) | 1.76 (1.52, 2.04) |
| Age, year | 1.02 (1.01, 1.03) | 1.03 (1.03, 1.04) |
| Living alone | 1.46 (1.17, 1.82) | 1.30 (1.12, 1.50) |
| Rural residence | 0.93 (0.65, 1.35) | 0.68 (0.52, 0.87) |
| Province | ||
| Ontario (ref) | - | - |
| Alberta | 1.63 (1.04, 2.54) | 1.45 (1.11, 1.90) |
| British Columbia | 2.23 (1.50, 3.32) | 2.67 (2.10, 3.39) |
| Manitoba | 1.71 (1.09, 2.68) | 1.17 (0.88, 1.56) |
| New Brunswick | 2.55 (1.40, 4.63) | 0.99 (0.62, 1.56) |
| Newfoundland and Labrador | 2.72 (1.63, 4.54) | 1.04 (0.71, 1.53) |
| Nova Scotia | 1.08 (0.65, 1.81) | 1.21 (0.90, 1.61) |
| Prince Edward Island | 3.39 (1.64, 7.02) | 1.48 (0.84, 2.61) |
| Quebec | 3.08 (2.20, 4.32) | 1.76 (1.41, 2.18) |
| Saskatchewan | 2.78 (1.31, 5.89) | 1.84 (1.08, 3.14) |
| Household Income | ||
| <50k | - | - |
| 50k to 100k | 1.03 (0.79, 1.33) | 1.90 (1.60, 2.27) |
| 100k+ | 0.77 (0.55, 1.07) | 1.60 (1.30, 1.97) |
| Savings and Investments | ||
| <50k | - | - |
| 50k to 100k | 1.09 (0.79, 1.51) | 1.68 (1.34, 2.10) |
| 100k − 1 mil | 1.03 (0.79, 1.35) | 2.11 (1.76, 2.53) |
| 1 mil+ | 0.77 (0.45, 1.34) | 2.38 (1.73, 3.28) |
| Health Status | ||
| Functional impairment | ||
| None | ||
| Mild | 1.89 (1.46, 2.46) | 1.32 (1.12, 1.55) |
| Moderate+ | 2.08 (1.56, 2.78) | 0.64 (0.52, 0.79) |
| Cognitive Impairment | 0.95 (0.52, 1.73) | 0.83 (0.55, 1.24) |
| Assistive device use (any) | 0.91 (0.73, 1.14) | 0.55 (0.48, 0.64) |
| CES-D score1 | 1.01 (0.99, 1.03) | 0.98 (0.97, 0.99) |
| Self-Reported Anxiety | 0.92 (0.68, 1.25) | 0.67 (0.54, 0.83) |
| Self-reported health | ||
| Excellent/very good | - | - |
| Good | 1.03 (0.78, 1.37) | 0.89 (0.74, 1.07) |
| Fair/poor | 1.19 (0.91, 1.56) | 0.63 (0.53, 0.76) |
| Count of chronic conditions 2 | 1.03 (0.96, 1.10) | 0.93 (0.89, 0.97) |
1. Center for Epidemiologic Studies Depression Scale, 10-item
2. Includes heart disease, lung disease, chronic kidney disease, stroke, dementia, diabetes, hypertension, cancer, rheumatoid arthritis, and osteoarthritis
Table 4.
Adjusted multinomial logistic regression models on home care payment group
| Variable | Paid Part vs. Paid None | Paid All vs. Paid None |
|---|---|---|
| aOR (95% CI) | aOR (95% CI) | |
| Sociodemographics | ||
| Sex, F | 1.85 (1.36, 2.53) | 1.96 (1.59, 2.41) |
| Age, year | 1.03 (1.02, 1.05) | 1.04 (1.03, 1.05) |
| Living alone | 1.10 (0.80, 1.52) | 1.25 (1.00, 1.57) |
| Rural residence | 0.86 (0.53, 1.39) | 0.58 (0.41, 0.82) |
| Province | ||
| Ontario (ref) | - | - |
| Alberta | 1.78 (1.02, 3.11) | 1.59 (1.11, 2.28) |
| British Columbia | 2.23 (1.35, 3.68) | 3.29 (2.40, 4.53) |
| Manitoba | 1.84 (0.94, 3.60) | 1.59 (1.02, 2.48) |
| New Brunswick | 1.52 (0.66, 3.50) | 1.32 (0.74, 2.35) |
| Newfoundland and Labrador | 3.75 (1.91, 7.34) | 1.65 (0.97, 2.80) |
| Nova Scotia | 1.08 (0.52, 2.27) | 1.51 (0.98, 2.33) |
| Prince Edward Island | 2.58 (0.90, 7.43) | 2.54 (1.22, 5.32) |
| Quebec | 3.21 (2.08, 4.95) | 2.27 (1.70, 3.04) |
| Saskatchewan | 2.74 (1.02, 7.34) | 1.96 (0.95, 4.06) |
| Household Income | ||
| <50k | - | - |
| 50k to 100k | 1.38 (0.98, 1.96) | 2.09 (1.64, 2.66) |
| 100k+ | 1.47 (0.93, 2.32) | 2.44 (1.81, 3.30) |
| Health Status | ||
| Functional impairment | ||
| None | - | - |
| Mild | 1.97 (1.38, 2.79) | 1.54 (1.21, 1.94) |
| Moderate+ | 1.96 (1.25, 3.06) | 1.03 (0.75, 1.42) |
| Cognitive Impairment | 0.72 (0.34, 1.55) | 0.80 (0.47, 1.37) |
| Assistive device use (any) | 0.62 (0.45, 0.85) | 0.50 (0.40, 0.62) |
| CES-D score1 | 1.03 (1.00, 1.06) | 1.01 (0.99, 1.03) |
| Self-Reported Anxiety | 0.82 (0.54, 1.25) | 0.79 (0.58, 1.06) |
| Self-reported health | ||
| Excellent/very good | - | - |
| Good | 0.80 (0.55, 1.16) | 0.84 (0.66, 1.07) |
| Fair/poor | 0.99 (0.66, 1.48) | 0.87 (0.66, 1.15) |
| Count of chronic conditions2 | 0.99 (0.91, 1.08) | 0.93 (0.88, 0.99) |
|
Hosmer-Lemeshow Χ²=15.105, df = 16, p-value = 0.517 AUC3: Paid all vs. Paid None = 0.72; Paid Part vs. Paid None = 0.70 | ||
1. Center for Epidemiologic Studies Depression Scale, 10-item
2. Includes heart disease, lung disease, chronic kidney disease, stroke, dementia, diabetes, hypertension, cancer, rheumatoid arthritis, and osteoarthritis
3. Area under the receiver operating curve
The Hosmer-Lemeshow test for the adjusted multinomial model did not reject the null hypothesis (p = 0.517), indicating no strong evidence of poor model fit. Area under the receiver operating curve values were 0.72 for paid all vs. paid none and 0.70 for paid part vs. paid none. All e-values can be found in Supplementary Material 3. The e-values for income categories in the paid all vs. paid none were around 2. Provincial associations that exceed an e-value of 2 included Newfoundland and Quebec vs. Ontario in for paid part vs. paid none and British Columba vs. Ontario for paid all vs. paid none.
Sensitivity analysis
Using savings and investment categories in place of income categories produced a similar dose-response relationship (Supplementary Material 4). Home care measures at Follow-Up 2 were also not meaningfully different than those at Follow-Up 1 (Supplementary Material 5).
Discussion
We used data from the Canadian Longitudinal Study on Aging (CLSA) to characterize home care users across categories based on how much of their care was paid for out-of-pocket: none, part or all. Overall, we found that roughly equal proportions of individuals paid for all (41%) or none (46%) or their care, with fewer (13%) paying for part of their care. We found significant differences in sociodemographic and health characteristics across payment categories. Higher income was associated with a higher likelihood to pay all the costs of home care, and the likelihood of paying for part or all of the costs of home care varied widely by province of residence.
The literature indicates that while paying out of pocket for home care services is nearly universal across healthcare systems, it is also highly heterogeneous [23, 24]. The degree of private payment has been shown to be related to the strength of social protection systems, affordability of care, and overall expenditure on home care on a system levels [25, 26]. Generally, the evidence suggests that Canada is in the middle of the pack of comparator countries in terms of likelihood of paying out of pocket for home care and overall public funding for home care [26, 27]. We can compare our results to a survey of home care users in the United States that used similar categories that found that 28% of users were solely publicly funded, 63% were personally financed, and 9% included both [6]. While some previous studies have reported overall proportions of private payment for home care in Canada, our study is novel in its examination of individuals characteristics of groups and comparison of payment categories across provinces with different funding policies.
We found meaningful differences in the health characteristics of individuals across home care payment categories, with those who reported paying for all of their care having the best reported health, those paying for part having the worst self-reported health, and those paying for none in the middle. This suggests that the group paying entirely out-of-pocket may be largely made up of lower needs of individuals who do not reach eligibility thresholds for publicly funded care and pay out-of-pocket to access care. Conversely, the group that paid partially out-of-pocket had the highest levels of impairment and may be receiving publicly funded home care in addition to purchasing services for unmet care needs.
We identified associations between income and province of residence and home care payment categories that grew more pronounced when we adjusted for health and sociodemographic variables. The positive association between income and the likelihood of paying for care is straightforwardly interpretable given that higher economic means are required to pay for private care. This finding also is in alignment with other research that has found higher income individuals in the United States were more likely to bear all the costs of home care [6]. The considerable variations in payment groups across provinces that we observed is also consistent with other research on home care across Canada [28] and reflects differences in provincial policies, particularly the use of financial testing. Ontario, which does not implement financial testing, had the highest proportion of users reporting no out-of-pocket payments for care. This may be contrasted with a province like British Columbia, which has a cost-sharing program for home support services in which the client pays up to maximum daily rate based on their after-tax household income [29]. Our results suggest that residents of British Columbia have 2–3 three times the odds of paying for home care than similar individuals living in Ontario.
These provincial differences reflect policy choices that provinces have made in terms of how they will allocate their scarce home care resources. Despite clear differences in funding structures, a cursory examination of nationally reported quality home care indicators does not suggest an apparent quality advantage for one province over the other. For example, compared to British Columbia, Ontario has higher rate of reports that home care helps people stay at home, but also higher reported rates of caregiver distress [30].
A steadily growing older population has increased the demand for home care worldwide, requiring healthcare systems to adapt to meet these changing needs. For example, in Nordic countries, typically noted for their universal approach to health care, home care has recently shifted towards prioritization, informalization, and privatization [31]. As home care systems transform, it is crucial that policymakers consider how they can support individuals with unmet home care needs who cannot afford private care. Universality is cornerstone of Canadian, and many other, healthcare systems, yet given the constrained resources available for home care, distribution of services irrespective of economic means to pay for private care may fail to meaningfully improve health equity [32, 33]. Proportionate universality may provide a more useful framework to maximize equity and quality [33]. In proportionate universality, provision of care is not allocated not just on need but also proportionate to the degree of disadvantage [34]. In the context of home care this would include tailoring the provision of public care based not only on means to pay for private care but also on availability of social support networks to provide informal care. Further research to identify which frameworks and programs reduce the health gradient and produce the best population level outcomes is needed to inform future health system transformation. The diversity of home care systems across provinces in Canada may provide a useful natural experiment that can be leveraged to investigate this question.
Limitations
While the CLSA is a nationally generalizable sample of middled-aged and older adults, the recruitment yielded a sample that was whiter, healthier and wealthier than the general Canadian population. While CLSA collects detailed data on racial and culture background, there were an insufficient number of non-White individuals to report further categories. Additionally, the exclusion of individuals with overt cognitive impairment in the baseline recruitment of CLSA may have yielded a sample of home care recipients with lower needs than a general home care population.
Conclusion
Our study highlights that private payment for home care services is common in Canada, with significant variation depending on the service type. Individuals without informal support networks or economic resources to purchase private care may be disadvantaged in their ability to age in their desired location. As demand for home care services continues to rise, strategic allocation of limited public home care resources will be crucial in ensuring equitable access for all Canadians.
Supplementary Information
Supplementary Material 1. Description of formal home care services included in Canadian Longitudinal Study On Aging.
Supplementary Material 2. Row percentages of formal home care services by home care payment groups, Canadian Longitudinal Study on Aging participants, 2015-0218.
Supplementary Material 3. Adjusted multinomial logistic regression models on home care payment group with e-values.
Supplementary Material 4. Adjusted multinomial logistic regression models on home care payment group, with savings and investment categories rather than household income.
Supplementary Material 5. Formal home care services among Canadian Longitudinal Study on Aging participants, 2018-2021.
Acknowledgements
This research was made possible using the data/biospecimens collected by the Canadian Longitudinal Study on Aging (CLSA). Funding for the Canadian Longitudinal Study on Aging (CLSA) is provided by the Government of Canada through the Canadian Institutes of Health Research (CIHR) under grant reference: LSA 94473 and the Canada Foundation for Innovation, as well as the following provinces, Newfoundland, Nova Scotia, Quebec, Ontario, Manitoba, Alberta, and British Columbia. This research has been conducted using the CLSA dataset Baseline Comprehensive v7.0, Baseline Tracking v4.0, Follow-up 1 Comprehensive v3.1, and Follow-up 1 Tracking v3.1, Follow-up 2 Comprehensive v1.0, and Follow-up 2 Tracking v1.0 under Application Number (23CA011). The CLSA is led by Drs. Parminder Raina, Christina Wolfson and Susan Kirkland. The opinions expressed in this manuscript are the author’s own and do not reflect the views of the Canadian Longitudinal Study on Aging.
Abbreviations
- ADL
Activities of daily living
- aOR
Adjusted odds ratio
- CES-D
Center for Epidemiologic Studies Depression Scale, 10-item
- CLSA
Canadian Longitudinal Study on Aging
- IADL
Instrumental activities of daily living
- OR
Odds ratio
Aaron Jones
is supported by the Schlegel Chair in Clinical Epidemiology and Aging at McMaster University and the Schlegel-UW Research Institute for Aging.
Authors’ contributions
AJ, SB, AC, PH, CM, JW, AM, HS, CS contributed to the design and conception of the work. JL was responsible for data analysis. AJ wrote the initial draft. All authors contributed to the interpretation of the data, reviewed it critically for important intellectual content, and gave final approval for publication.
Funding
This work was supported by a grant from the Canadian Institutes of Health Research (185769). The funder had no role in conceptualization, design, data collection, analysis, decision to publish, or preparation of the manuscript.
Data availability
Data are available from the Canadian Longitudinal Study on Aging (www.clsa-elcv.ca) for researchers who meet the criteria for access to de-identified CLSA data.
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with the Declaration of Helsinki. We received ethics approval from the Hamilton Integrated Research Ethics Board (16413). All participants of the Canadian Longitudinal Study on Aging provided informed consent to participate in the study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Health Canada. Home and community health care. 2016. https://www.canada.ca/en/health-canada/services/home-continuing-care/home-community-care.html. Accessed 29 Jul 2024.
- 2.Government of Ontario. Home and community care. https://www.ontario.ca/page/homecare-seniors#section-3. Accessed 4 May 2021.
- 3.Li J, Song Y. Formal and informal care. In: Gu D, Dupre ME, editors. Encyclopedia of gerontology and population aging. Cham: Springer International Publishing; 2019. pp. 1–8. [Google Scholar]
- 4.Tarricone R, Tsouros AD. Home care in europe: the solid facts. World Health Organization. Regional Office for Europe; 2008. https://iris.who.int/handle/10665/328766.
- 5.Chen L, Zhang L, Xu X. Review of evolution of the public long-term care insurance (LTCI) system in different countries: influence and challenge. BMC Health Serv Res. 2020;20:1057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Janus AL, Ermisch J. Who pays for home care? A study of nationally representative data on disabled older Americans. BMC Health Serv Res. 2015;15:301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Canada Health Act RSC. 1985, c C-6. https://laws-lois.justice.gc.ca/eng/acts/c-6/page-1.html.
- 8.National Institute on Ageing. Enabling the future provision of long-term care in Canada. Toronto; 2019. https://static1.squarespace.com/static/5c2fa7b03917eed9b5a436d8/t/5d9de15a38dca21e46009548/1570627931078/Enabling+the+Future+Provision+of+Long-Term+Care+in+Canada.pdf.
- 9.Statistics Canada. Home care use and unmet home care needs in Canada. 2021. https://www150.statcan.gc.ca/n1/daily-quotidien/220826/dq220826a-eng.htm.
- 10.Statistics Canada. Formal home care use in Canada. 2018. https://www150.statcan.gc.ca/n1/pub/82-003-x/2018009/article/00001/tbl/tbl04-eng.htm. Accessed 1 Mar 2022.
- 11.Wilson MG. Evidence brief: organizing a care system for older adults in Ontario. 2011. https://www.mcmasterforum.org/docs/default-source/Product-Documents/evidence-briefs/care-systems-for-older-adults-in-ontario-eb.pdf?sfvrsn=2.
- 12.MacDonald B, Wolfson M, Hirdes J. The future Co$t of Long-Term care in Canada. Toronto, ON: National Institute on Ageing; 2019. [Google Scholar]
- 13.Lee J, Watt JA, Mayhew A, Sinn C-L, Schumacher CL, Costa AP et al. Equity in home care use in Canada: a cross-sectional analysis of the Canadian longitudinal study on ageing. BMJ Public Health. 2024;2. 10.1136/bmjph-2023-000812. [DOI] [PMC free article] [PubMed]
- 14.Yakerson A. Home care in ontario: perspectives on equity. Int J Health Serv. 2019;49:260–72. [DOI] [PubMed] [Google Scholar]
- 15.Raina P, Wolfson C, Kirkland S, Griffith LE, Balion C, Cossette B, et al. Cohort profile: the Canadian longitudinal study on aging (CLSA). Int J Epidemiol. 2019;48:1752–j1753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Researcher Resources. Canadian Longitudinal Study on Aging (CLSA). https://www.clsa-elcv.ca/researcher-resources/. Accessed 4 Dec 2024.
- 17.CLSA. Derived Variables– Care Receiving 1/Formal Care (CR1) & Care Receiving 2/Informal Care (CR2). 2017. https://www.clsa-elcv.ca/sites/default/files/documents/dv_cr_4dec2017.pdf. Accessed 19 Sep 2023.
- 18.O’Connell ME, Kadlec H, Griffith LE, Wolfson C, Maimon G, Taler V, et al. Cognitive impairment indicator for the neuropsychological test batteries in the Canadian longitudinal study on aging: definition and evidence for validity. Alzheimers Res Ther. 2023;15:167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Radloff LS, The CES-D, Scale. A Self-Report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401. [Google Scholar]
- 20.Hosmer D, Stanley L, Rodney XS. Applied logistic regression. Wiley; 2013. https://www.wiley.com/en-us/Applied+Logistic+Regression%2C+3rd+Edition-p-9780470582473.
- 21.VanderWeele TJ, Ding P. Sensitivity analysis in observational research: introducing the E-Value. Ann Intern Med. 2017;167:268–74. [DOI] [PubMed] [Google Scholar]
- 22.R Core Team. R: A language and environment for statistical computing. 2020. https://www.r-project.org/.
- 23.Tur-Sinai A, Bentur N. Out-of-pocket expenditure on community healthcare services at end-of-life among decedents from cardiovascular disease in six European countries and Israel. Health Econ Rev. 2023;13:37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ailshire JA, Herrera CA, Choi E, Osuna M, Suzuki E. Cross-national differences in wealth inequality in health services and caregiving used near the end of life. eClinicalMedicine. 2023;58:101911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Angrisani M, Regalado JCO, Hashiguchi TCO. Financial social protection and individual out-of-pocket costs of long-term care in the USA and europe: an observational study. eClinicalMedicine. 2022;50:101503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Spending on long-term care. OECD. 2020. https://www.oecd.org/en/publications/spending-on-long-term-care_e1f00970-en.html. Accessed 4 Mar 2025.
- 27.Rice T, Quentin W, Anell A, Barnes AJ, Rosenau P, Unruh LY, et al. Revisiting out-of-pocket requirements: trends in spending, financial access barriers, and policy in ten high-income countries. BMC Health Serv Res. 2018;18:371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hébert PC, Morinville A, Costa A, Heckman G, Hirdes J. Regional variations of care in home care and long-term care: a retrospective cohort study. Can Med Assoc Open Access J. 2019;7:E341–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ministry of Health - Province of British Columbia. Home Support - Province of British Columbia. https://www2.gov.bc.ca/gov/content/health/accessing-health-care/home-community-care/care-options-and-cost/home-support. Accessed 24 Oct 2024.
- 30.Canadian Institute for Health Information. Your Health System: In Brief. https://www.cihi.ca/en/your-health-system-in-brief. Accessed 1 Mar 2022.
- 31.Rostgaard T, Jacobsen F, Kröger T, Peterson E. Revisiting the nordic long-term care model for older people—still equal? Eur J Ageing. 2022;19:201–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Fisher M, Harris P, Freeman T, Mackean T, George E, Friel S, et al. Implementing universal and targeted policies for health equity: lessons from Australia. Int J Health Policy Manag. 2022;11:2308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Carey G, Crammond B, De Leeuw E. Towards health equity: a framework for the application of proportionate universalism. Int J Equity Health. 2015;14:81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Moffatt H, Fish K, Schwenger S, Sankaran S. Let’s talk: universal and targeted approaches to health equity. National Collaborating Centre for Determinants of Health; 2013. https://nccdh.ca/images/uploads/Approaches_EN_Final.pdf.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1. Description of formal home care services included in Canadian Longitudinal Study On Aging.
Supplementary Material 2. Row percentages of formal home care services by home care payment groups, Canadian Longitudinal Study on Aging participants, 2015-0218.
Supplementary Material 3. Adjusted multinomial logistic regression models on home care payment group with e-values.
Supplementary Material 4. Adjusted multinomial logistic regression models on home care payment group, with savings and investment categories rather than household income.
Supplementary Material 5. Formal home care services among Canadian Longitudinal Study on Aging participants, 2018-2021.
Data Availability Statement
Data are available from the Canadian Longitudinal Study on Aging (www.clsa-elcv.ca) for researchers who meet the criteria for access to de-identified CLSA data.