Introduction
The Canada Health Act, the federal legislation for publicly funded health care insurance, is designed to ensure that all eligible residents of Canada have reasonable access to insured health services on a prepaid basis, without direct charges at the point of service for such services. The Act requires that medically necessary hospital and physician services be fully covered. The provisions apply also to certain aspects of long‑term residential care (nursing home intermediate care and adult residential care services) (Health Canada, 2008), but discretion is left to the provinces and territories to determine precisely what aspects are publicly funded. In consequence, the costs to residents of long-term care facilities differ substantially from one jurisdiction to another (Romanow, Reference Romanow2002, p. 5). What led to such differences in three provinces, and how they affect the experiences of some residents, is discussed by Stadnyk (Reference Stadnyk2009). Our purpose here is to document and analyse the extent to which costs differ for individuals who have similar socio-economic characteristics but who live in different parts of the country.
All provinces and territories pay a large portion of the costs of long-term care (LTC), but our concern is specifically with the private costs (i.e., those costs borne directly by residents and their families). Total expenditure on nursing homes and residential care facilities in 2006 in Canada is estimated at $15.5 billion, of which about $3.8 billion was from private sources (Canadian Institute for Health Information, 2008). Most of that would be from the fees paid by residents of nursing homes. As we show in what follows, there are very large differences in the fees that are paid, depending on where you live. That restricts the ability of older Canadians to move from one province to another, perhaps in order to be closer to their adult children or other family members, and it raises concerns about the fairness and equity of this important component of our health care system.
Approach
We used publicly available information to determine the private cost of LTC in each of the provinces and territories of Canada. We estimated the cost for a number of stylized individuals with characteristics that are representative of the residents of such facilities. In all provinces, admission to LTC facilities is based on both proof of residency and medical need for 24-hour care, as certified by a physician (Manulife Financial, 2009). We limited attention to typical individuals who meet the residence and medical need requirements for LTC.
In all provinces the amount they are charged depends on their marital status and, if married, on whether one or both spouses are in LTC. We therefore distinguished between individuals who were not married and those who were. In the latter category, we distinguished between whether one spouse was receiving care or both, and we assumed that there were no other dependents.
While remarkably little is known about the characteristics of residents of LTC institutions in Canada, from the National Nursing Home Surveys conducted in the United States we know that about 39 per cent of male residents and 16 per cent of female residents in the U.S. are married, the rest not (Ness, Ahmed, & Aronow, Reference Ness, Ahmed and Aronow2004). The proportions might be similar in Canada, but we have no way of telling.
Most jurisdictions also base the amount charged on the economic circumstances of each resident, a concept that is usually measured by their individual or family income but, in one province (Quebec), also by the value of assets held. We estimated the charges that would apply to individuals who differed in terms of their marital status, their income levels, and, where relevant, the assets that they owned as well as in their place of residence. All charges and income levels relate to 2008, the latest year for which we could assemble information.
The province of residence matters not only in terms of the direct charges that are incurred for LTC, but also because differences in provincial income tax rates affect the after-tax incomes of residents and hence the portion of their income that is available for other uses.
Information about the policies in each of the provinces and territories was collected from a variety of sources, and with considerable difficulty. Much of what was needed was provided in reports for each jurisdiction prepared by Manulife Financial (2009). In addition, government documents were reviewed for individual provinces, and telephone help lines were asked for clarification as needed. That was supplemented by e-mail correspondence with government officials in charge of long-term care. Manitoba and Quebec provide calculators on their websites (Manitoba, 2009; Quebec, 2009) that can be used to estimate how much individuals in various circumstances would be charged. We found them to be very helpful.
Cost of Long-term Care
The private annual costs that applied in 2008 in each of the provinces and territories are recorded in Table 1 for non-married individuals and in Table 2 for those married. The non-married category includes anyone without a living partner (i.e., those who never married or are divorced as well as those whose partners have died). We have opted to report the fees actually paid, without making allowance for medical tax credits. Fees are out-of-pocket and have to be paid when due. Someone who entered a nursing home at mid-year, as would happen on average, would have no benefit from the medical tax credit for almost a full year until after his or her income tax return had been filed and processed. A large fraction of residents would never realize the value of such benefits.
Table 1: Private cost of long-term care by region and income level, 2008: Non-married seniors
Note 1: The costs shown are the total cost paid by or on behalf of the resident, ignoring any medical credits.
Note 2: “Very low income” is defined as being in receipt of Old Age Security plus maximum Guaranteed Income Supplement, the annualized average of 2008 quarterly amounts.
“Low income” is defined as the 2008 low income cut-off for a one- or two-person household, as appropriate, resident in an urban area of between 100,000 and 500,000; values are drawn from Statistics Canada “Low income cut-offs for 2008 and low income measures for 2007”, Catalogue no. 75F0002M, No. 002; http://www.statcan.gc.ca/pub/75f0002m/75f0002m2009002-eng.pdf, p 25.
“Average income” is defined as the 2008 average income for elderly males or elderly couples, as appropriate; it is based on the published value for 2007 average total (before-tax) income (see http://www40.statcan.gc.ca/l01/cst01/famil05a-eng.htm), adjusted to 2008 levels by adding the CPI plus 1 percent.
Note 3: Quebec charges relate to 2009, since 2008 values were not available. Asset holdings are assumed to be zero in Quebec (1) and $48,700 in Quebec (2).
Table 2: Private cost of long-term care by income level, care requirements and region, 2008: Married seniors
Note 1: The costs shown are the total cost paid by the resident, ignoring medical credits.
Note 2: For calculations in which the separate incomes of the spouses matter, it is assumed that the low-income spouse has an income of $11,200 and the high-income spouse has the balance.
Note 3: Assets holdings in Quebec (1) are assumed to be 0 in all cases; in Quebec (2) they are $70,000 in all cases if both are in care; with one in care they are assumed to be $364,000 if income is “very low” and $352,200 if income is “low” or higher.
In assessing costs, we considered five levels of income ranging from very low to three times the average, all based on Canada-wide average values. The dollar values for each are reported in the last rows of the first two tables. At one extreme, very low income is defined as the sum of benefits received under the Old Age Security (OAS) and the Guaranteed Income Supplement (GIS) programs; in 2008 that represented a before-tax annual income level of $13,759 for non-married individuals and $22,394 for couples. Everyone over the age of 65 who has lived in Canada for at least 10 years is assured of this minimal level of income in the form of transfers from the federal government.
Low income is defined here as the so-called low-income cut-off (LICO) published by Statistics Canada People generally spend less of their income on necessities (defined as food, shelter, and clothing) at higher levels of income and more at lower levels. By construction, households below the LICO spend a larger-than-average share of income on necessities and, as such, might be deemed poor. The values used here assume residence in an urban community of between 100,000 and 500,000 and a family size of one (Table 1) or two (Table 2).
The values chosen for average income are specific to the older population. For a non-married person we projected the 2008 average income for elderly males; for those who are married we projected the value for an elderly couple with no other dependents. (Further details are provided in Note 2 to Table 1.) Higher levels of income are defined as two or three times greater than the average.
Some provinces base LTC charges on the before-tax income of residents, other provinces on after-tax income; such differences are reflected in the charges shown. Tables 3 and 4 report the after-tax income for each of the five levels of before-tax income. A married senior with before-tax income of $22,750 would have an after-tax income that would be between 5.2 per cent lower and 2.3 per cent higher, depending on the province or territory of residence.
Table 3: Levels of household income, before- and after-tax, 2008, by region: Non-married seniors
Note 1: Definitions of the income levels are provided in Table 1, Note 1.
Note 2: After-tax income is calculated using the Lorne S Marr Insurance Limited “Canadian Income Tax Calculator 2008”, available at http://lsminsurance.ca/calculators/canada/income-tax, and Revenue Canada’s “General Income Tax and Benefit Package for 2008”, http://www.cra-arc.gc.ca/formspubs/t1gnrl/menu-eng.html , retrieved March 14, 2009.
Table 4: Levels of household income, before- and after-tax, by region, 2008: Married seniors
Two sets of calculations are provided for Quebec. In that province alone, account is taken of assets holdings (specifically, the value of financial assets and real estate) in determining the cost of LTC. The row labelled “Quebec (1)” assumes asset holdings of zero, which triggers the maximum subsidy and hence the lowest charge for LTC, while “Quebec (2)” assumes a level of asset holdings that is just sufficient to require the resident to pay the maximum charge. (The dollar values for the assets are shown in Note 3 of the tables.)
Interpretation
As can be seen from Tables 1 and 2, the annual private cost of long-term care generally increases with the level of household income. The territories are exceptions. Not only are the charges levied in all three territories independent of the level of income, but they are the lowest in the entire country, and, in the case of Nunavut, the charge is zero. However, since only a minute fraction of LTC is provided in the territories, in what follows we focus attention mostly on the situation in the provinces.
Consider the case of those not married, for whom the costs are recorded in Table 1 and, for the provinces, plotted in Figure 1, Part A. It is clear that costs vary considerably from one province to another and, within province, by level of income. For those with very low incomes, the charges are lowest in Nova Scotia, but only slightly higher in Alberta and also in Quebec, at least for those with no assets. By contrast, the charges in Newfoundland and Prince Edward Island are about 25 per cent more than in neighbouring Nova Scotia.
Figure 1: Private costs of long-term care, by income level, marital status, care requirements and region, 2008.
Within each province, the charges for care generally either increase or remain the same at successively higher levels of income, but the gradients differ considerably from one province to another. For example, higher charges apply in most provinces for those with low rather than very low incomes. However, the charge is the same in British Columbia, and also in Quebec, for those with no assets. Still higher charges apply to those with incomes that are average or greater (except in Quebec, where the maximum charge applies to those with incomes that are low or greater).
In seven of the provinces, the charges are the same for those with greater than average incomes as for those whose incomes are average, and in British Columbia and Newfoundland the charges are higher still for those with twice the average income, but do not increase further for those with three times the average income.
In sum, the private cost for non-married residents of LTC facilities is very uneven across the country. Those with very low incomes pay about one quarter more for their care if they live in Newfoundland or Prince Edward Island rather than Nova Scotia, those with low incomes pay half again as much if they live in Ontario rather than British Columbia and two-thirds more if they live in Newfoundland or Prince Edward Island, while those with average incomes or higher pay approximately twice as much in five provinces (British Columbia, Manitoba, Nova Scotia, New Brunswick, and Prince Edward Island) as in Quebec, and close to three times as much in Newfoundland.
What of those married? As a comparison of Table 2 with Table 1 shows, all provinces charge less for a spouse in care than they do for a non-married person, with the reduced charges generally favouring those with incomes that are average or less. That presumably reflects the assumed greater needs of the partner still living in the community, perhaps allowing that spouse to continue to live in the family home.
However, the situation, as reported in Table 2, is more complicated than for a non-married person since the charge that is levied for residential care varies not only with the income of the couple but, depending on the province, also with whether one spouse is in care, or both, and on how the income is seen to be split between the partners. In Quebec it again varies with the assets owned by the couple.
Five provinces (British Columbia, Alberta, Saskatchewan, Ontario, and Quebec) take account of the income split between spouses. Their practice is to charge the high-income spouse more as might be expected. However, of the five, Saskatchewan imposes the higher charge only if one spouse is in care, and Quebec only if both. In our calculations, we assumed that if an income split matters, the low-income spouse has very low income (i.e., the amount of benefit entitlement received through the OAS and GIS programs); the high-income spouse was assigned the balance.
Figure 1, Part B, shows the charges for the high- and low-income partners in four provinces (British Columbia, Alberta, Saskatchewan, and Ontario) where the incomes of each of the spouses are treated separately. Given our assumption that the income of the low-income partner is very low, only the minimum charge would be levied if the low-income partner is in care, even if the joint income of the couple is very high. At the same time, the charge would be twice as high if it were instead the high-income partner in care. In Alberta, the situation is exaggerated even more; in that province the charge for the low-income partner actually declines somewhat as family income increases. That occurs because the charge is based on the after-tax income of the low-income spouse, which declines because the value of the benefit associated with the marital tax credit is reduced as the income of the higher income spouse increases.
Figure 1, Part C, shows the charges when both partners are in care. We retained the assumption about how the income is split in provinces where that has relevance. A number of features stand out. One is that lower income couples pay the least for care; that is true in all provinces. Another is that in most provinces, the charge rises relatively sharply as income increases. Those with average incomes are charged one-third more than those with very low incomes in two provinces (Alberta, Ontario), about twice as much in three provinces (British Columbia, Saskatchewan, and Manitoba), and even more in four others (New Brunswick, Nova Scotia, Newfoundland, and Prince Edward Island). Quebec is again an exception in that those with sufficient assets bear the same (relatively low) charge whatever their level of income; even among those with no assets, those with average income pay only 18 per cent more than those with the lowest income level.
In four provinces (Alberta, Ontario, Quebec, and Prince Edward Island) the maximum rate applies to the higher income spouse if the couple has average or greater (joint) income (or to both spouses, if both are in care); in the other six provinces the maximum rate applies only to those couples with twice the average income or higher. In five provinces (Alberta, Ontario, Quebec, Newfoundland, and Prince Edward Island) those with much higher income levels pay the same as those with average incomes, but in three other provinces (British Columbia, Saskatchewan, and New Brunswick) they pay about 10 per cent more, 20 per cent more in one (Manitoba), and almost 40 per cent more in another (Nova Scotia).
In sum, there are enormous differences in the amounts paid by married couples, depending on the province of residence and whether one spouse is in care, or both. When only one spouse is in care, the charge for those with very low incomes ranges from less than $2,500 in Nova Scotia to almost five times that much in Saskatchewan and Quebec (for those with sufficient assets). If both are in care, the charges would be at least twice as great in Manitoba and all of the Atlantic provinces as in Quebec (for those with no assets), and only 10 per cent more in Alberta.
Costs Relative to Income
As we have seen, the amounts charged for LTC vary across provinces for individuals in differing income and marital situations but so too does the amount of tax paid on a given level of income. Furthermore, some provinces base charges on the resident’s before-tax income, others on after-tax income.
The private costs of LTC relative to after-tax incomes are compared in Table 5 for non-married persons and Table 6 for those married with both spouses in care and in Figure 2 for both. A number of interesting features are apparent. First, we would expect to find that those with lower levels of income would spend a larger fraction of their incomes on care than those with higher incomes. That turns out to be true in all provinces for those with levels of income that are average or higher, whether married or not. However, it is not always the case. For example, those with very low incomes, whether married or not, spend a smaller fraction of their incomes on care than do those with low incomes in each of the four Atlantic provinces and also in Alberta and Ontario.
Figure 2: Private cost of long-term care by marital status, income level and province, 2008: Percent of after-tax income.
Table 5: Private cost of long-term care as percent of after-tax income, by income level and region, 2008: Non-married seniors
Table 6: Private cost of long-term care as percent of after-tax income, by income level, care requirements and region, 2008: Married seniors
Second, the fraction of income that is spent on care, at all levels of income, varies greatly from one province to another. For example, for those married with “very low” income levels and with one spouse in care, the fraction varies from 11 per cent (Nova Scotia and Quebec) to 50 per cent or more (British Columbia, Saskatchewan, and Quebec). Similar large differences exist across provinces for those at higher levels of income, and for the non-married.
Third, within each province the fraction that those who are married spend on care is virtually the same at low and very low levels of income (even though it varies a great deal from one province to another). By contrast, again within each province, the fraction spent by those not married differs greatly: those with very low incomes spend 5 per cent more of their incomes on care than do those with low incomes in four provinces and 5 per cent less in four others.
Concluding Remarks
The spirit and intent of the Canada Health Act, in part, is to provide Canadians with reasonably equal access to health care services whether they are rich or poor and wherever they live. However, as we have demonstrated, the charges incurred by residents of LTC facilities and their families differ greatly from one province or territory to another and, within most jurisdictions, depend on income, marital status, and in Quebec alone, also on asset holdings. Our calculations relate to representative individuals and couples, taking into account their marital status and income, the levels of which range from very low to quite high. In assessing the implications of alternative income levels, we have assumed that there are no dependents to take into consideration, other than possibly a spouse. Also, the lowest level of income that we consider is one that is assured to all over the age of 65 who have lived in Canada for at least 10 years. Those who immigrated shortly before reaching that age would not be assured of having even that minimal level of income. Exceptional circumstances are taken into account in some jurisdictions, but we have not taken them into consideration here.
However, for the representative individuals and couples considered, it is clear that almost all older people in need of LTC at the lower end of the income distribution range can afford access, in the sense that in almost all jurisdictions the amounts charged for their care is less than the benefit payments that they receive from the combined OAS and GIS programs. At the same time, they are left with much more room for discretionary spending in some jurisdictions than in others after paying their costs of care. At one extreme, no charge is imposed in Nunavut; that leaves those with very low incomes who are not married $13,039 to spend each year while those married would have $21,930 if both are in care. Among the provinces, individuals in similar circumstances, however, would be left with much less: a non-married person in Prince Edward Island would have only $20; married couples in Manitoba would have to come up with an additional $456 whether or not they have assets on which to draw; and in Quebec, those with sufficient assets would have to pay an additional $2,306. At the other extreme, similar married couples resident in Alberta would be left with $6,360 to spend in ways of their own choosing. The differences are greater at higher levels of income.
Such differences across jurisdictions might be regarded as unfair, but they raise other considerations as well. Among them, we note that residency requirements tend to keep those already in LTC facilities from moving, perhaps to be closer to family. But lower charges could also be an attraction: forward-looking individuals might consider moving from higher to lower cost areas to benefit in the future from the lower rates that apply elsewhere. Consider an average income couple, one of whom is in care. After paying for the cost of LTC such a couple would have discretionary spending of more than $43,000 in Alberta but less than $20,000 in Newfoundland.
We do not know if people actually move (or fail to move) for such reasons. Indeed, we know remarkably little at an aggregate level about the characteristics of those in LTC in any part of Canada. So far as we have been able to determine, there is little published or even publicly available information about this population. From the Statistics Canada Residential Care Facilities Survey, we have estimates of the total number of residents in such institutions and the numbers of men and women in each age group, but we know nothing of their marital status and prior place of residence, their health care needs, and what family and other social support they have. That represents an important gap in our knowledge and statistical systems, one that could be remedied by analysis of existing administrative data combined with appropriate on-going surveys.
