Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
The Canada Health Act, which was adopted in 1984, mandates universal rights of access to publicly funded medically necessary health care, free of financial or other barriers. No one may be discriminated against on the basis of factors such as income, age and health status.1
Among the models that have been devised to examine the association between the need for health care and the use of services is that proposed by Andersen,2,3 which assumes that three types of factors come into play when individuals seek care: the state of their health, their predisposition toward using services, and their ability to obtain services. These factors are categorized as: need, predisposing and enabling.
Need factors are the individual’s perceived illnesses and illnesses diagnosed by health care professionals. Predisposing factors are characteristics of the individual that existed before the onset of illness, such as age, sex and race. Enabling factors include education, income and access to health care providers and health facilities.
This article, based on the Andersen model, examines the use of general practitioners and specialists by Canadians aged 18 or older (see Methods). Because the factors that are important for seniors when they seek health care may differ from those that are important at younger ages, separate analyses were conducted for the 18-to-64 and the 65-or-older age groups.
Since the Canada Health Act came into effect, numerous studies have focused on socio-economic advantage or disadvantage in relation to the use of services.8-19 While this analysis, too, looks at associations between household income and physician consultations, it also examines variations by sex, age, racial/cultural group, language, having a regular doctor, and urban/rural residence. Emphasis is placed on determining if predisposing and enabling characteristics are associated with physician consultations, independent of need (chronic conditions and self-perceived general and mental health) (see Definitions).
Majority consulted a general practitioner
Majority consulted a general practitioner
Canadians’ initial contact with the health care system is frequently through a general practitioner GP; GPs are also the main gatekeepers for specialist services.
In 2005, 77% of people aged 18 to 64 (an estimated 15.8 million) reported having consulted a GP at least once in the previous year, and 25% of them had done so four or more times (Chart 1).
GP contacts were even more common among seniors. Almost 9 out of 10 people aged 65 or older (an estimated 3.4 million) reported having consulted a GP, and 44% had had four or more contacts.
Smaller proportions of the population reported specialist consultations. Just over one-quarter of people aged 18 to 64 and more than one-third of seniors had seen a specialist at least once in the previous year.
Strong association with need
As might be expected, the likelihood of consulting physicians was strongly related to the presence of chronic conditions and to self-perceived health. And indeed, this is in line with the intention of the Canada Health Act, which aimed to provide access to care based on health status or “need.”
Among people aged 18 to 64, 72% with no chronic conditions had consulted a GP in the previous year, compared with 94% of those with three or more conditions (Table 1). Similarly, about 75% who described their general or mental health as excellent or very good had been to a GP, whereas the figure was around 86% for those whose general health or mental health was fair or poor. Associations between health status and GP consultations were the same for seniors. As well, for people in both age ranges, the percentages reporting multiple GP visits or specialist consultations increased with the number of chronic conditions, and were greatest among those with fair or poor general or mental health.
Of course, the likelihood of having chronic conditions or of reporting fair or poor health is not the same for everyone. For example, the number of chronic conditions tends to rise with age, and fair or poor health is more prevalent among people in lower income households and in rural areas (Appendix Table A and Appendix Table B). As well, the prevalence of chronic conditions and fair or poor health is high among some visible minorities, notably Aboriginal people.
When sex, age, household income, residence and race (as well as language and having a regular family doctor) were taken into account, chronic conditions and self-perceived health continued to be potent predictors of doctor consultations (Table 2). However, the strength of the associations diminished — invariably, the odds that people with the greatest “need” (as indicated by the presence of chronic conditions and fair or poor health) would consult physicians were substantially reduced (see Methods). For instance, at ages 18 to 64, the unadjusted odds of a specialist consultation for an individual with at least three chronic conditions were four and a half times greater than the odds for someone with no chronic conditions. When the effects of the predisposing and enabling factors were controlled, the odds, while still greater, fell to about three times those of someone with no chronic conditions. Among seniors, the corresponding odds ratio dropped from 3.47 to 2.79.
The remainder of this analysis examines how these predisposing and enabling factors were related to the use of GPs and specialists in Canada, when controlling for need.
Consultations and age
Because advancing age is associated with declining health (Appendix Table A and Appendix Table B), physician consultations tended to increase at older ages (Appendix Table C and Appendix Table D). But when the level of need and the other characteristics were controlled, the relationship between age and physician consultations was less clear.
In fact, among 18- to 64-year-olds, the age gradient was no longer evident (Table 3). Compared with people aged 18 to 24, only 25- to 34-year-olds had high odds of reporting a GP consultation or multiple GP consultations, and this largely reflected frequent use of health care services by women around the time of childbirth. When women who were pregnant at the time of their CCHS interview and those who had given birth in the previous year were excluded from the analysis, 25- to 34-year-olds no longer had significantly high odds of a GP consultation or multiple GP consultations (data not shown).
The relationship between age and specialist consultations was different from that for GP consultations. Among people aged 18 to 64, 25- to 34-year-olds had significantly high odds of having consulted a specialist compared with 18- to 24-year-olds (Table 3). Even when women who were pregnant and those who had recently given birth were excluded, the odds were reduced, but remained significantly high. Among seniors, the odds of a specialist consultation were actually lower for those aged 75 or older, compared with 65- to 69-year-olds.
Higher among women
Women have consistently been found to use medical services more often than men do.12-14, 20,21 According to the results of the 2005 CCHS, even allowing for the effects of chronic conditions, self-perceived health and the other factors, the relationship between sex and GP consultations persisted at ages 18 to 64 (Table 3). Compared with men, women in this age range had high odds of reporting a GP consultation, multiple GP visits and a specialist consultation. Although the odds were reduced, these findings held when those who were pregnant or who had given birth in the previous year were excluded (data not shown).
By contrast, among seniors, when chronic conditions, self-perceived health and the other factors were taken into account, senior women’s odds of having consulted a GP or reporting multiple GP visits were statistically similar to the odds for senior men (Table 4). And the odds that elderly women had consulted a specialist in the previous year were significantly lower than the odds for elderly men.
Household income and education
Earlier studies have documented associations between the use of health care services in Canada and socio-economic factors, even after the introduction of universal health insurance.9-18 Data from the 2005 CCHS support these findings, at least with regard to physician consultations.
Univariate analyses indicate that people aged 18 to 64 in the highest income groups were more likely than those in the middle income group to have consulted a GP in the previous year, while those in the lowest income households were less likely (Appendix Table C). For seniors, the income gradient was not as strong; only those in the lowest income households had a significantly low rate of GP consultations (Appendix Table D). Associations between GP use and education were also evident in both age groups; people with less than secondary graduation were less likely to have consulted a GP, compared with those with postsecondary graduation.
In the multivariate model, which controlled for need and other factors, the relationship between household income and GP consultations persisted for 18- to 64-year-olds, and became even stronger for seniors (Table 3 and Table 4). Education was not considered in the multivariate analysis because of its high correlation with income.
In the univariate analyses, for both age groups, multiple GP consultations were most common among people in low income households. (The same was true for low education.) When need and the other factors were considered, the income gradient was no longer evident for seniors, but for 18- to 64-year-olds, those in both lower and upper income households were more likely than those in middle income households to report multiple GP consultations.
For specialist contacts, the relationship with household income was clear. When the effects of need and the other factors were taken into account, at ages 18 to 64, the odds of reporting a consultation were significantly high for people in upper-middle and highest income households, compared with those in middle-income households (Table 3). Among seniors, the odds of a specialist consultation were significantly high for people in higher income households, and significantly low for those in the lowest income households (Table 4).
At ages 18 to 64, the odds that members of visible minority groups would report a GP consultation were statistically similar to those for Whites when need and factors such as age and household income were taken into account (Table 3). However, the odds of multiple GP consultations were higher for Aboriginal people and other visible minorities, compared with Whites.
Specialist consultations were a different matter. Whether they were aged 18 to 64 or seniors, Aboriginal people, Blacks and other visible minorities had significantly low odds of having had a specialist consultation in the previous year.
Language has been cited as a potential barrier to the use of health care services,22 but according to the results of the 2005 CCHS, this was not the case for GP consultations. When need and the other factors were taken into account, at ages 18 to 64, the odds of consulting a GP were similar among those who could converse comfortably in English or French and those who could not. And people who could not converse in English or French had significantly high odds of reporting multiple GP consultations.
For seniors, the odds of a GP consultation and multiple GP contacts were not significantly related to language, but this was partly attributable to having “racial or cultural group” in the model. When that characteristic was excluded, the odds of a GP consultation and multiple GP consultations for seniors who could not converse in English or French were about twice those for seniors who could (data not shown).
When all the factors were considered, there was initially no relationship between specialist consultations and language. But when racial or cultural group was excluded, the odds of a specialist consultation were significantly low for 18- to 64-year-olds who could not converse in English or French (data not shown). The finding that language was not related to specialist service use among seniors persisted (data not shown).
The use of health care services has been shown to be associated with geographic location.23 Health care providers, especially medical specialists, tend to be concentrated in urban areas. For people in rural locales, access to such services is often inconvenient.24
The results of the 2005 CCHS show that rural residents were just as likely as urban dwellers to have GP consultations, even when need and the other factors were considered (Table 3 and Table 4). Moreover, rural residents in both age groups had significantly higher odds than did people in urban communities of having multiple GP consultations.
The use of specialist services, however, was lower among people in rural areas. Whether they were aged 18 to 64 or seniors, rural residents had significantly low odds of a specialist consultation, compared with people in urban areas.
Having a regular physician
In 2005, a substantial share of adult Canadians reported that they did not have a regular family doctor. At ages 18 to 64, the proportion was 16% (an estimated 3.3 million), and among the elderly, almost 5% (an estimated 186,000) (data not shown).
Not surprisingly, whether they were aged 18 to 64 or seniors, people without a family doctor were far less likely to report consultations with GPs, let alone specialists (Appendix Table C and Appendix Table D). However, these people also tended to be in better health — they were less likely than those who had a doctor to have three or more chronic conditions or to report fair or poor general or mental health (Appendix Table A and Appendix Table B). Yet even allowing for these need factors and the other characteristics, people who did not have a family doctor had significantly low odds of GP and specialist consultations.
According to results from the 2005 Canadian Community Health Survey (CCHS), individual health needs — as measured by chronic conditions and self-perceived general and mental health — were strong determinants of physician consultations. However, consistent with Andersen’s theory, when sex, age, race, language, household income, urban or rural residence and having a regular family doctor were taken into account, the strength of the associations between health need and physician consultations diminished. While chronic conditions and self-perceived health continued to be potent predictors, these other factors were independently related to the likelihood of going to the doctor, particularly specialists.
Some groups were relatively unlikely to consult specialists, even though such services are also covered by the provisions of the Canada Health Act. In a number of cases, these were the same groups who reported repeated visits to GPs. For instance, the odds of a specialist visit were significantly low for very old people, residents of low income households, visible minorities and rural residents. At the same time, very old people, other visible minorities, rural residents and people aged 18 to 64 who were Aboriginal or lived in low income households all had high odds of reporting four or more GP consultations.
About 3.5 million Canadian adults do not have a regular family doctor. While this group tended to be in relatively good health, even when that was taken into account, they were particularly unlikely to have had a physician consultation.
Twenty years after the introduction of the Canada Health Act, several factors beyond need were significantly associated with the likelihood of having seen a doctor. The results of this analysis indicate that socio-economic status remains a factor in the use of physicians’ services. In addition, several other factors — sex, age, race, language, and residence — were associated with individuals’ likelihood of consulting a doctor, independent of the state of their health.