Social determinants of health in Canada’s immigrant population: results from the National Population Health Survey
Introduction
As part of the Metropolis project — a large-scale investigation of immigration and integration, including well-being of immigrants in a number of areas of social life — in this paper we investigate the social determinants of health in Canada’s immigrant population using Canada’s National Population Health Survey (NPHS). There are a wide variety of potential reasons for studying the health of immigrants as a distinct group. Immigrants coming from countries with poorer health status may be of concern with respect to their future health care utilization, for instance. Usually, however, the opposite effect — a ‘healthy immigrant’ effect — is seen due to medical testing used to screen immigrants for entry. Yet another reason to study immigrant health is that it represents a ‘natural experiment’ that allows for the investigation of the effects of social environments on human health (see for example Marmot et al., 1975, Nichaman et al., 1975, Syme et al., 1975, Worth et al., 1975). In this paper, we undertake the following investigations: (a) examine differences in health status and health care utilization between immigrants and non-immigrants, immigrants of European and non-European origin, and immigrants of <10 years and >10 years residence in Canada; (b) examine the social determinants of health care utilization and health status in immigrants and non-immigrants, and (c) evaluate the utility of large-scale, national databases for these purposes.
In this paper, we specifically adopt a ‘population health’ or, alternatively, a ‘social determinants of health’ perspective. This is a perspective based on a synthesis of a diverse public health and social scientific literature, which suggests that the most important antecedents of human health status are not medical care inputs and health behaviours (smoking, diet, exercise, etc.). Rather, they are social and economic characteristics of individuals and populations (Evans, Barer & Marmot, 1994; Frank, 1995, Hayes and Dunn, 1998). The mechanisms believed to produce health inequalities which are summarized by socio-economic status can be expected to be influential across class, gender, race, ethnicity, language, and age differences, as well as socio-economic differences. For immigrants, however, the influence of socio-economic factors is not well-understood.
We use data from the National Population Health Survey (1994–95) to investigate the determinants of health status, chronic conditions, overnight hospitalization, and reporting of unmet needs for care, making comparisons between immigrants and non-immigrants. Explanatory variables considered include individual socio-economic characteristics, immigration characteristics, health behaviour indicators, and social support measures. The results are ambiguous. We find that the NPHS data are of limited utility for this purpose and that existing conceptual and theoretical development of the social determinants of immigrant health is equally constraining.
Section snippets
Past research on immigrant health status
Past literature on the social determinants of health in immigrants is sparse, and the results ambiguous. Although there is a vast literature on differences in health status of groups categorized according to ‘race,’ particularly in the United States, it is largely unhelpful for our purposes. First, it tends to conflate ‘race’ and immigrant status, and rests on an unexamined assumption that the differential distribution of health status could be reduced to biological and genetic factors (
The population health perspective: background and context
In recent years, Canadian health policy discourse has increasingly been characterized by a concern with the ‘social determinants of health,’ as evidenced by its strong presence in health policy rhetoric and academic research (see Bhatti and Hamilton, 1996, Hayes and Dunn, 1998; Hayes, 1994, Hayes, 1999; Evans et al., 1994; British Columbia, 1993, British Columbia, 1995, British Columbia, 1996, British Columbia, 1997; Hertzman and Weins, 1996, National Forum on Health, 1997; etc.). Analysis of
Materials and methods
The National Population Health Survey (1994–95) Public Release ‘Health File’ consists of data from telephone interviews with 17,626 Canadians, including 2400 immigrants. In addition to the two official languages of Canada, English and French, interviews were also completed in Spanish, Portugese, Italian, Chinese and Punjabi. Because it is not well-understood how social status differences manifest in the health of teenagers (West, 1997), and because of limitations of the public-use NPHS file, we
Descriptive results
Table 2 shows the relative distribution of explanatory variables between: the immigrant and Canadian-born samples; immigrants of ‘European’ (Europe, Australia, United States) and ‘non-European’ (Asia, Africa, South America) origins; and immigrants of <10 years and >10 years. The parameters for this table were selected for their theoretical importance as provided by the population health perspective and knowledge of immigrant integration. Past research has suggested that it is after a period of
Immigration sub-groups and health outcomes
Table 3 shows differences between immigrants and native-born Canadians on the four health outcomes of interest. The table shows the percentage of the samples reporting each outcome. The number in parentheses beside each proportion is the standardized residual, indicating the difference between the observed and expected proportion of respondents in each category. The column labelled ‘test statistic’ gives the value of Kendall’s tau-c for each comparison, with the significance level indicated by
Logistic regression analyses
Multiple logistic regression allows for the investigation of the simultaneous influence of a group of explanatory variables on a health outcomes. It is useful for describing the simultaneous relationship of a group of continuous and/or categorical explanatory variables and a dichotomous outcome variable (DeMaris, 1995, Streiner, 1994, Wrigley, 1985).
The results of the logistic regression analyses done for this investigation appear in Table 6, Table 7, Table 8, Table 9. Table 6 shows the results
Self-rated health status
Focusing on the relative odds ratios greater than 1.5 (or <0.66), Table 6 suggests results that are roughly consistent with a population health perspective on the social determinants of health. The likelihood of reporting fair or poor health increases with age in both immigrants and non-immigrants, although the gradient is steeper amongst immigrants. Amongst immigrants, those who originated in Asia, Africa or S. America were more likely to report poorer health, as were those immigrants who had
Chronic conditions
Again focusing on relative odds ratios greater than 1.5, the likelihood of reporting a chronic condition increased with age for both Canadian-born and immigrant respondents, but a very steep age gradient was shown for immigrants (Table 7). Amongst immigrants, those for whom 10 years or more had elapsed since immigration were more likely to report a chronic condition than those in Canada less than 10 years. Canadian-born respondents living in Montreal or Vancouver were significantly less likely
Overnight hospitalization in the previous year
The determinants of overnight hospitalization in the past year are roughly similar between the two respondent groups of interest, but they do show some unusual results (Table 8). In both groups, being aged 30–49, 50–64, or 65+ meant that one was less likely to report an overnight hospitalization in the past year than respondents aged 20–29. Place of birth and length of time since immigration showed little influence on the likelihood of reporting a hospital admission, while both immigrants and
Unmet needs for health care
Based on relative odds greater than 1.5, the logistic regression models for immigrants and Canadian-born respondents on the outcome ‘unmet need for health care in the past year’ show a strong influence of social factors for immigrants, and a smaller impact for Canadian-born respondents (Table 9). There were strong and significant effects of age for both respondent groups, with immigrants showing a sharp gradient that indicate a reduced likelihood of having an unmet need for care with age.
Summary
The results of this analysis of data from the National Population Health Survey 1994–95 suggest that socio-economic factors are important to self-rated health status and presence of chronic conditions for both immigrants and non-immigrants, but more so for immigrants. Descriptive results of socio-economic characteristics show that in general, immigrants, compared to non-immigrants, are older, more likely to live in a major metropolis, less likely to be wealthy, more likely to have no schooling
Discussion
In short, data from the National Population Health Survey, show no obvious pattern of association between socio-economic characteristics and immigration characteristics on the one hand, and health status on the other. This does not mean that socio-economic factors in Canada are not influential in shaping immigrants’ health status. In fact, the results of the logistic regression models calculated for immigrants and non-immigrants on four outcome variables in this study suggest that
Acknowledgements
This research was supported by the Vancouver Centre of Excellence for Research on Immigration and Integration in the Metropolis, which is in turn supported by funds from the Social Sciences and Humanities Research Council of Canada, Citizenship and Immigration Canada, Simon Fraser University, The University of British Columbia, and The University of Victoria. James R. Dunn was supported by a Doctoral Fellowship (♯ 752-94-1249) and a Postdoctoral Fellowship (♯ 756-98-0194) from the Social
References (54)
- et al.
Child development and long-term outcomes: A population health perspective and summary of successful interventions
Social Science and Medicine
(1996) - et al.
Self-rated health status as a health measure: The predictive value of self-reported health status on the use of physician services and on mortality in the working-age population
Journal of Clinical Epidemiology
(1997) - et al.
Socioeconomic inequalities in health: Some international comparisons
Journal Health Economics
(1997) Health inequalities in the early years: is there equalisation in youth?
Social Science and Medicine
(1997)Using ethnicity as a classification variable in health research: Perpetuating the myth of biological determinism, serving socio-political agendas, or making valuable contributions to medical sciences?
Ethnicity and Health
(1999)The idea of Chinatown: The power of place and institutional practice in the making of a racial category
Annals of the Association of American Geographers
(1987)Vancouver’s Chinatown: Racial Discourse in Canada, 1875–1980
(1991)- et al.
Immigrant and refugee children in Canada
Canadian Journal of Psychiatry
(1995) - et al.
Mental health of immigrants and refugees
- et al.
Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents
American Journal of Epidemiology
(1979)