The findings from this study on whether markers of social inequality matter in assessing the health of immigrants indicate that the answer to this question is yes. The more important inequity question to consider, however, is how do they matter? In responding to this inquiry, we turn to a discussion on how the healthy immigrant effect plays out vis-a-vis these markers. The healthy immigrant effect applies to midlife males. Specifically, recent - those who immigrated less than 10 years ago - immigrant men between the ages of 45 and 64 years have better self-rated health compared to the Canadian-born. And, upon further examination, the results suggest that there is a convergence in health differences between foreign- and Canadian-born men in midlife. Interestingly, the health advantage of recent immigrants is especially strong for visible minorities, and is not accounted for by differences in age, economic status, or health behaviors between the immigrant/visible minority groups. This contradicts the argument that a healthier immigrant population can be attributed to advantages arising from such factors. In contrast, the findings are not consistent with a HIE among midlife women. Foreign-born women ages 45-64, regardless of ethnicity, years since immigration, or controls for demographic, economic, and lifestyle factors, are disadvantaged in health compared to Canadian-born women. This disadvantage may reflect differences in the health status of immigrants who enter Canada under different classes; that is, the poorer health status of midlife women, particularly visible minority women (i.e., South Asian, Chinese), may be partially attributed to the fact that they are more likely to have entered the country as family-sponsored immigrants than men (who enter under independent, professional or skilled worker or business classifications). Family-sponsored immigrants come in as "dependents" and this may indicate or suggest that they may not be as "physically" resilient as their "independent" (male) sponsors. This initial vulnerability may endure over the course of their mid-life years and well into later life, especially if access to health care services and programs is (and continues to remain) an issue due to larger cultural (i.e., incongruence in health beliefs, family versus individual decision-making processes) and/or social structural (i.e., lack of appropriate and/or adequate policies and programs) issues.
A different picture emerges in old age. For older men, recent immigrants, particularly visible minorities, are more likely to be disadvantaged with regard to self-reported health even after controlling for key factors. On the other hand, recent visible minority immigrant women in the latter stages of the life course fare much better on self-reported health. This advantage, however, disappears when the data are adjusted for other differences.
Based on these findings, a discussion of the implications for health care policy and program planning for immigrant men and women in mid- to late adulthood - individuals that make up over one-half of the foreign-born adult population in Canada and increasingly larger proportions of the populations in the US, UK and Australia - is warranted. In particular, the findings underscore the necessity for policymakers in such immigrant-receiving countries to address the differential health care needs of immigrant adults by gender and age group. Recent immigrant visible minority men in midlife and, to a lesser extent, their later life female counterparts may have fewer needs for services and programs in the early years of their residency, while certain new immigrant sub-groups, namely older men and midlife women of color may actually have increased needs for services due to poor health status at migration. It should be underscored that this increased need is likely to continue for these women as they age, especially if they experience social isolation (due to geographic, language and/or cultural barriers) and/or under- or unemployment for long periods of time (due to discrimination, etc.). In response to this reality, it is important that policies and programs be developed at both the national and province/state levels, particularly in geographic areas (i.e., around urban centers) in which the majority of new immigrants often choose to reside, that: (a) target midlife immigrant and certain sub-groups of older immigrant women as they age over time; and (b) respond to the needs of an older immigrant male population from the outset.
Specific policy recommendations include the need to actively incorporate a health promotion framework in public health policy. To this end, policymakers must move beyond the funding of large-scale health promotion programs that mainly target children and adolescents in schools, i.e., Participaction, to developing programs that are relevant and accessible to the increasing number of adult Canadians, a significant proportion of whom are foreign-born, who are aging with or at significant risk of developing chronic disease. Complementing a dual focus on prevention and treatment, such a policy agenda calls attention to a broad range of social determinants of health and illness that differentially affect the health care utilization patterns and health status of immigrant men and women at various stages of the adult life course.
Finally, despite differences in demographic composition and policy frameworks in the immigration and health care domains of the UK, US, Canada and Australia, Kennedy, McDonald and Biddle (2006) find that there is "evidence of strong positive selection effects for immigrants from all regions of origin in terms of education" in their study of the HIE . This finding provides some empirical support for the cross-national application of the current study's findings and subsequent policy discussion to other large immigrant-receiving countries.
a. Limitations of the Study
Although the Canadian Community Health Survey (CCHS) provides information on the health status and health care needs of adult Canadians, there are a number of limitations in using these data for this study. First, despite the fact that its data allows for an examination of health status and health care utilization among immigrants, the survey does not collect information on immigrant status or on the reasons for immigrants' entry into Canada. Hence, a more detailed analysis of immigrant men and women's health is not possible; that is, important variations in health status among naturalized citizens, landed immigrants, refugees, and non-permanent men and women cannot be examined in this study.
Second, while CCHS respondents who could not understand English or French were interviewed in their own language, linguistic (as well as cultural) barriers faced by new immigrants may still prevent them from consulting health-care professionals, resulting in an under-diagnosis of health problems (Laroche, 2000) . Cultural factors like adherence to traditional values and beliefs may also influence an individual's willingness to report health problems (Ali 2002; Kopec, Williams, To, and Austin, 2001), since there may be differences in their fundamental conceptualizations of health and illness (Saldov, 1991) [17–19]. Subjective measures of health, like self-rated health, may be affected by differences in "thresholds" used by individuals or groups in assessing their health status (Franks, Gold, & Fiscella, 2003; Schnittker, 2005; Simon, De Boer, Joung, Bosma, & Mackenbach, 2005) [20–22]. It is not unreasonable to assume that the meaning, interpretation, and reporting of self-rated may change across age groups, cultures, and ethnicities.
The extent to which cultural and language differences in the Canadian population influence the interpretation and reporting of health problems is not well known. The magnitude of the differences in men's and women's health status between immigrant and Canadian-born populations reported here, however, make it unlikely that cultural factors exclusively may explain these results.
Third, despite the evidence provided in this study, longitudinal data are needed to verify a true convergence in health status between immigrants and native-born persons over time. It is not possible with the cross-sectional data used here to rule out a cohort effect, whereby differences in men's and women's health among immigrant groups are partly due to the country of birth of immigrants. Longer-term immigrants are more likely to be from Europe and recent immigrants from non-European regions, and both regions vary in terms of general population health - today's immigrants may make-up a healthier cohort than cohorts who immigrated earlier - and in the type and quality of health care systems. Health requirements for entry into Canada (as well as the US, UK and Australia) have also changed, i.e., become more stringent, over time (Perez, 2002) . It should also be noted that, as we are unable to acquire standardized health status and utilization data pre-immigration from all source countries, the validity of findings on the HIE may be called into question even if longitudinal data were collected post-immigration.
Finally, the CCHS data used for this study are limited in two ways. First, age is defined in five-year groups (e.g., 45-49 years) as opposed to respondents' actual age. Subsequently, some of the key variations between immigrants and non-immigrants may be due to small differences in the average age of respondents within each of their age cohort groups. Second, these data do not allow for the consideration of a key variable such as ethnicity (i.e., country of birth) as both a control and independent variable in the current analyses.