Health Reports
Hospitalization rates among economic immigrants to Canada

by Edward Ng, Claudia Sanmartin and Douglas G. Manuel

Release date: July 19, 2017

Health research recognizes that immigrants to Canada are not a homogeneous group and has revealed differences in outcomes by characteristics such as place of origin and period of arrival.Note 1Note 2 Owing to a lack of data, few studies have focused on health outcomes by immigrant category (family class, refugees, economic class), although it is an important dimension of immigrant research.Note 3Note 4 Family class immigrants and refugees are admitted for reunification and compassionate reasons. Economic class immigrants are selected based on a point system designed to assess their skills and potential contribution to the economy.Note 5

With the development of new data sources, it has been possible to examine health outcomes for specific immigrant categories, notably, refugees, who are perceived to be the most vulnerable.Note 6Note 7 Less attention has been paid to economic class immigrants, although they constitute the majority of recent arrivals (55% in 2006, compared with 13% for refugees).Note 8 An exception was a study that linked landing file information to provincial administrative data and showed substantial heterogeneity within the economic class in their use of medical services. However, the results were preliminary and represented only two provinces—Manitoba and British Columbia.Note 9

Economic immigrants comprise three subcategories: skilled workers, business immigrants, and live-in caregivers.Note 8 Skilled workers and business immigrants are selected for their ability to establish themselves in the labour market or through entrepreneurial activities. Under the Live-in Caregiver Program, temporary foreign workers are admitted to care for children, seniors, or people with disabilities in private households. Program participants may apply for permanent resident status within three years, upon completing two years as a live-in caregiver.Note 10

When they arrive, immigrants tend to be in good health relative to the Canadian-born—a phenomenon known as the “healthy immigrant effect.” The rate at which their health declines over time can vary.Note 1Note 11 For example, the financial investment required to be admitted as a business immigrant suggests that such individuals have considerable wealth, which may be associated with access to health care and a more favourable health profile.Note 9 By contrast, qualitative studies have reported that live-in caregivers may face employment conditions that pose health challenges,Note 12Note 13Note 14 although little quantitative evidence is available to compare their health with that of business immigrants and skilled workers.

Because immigration applications are screened mainly through principal applicants, the present study focuses on this group, who constitute about 40% of the economic class.Note 8 With information from linked databases, hospitalization rates of economic class principal applicants (ECPAs) aged 25 to 74 were calculated by subcategory and sex, and compared with those of the Canadian-born population and between the subcategories.

Hospitalization is an imperfect health indicator, in that it does not capture less severe illnesses that do not warrant hospital admission. Nonetheless, hospital costs account for the largest share of Canada’s health care spending (close to 30% of the total health expenditure of $219 billion in 2015).Note 15 It is, therefore, important to understand patterns of use by groups such as immigrants.

Data and methods

Data sources

The Immigrant Landing File (ILF), a national database provided to Statistics Canada annually by Immigration, Refugees and Citizenship Canada (formerly Citizenship and Immigration Canada), is a census of immigrants who arrived since 1980. The ILF includes date of entry, source country, admission category (economic class, family class, refugee), and principal applicant status.Note 16 Landing records for 1980 through 2008 were used for linkage.

The Discharge Abstract Database (DAD) is a census of discharges from all public hospitals (excluding Quebec), which is provided to Statistics Canada by the Canadian Institute for Health Information.Note 17 It contains demographic, administrative, and clinical data for about 3 million discharges annually. Hospital discharges that occurred from 2006/2007 through 2008/2009 were used for linkage.

The ILF-DAD linkage was performed through deterministic exact matching with the 2006 Census as a “bridge” file.Note 18Note 19 Of 2.6 million eligible ILF records, 10% were linked to at least one DAD record between 2006/2007 and 2008/2009. Validation indicated that the linked ILF-DAD file is representative of recent immigrants residing in Canada in 2006 and their hospital experiences.Note 20 Details about the linkage are available elsewhere.Note 21Note 22

The long-form census questionnaire is administered to 20% of the non-institutional population. Linked 2006 Census-DAD data were used to estimate hospitalizations among the Canadian-born population. Using hierarchical deterministic matching, about 4.6 million long-form respondents (excluding Quebec) were linked to the DAD for the three fiscal years from 2006/2007 through 2008/2009, based on birthdate, sex, and residential postal code. Validation concluded that the linked file is suitable for health-related research.Note 19

Each linkage was approved by Statistics Canada’s Executive Management Board.Note 22 Use of these linked data is governed by the Directive on Record Linkage.Note 23 Statistics Canada ensures respondent privacy during the linkage and subsequent use of linked files. Only employees directly involved in the process have access to the unique identifying information (such as names and sex) required for linkage and do not access health-related information. When linkage is complete, an analytical file is created from which identifying information has been removed. The present analysis used this de-identified file.

Study cohorts

This study is based on two cohorts representing the immigrant and Canadian-born populations. The immigrant cohort was created from the linked ILF-DAD data. Inclusion criteria were all EPCAs aged 25 to 74 residing outside of Quebec at the time of the 2006 Census. The cohort was composed of 507,405 ECPAs, of whom 54,445 were linked to hospital discharges during the three years of follow-up.

The Canadian-born cohort was created from the linked Census-DAD data. Inclusion criteria were all Canadian-born individuals aged 25 to 74 residing outside of Quebec. The cohort was based on a weighted population of 10,031,000 derived from the sample of 2,063,100 records, of which 517,350 were linked to hospital discharges during the three years of follow-up.


ECPAs were divided into skilled workers, business class immigrants, and live-in caregivers. A small number who had been nominated for admission by provinces under a new stream (provincial nominee program) were grouped with skilled workers.

Canada has long recruited temporary foreign workers for domestic employment.Note 10Note 24 On April 27, 1992, the Foreign Domestic Workers Program was replaced by the Live-in Caregiver Program. In the overall analysis, the term “live-in caregiver” includes those who arrived under the Foreign Domestic Workers Program. To differentiate between the two groups, landing year was categorized as arrival before 1993 or not. Because the vast majority of live-in caregivers are women, male live-in caregivers were not included in the detailed analysis.

Education was defined as having at least a bachelor’s degree (university graduation) versus less at the time of application Official language proficiency was reported knowledge of English and/or French at the time of application. World regions were: Caribbean, Central and South America, Western Europe, Eastern Europe, Southwest Asia and Africa, South Asia, Southeast Asia, East Asia, and other (including the United States and Oceania).Note 1

Based on the distribution of ECPAs by sex and subcategory, age was grouped as: 25 to 44, 45 to 54, and 55 to 74.

Statistical methods

Descriptive statistics were used to profile ECPAs by subcategory. The primary outcome was at least one inpatient acute care hospital discharge during the period from April 1, 2006 through March 31, 2009. Age-standardized hospitalization rates (ASHRs) were annualized and derived for all causes (excluding pregnancy) and for the leading causes, based on the most responsible diagnosis as defined by the International Classification of Diseases Version 10.Note 25 The leading causes were circulatory diseases (I00 to I93), digestive diseases (K00 to K93), and cancer (C00 to D48). The Canadian-born population was used as the reference for age-standardization and as the reference category for ASHR comparisons.

Statistical testing for differences in ASHRs was conducted using logarithmic transformation to adjust for the skewness in the distribution of the standardized rates in analysis of rare events.Note 26 Logistic regression was used to derive sex-specific age-adjusted and fully adjusted odds ratios for overall and cause-specific hospitalizations by ECPA subcategory, with skilled workers as the reference category. The covariates in the fully adjusted model were education, official language proficiency,Note 27 and world region.Note 1Note 2 To isolate differences between the female live-in caregiver and foreign domestic worker populations, logistic regression analyses were conducted by landing cohort.


Description of cohorts

The distributions of male and female ECPAs by subcategory differed. Most (88%) male ECPAs aged 25 to 74 were skilled workers. Another 12% were business immigrants; very few (0.3%) were live-in caregivers (Table 1).

Similarly, among female ECPAs, the majority (69%) were skilled workers, but more than a quarter (27%) were live-in caregivers. The remaining 4% were business class immigrants.

Except for business class immigrants (half of whom were aged 55 to 74), ECPAs tended to be younger than the Canadian-born population.

The percentage of ECPAs who were university graduates (bachelor’s degree or more) was highest among skilled workers—73% of men and 66% of women. As well, the percentages of business class immigrants and live-in caregivers with university degrees (36% to 42%) exceeded the figure for the Canadian-born population (about 20%).

Official language proficiency was highest among live-in caregivers, about 95% of whom reported knowledge of English and/or French. More than 80% of skilled workers reported knowing English and/or French; the percentage for business class immigrants was about 50%.

For male ECPAs, the leading world regions were East Asia (31%), South Asia (19%) and Southwest Asia and Africa (13%). The leading world regions for female ECPAs were Southeast Asia (31%), East Asia (25%) and Eastern Europe (10%). The first-place rank of Southeast Asia was due to live-in caregivers from the Philippines.

The majority of ECPAs landed after 1992 (71% and 66% of men and women, respectively).

ASHR lower than Canadian-born

The all-cause ASHR of male ECPAs was less than half that of Canadian-born men (301 versus 665 per 10,000) (Table 2). This pattern prevailed among skilled workers and business class immigrants, and for circulatory diseases and digestive diseases.

Among female ECPAs, the all-cause (excluding pregnancy) ASHR was significantly lower than that of Canadian-born women (310 versus 668 per 10,000). Again, the differences were significant for each ECPA subcategory and cause of hospitalization.

The rank order of the leading causes of hospitalization was the same among male ECPAs and Canadian-born men: circulatory diseases followed by digestive diseases. However, for female ECPAs, cancer ranked first, followed by digestive diseases, whereas for Canadian-born women, digestive diseases ranked first, followed by cancer.

Differences between subcategories

Among male ECPAs, business class immigrants were significantly less likely than skilled workers to be hospitalized (OR = 0.86, 95% CI = 0.82 to 0.90) (Table 3). Patterns were similar for circulatory disease and digestive disease hospitalizations, but when education, official language proficiency and world region were taken into account, differences between the two subcategories were no longer statistically significant.

Among female ECPAs, business class immigrants had significantly lower odds of all-cause hospitalization than did skilled workers (OR = 0.75, 95% CI = 0.68 to 0.84), while live-in caregivers had significantly higher odds (OR = 1.10, 95% CI = 1.05 to 1.11). Results were similar for digestive disease hospitalizations, but not for cancer, where no significant difference was found among the ECPA subcategories. When the models were fully adjusted, female business class immigrants continued to have lower odds of all-cause hospitalization than did skilled workers, but the odds for live-in caregivers were no longer statistically different. However, for digestive diseases, significantly higher odds of hospitalization among live-in caregivers persisted (OR = 1.22, 95% CI = 1.05 to 1.40).

Live-in caregivers

An analysis of ASHRs for female ECPAs by landing cohort indicates the influence of replacement of the Foreign Domestic Workers Program by the Live-in Caregiver Program in 1992 (Table 4). The odds of all-cause hospitalization for women who entered Canada under the Foreign Domestic Workers Program were not significantly different from the odds for female skilled workers who arrived in that period (OR = 1.03, 95% CI = 0.96 to 1.10), regardless of education, official language proficiency, and world region. This was also true for cancer hospitalizations. However, for digestive diseases, full adjustment resulted in marginally higher odds of hospitalization among Foreign Domestic Workers (OR = 1.24, 95% CI = 1.01 to 1.51).

Women who arrived after 1992 under the Live-in Caregivers Program had significantly higher odds of all-cause hospitalization and hospitalization for digestive diseases than did female skilled workers. When education, official language proficiency and world region were taken into account, the odds were no longer significantly different.


This is the first national study to examine hospitalization rates of economic immigrants by subcategory. Compared with the Canadian-born population aged 25 to 74, ECPAs had a low rate of hospitalization, although differences emerged between subcategories. Adjustment for education, official language proficiency and world region had differential impacts on the results―an indication of the importance of taking these factors into account.

The low hospitalization rates among ECPAs relative to the Canadian-born population are consistent with the “healthy immigrant effect.” The tendency for immigrants, especially new arrivals, to be healthier may be due to self-selection and to medical screening.Note 1Note 11 As well, the employability (or economic contribution) requirement for ECPAs may play a role, with the “healthy worker effect”Note 28 enhancing the healthy immigrant effect. To disentangle the healthy worker and healthy immigrant effects, further analysis might compare ECPAs and a Canadian-born subpopulation with similar employability.

The rank order of leading causes of hospitalization was the same for male ECPAs and Canadian-born men. By contrast, for female ECPAs, cancer ranked first and digestive diseases second, whereas the order was reversed among Canadian-born women. An earlier study based on the same linked data showed that female ECPAs had higher cancer-related ASHRs than did their family class and refugee counterparts.Note 21 That study also reported higher ASHRs for cancer than for digestive diseases among recent female immigrants from Asia, Eastern Europe, and the Caribbean. The present analysis found a high ASHR for cancer among female ECPAs, not only for live-in caregivers, but also for skilled workers and business class immigrants, many of whom were from East Asia, a group with somewhat higher cancer mortality among recent immigrants.Note 1

A number of studies have reported that immigrants may not receive necessary health care because of linguistic, cultural and other barriers.Note 29Note 30Note 31 Live-in caregivers may experience more of these barriers.Note 10Note 12Note 13Note 14Note 32Note 33Note 34Note 35 Also, according to the Immigration Refugee Protection Regulations of 2002, live-in caregivers who applied for immigration were exempt from the medical examination requirement, as they would have undergone a previous examination when applying for temporary residency.Note 36 The present analysis showed that, compared with skilled workers, live-in caregivers had higher age-adjusted odds of hospitalization overall and for digestive diseases, but adjustment for education, official language proficiency and world region yielded odds that were not significantly different. This suggests that the differences may be related to their origins in countries where digestive diseases are more prevalent. Alternatively, the possibility of gene-environment interactions for some digestive diseases has been suggested.Note 37Note 38 Further study could control for the fact that, owing to their temporary worker status before immigration, live-in caregivers tended to have been in Canada longer than their skilled worker and business class counterparts. Therefore, at least for those who immigrated after 2002, medical examinations were conducted more than two years earlier than those of other ECPAs.

Strengths and limitations

The major strength of this analysis is the use of linked data from the Immigrant Landing File and the Discharge Abstract Database. This linkage makes it possible to examine hospitalization rates among subcategories of ECPAs by leading causes, and compare these rates with those for the Canadian-born.

A limitation of the analysis is the inability to track subsequent emigration. Immigrants are more likely than the Canadian-born population to leave the country for extended periods; this may be especially true for business class immigrants.Note 39 The linkage approach ensured that ECPAs were in Canada on Census Day (May 16 2006), but it is not known if they remained in Canada throughout the follow-up period (2006/2007 to 2008/2009). To the extent that ECPAs had a greater likelihood of being out of the country, their ASHRs will be underestimated.

Another limitation is the lack of hospital data from Quebec, which meant that immigrants residing in that province were excluded.


This study demonstrates the feasibility of using linked ILF-DAD data to examine patterns of hospitalization among immigrant subgroups. As Canada continues to meet labour market needs through immigration, there is a need to determine health outcomes for immigrants and live-in caregivers and to compare them with the Canadian-born population.

Recent policy changes have affected the admission of economic class immigrants. For instance, the provincial nominee program, which complements the Federal Skilled Worker Program, has grown, and the live-in requirement for caregivers was removed.Note 40 As well, temporary foreign workers and foreign graduates with qualifying work experience can apply to become economic immigrants under the Canadian Experience Class.Note 41 The impact of such developments on health care utilization requires additional research.

Date modified: