COVID-19 vaccine coverage among immigrants and refugees in Alberta: a population-based cross-sectional study

Setting

Alberta is a western Canadian province of 4.5 million residents, 99% of whom are registered with the publicly funded Alberta Health Care Insurance Plan (AHCIP). While the Canadian government oversees vaccine approval and procurement, provinces are responsible for the roll-out of vaccination programs.(10,11) COVID-19 vaccination programs in Alberta began in December 2020 in a phased manner, with vaccine availability to the entire population aged 12 and above by mid-May 2021. Records for all COVID-19 vaccines administered in the province, regardless of provider, are submitted to the provincial immunization repository, known as Imm/ARI (with a few exceptions, noted below).

Study design, population, and data sources

This was a population-based cross-sectional study using administrative data held by the Alberta Ministry of Health and including all residents of Alberta aged 12 years and above. Albertans <12 years were excluded because they were not eligible to receive the COVID-19 vaccine during the study period. We used the AHCIP quarterly population registry for the last quarter of 2020 to identify residents of the province, irrespective of birthplace. We excluded First Nations residents of Alberta (since data was not consistently submitted to Imm/ARI), Lloydminster residents (since vaccines are delivered by the neighbouring province), and those who left the province or died during the study period. Figure A1 in the appendix provides further detail on exclusions. We used the provincial immigrant registry to identify foreign-born immigrants and refugees who migrated to Alberta from another country between 1983 and 2020. We considered interprovincial migrants as non-immigrants since it is impossible to differentiate between foreign-born immigrants and Canadian-born individuals moving to Alberta from another province. We used the Imm/ARI repository to determine COVID-19 vaccination status. We extracted data on November 29, 2021, then deterministically linked the databases using unique personal health numbers.

Outcome measure

We defined ‘COVID-19 vaccination coverage’ as the proportion of eligible Alberta residents who received at least one dose of a COVID-19 vaccine and ‘full COVID-19 vaccination coverage’ as the proportion who received at least two doses.

Exposure variables

We created a migration status variable with five categories, including non-immigrants and four categories of immigrants based on year since migration: (1) in or after Jan 2019 (i.e., in the past 2 years), (2) Jan 2011 to Dec 2018 (i.e., in the past 3-10 years), (3) Jan 2000 to Dec 2010 (i.e., in the past 11-20 years), (4) Jan 1983 to Dec 1999 (i.e., more than 20 years ago).

We grouped participants into six age categories based on the categories used to prioritize COVID-19 vaccine eligibility in Alberta: 12-17 years, 18-29 years, 30-49 years, 50-64 years, 65-74 years, and 75 years and above. Biologic sex at birth was categorized into male and female. Neighborhood income quintiles (Q1 indicating the poorest neighborhood and Q5 indicating the richest) were based on the 2016 Canadian census. Place of residence was categorized into three categories based on the 2016 census: metro and moderate metro, urban and moderate urban, and rural and remote rural. Place of origin among the immigrant population was divided into six continental categories: North America, South America, Europe, Asia, Africa, and Oceania.

Patient and public involvement

Patients and the public were not involved in the design, conduct and dissemination of this research.

Cohort characteristics relative to migration status

After excluding participants with missing data on time since migration (n=13), postal codes (n=117), and sex (n=8) from age-eligible Albertans, the final sample size was 3,931,698, of which 731,217 (18.5%) were international immigrants to Alberta. Those above 50 years of age comprised nearly 30% of the immigrant population and 40% among non-immigrants. Females comprised 51.8% of immigrants and 49.2% of non-immigrants. Immigrants from Asia comprised nearly half of the immigrant population in this study (48.3%).

COVID-19 vaccine coverage

Overall, COVID-19 vaccine coverage was slightly higher among immigrants (78.2%; 95% CI: 78.1%, 78.3%) compared to non-immigrants (76.0%; 95% CI:75.9%, 76.0%). Coverage among immigrants differed by continent of origin (Table 1). Asian immigrants had the highest overall vaccine coverage (83.3%) followed by African immigrants (79.3%). In contrast, immigrants from elsewhere in North America had the lowest overall coverage (62.9%). For the age categories between 12 and 74 years, immigrants from North America, Oceania, and Europe had the lowest coverage. For the oldest age group (75 years and older), immigrants from North America, Asia, and Europe had the lowest coverage.

The proportions of vaccinated individuals were significantly different (p<0.001) by age and migration status (Table 2). Vaccination coverage for immigrants generally trended downwards after 50 years of age, irrespective of time since migration. Among immigrants under 50 years of age, those who migrated within the past 20 years tended to have higher vaccination coverage than those who migrated more than 20 years ago. However, among older immigrants (50 years of age and older), those who migrated within the past 20 years had lower coverage than those who migrated more than 20 years ago. In non-immigrants, vaccination coverage increased substantially in older age categories; 70.9%-73.5%% of those 12-49 years old were vaccinated, increasing thereafter, with 86.3% of those 75 years and above having received at least one dose.

There was also variability by other sociodemographic characteristics and migration status. Vaccination coverage among female immigrants was slightly higher than males (except for those who migrated in the last 2 years), with coverage difference increasing slightly with increasing time since migration (Table 2). This difference between females and males was more pronounced in non-immigrants (77.8% vs 74.2%). By place of residence, metro areas had the highest vaccination coverage compared to urban and rural areas, regardless of migration status. There was a wider gap in immigrants who lived in rural areas compared to immigrants who lived in urban areas (76.4% vs 64.2% among those migrating in the last 2 years). Among non-immigrants, the gap between urban and rural residents was 3% (68.1% vs 71.2%). In immigrants, coverage was consistent across the income quintiles, while in non-immigrants, coverage ranged from 71.4% in the poorest quintile to 79.0% in the richest quintile.

Interaction between age and migration status

There was a significant interaction between age category and migration status (Table 3 and Figure 1). Among immigrants, younger age groups that immigrated most recently (10 years or less) had higher coverage than those who immigrated more than 10 years ago. This trend changes after 50 years of age, when coverage in more recent immigrants starts to drop below those who immigrated more than 10 years ago. Among non-immigrants, vaccine coverage continues to increase with age, peaking in those 75 years and above (86.3%). Multivariable logistic regression analysis showed that immigrants generally had significantly higher vaccine uptake for those aged 12 to 49 years old compared to non-immigrants. However, among those aged 50 years and above, immigrants had significantly lower vaccine uptake compared to non-immigrants. Immigrants who migrated in the last 2 years had nearly twice the likelihood of getting vaccinated compared to non-immigrants (OR = 1.73, 95% CI: 1.61, 1.86) among those aged 12-17 years. Conversely, among those aged 75 years and above, immigrants who migrated in the last 2 years had an 83% lower chance of getting vaccinated compared to non-immigrants (OR = 0.17, 95% CI: 0.14, 0.19).

• Download figure
• Open in new tab

Figure 1. Vaccination coverage in different age categories by migration status.

Summary of findings

This study reports on the status of COVID-19 vaccination coverage among immigrants in comparison to non-immigrants in Alberta. Overall, vaccine coverage among immigrant populations is slightly higher compared to non-immigrants and was significantly higher among immigrants under 50 years of age. Unlike the non-immigrant population, which had very high coverage in older adults, coverage decreased in older immigrants, regardless of time since migration.

Interpretation

Previous studies have shown lower uptake of routine vaccines among immigrant populations compared to non-immigrants.(6,12–14) However, COVID-19 vaccination intention in a study from California, USA was higher among immigrant adults (75%) compared to non-immigrant adults (68%).(15) Similarly, a Canadian web survey of participants aged 18 years and older also found that immigrants had a higher intention to receive the COVID vaccine (75%) compared to Canadian-born residents (69%). The difference in intention was the highest in Alberta (86% in immigrants versus 61% in Canadian-born).(16) The same survey reported that nearly 80% of the immigrants were satisfied with the federal government’s handling of the pandemic.(17) Immigrants in Alberta generally have a high trust towards government and health authorities regarding vaccination decisions.(18) Additionally, immigrants are more likely to be employed in essential occupations where there are higher chances of exposure to COVID-19 infection,(19) which could have motivated immigrants to get vaccinated.(20) Some employers instituted vaccine mandates for workers in essential occupations at the beginning of the pandemic. As well, the availability of vaccines free-of-charge through public health centers, pharmacies, and physician clinics irrespective of migration status might have reduced barriers. Although studies have not further explored COVID-19 vaccine hesitancy between immigrants and non-immigrants, differing concerns about safety, efficacy, and perception of mild disease may also explain the differences.(21)

Immigrants from Europe, Oceania, and North America had lower vaccine coverage for those aged 12 to 74 years, compared to immigrants from Asia. Older immigrants (75 years and above) had lower vaccine coverage than middle-aged (30-49 years) immigrants for all continents of origin. Consistent with our findings of higher overall coverage in Asian immigrants, an earlier study from Norway found that immigrants from Asian countries (Vietnam, Sri Lanka, Thailand) had COVID-19 vaccine coverage similar to Norwegian-born individuals (>90%).(22) They also found that vaccine coverage among immigrants from other Asian countries (India, Philippines, Iran) and from Northern Europe (Denmark, UK, Sweden) was much higher than immigrants from Eastern Europe (Latvia, Bulgaria, Poland, Romania, and Lithuania, and proposed that vaccine hesitancy translated into lower vaccine uptake.(22) Vaccine hesitancy in these countries is believed to be associated with a history of authoritarian governments.(23,24)

Older immigrants (65 years and older) had significantly lower coverage compared to older non-immigrants and this difference was higher among recent immigrants. This finding is particularly relevant as older age groups were one of the first priority groups for COVID-19 vaccination in Alberta. A study from Ontario, Canada also showed that COVID-19 vaccine coverage was significantly lower among immigrants and refugees aged 70 years and over, compared to their Canadian-born counterparts.(25) Lower uptake of pneumococcal vaccine among older immigrants was also reported in the US.(14) Possible reasons for lower uptake among older immigrants may be related to cultural and language barriers, higher exposure to misinformation from informal sources and social media, distance to clinics/pharmacies, opening hours, and challenges with online booking systems.(26) Additionally, residents in long term care (LTC) and designated supported living (DSL) facilities were the highest priority groups for COVID-19 vaccination in Canada.(27) Lower admission rate among older immigrants in LTC and DSL facilities due to their ‘strong preference to grow old at home’ and LTC-related barriers might have resulted in lower coverage.(28)

We found that vaccination coverage by migration status showed an interaction with age category. Vaccination coverage among immigrants who migrated in the last 10 years started to drop after 50 years of age. Among non-immigrants, vaccine coverage rose after 50 years of age. Interestingly, immigrants who migrated more than 20 years ago followed a pattern similar to non-immigrants, i.e., increasing coverage after 50 years of age, until the oldest age groups, when coverage started to decline. Since views and beliefs about vaccines are influenced by interpersonal relationships, as well as an individual’s historical and social context,(29) it is possible that migrating later in life means that early lifetime influences of older immigrants continue to affect their vaccination behaviour to present day. Among younger immigrants, however, higher educational status before migration, and smooth acclimatization and integration into Canadian society may have instilled positive views towards vaccines. It can thus be suggested that factors such as time since migration, age at migration, and acculturation window determine the occurrence of the healthy immigrant effect in terms of COVID-19 vaccination, and it does not hold for older immigrants.

Future directions

Understanding how COVID-19 vaccine uptake varies in Canada’s immigrant population is important to design strategies for vaccine delivery and messaging.(12) This is particularly true for older immigrants who had disturbingly low coverage levels, despite their high risk from COVID-19 infection. Our findings suggest that COVID-19 vaccination strategies need to be tailored to the needs of diverse immigrant populations. This could include translating educational materials about vaccines into the language they understand.(16) Strategies may vary for immigrants from North America, Oceania, and Europe, who may not face language barriers, but nonetheless had lower coverage. Moreover, rural-residing immigrants might need additional strategies to receive the COVID-19 vaccine. These might include co-designing tailored approaches according to local needs, leveraging the role of social media, reconsidering current models for vaccine distribution,(30) and partnering with local community-based organizations,(26) faith leaders, and underserved communities, including immigrant communities either directly or through local partners.(31) Communication through text messages, mass media, or telephone, as well as offering vaccines on a walk-in basis can attract diverse and hard-to-reach populations.(32)

Strengths and limitations

This is one of the few studies in Canada to investigate vaccine coverage across different age groups of immigrant populations in comparison to non-immigrant populations. We used population-based immunization, immigrant, and resident registries to derive our coverage estimates, ensuring a large and complete dataset for analysis.

Our study has some limitations. Our dataset likely included individuals who moved out of Alberta without reporting this to the provincial health insurance registry, which may have resulted in an underestimation of vaccine coverage in all groups. Thus, our denominator is larger than reported by Alberta Health,(33) which applies aggregate weights to offset the population inflation, an approach that was not possible for our individual-level data. We did attempt to remove non-residents from our denominator through other approaches. For instance, we tried excluding individuals aged 50 years and over who had not used any of the health services in seven domains in 2020 and 2021 (ambulatory care, inpatient care, laboratory data, COVID-19 test data, long term care, pharmacy data, physician claims), assuming that if they did not use any of these health services then they were probably living out of the province. However, although vaccine coverage by immigration status showed a similar pattern as was seen in the full sample, we lost a significant number of individuals who received COVID-19 vaccines in Alberta in the past year. Therefore, we decided to conduct analysis on the full sample.

In addition, since interprovincial migration was included in the non-immigrant population, there may have been some misclassification of international immigrants as non-immigrants, thereby overestimating the non-immigrant coverage levels. Furthermore, since children born to immigrants in Canada are included in the non-immigrant populations, it might have also overestimated the vaccine coverage among non-immigrants. The findings from this study may not be generalizable to other jurisdictions due to differences in immigration trends, and immigrant characteristics, among others. Finally, COVID-19 vaccines received outside of Alberta may not have been captured in Alberta’s immunization repository.

Conclusion

Public health interventions should focus on older immigrants, immigrants living in rural areas, and immigrants from specific continental backgrounds to improve COVID-19 vaccination coverage. Future research should examine COVID-19 vaccination uptake in children of immigrants, specifically the newly eligible 5- to 11-year-old age group.