Abstract
Objective
We sought to examine whether asthma risk is lower in second-generation immigrants (i.e., Canadian-born children with at least one foreign-born parent) and first-generation immigrants (i.e., foreign-born children) compared with non-immigrants (i.e., Canadian-born children to Canadian-born parents).
Methods
Data were obtained from the Canadian National Longitudinal Survey of Children and Youth from 1994 to 2008, which measured child health and developmental factors from birth to early adulthood. The sample included 15,799 participants aged 2–26 years. Asthma was defined as diagnosis by a health professional as having asthma, having wheezing or whistling in the chest, or use of medication for asthma.
Results
Prevalence of asthma (defined as a combination of any three factors) was lower in first-generation (32%) and second-generation (34%) immigrants compared with non-immigrants (46%). After controlling for covariates, first- and second-generation immigrants had 0.21 (AOR = 0.21; 95% CI = 0.07–0.67) and 0.19 (AOR = 0.19; 95% CI = 0.09–0.39) lower odds of reporting asthma compared with non-immigrants, respectively. For every year the parent(s) of second-generation immigrants resided in Canada, the odds for asthma increased by 5% (AOR = 1.05; 95% CI = 1.02–1.06).
Conclusion
Immigrant children and youth in Canada, regardless of whether they are first- or second-generation, have lower odds for asthma compared with non-immigrants.
Electronic supplementary material
The online version of this article (10.17269/s41997-019-00201-8) contains supplementary material, which is available to authorized users.
Keywords: Asthma, Wheezing, Allergies, Immigrants, First-generation immigrants, Second-generation immigrants
Résumé
Objectif
Nous avons voulu déterminer si le risque d’asthme est plus faible chez les immigrants de deuxième génération (enfants nés au Canada ayant au moins un parent né à l’étranger) et les immigrants de première génération (enfants nés à l’étranger) que chez les non-immigrants (enfants nés au Canada de parents nés au Canada).
Méthode
Nos données provenaient de l’Enquête longitudinale nationale sur les enfants et les jeunes (1994 à 2008), qui mesurait la santé des enfants et les facteurs développementaux de la naissance au début de l’âge adulte. L’échantillon comptait 15 799 participants âgés de 2 à 26 ans. La définition de l’asthme comprenait un diagnostic d’asthme par un professionnel de la santé, une respiration bruyante ou sifflante, ou l’utilisation de médicaments contre l’asthme.
Résultats
La prévalence de l’asthme (définie comme une combinaison de ces trois facteurs) était plus faible chez les immigrants de première génération (32 %) et de deuxième génération (34 %) que chez les non-immigrants (46 %). Compte tenu des covariables, la probabilité de déclarer de l’asthme était plus faible de 0,21 (rapport de cotes ajusté [RCa] = 0,21; IC de 95% = 0,07-0,67) chez les immigrants de première génération et de 0,19 (RCa = 0,19; IC de 95% = 0,09-0,39) chez les immigrants de deuxième génération que chez les non-immigrants. Chez les immigrants de deuxième génération, pour chaque année de résidence des parents au Canada, la probabilité d’asthme augmentait de 5 % (RCa = 1,05; IC de 95% = 1,02-1,06).
Conclusion
Les enfants et les jeunes immigrants au Canada, qu’ils soient de première ou de deuxième génération, présentent une probabilité d’asthme plus faible que les non-immigrants.
Mots-clés: Asthme, Respiration sifflante, Allergie, Immigrants, Immigrants de première génération, Immigrants de deuxième génération
Introduction
The highest prevalence of asthma symptoms is found in developed countries (e.g., United States, United Kingdom, Australia, and Canada) and lowest in more developing countries (e.g., China, India, and Ethiopia) (Beasley et al. 2003). In Canada, 15.2% of children and youth (aged 1–19 years) have asthma (Betancourt et al. 2014). Though the specific aetiology of asthma is unknown, many studies have shown that risk factors for asthma are multifaceted, involving the individuals’ genetics, behaviours, and environment (Subbarao et al. 2009).
Among immigrants, the risk for asthma in the host country has been shown to vary based on their country of origin, age of immigration, duration of stay in the host country, and their predisposition to allergies (Cabieses et al. 2014). Migrant studies, mainly from the US and Europe, show that children born in a country with low asthma prevalence continue to have lower risk for asthma even after moving to the higher-prevalence country compared with their native-born peers (Cabieses et al. 2014). A systematic review by Cabieses et al. showed that immigrants are 40% less likely to report asthma compared with their host population (Cabieses et al. 2014). Canadian literature on this topic is limited and has been conducted in smaller jurisdictions and immigrants from a specific country of origin or ethnicity (Wang et al. 2008; Stanojevic et al. 2014; Benchimol et al. 2015). Asthma symptoms among Chinese adolescents were highest for those born in Canada, intermediate for those children who were born in China and later immigrated to Canada, and lowest for those children living in China (Wang et al. 2008). These findings suggest that environmental exposures in Canada and other countries may increase the risk of developing asthma in immigrant populations.
Migration involves changes in exposure to pollutants, allergens, socio-economic status (SES), diet, and accessibility to medical services. Studies of immigrant children present a unique opportunity to differentiate between genetic and environmental causes of asthma because, relative to the native population, immigrants undergo environmental and lifestyle changes in a short time frame (Kuehni 2011). Increased knowledge regarding modifying factors that contribute to the development of asthma in immigrants may help to reduce the risk of asthma in both the native and immigrant populations.
This study aims to utilize a population-level survey of Canadian children and youth to examine the association between asthma and first-generation immigrants (foreign-born children), second-generation immigrants (Canadian-born children with foreign-born parents), and non-immigrant children. To the best of our knowledge, no other study in Canada has examined associations between immigration background and asthma in a nationally representative sample of Canadian children and youth. The moderating effects of duration of stay in Canada, region of origin, and allergies on the association between immigration and asthma were also examined.
Methods
Sample
Data for this study were obtained from the National Longitudinal Survey of Children and Youth (NLSCY), a longitudinal survey conducted by Statistics Canada. NLSCY began in 1994 with a cohort of children aged 0 to 11 years old. Participants were followed at 2-year intervals up to eight survey cycles until 2008/2009. At each cycle, a new cohort of children between the ages of 0 and 8 were brought in and were subsequently followed up to obtain better estimates on early childhood development (Supplementary Figure 1). The survey collected information on child’s social, emotional, and behavioural developments from birth to early adulthood. The majority of information was collected from parents on behalf of their children through in-person and telephone interviews. After the children turned 10 years of age, some information was directly collected from the child through self-report questionnaires or interviews. Further details regarding the NLSCY have been published elsewhere (Statistics Canada 2010).
In total, 66,826 participants completed at least one survey cycle. This study restricted the sample to participants who were introduced in an earlier cycle and participated in the final cycle in 2008/2009 (n = 20,329). This was carried out to utilize the non-funnel longitudinal weights provided by Statistics Canada. The survey weights account for the probability of a child being selected by the NLSCY sample with adjustment for non-response and non-coverage of population by age, sex, and province. Only one child per household was chosen to avoid clustering of participants under families. Thus, of the 1824 households with two participants, the participant with the most number of observations was chosen from each household. This study further restricted the sample to participants who can be classified as first-generation immigrant, second-generation immigrant, or non-immigrant. Canadian-born children with one Canadian-born parent and unknown birthplace for the second parent were excluded from the study (n = 2706) to avoid misclassification by immigrant status. Proposal for this study was approved by the Social Sciences and Humanities Research Council and Statistics Canada.
Measures
Asthma
A dichotomous scale was used to classify the presence of asthma in each participant at every survey cycle. Due to the changes in asthma-related questions between NLSCY survey cycles, the definition of asthma was constructed to reflect any differences in asthma questions between cycles. Participants, aged 0–15 years, were classified as having asthma using a three-item measure: diagnosis by a health professional as having asthma in their lifetime, wheezing or whistling in the chest at any time in the past 12 months, or took asthma medication (i.e., Ventolin, inhalers, or puffers) on a regular basis in the past 12 months. For participants aged 16 years and older, the number of asthma-related survey questions varied over NLSCY cycles, affecting the definition of asthma. At cycles 4 and 5, they were classified as having asthma using the same three-item measure as those aged 0–15 years. In cycles 6, 7, and 8, only the health professional diagnosis of asthma in their lifetime was used to classify them as having asthma.
Asthma-related questions were asked for all participants at their initial interview, but certain age groups were skipped out of the questions during some follow-up cycles to reduce survey burden. For those age groups, data were imputed using the last observation carried forward (LOCF) (Engels and Diehr 2003) procedure to reduce missing data. The imputation procedure was believed to be suitable because the majority of participants with asthma initially exhibit symptoms at an earlier age (Stern et al. 2008). Once the participant was identified as having asthma, they were classified as asthmatic for all subsequent cycles.
Asthma diagnosis
A dichotomous variable was used to classify whether or not a participant had ever been diagnosed by a health professional as having asthma. This variable was used as another outcome variable in this study to compare findings with those of the aforementioned asthma variable. The LOCF method was used to impute values for those who were skipped out of the questions.
Immigrant generational status
This is a three-level variable for immigrant generational status: (1) first-generation immigrants were classified as foreign-born participants; (2) second-generation immigrants were classified as Canadian-born participants with at least one foreign-born biological parent; and (3) non-immigrants were classified as Canadian-born participants to Canadian-born biological parents.
Allergies
Participants with allergies were either those diagnosed by a health professional or those reported as having long-term allergies. Survey questions regarding allergies changed periodically. In earlier cycles (Beasley et al. 2003; Betancourt et al. 2014; Subbarao et al. 2009), only one response category was offered (“allergies”). Afterward, three response categories were offered on allergies: food or digestive allergies; respiratory allergies (e.g., hay fever); and any other allergies. An overall allergy status was assigned to participants when food allergies, respiratory allergies, or any other allergies were reported. The LOCF procedure was used to impute values for participants who were skipped out of the questions in follow-up interviews (Burke et al. 2003).
Parent(s) region of birth
Biological parent’s region of birth was categorized as 1 = Canada; 2 = US; 3 = Europe (i.e., UK, France, Germany, and Netherlands); 4 = Asia (i.e., China, India, and Philippines); and 5 = other (i.e., Jamaica and Guyana). In cases where the mother and father were born in different regions, the region of birth with the assigned maximum value was chosen to reflect a region that is different in social and physical environments from Canada.
Parent’s time in Canada
For immigrant participants, parent’s time in Canada was assigned as the minimum number of years after immigration to Canada by the parent(s). For non-immigrants, minimum age of the parents was defined as their time in Canada. This was calculated to vary as the participants proceeded through the survey.
Parental history of asthma
Biological parent(s)' reported history of asthma as one of their “long-term conditions” was used to assess the participant’s genetic predisposition to asthma. Positive family history has been shown to increase the risk of asthma in children in many studies (Burke et al. 2003).
Parental smoking
Parental smoking was included as proxy measure for exposure to secondhand smoking in participants. Parental smoking has been identified to increase the risk of asthma and other respiratory symptoms in children (Cook and Strachan 1999). Smoking was defined as daily or occasional smoking and it was assessed from the question “At the present time do you smoke cigarettes daily, occasionally or not at all?”
Other measures included child’s age at each cycle and maternal age at delivery. Younger maternal age at delivery has been reported to be a risk factor for asthma in participants even after controlling for the effects of maternal smoking, SES, and household size (Schwartz et al. 1990; Laerum et al. 2007).
Analysis
The estimates were weighted using the non-funnel longitudinal weights at cycle 8 to ensure that the estimates are unbiased with respect to the survey design. Logistic regression was used to examine overall associations between asthma and the predictors using non-funnel longitudinal weights, and variation estimations were performed using the 1000 bootstrap weights per child. Generalized estimation equation (GEE) regression models were used to account for repeated measures from participants throughout the survey. According to the recommendations by Statistics Canada, the non-funnel longitudinal weights were normalized to reflect the actual sample size of the study by dividing each participant’s survey weight by the average survey weight for all participants in the sample [16]. Quasi-likelihood information criterion (QIC), the goodness of fit criterion for GEE, was used to assess the best-fit model. Analyses were conducted using SAS (SAS Institute Inc. 2014).
Results
The final sample included 15,799 participants aged 2–26 years. Of those, 46% (n = 7207) joined the survey at cycle 1 and the remaining 54% joined the survey in three separate cohorts: cycle 5 (15%); cycle 6 (18%); and cycle 7 (22%). The sample comprised 1% of first-generation immigrants, 17% of second-generation immigrants, and 82% of non-immigrants. The distribution of immigrants in the final sample was similar to that of the total NLSCY sample (1%, 15%, and 66%, respectively).
Presence of any allergies (odds ratio (OR) = 2.83; 95% CI = 2.56–3.12) and parental history of asthma (OR = 2.36; 95% CI = 2.0–2.78) showed the highest odds for asthma compared with other characteristics (Table 1). Parental smoking increased the participant’s odds for asthma by 1.39 (95% CI = 1.24–1.55) (Table 1). First-generation immigrants and second-generation immigrants reported a lower prevalence of asthma (32% and 34%, respectively) compared with the non-immigrant participants (46%) throughout the survey (Table 2). In the logistic regression analysis, first-generation immigrants and second-generation immigrants had 0.57 (95% CI = 0.34–0.95) and 0.61 (95% CI = 0.53–0.70) (respectively) lower odds of reporting overall asthma compared with non-immigrant participants. In the GEE analysis using repeatedly measured asthma status from all cycles, second-generation immigrants reported 0.65 lower odds (95% CI = 0.56–0.75) for asthma compared with non-immigrants, but the findings for first-generation immigrants were not statistically significant (Table 1). Second-generation immigrants had 0.80 lower odds of reporting asthma diagnosis by a health professional (95% CI = 0.66–0.96) and 0.58 lower odds of reporting wheezing (95% CI = 0.51–0.68) compared with non-immigrants (Table 2). Use of asthma medication was significantly lower in second-generation immigrants compared with non-immigrants (OR = 0.79; 95% CI = 0.64–0.97).
Table 1.
Bivariate repeated measure models examining the association of asthma with select characteristics
| Characteristics | Unweighted | Asthma2 | Asthma diagnosis | ||
|---|---|---|---|---|---|
| n 1 | OR | 95% CI | OR | 95% CI | |
| Immigrant status | |||||
| First-generation immigrants | 190 | 0.79 | (0.46–1.34) | 1.08 | (0.57–2.06) |
| Second-generation immigrants | 2645 | 0.65** | (0.56–0.75) | 0.9 | (0.74–1.11) |
| Non-immigrants | 12,965 | Ref | Ref | ||
| Parent(s) region of birth | |||||
| US | 325 | 0.94 | (0.63–1.38) | 1.38 | (0.52–1.64) |
| Europe | 790 | 0.80 | (0.63–1.01) | 1.01 | (0.9–1.62) |
| Asia | 510 | 0.61* | (0.43–0.86) | 0.86 | (0.48–1.3) |
| Other | 1160 | 0.54** | (0.42–0.68) | 0.68 | (0.54–1.03) |
| Canada | 12,995 | Ref | Ref | ||
| Allergies (combined) | |||||
| Yes | 4950 | 2.83** | (2.56–3.12) | 3.12** | (3.09–3.88) |
| No | 10,845 | ref | Ref | ||
| Food allergies | |||||
| Yes | 1540 | 1.93** | (1.6–2.32) | 2.32** | (1.93–2.88) |
| No | 14,080 | Ref | Ref | ||
| Respiratory allergies | |||||
| Yes | 1830 | 3.33** | (2.81–3.94) | 3.94** | (3.64–5.12) |
| No | 13,790 | Ref | Ref | ||
| Any other allergies | |||||
| Yes | 2595 | 2.30** | (2.0–2.65) | 2.65** | (2.01–2.7) |
| No | 13,025 | Ref | Ref | ||
| Parental asthma | |||||
| Yes | 2280 | 2.36** | (2.0–2.78) | 2.78** | (2.44–3.55) |
| No | 13,490 | Ref | Ref | ||
| Parental smoking | |||||
| Yes | 6950 | 1.39** | (1.24–1.55) | 1.55** | (1.24–1.55) |
| No | 8845 | Ref | Ref | ||
| Parent’s time in Canada | 15,550 | 1.00 | (1.0–1.01) | 1.01 | (0.99–1.01) |
| Age at the time of interview | 15,800 | 1.02** | (1.02–1.03) | 1.03** | (1.06–1.08) |
| Maternal age at delivery | 15,710 | 1.00 | (0.99–1.01) | 1.01 | (0.98–1.01) |
1Numbers have been rounded to the nearest 5. 2Asthma is defined as diagnosis by a health professional as having asthma, having wheezing or whistling in the chest, or taking Ventolin, inhalers, or puffers for asthma in their lifetime. *p value < 0.05, **p value < 0.001
Table 2.
Association between ever reporting of asthma-related symptoms by generation of immigrants using logistic regression analyses
| Characteristics | First-generation immigrant (n = 190) | Second-generation immigrant (n = 2645) | Non-immigrants (n = 12,965) | Total (n = 15,799) | First-generation immigrants vs. non-immigrants | Second-generation immigrants vs. non-immigrants | ||
|---|---|---|---|---|---|---|---|---|
| % | % | % | OR | 95% CI | OR | 95% CI | ||
| Asthma1 | ||||||||
| Yes | 32 | 34 | 46 | 42 | 0.57 | (0.34–0.95) | 0.61 | (0.53–0.70) |
| No | 68 | 66 | 54 | 58 | ||||
| Asthma diagnosis | ||||||||
| Yes | 22 | 19 | 22 | 21 | 0.99 | (0.54–1.82) | 0.80 | (0.66–0.96) |
| No | 78 | 81 | 78 | 79 | ||||
| Wheezing | ||||||||
| Yes | 30 | 28 | 40 | 37 | 0.64 | (0.38–1.09) | 0.58* | (0.51–0.68) |
| No | 70 | 72 | 60 | 63 | ||||
| Use of asthma medication2 | ||||||||
| Yes | 17 | 12 | 14 | 14 | 1.23 | (0.62–2.43) | 0.79* | (0.64–0.97) |
| No | 83 | 88 | 86 | 86 | ||||
| Allergies | ||||||||
| Yes | 39 | 34 | 38 | 37 | 1.05 | (0.64–1.72) | 0.85 | (0.74–0.97) |
| No | 61 | 66 | 62 | 63 | ||||
1Asthma is defined as diagnosis by a health professional as having asthma, having wheezing or whistling in the chest, or taking Ventolin, inhalers, or puffers for asthma in their lifetime. 2Medications included Ventolin and inhalers or puffers for asthma
n, unweighted number of participants; %, weighted percent; *p value < 0.05
First-generation immigrants (adjusted odds ratio (AOR) = 0.12; 95% CI = 0.07–0.23) and second-generation immigrants (AOR = 0.12; 95% CI = 0.04–0.34) had significantly lower odds for asthma compared with non-immigrants in the multivariable analysis controlling for the effects of participant’s age, parent’s time in Canada, and interaction between generation and parent’s time in Canada (Table 3, model 1). When the model was adjusted for additional covariates (allergies, parental history of asthma, parental smoking, participant’s age, and maternal age at delivery), excluding the interaction between generation and parent’s time in Canada, the odds of reporting asthma increased to 0.55 (95% CI = 0.24–1.28) in first-generation immigrants and 0.61 in second-generation immigrants (95% CI = 0.45–0.82) (model 2). The final model, which included the interaction between generation and parent’s time in Canada, showed first-generation (AOR = 0.21; 95% CI = 0.07–0.67) and second-generation immigrants (AOR = 0.19; 95% CI = 0.09–0.39) have significantly lower odds for asthma compared with non-immigrants. The odds of a second-generation immigrant reporting asthma increased by 1.05 (95% CI, 1.02–1.06) for every year the participant’s parent(s) lived in Canada. Comparison of QIC values for all multivariable models revealed the full model for immigrant status with the interaction term as the best-fit model (model 3).
Table 3.
Multivariable repeated measure models examining association between generation of immigrants and asthma
| Characteristics | Asthma1 | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | ||||
| AOR | 95% CI | AOR | 95% CI | AOR | 95% CI | |
| Generation | ||||||
| First-generation immigrants | 0.12* | (0.07–0.23) | 0.55 | (0.24–1.28) | 0.21* | (0.07–0.67) |
| Second-generation immigrants | 0.12* | (0.04–0.34) | 0.61* | (0.45–0.82) | 0.19* | (0.09–0.39) |
| Non-immigrants | Ref | Ref | Ref | |||
| Parent(s) region of birth | ||||||
| Asia | 0.64* | (0.43–0.96) | 0.73 | (0.48–1.10) | ||
| Other | 0.61 | (0.36–1.06) | 0.74 | (0.42–1.31) | ||
| Canada/US/Europe | Ref | Ref | ||||
| Parent’s time in Canada2 | 0.95* | (0.94–0.97) | 0.98* | (0.97–0.99) | 0.96* | (0.94–0.97) |
| Generation * parent’s time in Canada | ||||||
| First-generation * parent’s time in Canada | 1.07 | (0.96–1.18) | 1.04 | (0.94–1.17) | ||
| Second-generation * parent’s time in Canada | 1.05* | (1.03–1.07) | 1.05* | (1.02–1.06) | ||
| Non-immigrants * parent’s time in Canada | Ref | Ref | ||||
| Allergies (combined) | ||||||
| Yes | 2.70* | (2.40–3.03) | 2.65* | (2.36–2.98) | ||
| No | Ref | Ref | ||||
| Parental asthma | ||||||
| Yes | 2.30* | (1.86–2.84) | 2.29* | (1.85–2.85) | ||
| No | Ref | Ref | ||||
| Parental smoking | ||||||
| Yes | 1.43* | (1.24–1.65) | 1.41* | (1.22–1.63) | ||
| No | Ref | Ref | ||||
| Child’s age at the time of interview | 1.04* | (1.03–1.04) | 1.03* | (1.02–1.03) | 1.03* | (1.02–1.04) |
| Maternal age at delivery | 1.00 | (0.99–1.02) | 1.00 | (1.00–1.02) | ||
| Quasi-likelihood information criterion | 77,589 | 71,161 | 71,062 | |||
1Asthma is defined as diagnosis by a health professional as having asthma, having wheezing or whistling in the chest, or taking Ventolin, inhalers, or puffers for asthma in their lifetime. 2For immigrant participants, parent’s time in Canada was assigned as the minimum number of years after immigration to Canada; for non-immigrants, minimum age of the parents was defined as their time in Canada. *p value < 0.05
Similar to the findings for asthma, analysis of health professional diagnosis showed that presence of allergies and parental history of asthma are the biggest risk factors for asthma (Table 1). Parental smoking and age at time of interview also showed significantly higher odds for asthma diagnosis. Parental smoking increased the participant’s odds for asthma diagnosis by 55%. Unlike the findings for asthma, participants with parents born in Asia and other countries did not show any significant difference for asthma diagnosis (Table 1).
The rate of asthma diagnosis was lower in second-generation immigrants compared with non-immigrant participants (19% vs. 22%; 95% CI 0.66–0.96); however, there were no significant differences in rates between first-generation immigrants and non-immigrants (Table 2). Multivariable analysis (Table 4) showed first-generation immigrants (AOR = 0.28; 95% CI = 0.08–0.93) and second-generation immigrants (AOR = 0.27; 95% CI = 0.13–0.55) had significantly lower odds for health professional diagnosis of asthma compared with non-immigrants (model 1). The odds of a second-generation immigrant having a diagnosis for asthma increased 4% for every year the participant’s parent(s) lived in Canada (model 1). When the model was adjusted for additional covariates, the association between immigrant generation status and asthma diagnosis was no longer statistically significant (model 3).
Table 4.
Multivariable repeated measure models examining association between generation of immigrants and asthma diagnosis by health professional
| Characteristics | Asthma diagnosis | |||||
|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | ||||
| AOR | 95% CI | AOR | 95% CI | AOR | 95% CI | |
| Generation | ||||||
| First-generation immigrants | 0.28* | (0.08–0.93) | 0.94 | (0.35–2.51) | 0.66 | (0.16–2.75) |
| Second-generation immigrants | 0.27* | (0.13–0.55) | 1.07 | (0.74–1.56) | 0.55 | (0.24–1.29) |
| Non-immigrants | Ref | Ref | Ref | |||
| Parent(s) region of birth | ||||||
| Asia | 0.70 | (0.35–1.39) | 0.77 | (0.38–1.58) | ||
| Other | 0.74 | (0.46–1.18) | 0.79 | (0.49–1.27) | ||
| Canada/US/Europe | Ref | Ref | ||||
| Parent’s time in Canada2 | 0.97* | (0.95–0.98) | 0.99 | (0.98–1.01) | 0.98* | (0.96–1.00) |
| Generation * parent’s time in Canada | ||||||
| First-generation * parent’s time in Canada | 1.02 | (0.90–1.17) | 0.98 | (0.96–1.00) | ||
| Second-generation * parent’s time in Canada | 1.04* | (1.01–1.07) | 1.02 | (1.00–1.05) | ||
| Non-immigrants * parent’s time in Canada | Ref | Ref | ||||
| Allergies (combined) | ||||||
| Yes | 2.79* | (2.46–3.18) | 2.77* | (2.43–3.34) | ||
| No | Ref | Ref | ||||
| Parental asthma | ||||||
| Yes | 2.98* | (2.37–3.75) | 2.97* | (2.36–3.76) | ||
| No | Ref | Ref | ||||
| Parental smoking | ||||||
| Yes | 1.28* | (1.08–1.53) | 1.27* | (1.06–1.51) | ||
| No | Ref | Ref | ||||
| Child’s age at the time of interview | 1.06* | (1.05–1.07) | 1.05* | (1.04–1.06) | 1.05* | (1.04–1.06) |
| Maternal age at delivery | 1.00 | (0.98–1.01) | 1.00 | (0.98–1.02) | ||
| Quasi-likelihood information criterion | 60,408 | 54,533 | 54,551 | |||
1For immigrant participants, parent’s time in Canada was assigned as the minimum number of years after immigration to Canada; for non-immigrants, minimum age of the parents was defined as their time in Canada. *p value < 0.05
Discussion
Our analysis of over 15,000 participants found that immigrants from all regions (including other industrialized nations) have a lower risk for asthma in Canada compared with Canadian-born participants with Canadian-born parents. First- and second-generation immigrants have lower odds for asthma compared with non-immigrants, even after adjusting for possible confounders like parent’s age, history of asthma, and smoking. The risk for asthma is relatively similar between first- and second-generation immigrants compared with non-immigrant participants. Finally, parent’s length of time living in Canada increases the risk of acquiring asthma in immigrant participants, but only at a very slow rate.
We found that a parent born in a foreign country is a protective factor for asthma in Canadian-born participants. This suggests that the protective effect of being a foreign-born is passed onto the next generation by parents. A study from Sweden also reported a protective effect for asthma in second-generation immigrants. In the study, Swedish-born participants with foreign-born mothers showed lower risk of being admitted to the hospital for asthma compared with participants of Swedish-born mothers (Hjern et al. 1999). Our study showed no significant difference in odds for asthma between first- and second-generation immigrants. However, review of five studies by Cabieses et al. showed first-generation immigrants have lower odds for asthma (pooled OR, 0.37; 95% CI, 0.25–0.58) compared with second-generation immigrants (Cabieses et al. 2014). Several studies that examined the risk of asthma by ethnicity, an indicator of immigrants from many generations, have also reported a protective effect for asthma in ethnic minorities in the US and UK (Akinbami et al. 2009; Netuveli et al. 2005; Whitrow and Harding 2010). Together, being an immigrant is a protective factor for asthma, regardless of whether it was the child or the parent who immigrated to Canada.
Finally, we showed a gradual increase in the probability of asthma in participants as their parent(s) lived longer in Canada. This is the first Canadian study to examine the relationship by parent’s length of stay. Other international studies that explored the association by participants’ length of stay in the host country also reported a negative correlation between length of stay and asthma (Cabieses et al. 2014). This association is often explained using the hygiene hypothesis, which suggests that foreign-born immigrants have better immunity against the development of asthma in their host country due to their early childhood exposure to infections in their country of birth (Rottem et al. 2005). Our findings on the rate at which the protective effects change in immigrants varied from findings in the literature. Other studies have shown a relatively accelerated development of asthma in immigrants than what was seen in this study (Eldeirawi et al. 2009; Migliore et al. 2007). Mexican-born participants who lived in the US for more than 10 years reported a significant increase (2.3 odds) in development of asthma compared with their counterparts who lived in the US for less than 10 years (Eldeirawi et al. 2009). For every additional year immigrant children resided in Italy, there was a 12% increase in the odds for asthma (Migliore et al. 2007). The reason for discordant findings may include the differences in the use of child's vs. parent’s time in Canada, sampling frame for immigrant population, and the environment in the host country.
One of the limitations of this analysis is around the broad definition of asthma used (physician diagnosis or symptoms or medication use) to classify participants. There is no gold standard for detecting asthmatic participants in epidemiological studies (Pekkanen and Pearce 1999). Many studies have traditionally used professional diagnosis to define asthma (Lara et al. 2006; Yu et al. 2004). The broad definition used in this study showed much higher prevalence of asthma (43%) compared with the nationally reported prevalence (15.2%) (Betancourt et al. 2014) and is likely due to inclusion of wheezing in our definition. This definition of asthma was used to capture all asthmatic participants who may or may not have been diagnosed, including recent immigrants who may not have had the opportunity to be diagnosed by a health professional. This may have resulted in some misclassification of participants who wheeze but do not have asthma. Other studies examining the validity of asthma measures have indicated that inclusion of wheezing is an appropriate measure for studying asthma (Yu et al. 2004; Thomas 2010; Remes et al. 2002; Hederos et al. 2007; Pekkanen et al. 2005). As a single measure, wheezing was shown to have 78% sensitivity and 97% specificity with clinically diagnosed asthma (Remes et al. 2002). In order to overcome some of these limitations and for comparability with other studies, analysis by asthma diagnosis was carried out, but due to smaller number of participants in this group, many findings were not statistically significant.
Use of the LOCF method to reduce missing data and make corrections to discrepant reporting of asthma may have resulted in misclassification of participants with asthma. Assuming that report of asthma at an earlier cycle was more accurate than denial of asthma at a later cycle may have led to non-random misclassification of participants. Sensitivity analyses were carried out on the LOCF method by comparing models presented in this paper to models using two other approaches: (1) carried forward responses for those who were skipped out of the asthma questions, but with no changes for discrepant reporting of asthma, and (2) no adjustment to asthma outcomes. Comparison of bivariate models showed similar effects for first- and second-generation immigrants across all three approaches. For multivariable models, the protective effect for asthma and asthma diagnosis among second-generation immigrants was still significant after controlling for age, parent's length of stay in Canada, and the interaction between generation and parent’s time in Canada (model 1 in Tables 3 and 4). After controlling parent’s place of birth, presence of allergies, parental history of asthma, and parental smoking (models 2 and 3 in Tables 3 and 4), the protective effect was no longer significant for both approaches, which may have been attributable to the loss of observations due to deletion of missing values across all variables. Since the bivariate and initial multivariable models showed similar associations and the effect size presented in the study is similar or lower compared with the other two sensitivity approaches, we believe there is minimal bias associated with using the LOCF method.
Another limitation of this study may be an underrepresentation of recent immigrants. The original cohort was developed to be representative of the population, but over time, the sample design did not specifically ensure an adequate sample of new immigrants. As the original cohort grew older, the NLSCY topped up the sample with young children and these may have included some recent immigrants.
One of the major strengths of this study is that as a population-level study across Canada, this study provides more generalizable findings of Canadians. Unlike other studies, the large sample (n > 18,000) allowed for adjustments for many covariates—including parent’s region of birth. The use of a common definition for asthma, health professional diagnosis, allows for greater comparability across studies. In addition, this study contributes to the limited number of studies that have examined the association between second-generation immigrants and asthma. The slower rate of increase in asthma development seen among immigrants in this study compared with that in the US indicates the need for consideration of social context in the relationship between asthma and generation of immigrants.
Future studies should examine this association closer through the use of objective measures for asthma. This study also emphasizes the need for prospective research on factors influencing the change in immunity among immigrant participants. Soon after migration, many immigrants experience decline in SES and acculturation challenges. Many often live in low-income neighbourhoods that may expose them to neighbourhood-level risk factors. Better knowledge of the contributing factors for the development of asthma in immigrant population will protect against environmental factors (food, culture, and physical environment) in both immigrant and non-immigrant populations.
Electronic supplementary material
(DOCX 14 kb)
(XLSX 36 kb)
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- Akinbami LJ, Moorman JE, Garbe PL, Sondik EJ. Status of childhood asthma in the United States, 1980–2007. Pediatrics. 2009;123(Supplement):S131–S145. doi: 10.1542/peds.2008-2233C. [DOI] [PubMed] [Google Scholar]
- Beasley R, Ellwood P, Asher I. International patterns of the prevalence of pediatric asthma the ISAAC program. Pediatric Clinics of North America. 2003;50(3):539–553. doi: 10.1016/S0031-3955(03)00050-6. [DOI] [PubMed] [Google Scholar]
- Benchimol EI, Manuel DG, To T. Mack DR, Nguyen GC, Gommerman JL, et al. Asthma, type 1 and type 2 diabetes mellitus, and inflammatory bowel disease amongst South Asian immigrants to Canada and their children: a population-based cohort study. PLoS One. 2015;10(4):1–13. doi: 10.1371/journal.pone.0123599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Betancourt MT, Roberts KC, Bennett TL, Driscoll ER, Jayaraman G, Pelletier L. Monitoring chronic diseases in Canada: the chronic disease indicator framework. Chronic Diseases and Injuries in Canada. 2014;34:1–31. [PubMed] [Google Scholar]
- Burke W, Fesinmeyer M, Reed K, Hampson L, Carlsten C. Family history as a predictor of asthma risk. American Journal of Preventive Medicine. 2003;24(2):160–169. doi: 10.1016/S0749-3797(02)00589-5. [DOI] [PubMed] [Google Scholar]
- Cabieses, B., Uphoff, E., Pinart, M., Anto, J. M., Wright, J. (2014). A systematic review on the development of asthma and allergic diseases in relation to international immigration: the leading role of the environment confirmed. PLoS One, 9(8). [DOI] [PMC free article] [PubMed]
- Cook DG, Strachan DP. Summary of effects of parental smoking on the respiratory health of children and implications for research. Thorax. 1999;54(4):357–366. doi: 10.1136/thx.54.4.357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eldeirawi K, McConnell R, Furner S, Freels S, Stayner L, Hernandez E, et al. Associations of doctor-diagnosed asthma with immigration status, age at immigration, and length of residence in the United States in a sample of Mexican American School Children in Chicago. The Journal of Asthma. 2009;46(8):796–802. doi: 10.1080/02770900903114572. [DOI] [PubMed] [Google Scholar]
- Engels JM, Diehr P. Imputation of missing longitudinal data: a comparison of methods. Journal of Clinical Epidemiology. 2003;56(10):968–976. doi: 10.1016/S0895-4356(03)00170-7. [DOI] [PubMed] [Google Scholar]
- Hederos CA, Hasselgren M, Hedlin G, Bornehag CG. Comparison of clinically diagnosed asthma with parental assessment of children’s asthma in a questionnaire. Pediatric Allergy and Immunology. 2007;18(2):135–141. doi: 10.1111/j.1399-3038.2006.00474.x. [DOI] [PubMed] [Google Scholar]
- Hjern A, Haglund B, Bremberg S, Ringbäck-Weitoft G. Social adversity, migration and hospital admissions for childhood asthma in Sweden. Acta Paediatrica. 1999;88(10):1107–1112. doi: 10.1111/j.1651-2227.1999.tb00997.x. [DOI] [PubMed] [Google Scholar]
- Kuehni CE. Do migrant studies help to identify causes of asthma? Clinical and Experimental Allergy. 2011;41(8):1054–1058. doi: 10.1111/j.1365-2222.2011.03815.x. [DOI] [PubMed] [Google Scholar]
- Laerum BN, Svanes C, Wentzel-Larsen T, Gulsvik A, Torén K, Norrman E, et al. Young maternal age at delivery is associated with asthma in adult offspring. Respiratory Medicine. 2007;101(7):1431–1438. doi: 10.1016/j.rmed.2007.01.020. [DOI] [PubMed] [Google Scholar]
- Lara M, Akinbami L, Flores G, Morgenstern H. Heterogeneity of childhood asthma among Hispanic children: Puerto Rican children bear a disproportionate burden. Pediatrics. 2006;117(1):43–53. doi: 10.1542/peds.2004-1714. [DOI] [PubMed] [Google Scholar]
- Migliore, E., Pearce, N., Bugiani, M., Galletti, G., Biggeri, A., Bisanti, L., et al. (2007). Prevalence of respiratory symptoms in migrant children to Italy: the results of SIDRIA-2 study. Allergy, 293–300. [DOI] [PubMed]
- Netuveli G, Hurwitz B, Sheikh A. Ethnic variations in incidence of asthma episodes in England & Wales: national study of 502,482 patients in primary care. Respiratory Research. 2005;6:120. doi: 10.1186/1465-9921-6-120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pekkanen J, Pearce N. Defining asthma in epidemiological studies. The European Respiratory Journal. 1999;14(4):951–957. doi: 10.1034/j.1399-3003.1999.14d37.x. [DOI] [PubMed] [Google Scholar]
- Pekkanen J, Sunyer J, Anto JM, Burney P, Abramson M, Kutin J, et al. Operational definitions of asthma in studies on its aetiology. The European Respiratory Journal. 2005;26(1):28–35. doi: 10.1183/09031936.05.00120104. [DOI] [PubMed] [Google Scholar]
- Remes ST, Pekkanen J, Remes K, Salonen RO, Korppi M. In search of childhood asthma: questionnaire, tests of bronchial hyperresponsiveness, and clinical evaluation. Thorax. 2002;57(2):120–126. doi: 10.1136/thorax.57.2.120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Rottem M, Szyper-Kravitz M, Shoenfeld Y. Atopy and asthma in migrants. International Archives of Allergy and Immunology. 2005;136:198–204. doi: 10.1159/000083894. [DOI] [PubMed] [Google Scholar]
- SAS Institute Inc.(2014). SAS software 9.4 (pp. 1–25). SAS Institute Inc. MarketLine Company.
- Schwartz J, Gold D, Dockery DW, Weiss ST, Speizer FE. Predictors of asthma and persistent wheeze in a national sample of children in the United States. Association with social class, perinatal events, and race. The American Review of Respiratory Disease. 1990;142(3):555–562. doi: 10.1164/ajrccm/142.3.555. [DOI] [PubMed] [Google Scholar]
- Stanojevic S, Shah BR, Anand SS, Sears MR, Su J, Subbarao P. The burden of asthma among South Asian and Chinese populations residing in Ontario. Canadian Respiratory Journal. 2014;21(6):346–350. doi: 10.1155/2014/160476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Statistics Canada (2010). National Longitudinal Survey of Children and Youth. Survey overview for the 2008/2009 data collection. Cycle 8. Ottawa, ON. Available at: http://www23.statcan.gc.ca/imdb-bmdi/document/4450_D2_T9_V4-eng.pdf. Accessed 11 June 2018.
- Stern DA, Morgan WJ, Halonen M, Wright AL, Martinez FD. Wheezing and bronchial hyper-responsiveness in early childhood as predictors of newly diagnosed asthma in early adulthood: a longitudinal birth-cohort study. Lancet. 2008;372(9643):1058–1064. doi: 10.1016/S0140-6736(08)61447-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Subbarao, P., Mandhane, P. J., Sears, M. R. (2009). Asthma: epidemiology, etiology and risk factors. Canadian Medical Association Journal, 181(9). [DOI] [PMC free article] [PubMed]
- Thomas EM. Recent trends in upper respiratory infections, ear infections and asthma among young Canadian children. Health Reports. 2010;21(4):47–52. [PubMed] [Google Scholar]
- Wang HY, Wong GWK, Chen YZ, Ferguson AC, Greene JM, Ma Y, et al. Prevalence of asthma among Chinese adolescents living in Canada and in China. Canadian Medical Association Journal. 2008;179(11):1133–1142. doi: 10.1503/cmaj.071797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Whitrow MJ, Harding S. Asthma in Black African, Black Caribbean and South Asian adolescents in the MRC DASH study: a cross sectional analysis. BMC Pediatrics. 2010;10:18. doi: 10.1186/1471-2431-10-18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yu ITS, Wong TW, Li W. Using child reported respiratory symptoms to diagnose asthma in the community. Archives of Disease in Childhood. 2004;89(6):544–548. doi: 10.1136/adc.2003.033688. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(DOCX 14 kb)
(XLSX 36 kb)