The Association between Migration and Prevalence of Allergic Diseases: A Systematic Review and Meta-Analysis

Abstract

Introduction

Allergic diseases remain of concern due to their increasing prevalence worldwide. Intrinsic and environmental risk factors have been implicated in the pathogenesis of allergic disease. Among the possible risk factors, migration has been associated with the manifestation of allergic diseases. We aimed to consolidate the existing evidence, review the hypotheses for the relationship between environmental factors and allergic disease, and provide a direction for future work.

Methods

This systematic review and meta-analysis complied with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The Web of Science database was searched in September 2023 to retrieve publications investigating the relationship between allergic rhinitis (AR), atopic dermatitis (AD), or asthma and the following factors: (i) migrant status (i.e., migrants vs. natives) or (ii) duration since migration among migrants. Risk of bias was assessed using the JBI critical appraisal tool. Details and findings from the included studies were also summarized and meta-analyses were conducted where appropriate.

Results

Fifty studies encompassing an estimated 3,755,248 individuals were reviewed. Articles investigated asthma (n = 46), AR (n = 16), and AD (n = 14). A variety of migration-related factors were also studied: movement of individuals across regions (n = 40), duration since immigration (n = 12), age at immigration (n = 9), and acculturation (n = 2). Migration status was not significantly associated with AD (pooled odds ratio [pOR] = 0.68, 95% confidence interval (CI) = 0.31, 1.49). Although AR prevalence was lower among immigrants than natives (pOR = 0.58, 95% CI = 0.45, 0.74), immigrants who had resided at least 10 years in the destination country had a higher risk of AR than immigrants with a duration of residence of less than 10 years (pOR = 8.36, 95% CI = 4.15, 16.81). Being an immigrant was also associated with a decreased risk of asthma (pOR = 0.56, 95% CI = 0.44, 0.72). Among immigrants, residing in the host country for at least 10 years was associated with increased asthma manifestation (pOR = 1.85, 95% CI = 1.25, 2.73). Immigrants who migrated aged 5 and below did not exhibit a significantly higher likelihood of asthma than migrants who immigrated older than 5 years (pOR = 1.01, 95% CI = 0.68, 1.50).

Conclusion

This review was limited by the primarily cross-sectional nature of the included studies. Objective diagnoses of allergic disease, such as using the spirometry of bronchodilator reversibility test for asthma rather than questionnaire responses, could add to the reliability of the outcomes. Furthermore, immigrant groups were mostly nonspecific, with little distinction between their country of origin. Overall, migration appears to be a protective factor for allergic diseases, but the protection subsides over time and the prevalence of allergic diseases among the immigrant group approaches that of the host population.

Introduction

Background

The prevalences of allergic diseases (i.e., allergic rhinitis [AR], atopic dermatitis [AD], asthma) have remained in flux for the past two centuries and have grown to the forefront of attention due to their drastic increase [1]. Multiple environmental factors (e.g., microbial exposure, air pollution) and mechanisms of action have been put forward as culprits in the pathogenesis of allergic disease, with varying degrees of evidential support [2]. Indeed, a plethora of epidemiological studies have been published, implicating numerous risk factors both non-modifiable (e.g., genetics, gender) and modifiable (e.g., diet, microbes, pollution) in the manifestation of allergic diseases [3]. Notably, migration has also been demonstrated to be associated with the manifestation of allergic disease. The healthy immigrant effect (HIE) refers to the phenomenon where immigrants exhibit lower prevalences of disease than the native population of the destination country [4]. Epidemiological evidence has been reported for the association of first-generation migrant status with a lowered risk of allergic diseases [5, 6].

Aims and Objectives

We aimed to systematically review the current literature on allergic diseases and migration. Our review investigated three specific allergic diseases – AR, AD, and asthma – and the changes in their prevalence associated with movement across different geographic regions or time spent in the new region following movement, both of which constituted acts of migration. Additionally, we aimed to update a previous systematic review of asthma and migration while expanding upon a previous narrative review of migration and allergic diseases [5, 6]. Finally, we address current hypotheses for the association of migration with allergic diseases and provide direction for future studies.

Methods

Search Strategy

The current systematic review was conducted under updated guidelines established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 statement and was neither funded nor registered [7]. Per PRISMA 2020 guidelines, this review was completed against the PRISMA checklist, and the search process was documented with the PRISMA 2020 flow diagram (Fig. 1, online suppl. Tables 1, 2; for all online suppl. material, see https://doi.org/10.1159/000539382). To obtain a comprehensive record of articles examining any of the three allergic diseases – AD, AR, and asthma – and migrant status, we developed a search term that selected for articles wherein the title or abstract fulfilled two criteria: (1) used keywords or terms relevant to any of the three allergic diseases (i.e., AR, AD, asthma); (2) included terms referring to migration. The Web of Science Core Collection database was searched in September 2023 using the following search term: TI = ([migran* OR migrat* OR immigra*] AND [epidemiology OR prevalence] AND [allergic disease OR atopic dermatitis OR allergic dermatitis OR allergic rhinitis OR atopic rhinitis OR asthma]) OR AB = ([migran* OR migrat* OR immigra*] AND [epidemiology OR prevalence] AND [allergic disease OR atopic dermatitis OR allergic dermatitis OR allergic rhinitis OR atopic rhinitis OR asthma]).

Fig. 1.

PRISMA 2020 flow diagram documenting the literature search and screening process.

The default publication date filter was applied – i.e., articles published between 1900 and 2023 were searched; no other filters were used. To ensure a thorough overview of the published literature, a secondary search was conducted, whereby citations of articles returned by the primary search were screened per our eligibility criteria and included in our review where appropriate. This process was repeated for citations of articles from the secondary search. The citations from secondary search articles were largely already included in our review, indicating that our literature search was sufficiently comprehensive.

Eligibility Criteria

Records from our primary search were subjected to screening based on the title and abstract. Articles that addressed any of the three allergic diseases as the outcome and migration as the exposure were included. We defined migration as the movement of individuals from one region to another; hence, articles comparing migrants to natives or comparing migrants based on duration since migration were included. Full-text articles available in English were then retrieved and rescreened according to the eligibility criteria. Excluded records were frequently of irrelevant topics (n = 100) or review articles (n = 27). The screening process was conducted by the first reviewer (Q.Y.A.W.) and reviewed by the second reviewer (F.T.C.). Finally, quality assessments via the JBI critical appraisal tool were performed for studies that fulfilled the eligibility criteria [8].

Data Extraction

Information relevant to the study setting, sample size, sample demographics, country of study, and study type were extracted. Additionally, allergic disease outcomes and exposures relating to the movement of individuals were recorded. Due to variations in terminology between studies, outcomes were categorized as either AR (e.g., of terminology: hay fever, rhinoconjunctivitis), AD (e.g., of terminology: eczema), or asthma (e.g., of terminology: wheezing). Finally, where reported, regression model types, the resultant effect size estimates and standard error or confidence intervals (CIs), and confounding factors adjusted for in the model were abstracted. Only adjusted effect sizes were extracted from articles reporting both crude and adjusted effect sizes. For the proceeding meta-analyses, an abbreviated version of our dataset comprising the following information was used: author and publication year, outcome, exposure and reference group, effect size, and CIs (or standard error, where reported).

Risk of Bias Assessment

The risk of bias for included articles we assessed using the checklist for analytical cross-sectional studies from critical appraisal tools for use in JBI systematic reviews [8]. Eight study criteria were examined to assess the study risk of bias: (1) were the criteria for inclusion in the sample clearly defined, (2) were the study subjects and the setting described in detail, (3) was the exposure measured in a valid and reliable way, (4) were objective, standard criteria used for measurement of the condition, (5) were confounding factors identified, (6) were strategies to deal with confounding factors stated, (7) were the outcomes measured in a valid and reliable way, (8) was appropriate statistical analysis used. Each criterion was checked as either “yes” where explicitly fulfilled, “no” where lacking, or “unclear” where ambiguity was involved. The number of “yes” to a given fulfilled criteria was tallied and converted to a percentage denominated against the eight criteria. Studies that scored less than 50% were considered to have a high risk of bias.

Statistical Analysis

Random effects meta-analysis using the DerSimonian-Laird method was conducted for common outcomes where the odds ratios and their accompanying 95% CI or standard error were reported [9]. For studies reporting 95% CIs, the standard error was calculated using the formula: (upper CI−lower CI) ÷ 3.92; the standard error was necessary for computing the pooled odds ratio (pOR) in the meta-analysis. pORs were calculated where appropriate, with consideration taken to ensure that study subjects did not overlap between studies and that exposure and reference categories were comparable. Among studies reporting odds ratios, exposure groups were primarily immigrants while reference groups were nonimmigrants; however, some studies assigned the immigrant group as a reference instead – for our analyses, the comparison was reversed by taking the inverse of the odds ratio. Where relevant, statistical significance was determined using the alpha value of 0.05. Heterogeneity was assessed using the I² index, whereby I² > 50% and an accompanying p value <0.05 was indicative of significant heterogeneity. Finally, for each pOR, publication bias was assessed using Begg’s funnel plots and Egger’s test.

Results

Literature Search

Our search process was documented using the PRISMA 2020 flow diagram (Fig. 1). Searching the Web of Science databases yielded 201 records, of which 13 were excluded before screening due to non-English language of publication or article type. Of the remaining 188 articles, the title and abstract of 57 were deemed not to meet our eligibility criteria, while another 26 were review articles. Full texts of 105 articles were sought for retrieval, of which 4 were inaccessible, 56 were irrelevant, and 1 was a review article. References of the 44 relevant articles were hand searched, yielding a further 6 pertinent original research publications. Overall, a total of 50 articles were included in the current systematic review (online suppl. Table 3). Forty-eight studies fulfilled at least 4 of the criteria and had a score of at least 50% in our risk of bias assessment. The remaining two studies were included upon referring to their citations to verify their methods (online suppl. Table 4).

Overview of Studies

Articles examined herein were published across the years 1994–2022 and originated from at least 14 distinct countries (Table 1). Moreover, two articles were multicenter studies, including a European Community Respiratory Health Survey (ECRHS) publication analyzing data from Europe, the USA, Australia, and New Zealand, and an International Study of Asthma and Allergies in Childhood (ISAAC) publication spanning 48 countries [10, 11]. Studies were preponderantly of the cross-sectional nature (n = 43), with the remaining studies being prospective cohort studies (n = 2), longitudinal studies (n = 4), and one retrospective cohort study. Overall, an estimated total of 3,755,248 individuals were included in 48 studies, barring two studies that did not report their sample size [12, 13]. Asthma was the most extensively studied of the three allergic diseases, with 46 articles investigating a variety of asthma-related outcomes including wheezing and chronic cough (online suppl. Table 5). Notwithstanding, 16 articles relevant to AR and 14 articles relevant to AD were retrieved. Allergic disease outcomes were frequently classified using questionnaires (n = 41), with notable standardized examples including the ISAAC and ECRHS questionnaires. The minority of studies involved the examination of medical records (n = 6), collected data via telephone interviews (n = 2), or conducted an interview accompanied by physical examination in a mobile health unit (n = 1). A summary of study details can be found in Table 1.

Table 1.

A summary of studies reviewed in this article, including study sample details and their respective findings

Author and year	Study dates and name (where applicable)	Age and race (where reported)	Sample size1	Country	Study type	Data collection methods	Association of immigrant status and allergic disease (comparison: immigrants vs. nonimmigrants, unless stated otherwise)
Alsowaidi et al. [59], 2010	2007–2008	8–93	6,543	United Arab Emirates	Cross-sectional	Modified ISAAC questionnaire	AR: aOR = 0.59, 95% CI = 0.46–0.76
							Asthma: aOR = 0.55, 95% CI = 0.39–0.79
Andiappan et al. [60], 2014	Ongoing	9–65, Chinese	7,949	Singapore	Cross-sectional, ongoing	Questionnaire collecting information on demographics and medical history, and based on ARIA and ISAAC guidelines	Prevalence of AR, asthma, and AD increases significantly among immigrants from China as their duration of stay increases
Barr et al. [61], 2016	Not reported [Hispanic Community Health Study/Study of Latinos (HCHS/SOL)]	18–74, Hispanics	16,415	USA	Cross-sectional, population-based	Standard respiratory questionnaire from Ferris [62], 1978	(i) Physician diagnosed asthma and current asthma prevalence significantly decreased from those born in the USA, to those who immigrated 0–15 years old, to those immigrated at least 15 years old
							(ii) Birth in the USA and immigration to the USA as a child, compared to immigration to USA as an adult, was associated with elevated asthma risk
Brugge et al. [63], 2007	2005	4–18, Asians	204	USA	Cross-sectional	Questionnaire administered verbally by trained undergraduates	Born in the USA versus born in China, Taiwan, Hong Kong, or other parts of Asia
							Asthma: aOR = 4.48, 95% CI = 1.68–11.94 (calculated from multivariate logistic regression results)
Brugge et al. [64], 2008	2005–2006	4–84, self-identified Black/African American	447	USA	Cross-sectional	Orally administered questionnaire based on BPAS, Bai et al. [65], 1999, and IUALTD	Among adult Black people, being born outside the USA was a protective factor against asthma
							Asthma: aOR = 0.34, 95% CI = 0.15–0.79 (calculated from multivariate logistic regression results)
Chen et al. [66], 1999	1994–1995 (National Population Health Survey [NPHS])	12 and above	17,605	Canada	Cross-sectional	Questionnaire	Canadian native versus immigrant
							Asthma (males): aOR = 1.6, 95% CI = 1.14–2.25
							Asthma (females): aOR = 1.25, 95% CI = 0.91–1.72
Cohen et al. [67], 2007	2001–2003	5–13, Puerto Ricans	2,491	USA	Population-based prospective cohort study	Questionnaire which was part of the Service Utilization and Risk Factors interview that was developed for the Epidemiology of Childhood and Mental Disorders Study, administered by trained interviewers	Living in Puerto Rico versus living in the South Bronx
							Asthma: prevalence = 41.3% versus 35.3% (significant)
Corlin et al. [68], 2014	2009–2012 (Community Assessment of Freeway Exposure and Health [CAFEH])	40–91, Chinese and Whites	453	USA	Cross-sectional	Questionnaire	Asthma: aOR = 0.86, 95% CI = 0.77–0.98
Eldeirawi and Persky [53], 2006	1999–2000 and 2001–2002 (NHANES)	12–19, Mexican Americans	1,734	USA	Cross-sectional	Home interview and examination in a mobile unit	US-born versus Mexican immigrants
							Asthma: aOR = 1.68, 95% CI = 1.06–2.66
							AR: prevalence = 10.47% versus 5.07% (significant)
							High acculturation versus low acculturation among Mexican immigrants
							Asthma: aOR = 2.63, 95% CI = 1.17–5.93
Eldeirawi and Persky [69], 2007	2004–2005 (Chicago Asthma School Study)	School children (age range not reported), Mexican descent	10,106	USA	Cross-sectional	ISAAC-based questionnaire	US-born versus Mexican immigrants
							Asthma: aOR = 2.24, 95% CI = 1.78–2.83
							At least 10 years of residence in the USA among Mexican immigrants
							Asthma: aOR = 1.93, 95% CI = 1–3.73
Eldeirawi et al. [70], 2005	1988–1994 (NHANES III)	2 months–16 years, Mexican Americans	4,121	USA	Cross-sectional	NHANES III questionnaire	US-born versus Mexican-born
							Asthma: aOR = 1.83, 95% CI = 0.92–3.65
							AR: prevalence rate = 3.65% versus 0.32% (significant)
Farfel et al. [12], 2007	nr	17–18	Not reported	Israel	Cross-sectional	Data obtained during health checkup and from records	Asthma: Israel – 119.3; Western – 103.9; Soviet Union – 83.9; Asian – 95.9; African – 27; Ethiopia – 27.8 (per 1,000 males)
							AR: Israel – 92.1; Western – 91.4; Soviet Union – 130.5; Asian – 68.5; African – 27; Ethiopian - 14.8 (per 1,000 males)
Garcia-Marcos et al. [10], 2014	nr (ISAAC)	6–7, 13–14	535,214	48 countries (adolescents), 31 countries (children)	Multicenter study	ISAAC Phase Three questionnaire	(i) There is generally a lower prevalence of allergic disease in the country of origin as compared to the destination country, and this association applies mainly to those migrating to affluent countries. (ii) The difference in prevalence rate between migrants and nonmigrants is greater among migrants who have spent fewer years in the host country. (iii) Immigration to nonaffluent countries does not appear to be associated with a change in allergic disease prevalence. (iv) For immigrants to show protection, they have to migrate to a country with high allergic disease prevalence
Gibson et al. [71], 2003	1997–1998	Secondary school students (age range not reported)	211	Australia	Cross-sectional	ISAAC video questionnaire	Per year increase in duration resided in Australia
							Asthma symptoms: aOR = 1.11, 95% CI = 1.01–1.21
Greenfield et al. [72], 2005	2004	5–18, Chinese	152	USA	Cross-sectional	BPAS questionnaire	US-born versus foreign-born
							Asthma: OR = 5.41, 95% CI = 2.29–12.8
Hjern et al. [73], 1999	1991–1995	18	14,630	Sweden	Cross-sectional	Physician diagnoses and register	(i) Conscripts born in Africa, Asia, Latin America, or the Mediterranean had a significantly lower risk for asthma and AR than Swedish-born conscripts. (ii) The risk of atopic disorder among the foreign-born conscripts increased with time of residency in Sweden
Hjern et al. [74], 1999	1991–1995	18	1,901	Sweden	Cross-sectional	Health questionnaire and register data	(i) The prevalence of allergic diseases were higher among males who migrated to Sweden before 2 years of age, as compared to those who immigrated at least 2 years old. (ii) Being born in the far east, as compared to being born in Latin America or Indian subcontinent, was a significant risk factor for AR and AD.
Holguin et al. [75], 2005	1988–1994 (NHANES III), 1997–2001 (NHIS)	17 and above, Mexican Americans	30,534	USA	Cross-sectional	Questionnaire and interview	US-born versus Mexican immigrants
							Asthma (NHANES): aOR = 2.12, 95% CI = 1.38–3.28
							Asthma (NHIS): aOR = 2.73, 95% CI = 1.63–5.5
							More than 10 years of residence in the USA
							Asthma: aOR = 1.60, 95% CI = 0.74–3.07
Jerschow et al. [76], 2017	2008–2011	18–74, Hispanics/Latinos	15,573	USA	Cross-sectional	Interview questionnaire	(i) US nativity and residence were not consistently associated with asthma among various US Hispanic/Latino groups
							(ii) Among Mexican and Dominican immigrants to US, asthma risk approaches of US natives over time
Kabesch et al. [77], 1999	1989–1990	9–11	6,490	Germany	Cross-sectional	Self-administered questionnaire according to international standards and skin prick test	Turkish Immigrants versus German natives
							Asthma: aOR = 0.53, 95% CI = 0.3–0.94
Kuehni et al. [78], 2007	1998 and 2003	Mean 30.9 (South Asians), 31.2 (Whites) (age range not reported); South Asians and Whites	6,560	UK	Cross-sectional	Self-completed mail questionnaire	Born in the UK or immigrated before the age of 5
							Asthma: aOR = 0.38, 95% CI = 0.23–0.64
							Asthma prevalence was similar between those who immigrated to the UK before the age of 5 and those born in the UK
Kutzora et al. [79], 2022	2016–2017 (Gesundheits-Monitoring-Einheiten)	Mean 5.4 (age range not reported)	4,767	Germany	Cross-sectional	Health Monitoring Unit survey	Migrant versus nonmigrant
							Asthma symptoms: aOR = 1.54, 95% CI = 1.1–2.15
							Asthma (physician diagnosed): aOR = 1.49, 95% CI = 0.72–3.09
Leung et al. [80], 1994	nr	Mean 31.6 (age range not reported)	636	Australia	Cross-sectional	Telephone interview	Immigrated to Australia aged at least 20
							AR: aOR = 2.4, 95% CI = 1.1–5.0
							Asthma: aOR = 1.6, 95% CI = 0.7–3.9
							At least 5 years resided in Australia
							AR: aOR = 3.7, 95% CI = 1.8–7.4
							Asthma: aOR = 1.4, 95% CI = 0.7–30
Lien [81], 2008	1999–2001 (Oslo Health Study)	15–16	7,343	Norway	Cross-sectional	Questionnaire	Immigrants versus nonimmigrants
							Asthma (boys): prevalence = 10 versus 13.7 (significant)
							Asthma (girls): prevalence = 9.5 versus 15.3 (significant)
							AD (boys): prevalence = 16.1 versus 25.5 (significant)
							AD (girls): prevalence = 22.7 versus 37.6 (significant)
Lombardi et al. [82], 2011	2009	18–67	1,557	Italy	Cross-sectional	Physician-diagnosis and medical records	Italians versus immigrants
							AD: prevalence = 14 versus 7.4 (significant)
							AR: prevalence = 95 versus 93 (ns)
							Asthma: prevalence = 47 versus 55 (significant)
Lombardi et al. [83], 2014	2010	1–15	402	Italy	Cross-sectional	Questionnaire	Between immigrant children born abroad: (i) asthma prevalence was not significantly different, (ii) rhinitis was significantly higher among Italians, (iii) AD was not significantly different. Among those who migrated at less than 4 years of age versus at least 4 years of age: (i) asthma was not significantly different, (ii) rhinitis was significantly higher, (iii) AD was significantly higher
Maheswaran et al. [14], 2018	2001–2010 (SHAIPS-2 and a separate analysis of a primary care dataset)	25 and above	10,185	UK	Cross-sectional (SHAIPS-2) and population cohort study (unique)	Postal questionnaire	Migrants to affluent area versus those remaining in deprive areas
							Asthma: aOR = 0.7, 95% CI = 0.53–0.93
							Migrants to deprived areas versus those remaining in affluent areas
							Asthma: aOR = 1.71, 95% CI = 1.25–2.35
Marcon et al. [84], 2011	2006 (Viadana study)	3–14	3,854	Italy	Cross-sectional	Questionnaire	Born abroad versus born in Italy
							Asthma symptoms: IRR = 0.47, 95% CI = 0.26–0.82
							AR: IRR = 1.17, 95% CI = 0.61–2.24
							AD: IRR = 0.43, 95% CI = 0.23–0.83
Martin et al. [85], 2013	2008–2011 (HealthNuts study)	11–15 months	4,972	Australia	Cross-sectional, population-based	Parent-completed questionnaire	Immigrant versus Australian native
							AD: aOR = 1, 95% CI = 0.7–1.2
Migliore et al. [86], 2007	2002 (SIDRIA-2)	6–7, 13–14	20,016	Italy	Cross-sectional	Standardized self-administered questionnaire	Duration of residence in Italy <5 years
							Asthma: aOR = 0.39, 95% CI = 0.23–0.66
							Duration of residence in Italy at least 5 years
							Asthma: aOR = 0.93, 95% CI = 0.59–1.46
Moyes et al. [87], 2012	1992–1993 (ISAAC phase I), 2001–2003 (ISAAC Phase III)	6–7, 13–14	24,190	New Zealand	Cross-sectional	ISAAC questionnaire	Born in New Zealand versus born abroad
							AR (6–7 years): aOR = 1.67, 95% CI = 1.29–2.15
							AR (13–14 years): aOR = 1.35, 95% CI = 1.12–1.62
Netuveli et al. [88], 2005	1991–1992 [Fourth National Study of Morbidity Statistics in General Practice (MSGP4)]	nr	415,528	UK	Prospective cohort study	Questionnaire	Born outside the UK versus born in the UK
							Asthma: aOR = 0.93, 95% CI = 0.82–1.06
Newbold [13], 2006	1994/95, 1996/97, 1998/99, and 2000/01 (National Population Health Survey [NPHS])	12 and above; 35 and above used for analysis	Not reported	Canada	Longitudinal	Questionnaire	Immigrants versus Canadian natives
							Asthma (1994/95): prevalence = 3.6% versus 4.7%
							Asthma (2000/01): prevalence = 4.1% versus 8.1% (significant)
Nisar et al. [89], 2021	2007–2016 (HABITAT [How Areas in Brisbane Influence HealTh and AcTivity] study)	40–65	11,035	Australia	Longitudinal, multilevel study	Self-administered mail survey	Immigrant from high-income country versus Australian-born
							Asthma: prevalence ratio = 0.98, 95% CI = 0.96–0.99
							Immigrant from low-middle-income country versus Australian-born
							Asthma: prevalence ratio = 0.95, 95% CI = 0.94–0.95
Ormerod et al. [90], 1999	1990–1991	Age range not reported, Asians	1,783	UK	Cross-sectional	Questionnaire	(i) Asians born in the UK were more likely to report asthma symptoms and be on treatment for asthma. (ii) The rates of symptoms and medication use increased among those who were born abroad as duration of residence in the UK increased. (iii) classifying Asians as asthma cases by symptoms rather than diagnosis resulted in a substantial rise in asthma prevalence, suggesting an underdiagnosis
Pereg et al. [91], 2008	1980–2004	17	1,466,654	Israel	Cross-sectional	Medical records	(i) Prevalence of asthma among Ethiopian immigrants was significantly lower than that of Israel. However, this difference became less pronounced the earlier Ethiopian immigrants came to Israel. (ii) Prevalence of asthma among FSU immigrants was significantly lower than that of Israel. however, the longer the residence in Israel, the higher the prevalence
Philipneri et al. [92], 2019	1994–2009 (National Longitudinal Survey of Children and Youth [NLSCY])	2–26	15,799	Canada	Longitudinal	Questionnaire	First-generation immigrants versus nonimmigrants
							Asthma: aOR = 0.21, 95% CI = 0.07–0.67
Powell et al. [93], 1999	nr	13–19	9,794	Australia	Two stage, stratified, cross-sectional survey	Self-administered questionnaire	Duration of residence in Australia above 5 years
							Asthma symptoms: aOR = 2.1, 95% CI = 1.1–4.0
Rodriguez et al. [94], 2017	2007–2010	5–16	2,510	Ecuador	Cross-sectional	ISAAC Phase II questionnaire	Migrant versus non-migrant, within Ecuador
							Asthma symptoms: aOR = 1.25, 95% CI = 0.91–1.71
							Rural to urban migration, within Ecuador
							Asthma symptoms: aOR = 2.01, 95% CI = 1.3–3.01
							Urban to urban migration, within Ecuador
							Asthma symptoms: aOR = 0.91, 95% CI = 0.61–1.63
Rosenberg et al. [95], 1999	nr	Mean 42.7 (age range not reported)	906	Israel	Cross-sectional	Patient file, reviewed by 2 physicians	Ethiopian origin versus Israeli
							Asthma: prevalence rate = 0.17 versus 0.058 (significant)
							Over time, the prevalence rate of asthma among immigrants approached that of the non-Ethiopian population in the same geographic region
Silverberg et al. [96], 2013	2007–2008 (National Survey of Children’s Health [NSCH])	0–17	79,667	USA	Cross-sectional	Telephone interview	Foreign-born versus US-born
							Asthma: aOR = 0.35, 95% CI = 0.24–0.52
							AD: aOR = 0.45, 95% CI = 0.3–0.69
							AR: aOR = 0.34, 95% CI = 0.22–0.52
Stoecklin-Marois et al. [54], 2015	2006–2007 (MICASA study)	Mean 37.7 (age range not reported), Latinos	702	USA	Cross-sectional, population-based	Survey form	Higher acculturation and residing for at least 15 years in the USA were associated with an increased risk of asthma
Subramanian et al. [97], 2009	2000–2001 (Project on Human Development in Chicago Neighborhoods study, wave 3)	14–53	2,209	USA	Multilevel, multimethod longitudinal study	ISAAC survey	Foreign-born versus US-born
							Asthma: aOR = 0.26, 95% CI = 0.17–0.39
Svendsen et al. [98], 2009	2001	3–8	6,396	USA	Cross-sectional	Survey form	Lifelong residence in El Paso versus Late Immigrant to El Paso (Since after entering 1st grade)
							Asthma: aOR = 1.75, 95% CI = 1.24–2.46
Tobias et al. [11], 2001	1991–1993 (European Community Respiratory Health Survey [ECRHS])	20–45	20,097	Europe, USA, Australia, and New Zealand	Cross-sectional	ECRHS questionnaire	Immigrants versus nonmigrants
							Asthma: aOR = 1.21, 95% CI = 1–1.51
							Emigrants versus nonmigrants
							Asthma: aOR = 1.31, 95% CI = 0.96–1.78
Ventura et al. [99], 2004	2001	20–44, Albanian	152	Italy	Cross-sectional	Questionnaire based on the ECRHS	Duration of residence in Italy >7 years
							AR: prevalence rate = 20.4 versus 2.5 (significant)
Volodina et al. [100], 2011	1990–2005	20–70	114	Germany	Cross-sectional	Self-administered questionnaire and phone interviews	Migrants from former Soviet Union versus German natives
							Asthma: 1-year period prevalence = 7% versus 6%
Wohl et al. [101], 2014	1998–2008	17	845,326	Israel	Retrospective cohort study	Database records, physician examination	Being an immigrant to Israel was a risk factor for AD
							Immigrant after 7 years of age
							AD: aOR = 1.88, 95% CI = 1.7–2.04
Wong et al. [102], 2007	nr	2–6, Chinese	3,089	Hong Kong	Cross-sectional	ISAAC phase II questionnaire	Born in Hong Kong versus born in Mainland China
							Asthma symptoms: aOR = 2.86, 95% CI = 1.39–5.90
Yao and Sbihi [103], 2016	2005 (Canadian Community Health Survey [CHS Cycle 3.1])	12 and above	116,232	Canada	Cross-sectional	Questionnaire	Duration of residence in Canada <10 years
							Allergic disease: aOR = 0.35, 95% CI = 0.3–0.4

AD, atopic dermatitis; AR, allergic rhinitis; ARIA, Allergic Rhinitis Impact on Asthma; aOR, adjusted odds ratio; BPAS, Brief Pediatric Asthma Screen; ECRHS, European Community Respiratory Health Survey; IRR, incidence rate ratio; ISAAC, International Study of Asthma and Allergies in Childhood; NHANES, National Health and Nutrition Examination Survey.

¹Where reported in the original research article, the number of subjects included in the statistical analyses was considered as the study sample size and indicated in italics in the table above.

Overview of Measures of Migration

Forty articles studied the movement of individuals across regions, with 39 evaluating cross-border movement between countries and 1 assessing the movement of individuals between affluent and deprived areas within the UK (online suppl. Table 6). Twelve articles assessed the relationship between duration of residence in the destination country with allergic disease among immigrants, 9 recorded the age at immigration, while 2 assessed acculturation which could be interpreted as a proxy measure of duration of residence. Most studies were conducted from the standpoint of an affluent region, wherein immigration entailed the movement of individuals from less affluent or equally affluent regions to an affluent region. Only 1 study analyzed the migration of individuals to a less affluent area [14]. Furthermore, while the destination of migration was often defined, the countries of origin were not usually reported with immigrants possibly originating from a variety of countries. Among studies that specified the country of origin, Mexican migrants to the USA (n = 5) and Chinese migrants (n = 5) were the most well represented, Ethiopians and migrants from the former Soviet Union (i.e., Russia, Ukraine, Georgia, Kazakhstan) were reported by two studies each, while Albanians, Malaysians, and Turkish were reported by one study each.

Associations between Migration and Allergic Disease

Despite the heterogeneity between studies with respect to migrant ethnicities, countries of origin, individuals categorized into the immigrant group, and country of destination, several unifying conclusions could be achieved. First, immigrants to regions variously referred to as affluent, high-income, or developed generally exhibited lower prevalences of AR, AD, and asthma as compared to the native population. This observation was consistent regardless of whether the migration event occurred between countries or within countries. A single noteworthy study compared migrants to deprived areas against those remaining in affluent areas and found that there was an increased likelihood of asthma among the migrant group (adjusted odds ratio = 1.71, 95% CI = 1.25–2.35). Second, as the duration of residence in the destination country increased, the prevalence of allergic diseases among the immigrant group increased – in most studies where the host country population exhibited a higher prevalence of allergic diseases than the immigrant population, the prevalence of allergic diseases among immigrant population approached that of the host region and stabilized. Third, the earlier the age of immigration to a country wherein the prevalence of allergic diseases was comparatively higher, the lower the magnitude of difference between the prevalence among immigrants and that of the host population. However, this finding could have been confounded by the possibility that at the time of data collection, individuals who immigrated at a younger age would have resided for a longer duration in the host region than those who immigrated later. Findings from the various studies are summarized in Table 1.

pORs were calculated for comparable outcomes and exposure-reference group pairs. Migration was not significantly associated with AD prevalence (pOR = 0.68, 95% CI = 0.31, 1.49, I² = 89.95%; Fig. 2a). A lower odds of AR was observed among immigrants when compared to nonmigrants (pOR = 0.58, 95% CI = 0.45, 0.74, I² = 73.36%; Fig. 2b), but residing at least 10 years in the destination country among immigrants compared to a residence duration of less than 10 years was associated with an increased likelihood of AR (pOR = 8.36, 95% CI = 4.15, 16.81, I² = 0.00%; Fig. 2c). Being a migrant also significantly decreased one’s odds of asthma (pOR = 0.56, 95% CI = 0.44, 0.72, I² = 86.07%; Fig. 3a), but residing in the host country for at least 10 years significantly associated with an increase in asthma among immigrants (pOR = 1.85, 95% CI = 1.25, 2.73, I² = 0.00%; Fig. 3b). Immigrating aged 5 and below compared to immigrating after the age of 5 was not a significant risk factor (pOR = 1.01, 95% CI = 0.68, 1.50, I² = 82.62%; Fig. 3c). Of note, there was significant heterogeneity in almost all the pORs calculated herein, except for the associations of 10 years of residence in the host country with AR and AD, respectively. While there appeared to be publication bias for AD and AR studies, funnel plots and Egger’s test p value indicated nonsignificant publication bias for asthma studies (online suppl. Fig. 1).

Fig. 2.

Forest plots and pORs obtained from random effects meta-analyses. a Forest plot for AD and migration. b Forest plot for AR and migration. c Forest plot for AR and duration of residence in destination country.

Fig. 3.

Forest plots and pORs obtained from random effects meta-analyses. a Forest plot for asthma and migration. b Forest plot for AR and duration of residence in destination country. c Forest plot for AR and age at immigration.

Discussion

Immigrants Exhibit Significantly Lower Odds of AR and Asthma

Results from our meta-analysis showed that immigrants, when compared to natives from their destination country, have a significantly lowered odds of AR (pOR = 0.58) and asthma (pOR = 0.56). These findings are reminiscent of the putative HIE, alternatively referred to as the healthy migrant hypothesis, which postulates that migrants exhibit better health than nonmigrants [15]. This trend has been observed in among migrants to Australia, Canada, the UK, and the USA – chronic diseases (comprising cancer, cardiac disease, diabetes, ulcers, arthritis, hypertension, bronchitis or emphysema, and asthma) were less prevalent among immigrants as compared to natives [16]. Additionally, the HIE has been found to be stronger for migrants from developing countries than those from developed countries [16].

The HIE has popularly been attributed to possible selection biases such as the salmon-bias or reverse-migration hypothesis where migrants with poorer health are more likely to return home, or a disparity in access to healthcare resulting from language barriers and low socioeconomic status [17–19]. The former speculation has been dispelled more than once [18, 20]. Furthermore, studies included in this review found an increase in the prevalence of AR and asthma among immigrants who had resided in their host countries for at least 10 years. On the latter count, a study of undocumented immigrants to the USA who are frequently of lower socioeconomic status and lack access to healthcare showed the persistence of the HIE [21]. Thus, while healthcare access and sociocultural factors may modify the effects of immigrant status on health, they do not fully account for the HIE [18].

The Human Microbiome and Allergic Disease

The human microbiome refers to the genetic material of microbiota which includes bacteria, fungi, and viruses [22]. While most of the human microbiota lives in the gut, various body sites such as the nasal cavity, skin, respiratory tract, and lungs are also hosts to microbes [22, 23]. Within their local sites, resident microbes have been implicated in various disorders – for example, the skin microbiota comprises Propionibacterium acnes which is associated with acne, while the fungi species Malassezia spp. has been implicated in seborrheic dermatitis [24]. Further examples of local disorders associated with their resident microbiota include inflammatory bowel disease, which may be exacerbated due to contact of bacterial antigens with immune cells as a result of a defective mucosal barrier [25]. Regardless of the site of colonization, the human microbiota also appears to be able to exert systematic effects on the human body [26]. Increasingly, evidence supporting the essential role of the microbiota on human health is being revealed, such as the regulation of the immune system by commensal intestinal bacteria [27]. Additionally, the gut microbiome has also been implicated in the immune response and allergic disease, where children suffering from allergic diseases had fewer microaerophilic bacteria and increased aerobic microorganisms among their intestinal microflora [28].

Indeed, there has been an emergence of evidence associating the microbiome with allergic diseases [29]. Asthmatic individuals exhibited increased proteobacteria such as Haemophilus spp. in the bronchi as compared to non-asthmatics, notwithstanding a dearth of causal evidence [30]. Further studies established that bacterial diversity, community composition, and relative abundance of specific phylotypes are associated with the degree of bronchial hyperresponsiveness among asthmatics, suggesting a possible role for microbes in asthma severity [31]. With AD, disease states were characterized by changes in microbial diversity and composition of Staphylococcus spp., where untreated flares were associated with reduced bacterial diversity and high Staphylococcus proportions [32]. Additionally, dysbiosis of Faecalibacterium prausnitzii in the gut has been shown to impair the gut epithelial barrier, releasing material from the gut into systematic circulation, which triggers TH2-type immune responses to allergens in the skin [33]. As such, the importance of the human microbiota and the maintenance of its balance are appearing increasingly paramount.

However, urbanization appears to be disruptive to the human microbiome. Westernization of diet, air pollution, increased antibiotic use, and improved hygiene were notable factors known to affect human microbiota [34]. Across different regions, urbanization has been shown to exert a consistent effect on the microbiome – a study of an urbanizing Chinese city found that the microbial composition of the Chinese population converged with that of American populations [35]. Notably, migration to developed countries has also been implicated as a contributing factor to the changes in the human microbiome, with multiple studies highlighting the changes in human microbiota following migration [36]. As such, exposure to an urbanized environment following migration could affect the human microbiota, which in turn mediates immune response and allergy, leading to a change in allergic disease manifestation.

Parasitic Infections and Allergic Disease

The relationship between helminthic infection and allergy has not gone unnoticed in Israel, which has received a large number of Ethiopian immigrants on whom studies have been conducted and reviewed herein [37]. In 2006, a meta-analysis of intestinal parasite infection and asthma reported that while there was a nonsignificant association of any parasite infection with asthma, stratifying the meta-analysis by species showed that Ascaris lumbricoides was significantly associated with an increased odd of asthma while hookworm infection was associated with a significantly decreased odd of asthma in a dose-response manner [38]. Later, a 2022 meta-analysis found no overall association between helminth infections and allergic diseases, but A. lumbricoides infections were associated with an increased risk of bronchial hyperresponsiveness in children and atopy in adults [39]. Concerning eczema, epidemiological studies have found that the association with helminth infection varies from positive to negative [40]. Helminth infections were associated with decreased skin test reactivity in several epidemiological studies; on the other hand, studies evaluating anthelmintic treatment and atopy showed that anthelminthic therapy was associated with an increased incidence of skin reactivity to house dust mites [40]. Interestingly, yet another meta-analysis has shown that helminth infection is associated with an increased gut bacterial diversity, which has been shown to regulate immune response, as discussed above [41].

Overall, confounding factors added levels of complexity to the analysis of the association between helminth infection and outcomes of interest, owing to the possibility of infection with more than one species of parasite, formerly infected individuals being erroneously classified as currently infected individuals due to testing methodology, and the unlikelihood of said infection being the sole mediator of studied outcomes. Taken together, it is possible that the effect of parasitic infections on allergic diseases is confined to specific species, and the association is confounded by other factors, environmental and otherwise. Indeed, studies have demonstrated that IL-10 induced by helminth infection downregulates the production of IgE, but helminth infection is also associated with increased IgE cross-reactivity to various allergens, including house dust mites [42]. As such, there appears to be a highly complex interaction involving multiple environmental and biomolecular processes, rendering this topic a complicated one to study.

Nonetheless, helminth infections and their treatment following migration could be a contributing factor to the changes in the prevalence of allergic diseases among immigrants as compared to populations in destination countries. Moreover, helminth infections remain relevant in many developing parts of the globe [43].

Mediation of Allergic Diseases by Other Environmental Factors

Diet

Diet has been shown to modulate immune functions, with studies showing the mechanisms of the effect of various macro- and micronutrients on immune regulation [44]. Moreover, diet and nutrition have been implicated in the pathogenesis and pathophysiology of allergic diseases – dietary components may directly affect innate immune responses or adaptive immune responses via nutrients and metabolites, while their effects on the gut microbiota and the resulting bacterial metabolites may also modulate immune function [45]. Epidemiological evidence supporting the association of different diets with the risk of allergic diseases has been published. Western dietary patterns that are high in energy, saturated fat, and processed foods but low in fiber consumption were associated with increased asthma and AR risk, while diets based on high-fiber whole foods – characteristic of a Mediterranean diet – were associated with lower asthma and AR risk [46]. Sufficient fiber and micronutrient intake were also associated with a reduction in the symptoms of allergic diseases [47, 48].

Air Pollution

There is increasing epidemiological evidence for the relationship between air pollution and allergic diseases [49]. NO₂ and O₃ have been found to be associated with asthma exacerbations and enhanced allergic responses to aeroallergens [49]. Additionally, high traffic density resulting in traffic-related air pollution was associated with increased incidence of asthma symptoms and bronchodilator use [50]. D Diesel exhaust particles have been found to induce proinflammatory responses in lung epithelial cells and aggravate TH2 immune responses which have been implicated in asthma [51]. Besides respiratory disorders, exposure to traffic-related pollution has been linked to the manifestation and exacerbation of AD, possibly due to oxidative stress of the skin or genetic mutation induced by air pollutants, leading to skin barrier dysfunction and immune dysregulation [52].

Knowledge and Education

Increased access to healthcare and better education could also explain the increased diagnoses of allergic diseases. Moreover, the initial language barrier following immigration could result in poor knowledge of allergic diseases and a greater barrier to health care services. Following adaptation to the region, a better grasp of language, and a changing perception of one’s health relative to their native counterparts, there would be a likely increase in healthcare-seeking behavior. Indeed, studies measuring acculturation, which includes gauging the individual’s usage of the local language, have found an increased risk of asthma among highly acculturated subjects [53, 54].

Inevitably, the movement of individuals from one region to another would result in a change in overall environmental exposure due to the state of development, differing availability of food, and social situation in the destination region. The effect of migration on allergic diseases is unlikely to be due to a single cause but a multitude of contributing factors working in conjunction with each other.

The Epigenetics of Migration and Allergic Disease

With evidence listing a variety of environmental factors that potentially work in complex interactions to influence allergic disease, discerning the risk factors responsible for the observed association between migration and allergic diseases is challenging. Moreover, it has been well established that genetics form a strong basis for the manifestation of allergic diseases [55]. As such, migration and its effect on allergic diseases further involve a complex gene-environment interaction. We propose that DNA methylation studies can help further the understanding of the migration-allergy relationship. DNA methylation refers to the attachment of a methyl group to the C5 position of the cytosine ring in DNA, catalyzed by DNA methyltransferase [56]. Notably, regulation of gene expression by the environment has been proposed to be via the process of DNA methylation [57]. Associations for DNA methylation with air pollution, microbial composition, and even mental state have been discovered and linked to the risk of asthma or AR [58]. Although this field remains to be thoroughly explored, the possibility of novel studies targeting other environmental exposures associated with allergic diseases via DNA methylation gives exciting hope to better understand the influence of migration on allergic diseases.

Conclusions

The findings of this review were limited by the primarily cross-sectional nature of the included articles. In an ideal setting, a longitudinal study would give a better idea of the HIE on allergic diseases, particularly with regards to whether age at immigration influences the risk of allergic diseases, and where reference to the population remaining in the origin country needs to be made. Additionally, allergic disease outcomes were frequently classified using questionnaire data and could be subject to the perception of the respondents, predisposing the studies to a potential source of bias and diminishing the reliability of this review. Objective diagnoses of allergic disease, such as using the spirometry of bronchodilator reversibility test for asthma, could add to the reliability of the outcomes. Furthermore, immigrant groups were highly nonspecific, with little distinction between their country of origin. Nonetheless, the current review was able to obtain 50 studies relevant to migration and allergic disease, with most studies having a low overall risk of bias. Additionally, these studies were identified via a justifiably comprehensive search of the literature. Despite the heterogeneity between studies, a pOR could be obtained to provide readers with an effect size for the association between migration and the allergic diseases.

In conclusion, we have systematically reviewed the existing literature on migration and allergic disease. A reasonably comprehensive summary of the literature was provided, and where appropriate, a meta-analysis was conducted. pORs obtained from the meta-analysis indicated that migration was significantly associated with a decreased risk of asthma and AR, but not with AD. Likewise, residing for more than 10 years in the destination country was associated with an increased risk of AR and asthma. Conversely, immigrating before the age of 5 was not significantly associated with asthma manifestation. We have also discussed the prevailing hypotheses and existing evidence. Finally, we posit that DNA methylation studies have the potential to help expand our understanding of migration and allergic diseases.

Acknowledgment

The authors would like to thank Avinaash Subramaniam for proofreading this review and providing valuable inputs to the manuscript.

Statement of Ethics

An ethics statement is not applicable because this study is based exclusively on published literature.

Conflict of Interest Statement

F.T.C. reports grants from Singapore Ministry of Education Academic Research Fund, Singapore Immunology Network, National Medical Research Council (Singapore), Biomedical Research Council (Singapore), National Research Foundation (NRF) (Singapore), Singapore Food Agency (SFA), and the Agency for Science Technology and Research (Singapore), during the conduct of the study; and has received consultancy fees from Sime Darby Technology Centre, First Resources Ltd, Genting Plantation, Olam International, and Syngenta Crop Protection, outside the submitted work. The other authors declare no other competing interests.

Funding Sources

F.T.C. received grants from the National University of Singapore (N-154-000-038-001), Singapore Ministry of Education Academic Research Fund (R-154-000-191-112; R-154-000-404-112; R-154-000-553-112; R-154-000-565-112; R-154-000-630-112; R-154-000-A08-592; R-154-000-A27-597; R-154-000-A91-592; R-154-000-A95-592; R154-000-B99-114), Biomedical Research Council (BMRC) (Singapore) (BMRC/01/1/21/18/077; BMRC/04/1/21/19/315; BMRC/APG2013/108), Singapore Immunology Network (SIgN-06-006; SIgN-08-020), National Medical Research Council (NMRC) (Singapore) (NMRC/1150/2008; OFIRG20nov-0033), National Research Foundation (NRF) (Singapore) (NRF-MP-2020-0004), Singapore Food Agency (SFA) (SFS_RND_SUFP_001_04; W22W3D0006), and the Agency for Science Technology and Research (A*STAR) (Singapore) (H17/01/a0/008; and APG2013/108). The funding agencies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author Contributions

F.T.C. conceived, supervised, and reviewed the current research study. Q.Y.A.W. conducted the literature review, collected, and analyzed the data, and wrote the manuscript. Both authors read and approved the final manuscript. All authors have read and consented to the publication of this manuscript.

Funding Statement

F.T.C. received grants from the National University of Singapore (N-154-000-038-001), Singapore Ministry of Education Academic Research Fund (R-154-000-191-112; R-154-000-404-112; R-154-000-553-112; R-154-000-565-112; R-154-000-630-112; R-154-000-A08-592; R-154-000-A27-597; R-154-000-A91-592; R-154-000-A95-592; R154-000-B99-114), Biomedical Research Council (BMRC) (Singapore) (BMRC/01/1/21/18/077; BMRC/04/1/21/19/315; BMRC/APG2013/108), Singapore Immunology Network (SIgN-06-006; SIgN-08-020), National Medical Research Council (NMRC) (Singapore) (NMRC/1150/2008; OFIRG20nov-0033), National Research Foundation (NRF) (Singapore) (NRF-MP-2020-0004), Singapore Food Agency (SFA) (SFS_RND_SUFP_001_04; W22W3D0006), and the Agency for Science Technology and Research (A*STAR) (Singapore) (H17/01/a0/008; and APG2013/108). The funding agencies had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability Statement

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.