A systematic review on the relationship between international migration and antimicrobial resistance

doi:10.21203/rs.3.rs-39567/v1

Introduction: Antimicrobial drug resistance (AMR) is considered a serious threat to public health worldwide. The relation between AMR and human mobility, particularly international migration, has drawn attention from the scientific community in recent years. However, several aspects about this relation remain unclear. Therefore, we aimed at expanding and updating previous systematic review studies, with a novel focus on the AMR prevalence in migrants compared to the local population of the host country, to examine external validity of previous findings.

Methods: We searched in Ovid MEDLINE all types of observational studies, without language or year of publication restrictions. We aimed at exploring differences in countries´ bacterial drug resistance rates based on immigration rates. All types of AMR were included, except for those related to HIV/AIDS and Tuberculosis. The comparator group of interest was the local population of the host country. The study protocol is registered with PROSPERO, number CRD42018114436.

Results: After screening of 322 articles, 15 papers were selected for data extraction, including 1930 migrants. Compared to the local population, higher rates of methicillin-resistant Staphylococcus aureus, Panton-Valentine leucocidin positive strains, multidrug-resistant Gram-negative bacteria, vancomycin-resistant Enterococcus, and having at least one multidrug-resistant organism were found in migrants in 12 of the 15 papers. Rates of AMR did not differ significantly in two studies and only one of them reported a lower burden of AMR in migrants.

Conclusions: Higher prevalence of AMR in migrants were presented in the majority of the included articles, addressing the emerge of the circulation of resistant strains within this group. More detailed descriptions, including time span and route taken by migrants to arrive to country of destination and length of stay by the time of inclusion are essential to gain a deeper understanding of the relation in between AMR and migration. Countries with high migration rates outside Europe should be encouraged to implement strategies for screening of both local population and migrants in countries.

Infectious Diseases

General Microbiology

Antimicrobial Drug Resistance

Human Migration

Emigration and Immigration

Health Policy

Infectious Disease Transmission

systematic review

Antimicrobial resistance (AMR) is an adaptive phenomenon of microorganisms to survive in the presence of antimicrobials, which has been documented ever since the discovery of antibiotics [1–3]. The inappropriate use of antimicrobials facilitates the emergence and spread of multidrug-resistant microorganisms (MDROs). The process has dramatically accelerated in recent decades, and nowadays it is not uncommon to encounter infections against which there are virtually no effective antimicrobial regimens [4, 5]. According to the World Health Organization (WHO), AMR is considered one of the most serious threats to public health worldwide [6, 7]. AMR is a global phenomenon, caused by several factors, including the massive use of antimicrobials in human and veterinary medicine, lack of prevention and control measures for healthcare-associated infections, and inadequate WASH infrastructure (water, sanitation, and poor hygiene). AMR not only results in increased mortality, morbidity, and health expenditures, but also affects the economic development [6].

According to the AMR report from the US Centers for Disease Control and Prevention (CDC), carbapenem-resistant Acinetobacter and carbapenem-resistant Enterobacteriaceae are considered urgent threat bacteria with over 8,500 and 13,100 infections in the USA in 2019, respectively [8]. Methicillin-resistant Staphylococcus aureus (MRSA), ranked as serious threat bacteria, causes a wide range of organ-specific infections including skin and subcutaneous tissue, bone and joint infections and pneumonia [9]. Physicians classify the virulence of MRSA strains based on a Panton-Valentine Leucocidin (PVL) factor which is associated to community-level acquired MRSA [10]. Vancomycin-resistant Enterococci (VRE), also classified as a serious threat bacteria, caused 54,500 cases and 5,400 deaths in the USA in 2019. VRE have raised concern due to its fast spread and clinical complications usually associated with gut diseases [11]. The genital bacteria Ureaplasma urealyticum and Mycoplasma hominis are not listed as bacterial threats, but are included in the watch list due to their high colonization capabilities and increasing resistance. They can cause severe complications like pleural effusion, pneumopathy, adenopathy, abscess, or systemic sepsis [12].

A dimension of increasing interest is the relationship between AMR and human mobility of all kinds, particularly international migration. International migration, the crossing of the political border to a country other than that of origin with the intention of settlement [13], is a complex and evolving phenomenon with several public health implications. As a global social process, international migration is directly influenced by international labour stratification, country inequalities, conflicts, poverty, climate change and natural disasters [14]. In all these scenarios, AMR arises as a public health problem of urgent attention, including in migratory contexts.

Two systematic reviews have examined the relation between international migration and AMR, one largely dedicated to drug-resistant tuberculosis and the other focused in European countries. Smalen et al. [15] aimed to identify the burden of AMR acquisition and transmission among refugees and asylum seekers, including Mycobacterium tuberculosis, Escherichia coli, Klebsiella pneumonia, Klebsiella oxytoca, Shigella spp., Staphylococcus aureus, Enterococcus faecium, and Acinetobacter baumannii. The review found a high percentage of resistant strains have been circulating among refugees and asylum seekers. Focusing in Europe, Nellums et al. [16] analysed 23 observational studies that reported on migration and AMR in 2319 migrants, including methicillin-resistant Staphylococcus aureus and antibiotic-resistant Gram-negative bacteria. The results showed a combined prevalence of any infection by resistant bacteria in migrants of 25.4% (I2 = 98%), with refugees and asylum seekers having a significantly higher risk than other groups of migrants and the general population (33% vs. 6%, I2 = 92% and 98%, respectively), with no evidence of AMR transmission from migrants to host populations. The review suggests migrants, and refugees and asylum seekers in particular, were exposed to conditions, such as inadequate sanitation, overcrowding, and restricted access to health services, that favoured the emergence of AMR during transit and upon arrival to host countries. Overall, findings highlight the need to improve living conditions, access to healthcare, and adequate treatment for AMR infections among these vulnerable populations.

Previous research has intended to clarify the relation in between migration and drug-resistance, however several important gaps remain in the literature. First, since AMR infection is due to several causes related to the systems and processes of care provision, economic limitations on systems and countries and human behaviour, difficulties has been encountered to identify where the acquisition of AMR specifically takes place and which conditions and factors play a role in the increase of incidence [17]. Second, most studies have been carried out in high-income countries where evidence-based measures for the prevention and control of infection are established. However, the highest percentage of refugees worldwide are hosted by developing countries, which do not have the resources like surveillance systems providing reliable data to respond and describe the burden that AMR causes, resulting in an information gap [18]. With this systematic review we aimed at expanding and updating previous systematic review studies dedicated to analyse the relationship between international migration and AMR.

Study design

We conducted a systematic review following international guidelines [19]. The review protocol was made available at the Prospero Web Page before the study was conducted (CRD42018114436) [20]. This manuscript details the review developed for the following research question: According to current international evidence, is there a relationship between AMR and international migration rates and immigrants´ profile (country of origin, legal status, length of stay) at country level? Our systematic review summarises primary studies that include a wide range of international migrants and AMR phenotypes. We followed general principles for conducting systematic reviews, in terms of search strategy, screening of studies, and appraisal of the methodological quality of included primary studies [21].

Types of studies

The review included original research articles written in English, Spanish, Portuguese, French, Dutch, and German. When available, systematic reviews were included as background information, but not in the actual review. We included all types of observational studies (cross-sectional; case-control; prospective, retrospective or mixed cohort designs) and then objectively assessed by the quality of evidence provided, using the quality assessment – Risk of Bias (Strobe v.4 & MMAT v.2011).

Types of participants

For this review, the concept of international migrant was defined as a person who crosses a national border to reside in another country for a significant period of time (i.e. at least one year) [22, 23]. In addition, asylum seekers and refugees were included as specific migrant groups. Studies conducted with ethnic minorities that did not clearly state to be international migrants were set aside and not included in this analysis.

Inclusion criteria: The review was ample, including studies focusing both at general population and specific groups like children or adults of any age group. No sex, ethnic or geographic limitations were made. We were aware that some articles might have used different subpopulations; therefore, we recorded these variations included in each study, in order to describe possible differences in results.

Exclusion criteria: We excluded articles dealing with non-human samples such as animals and environmental specimens. This review was limited to studies on human subjects only.

Types of AMR

We aimed at exploring differences in countries´ bacterial AMR rates based on immigration patterns. We considered the definition of “multidrug resistance” as any organism exhibiting resistance to at least one representative of three or more families of antimicrobials [6]. In this review, we included all types of bacterial AMR, except for those related to tuberculosis, as they show a distinctive pattern of AMR that require a separate in-depth analysis.

Comparator

The comparator group of interest in this review was the local population at every country were the study was conducted. Studies comparing the prevalence of AMR in migrants with the local population using national prevalence numbers were excluded.

Identification of studies

Given its relevance to biomedical science, we conducted our search in Ovid MEDLINE (1946 onwards). The search strategy was the following: ("Transients and Migrants"[Mesh] OR "Emigrants and immigrants"[Mesh] OR "Refugee"[All Fields] OR "Migration background"[All Fields] OR "Immigrant background"[All Fields] OR "Migrant"[All Fields] OR "Migrants"[All Fields] OR "Immigrant"[All Fields] OR "Immigrants"[All Fields] OR "Ethnic minority"[All Fields]) AND ("Drug Resistance, Microbial"[Mesh] OR "Drug Resistance, Bacterial"[Mesh] OR “antimicrobial resistance” OR “antibiotic resistance” OR “multidrug resistance”)

Selection of studies

Two authors independently screened titles and abstracts as per inclusion criteria. We obtained full-text manuscripts for all titles selected during the screening process, contacting study authors if necessary. Full-text articles were then reviewed by two authors independently; disagreements were solved through discussion, with a third review author to arbitrate if necessary. The selection of reviews was based on the inclusion and exclusion criteria relating to types of studies, participants, and interventions. We included primary studies regardless of reported outcomes, date and language of publication, and study quality. In addition, references of included studies were extracted to create a matrix to report the unique primary studies identified in this systematic review. This allowed us to assess overlap in the primary studies reported by various reviews.

Dealing with lack of information: When selected papers could not be obtained in their full text format or when the study did not provide enough information to ascertain suitability for inclusion (i.e. classified as “unclear”), we contacted the authors of the study by email at least three times within 30 days asking for the complete article or additional information. In case of failure to communicate with the primary investigators, the study was excluded. We did not need to deal with data duplication, as we only searched the Ovid MEDLINE electronic database.

Data extraction and management

We created a data extraction sheet in Microsoft Excel to collect data from included primary studies, and we piloted the data extraction sheet by entering data from the first ten included studies, adjusting as necessary. Data entry was performed independently by two trained investigators; a third researcher conducted a random check (n = 3) during data extraction, finding no relevant variation during this internal review.

We recorded the following information: full publication reference, first author, year of publication, journal´s impact factor (if available), journal´s country, study objective, study design, level of analysis (individual, aggregated), study representativeness (nation-wide, local), type of sampling strategy (random, purposive, convenience), recipient country, recipient continent, country of origin of migrants, continent of origin of migrants, type of migrants (no information, mixed/general, refugees/asylum seekers, labour migrants, other), generation of migrants (first or second generation), length of stay of migrants in host country in years, migration status (regular, irregular, no information), comparison group (yes/no), description of comparison group (if available), final sample size of migrants for analysis, final sample size of comparator (locals) for analysis, final age range of study populations (if available), data collection (in months), main study outcome, type of indicator of main outcome (proportion, odds ratio, incidence rate ratio, hazard ratio, prevalence, incidence, other), secondary outcome, type of indicator of secondary outcome (proportion, odds ratio, incidence rate ratio, hazard ratio, prevalence, incidence, other), tertiary outcome, type of indicator of tertiary outcome (proportion, odds ratio, incidence rate ratio, hazard ratio, prevalence, incidence, other), data source (surveillance screening in hospital, screening of high risk population, laboratory records, screening of patients with symptoms), type of data (self-reported, parent-reported, healthcare register, other), confounding variables, main result, and study main conclusion (migrants higher resistance, migrants lower resistance, no difference with locals, nor reported, unclear) and type of resistance test.

Assessment of methodological quality of included reviews

A detailed quality assessment of each selected study was performed and a separate Microsoft Excel spreadsheet was used to record the evaluation. We based our assessment on the items identified in the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) initiative and the Mixed Methods Appraisal Tool Version 2011 (MMAT-Version 2011). The following quality assessment information was recorded (yes = 1/no = 0 answers; 20 items): Is the research question well-justified / sound? Are objectives explicit? Is there a clear hypothesis? Is the study design declared? Is the study setting clear? Are study participants described? Are all relevant study variables described? Is source of data clear? Are data sources adequate to respond the hypothesis? Is analysis adjusted by potential confounders? Is sample size clear? Is the sample representative of the population of interest? Has the sample size enough power for analysis? Is data analysis clearly described? Is data analysis adequate to respond to the hypothesis? Are there sound descriptive stats? Are there clear results for the main outcome? Are there clear results for any secondary outcome? Is there a discussion of study limitations? Are figures/stats well presented? Are results generalizable? Is there a clear source of funding of this study? Is there a clear disclosure of the authors?

From these questions, we created a quality-assessment score of a 0–20 range, which was categorised into four categories as follows: low (1–5 score), medium-low (6–10 score), medium high (11–15 score), and high quality (16–20 score).

Data analysis

We found great heterogeneity in types of migrants, local populations for comparison, and types of AMR in selected studies. Therefore, we report results as a narrative synthesis of the characteristics of included primary studies. We synthesised large epidemiological patterns of AMR between different types of international migrants and the local population in each country.

Ethical considerations

The present study follows the ethical standards of the institutional research committee and the 1964 Declaration of Helsinki and its later amendments. We followed international guidelines on the conduction of systematic reviews [21].

The search was conducted on September 24^th, 2018. Two independent reviewers screened titles and abstracts of 322 records; 166 were included for full-text screening (Figure 1, Study selection).

The most common reasons for exclusion were: i) study not conducted in humans, ii) non-primary studies (e.g. editorials, commentaries, debates), and iii) absence of a direct comparison with the local population. All excluded studies with their respective reason for exclusion are listed in Table S1 (Supplementary material 1, Data hits selection). Among 166 selected papers, we were unable to obtain the full-text format of 12. After fully analysing the remaining 154 manuscripts, 139 were excluded for failure to fulfil the inclusion criteria. 15 studies were selected for data extraction [Reference list, Selected Papers ]. All data were extracted and studies were assessed on their quality as described in the review protocol [18]. A detailed description of the data extraction and the quality assessment of the 15 selected papers are provided in Table S2 (Supplementary material 2, Data extraction of selected papers) and Table S3 (Supplementary material 3, Quality assessment of select papers). All studies were published in English, except for one article that was written in Spanish.

Description of included studies

We included 15 articles, all of which analysed the burden of AMR in migrants as compared to the local population of the host country. A summary of the findings is provided in Table 1.

Table 1 Summary of findings

Topic	Nr
Studies with a direct comparison group	15
Study main conclusion
migrants > resistance	12
migrants < resistance	1
no difference	2
Migration status
regular	2
no information	13
Type of migrants
no information	8
mixed	1
refugees	5
labour	-
other	1
Data source / sample recollection
Surveillance screening in hospital/health care centrum	7
Screening of high risk population	4
Laboratory records / Screening for positive strains or high resistance profiles	3
Screening of patients with symptoms	1
Type of study
Retrospective	4
Prospective	10
Mixed	1

In 13 of the 15 articles, there was no specific data on the migration status of participants; the remaining 2 reported regular migrants. In terms of type of migrants, general migrant population was reported in one manuscript and 5 reported refugees. No details were specified in 9/15 studies and none of the reports focused specifically in labour migrants. Most studies came from developed countries located in the northern hemisphere, particularly Europe and the US. In terms of sample collection, 7 studies included patients using surveillance screening in hospitals or health care centres, 3 accessed laboratory records for MDR strains, 4 screened high-risk population and one study included patients with symptoms compatible with infections. The study design was retrospective and prospective in 4 and 10 studies, respectively; one of them mixed results from retrospective and prospective data.

Study participants

Sample sizes ranged from 4 to 973 migrants and from 4 to 12989 natives. International migrants came from a range of countries and continents of origin. Five of them reported migrants coming from all continents and the rest included migrant population originating mostly from Asia, Africa and Latin America. Countries of origin were multiple in all studies.

Comparators

The description of the local population as a comparator was often brief but ranged from general hospital population to particular subsets of hospital registries from local natives. In general, there was little description of the comparison groups profile and characteristics, as well as poor justification regarding its selection for each study.

Quality assessment

The quality of the most papers ranged between medium-high to high with an average quality assessment (QA) score of 17,3. Dimensions particularly underreported were adjustment for potential confounders (only 8/15 considered confounding in the analyses) and detailed description of the population (6/15 did not appropriately describe the population). The best reported dimensions were whether the setting and the analysis were clearly described (15/15), followed by whether the research questions were well justified (14/15) and data source adequate to research question (14/15). Table 2 describes a QA summary of selected papers. [INSERTION TABLE 2] Finally, most studies were conducted in developed, high-income countries in North America and Europe, which speaks to the lack of research on AMR and international migration in the developing world. The top-10 countries receiving the largest numbers of migrants and those with highest drug resistance index (DRI) are presented in table 3 to underline this problem [21,22].

Table 3 Top-10 countries hosting migrants versus countries with the highest DRI (2020)

Host country	Number of inmigrants (millions)	Country	Drug Resistance Index
USA	50.7	India	71 (65-78)
Germany	13.1	Thailand	60 (55-66)
Saudi Arabia	13.1	Ecuador	60 (43-77)
Russia Federation	11.6	Vietnam		59 (36-82)
United Kingdom	9.6	Romania	57 (41-74)
United Arab Emirates	8.6	Serbia	56 (42-70)
France	8.3	Turkey	55 (50-60)
Canada	8	Taiwan	55 (41-69)
Australia	7.5	South Africa	54 (50-59)
Italy	6.3	Venezuela	53 (27-78)

Main outcomes of the studies

Among the 15 papers included, results of 12 suggested that migrants presented higher AMR frequency than the local population. Rates of AMR did not differ significantly in two studies and only one of them reported a lower burden of AMR in migrants. Table 4 presents the data extraction of the 15 articles by AMR strains, showing the prevalence rates of the migrant groups and the local population. [INSERTION TABLE 4]

Types of MDROs included in the studies

Methicillin-resistant Staphylococcus aureus

Eight studies analysed MRSA as part of their outcomes (Table 4A), with 6 of them reporting a higher MRSA prevalence in migrants as compared to the local population (differences ranged from 4.4% to 65.7%). In contrast, Piper et al. (2016) studied rates of community-acquired MRSA carriage in nares and wound infections and reported lower prevalence in the migrant population as compared to locals (differences ranged from 5.4% in nasal infections and 9.5% in wound infections respectively). The article by Frick et al. (2010) did not find significant prevalence differences.

Finally, 3 out of 4 articles specifically searching for Panton-Valentine leucocidin-producing MRSA strains, found higher rates of such isolates in migrants as compared to the local population (Table 4B).

Multidrug-resistant Gram-negative bacteria(MDRGN)

All five articles reporting on MDRGN bacteria found a higher prevalence among migrants, with differences ranging from 16.3% up to 66.7% compared to the local population (Table 4C). Of note, all such studies performed screening of high-risk population, including asylum seekers and refugee patients in the hospital setting. Sánchez-Montalvá et al. (2015) studied subjects with signs of infection and reported a higher prevalence of imported drug-resistant Salmonella typhi or Salmonella paratyphi among immigrants arriving to Spain within 4 weeks of symptom onset (75% vs. 8.3%, p=.001). Similarly, Banatvala et al. (1994) prospectively studied patients attending for routine diagnostic upper gastrointestinal endoscopy and found higher rates of metronidazole-resistant Helicobacter pylori in migrants as compared to subjects born in the UK (43% vs. 17%, p=.001).

Vancomycin-resistant Enterococci

Oelmeier et al. (2017) studied anorectal colonization with VRE in pregnant refugees and pregnant residents upon admission at the clinic and found a higher rate among the refugees women (1.8% vs. 0%, respectively) (table 4D).

At least one MDRO

Costa et al. (2017) reported on the rates of intestinal colonization with at least one MDRO on hospital admission and found significantly higher rates of colonization among non-Italian vs. Italian children (Table 4E). A difference of 37.1% was found for MDRO carriage including MRSA, extended spectrum beta-lactamase (ESBL)-producing and carbapenem-resistant Enterobacteriaceae, and VRE. Perniciaro et al. (2018) compared cases of invasive pneumococcal disease, defined as Streptococcus pneumoniae isolated from a normally sterile site, in refugee children with Germany-born children. A higher percentage of MDR pneumococcal isolates was found in the refugee children group (38% vs. 2%, respectively).

Ureaplasma urealyticum & Mycoplasma hominis

Leli et al. (2012) reported on differences in prevalence and antimicrobial susceptibility of U. urealyticum and M. hominis in a population of Italian and immigrant outpatients reporting symptoms of urethritis (Table 4F). Immigrants showed to have a 6.3% higher prevalence of U. urealyticum and 1.6% higher prevalence of M. hominis. Susceptibility for both bacteria was tested to eight different antibiotics, but these results were not shown in between groups.

Summary of main findings

This systematic review shows that the majority of the included studies analysing the burden of AMR on migrants as compared to the local population of the host country, present a higher prevalence in the migrant population. These results confirm the emerge of the circulation of resistant strains among migrant groups being highlighted in previous research and reports of international health organisations like the WHO [15, 16, 22–24].

Quality of the evidence

All papers included in this systematic review were categorized as medium-high to high quality according to the QA 20-items scale, with an average QA score of 17,3. However, we acknowledge that the evidence has some key limitations. In 5 of the 15 articles screening was performed on refugee patients in hospital settings or asylum seekers. Since these specific groups are known for being high-risk population, the results might give an overestimation of resistance prevalence [25, 26]. None of the 15 articles described the time span and the route taken by the migrants to arrive to the country of destination, neither about the length of stay from the moment of arrival to the date of inclusion. While knowing that long complicated routes and staying in high-risk areas such as detention centres can have a major impact on the health of migrants, this additional information could be of big value in understanding the relation between migrants and antimicrobial resistance [26–29]. A more diverse migrant group, including detailed description would give a better representation of the migrant population and therefore result in generalizable evidence.

A second dimension that affects the quality of evidence is the strategy chosen for the inclusion of articles. For this review, the decision was to only include articles making a direct comparison between migrants and the local population. Studies comparing the prevalence of AMR in migrants with the local population using national prevalence numbers were excluded, hence, resulting in a smaller total inclusion. Furthermore, we observed that the current state of the literature greatly underrepresents some geographical regions, particularly developing countries, which play a major role in migration worldwide and are largely affected by AMR. Therefore, future studies providing information from these regions are urgently needed to draw appropriate conclusions in this matter. In addition, it is important to highlight that other variables play an important role in determining the risk of MDRO acquisition and spread in the population, including overcrowding, antibiotic usage, and proper healthcare access, all of which are potentially relevant in the migrant population.

This systematic review provides an update of the evidence available to understand the relationship between AMR and international migration. This study shows the importance of considering international migration as a relevant dimension in clinical practice for AMR, as it might be associated with higher chances of AMR in some subgroups of international migrants like refugees. Higher AMR does not imply that migrants import them from countries of origin but could also reflect upon risks they experience during transit and arrival. In countries that are averse to international migration, this must be carefully considered to prevent stigma and discrimination towards migrant populations. Human rights protection of international migrants during transit and arrival should become a global priority, not only for AMR prevention in the case of migrants in adverse social and political environments, but also for the development of inclusive, respectful societies.

Future research could improve our systematic review by updating it and including additional electronic databases for articles search. Based on our findings, primary studies in future should significantly improve the description and justification of selected migrant populations and local populations used for comparison. A more detailed report of the study would allow adequate pooled estimations of risk or prevalence of AMR in different populations. Research in other continents besides North America and Europe is urgently required. Given the complex dynamics of AMR and its geographically patterned features, a global comprehension of its true importance for prevention and control needs more evidence from other regions like Latin America and the Caribbean and Asia.

AMR – Antimicrobial drug resistance

CDC – Centers for Disease Control and Prevention

MDRGN – Multidrug-resistant Gram-negative bacteria

MDRO – Multidrug-resistant microorganism

MRSA – Methicillin-resistant Staphylococcus aureus

PVL – Panton-Valentine Leucocidin

QA – Quality assessment

STROBE - Strengthening the Reporting of Observational Studies in Epidemiology

VRE – Vancomycin-resistant Enterococci

WHO – World Health Organization

Ethics approval and consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and materials

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Competing interests

The authors declare that they have no competing interests

Funding Millenium Initiative for Collaborative Research on Bacterial Resistance (MICROB-R), Iniciativa Cientifica Milenio, Chile. Fondecyt Regular 1201461 and 1171805, CONICYT, ANID, Ministerio de Ciencia, Tecnología, Conocimiento e Innovación, Gobierno de Chile.

Authors’ contributions

AP, BC, and JM were responsible for the design of the research. EU did the database search. AP and BC did the screening, quality assessment, data extraction, analysis, and contributed to the writing of the manuscript, in collaboration with OR, MA, SA, RA and JM. All authors read and approved the final manuscript.

Acknowledgements

Not applicable

Wise R., Hart T., Cars O., Streulens M., Helmuth R., Huovinen P. et al. (1998). Antimicrobial resistance. BMJ, 317,609
Hart CA., Kariuki S. (1998). Antimicrobial resistance in developing countries. BMJ, 317, 647
Cohen ML. (1992). Epidemiology of Drug Resistance: Implications for a Post—Antimicrobial Era. Science, 257, 1050-1055
(2009). Medicine use in primary health care in developing and transitional countries: factbook summarising results from studies reported between 1990 and 2006, WHO/EMP/MAR/2009.3, Geneva
Holloway KA., van Dijk L., The World Medicines Situation 2011 (Third Edition), Rational Use of Medicines. WHOb/EMP/MIE/2011.2.2 World Health Organization, Geneva, Switzerland: 2011. Available from: http://www.who.int/medicines/areas/policy/world_medicines_situation/en/index.html
Organización Mundial de la Salud. Organización Panamericana de la Salud. Plan de acción sobre la resistencia a los antimicrobianos. 54.^o consejo directivo. 67.^a sesión del comité regional de la OMS para las américas. Washington, D.C., EUA, del 28 de septiembre al 2 de octubre del 2015.
Quiñones Pérez D. (2017). Resistencia antimicrobiana: evolución y perspectivas actuales ante el enfoque "Una salud". Rev Cubana Med Trop, 69(3), 1-17.
Antibiotic Resistance Threats in the United States. (2019). Atlanta, GA: U.S. Department of Health and Human Services.
Siddiqui AH., Koirala J. (2020). Methicillin Resistant Staphylococcus Aureus (MRSA) [Updated 2019 Dec 22]. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; Available from: https://www.ncbi.nlm.nih.gov/books/NBK482221/
Jun KANEKO & Yoshiyuki KAMIO(2004) Bacterial Two-component and Hetero-heptameric Pore-forming Cytolytic Toxins: Structures, Pore-forming Mechanism, and Organization of the Genes, Bioscience, Biotechnology, and Biochemistry, 68:5, 981-1003, DOI: 1271/bbb.68.981
Gutiérrez-Gutiérrez B., Sojo-Dorado J., Bravo-Ferrer J.for the EURECA project team, et al. (2017). EUropean prospective cohort study on Enterobacteriaceae showing REsistance to CArbapenems (EURECA): a protocol of a European multicentre observational study BMJ Open, doi: 10.1136/bmjopen-2016-015365
Aujard Y., Maury L., Doit C., et al. (2005). [Ureaplasma urealyticum and Mycoplasma hominis infections in newborns: personal data and review of the literature]. Archives de Pediatrie : Organe Officiel de la Societe Francaise de Pediatrie. Suppl 1:S12-8. DOI: 10.1016/s0929-693x(05)80004-1.
Munita JM., Peters A., Cabieses B. (2018). Migración internacional y resistencia antibacteriana: ¿qué sabemos? Rev Cuadernos médicos sociales, 58(4).
Cabieses B., Bernales M., McIntyre, AM. (2017). International migration as a social determinant of health in Chile: Evidence and recomendations for public policies. Ediciones Universidad del Desarrollo. Available from: http://hdl.handle.net/11447/978
Smalen, A. W., Ghorab, H., Abd El Ghany, M., & Hill-Cawthorne, G. A. (2017). Refugees and antimicrobial resistance: A systematic review. Travel Medicine and Infectious Disease, 15, 23-28.
Nellums, L. B., Thompson, H., Holmes, A., Castro-Sánchez, E., Otter, J. A., Norredam, M. et al. (2018). Antimicrobial resistance among migrants in Europe: a systematic review and meta-analysis. The Lancet. Infectious Diseases, 18(7), 796-811.
Moher D., Liberati A., Tetzlaff J., Altman DG. (2009). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of Internal Medicine, 151(4):264-9.
Cabieses, B., Munita, J., Peters, A., Uphoff, N., & Astorga, S. (2018). Systematic review: is there a relationship between bacterial drug resistance and global social conditionants like poverty, socioeconomic inequality and international migration? PROSPERO Natural Institute for Health Research website: http://www.crd.york.ac.uk/PROSPERO/display_record.php?ID=CRD42018114436
Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, et al. (2009) The PRISMA Statement for Reporting Systematic Reviews and Meta-Analyses of Studies That Evaluate Health Care Interventions: Explanation and Elaboration. PLOS Medicine 6(7): e1000100. https://doi.org/10.1371/journal.pmed.1000100
United Nations. World Migration Managing Migration - Challenges and Responses for People on the Move 2003. Available from: un.org.
United Nations, Department of Economic and Social Affairs. United Nations Population Division 2020. Available from: https://www.un.org/en/development/desa/population/migration/data/estimates2/estimatesgraphs.asp?2g2 (accessed April 19, 2020)
The Center for Disease Dynamics, Economics & Policy. ResistanceMap - Drug Resistance. Available from: https://resistancemap.cddep.org/DRI.php (accessed April 19, 2020)
Migration and health: key issues. 2016 http://www.euro.who.int/en/health-topics/health-determinants/migrations-and-halty/migrant-health-in-the-european-region/migration-and-health-key/dissues#page-wrage (accessed April 17, 2020)
Maltezou HC. (2016). Antibiotic resistance and the refugee crisis in Europe–preemptive action is indicated. Travel Med Infect Dis. 14, 69–70.
MacPherson DW., Gushulak BD., Baine WB., et al. (2009) Population mobility, globalization, and antimicrobial drug resistance. Emerg Infect Dis, 15, 1727-32
Memish ZA., Venkatesh S., Shibl AM. (2003). Impact of travel on international spread of antimicrobial resistance. Int J Antimicrob Agents, 21, 135-42
Rogers BA., Aminzadeh Z., Hayashi Y., et al. (2011). Country-to-country transfer of patients and the risk of multi-resistant bacterial infection. Clin Infect Dis, 53, 49-56

Oelmeier de Murcia, K., Glatz, B., Willems, S., Kossow, A., Strobel, M., Stühmer, B., … Möllers, M. (2017). Prävalenz multiresistenter Keime unter Zufluchtsuchenden: eine prospektive Fall-Kontroll-Studie in einem geburtshilflichen Kollektiv. TT - Prevalence of Multidrug Resistant Bacteria in Refugees: A Prospective Case Control Study in an Obstetric Cohort. Z Geburtshilfe Neonatol, 221(3), 132–136. Retrieved from http://dx.doi.org/10.1055/s-0043-102579
Ravensbergen, SJ., Berends, M., Stienstra, Y., & Ott, A. (2017). High prevalence of MRSA and ESBL among asylum seekers in the Netherlands. PLoS ONE, 12(4), 1–10. https://doi.org/10.1371/journal.pone.0176481
Reinheimer, C., Kempf, VAJ., Jozsa, K., Wichelhaus, TA., Hogardt, M., O’Rourke, F., & Brandt, C. (2017). Prevalence of multidrug-resistant organisms in refugee patients, medical tourists and domestic patients admitted to a German university hospital. BMC Infectious Diseases, 17(1), 4–11. https://doi.org/10.1186/s12879-016-2105-y
Piper Jenks, N., Pardos de la Gandara, M., D’Orazio, BM., Correa da Rosa, J., Kost, RG., Khalida, C., … Tobin, JN. (2016). Differences in prevalence of community-associated MRSA and MSSA among U.S. and non-U.S. born populations in six New York Community Health Centers. Travel Medicine and Infectious Disease, 14(6), 551–560. https://doi.org/10.1016/j.tmaid.2016.10.003
Reinheimer, C., Kempf, VAJ., Göttig, S., Hogardt, M., Wichelhaus, TA., O’Rourke, F., & Brandt, C. (2016). Multidrug-resistant organisms detected in refugee patients admitted to a university hospital, Germany June‒December 2015. Eurosurveillance, 21(2), 1–5. https://doi.org/10.2807/1560-7917.ES.2016.21.2.30110
Sánchez-Montalvá, A., Martínez-Pérez, Á., Pérez-Molina, JA., González-López, JJ., Lopez-Vélez, R., Salvador, F., … Molina, I. (2015). Clinical and Microbiological Profile of a Retrospective Cohort of Enteric Fever in 2 Spanish Tertiary Hospitals. Medicine (United States), 94(21), 1–6. https://doi.org/10.1097/MD.0000000000000791
Casado-Verrier, B., Gómez-Fernández, C., Paño-Pardo, JR., Gómez-Gil, R., Mingorance-Cruz, J., Moreno-Alonso De Celada, R., & Herranz-Pinto, P. (2012). Prevalencia de infecciones de piel y tejidos blandos producidas por Staphylococcus aureus resistente a Meticilina Comunitario en Madrid. Enfermedades Infecciosas y Microbiologia Clinica, 30(6), 300–306. https://doi.org/10.1016/j.eimc.2011.11.011
Frick, MA., Moraga-Llop, FA., Bartolom, R., Larrosa, N., Campins, M., Roman, Y., … Figueras, C. (2010). Infecciones por Staphylococcus aureus resistente a meticilina adquirido en la comunidad en niños. Enfermedades Infecciosas y Microbiologia Clinica, 28(10), 675–679. https://doi.org/10.1016/j.eimc.2010.01.007
Böcher, S., Gervelmeyer, A., Monnet, DL., Mølbak, K., & Skov, RL. (2008). Methicillin-resistant Staphylococcus aureus: Risk factors associated with community-onset infections in Denmark. Clinical Microbiology and Infection, 14(10), 942–948. https://doi.org/10.1111/j.1469-0691.2008.02055.x
Manzur, A., Dominguez, AM., Pujol, M., Limon, E., Hornero, A., Gonza, MPM., … Bellvitge, De HU. (2008). Community-acquired methicillin-resistant Staphylococcus aureus infections: an emerging threat in Spain. Clinical Microbiology and Infection, 14(4)
Cercenado, E., Cuevas, O., Marín, M., Bouza, E., Trincado, P., Boquete, T., … Vindel, A. (2008). Community-acquired methicillin-resistant Staphylococcus aureus in Madrid, Spain: transcontinental importation and polyclonal emergence of Panton-Valentine leucocidin-positive isolates. Diagnostic Microbiology and Infectious Disease, 61(2), 143–149. https://doi.org/10.1016/j.diagmicrobio.2008.01.001
Banatvala, N., Davies, GR., Abdi, Y., Clements, L., Rampton, DS., Hardie, JM., & Feldman, RA. (1994). High prevalence of Helicobacter pylori metronidazole resistance in migrants to east London: Relation with previous nitroimidazole exposure and gastroduodenal disease. Gut, 35(11), 1562–1566. https://doi.org/10.1136/gut.35.11.1562
Perniciaro, S., Imöhl, M., & van der Linden, M. (2018). Invasive pneumococcal disease in refugee children, Germany. Emerging Infectious Diseases, 24(10), 1934–1936. https://doi.org/10.3201/eid2410.
Costa, E., Tejada, M., Gaia, P., Cornetta, M., Moroni, A., Carfora, E., … Frigiola, A. (2018). Prevalence of multidrug-resistant organisms in migrant children admitted to an Italian cardiac surgery department, 2015–2016. Journal of Hospital Infection, 98(3), 309–312. https://doi.org/10.1016/j.jhin.2017.11.
Leli, C., Mencacci, A., Bombaci, JC., D’Alò, F., Farinelli, S., Vitali, M., … Bistoni, F. (2012). Prevalence and antimicrobial susceptibility of Ureaplasma urealyticum and Mycoplasma hominis in a population of Italian and immigrant outpatients. Infezioni in Medicina, 20(2), 82–87.

Table 2 Quality Assessment summary (n=15)

			QUALITY ASSESSMENT - RISK OF BIAS (Strobe v.4 & MMAT v.2011)																				QA	QA
No.	First author	Year	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	score	category
1	Oelmeier de Murcia	2017																					16	high
2	Ravensbergen	2017																					15	medium-high
3	Reinheimer	2017																					17	medium-high
4	Jenks	2016																					20	high
5	Reinheimer	2016																					20	high
6	Sánchez-Montalva	2015																					16	high
7	Casado-Verrier	2012																					16	high
8	Frick	2010																					17	high
9	Bocher	2008																					19	hgh
10	Manzur	2008																					13	medium-high
11	Cercenado	2008																					12	medium-high
12	Banatvala	1994																					20	high
13	Perniciaro	2018																					20	high
14	Costa	2017																					20	high
15	Leli	2012																					19	high

Legend of QA items:

1	Question well justified?
2	Clear objectives?
3	Clear hypothesis?
4	Clear study design?
5	Clear setting?
6	Participants well described?
7	Variables well described?
8	Data source well described?
9	Adjustment by confounders?
10	Adequate sample size?
11	Analysis well described?
12	Clear descriptive stats?
13	Main outcome clearly described?
14	Limitations discussed?
15	Results well interpreted?
16	Results generalizable?
17	Financial disclosure?
18	Conflict of interest?
19	Data source adequate to research question?
20	Sample representative of population of interest?

Legend of QA categories:

1	low (1-5)
2	medium-low (6-10)
3	medium high (11-15)
4	high (16-20)

Table 4A. Differences in the prevalence of methicillin-resistant Staphylococcus aureus in between migrants and local population

Study	Country	N° Migrants	N° Population	Type resistance	Prevalence Migrants	Prevalence Population	Difference
Bocher et al (2008)	Denmark	9	112	MRSA	88.9%	23.2%	↑65.7%
Casado-Verrier et al (2012)	Spain	16	23	CA-MRSA	50%	21.7%	↑28.3%
Reinheimer et al (2017)	Germany	117	495	MRSA	10.3%	1.4%	↑8.9%
Ravensbergen et al (2017)	Netherlands	973	12989	MRSA	10.3%	2.1%	↑7.9%
Oelmeier et al (2017)	Germany	50	50	MRSA	6.0%	0	↑6%
Reinheimer et al (2016)	Germany	143	1489	MRSA	5.6%,	1.2%	↑4.4%
Frick et al (2010)	Spain	6	6	MRSA	100%	100%	0%
Piper et al (2016)	US	40	77	wound infections CA-MRSA Nasal infections CA-MRSA	37.5% 10%	42.9% 19.5%	↓5.4% ↓9.5%

Table 4B. Differences in the prevalence of PVL (+) strains in between migrants and local population

Study	Country	N° Migrants	N° Population	Type resistance	Prevalence Migrants	Prevalence Population	Difference
Manzur et al (2008)	Spain	15	4	PLV(+) strains	78.9%	21.1%	↑57.8%
Ravensbergen et al (2017)	Netherlands	973	12989	PVL(+) strains	42%	17.8%	↑24.2%
Cercenado et al (2008)	Spain	7	6	PLV(+) strains	53.8%	46.2%	↑7.6%
Frick et al (2010)	Spain	6	6	PVL(+) strains	100%	100%	0%

Table 4C Differences in the prevalence of multidrug-resistant Gram-negative bacteria in between migrants and local population

Study	Country	N° Migrants	N° Population	Type resistance	Prevalence Migrants	Prevalence Population	Difference
Sánchez-Montalvá et al (2015)	Spain	4	12	Salmonella Typhi/Salmonella Paratyphi	75%	8.3%	↑66.7%
Reinheimer et al (2017)	Germany	117	495	MDRGN	52.1%	7.9%	↑44.2%
Reinheimer et al (2016)	Germany	143	1489	MDRGN	60.8%	16.7%	↑44.1%
Banatvala et al (1994)	England	54	46	H. pylori metronidazole resistant strains	43%	17%	↑26%
Ravensbergen et al (2017)	Netherlands	973	12989	ESBL MDRE	20.3% 21.4%	3.2% 5.1%	↑17.1% ↑16.3%

Study

Country

N°

Migrants

N°
Population

Type resistance

Prevalence Migrants

Prevalence

Population

Difference

Oelmeier et al (2017)

Germany

50

VRE

1(1.8%)

0

↑1.8%

Table 4D Differences in the prevalence of vancomycin-resistant Enterococcus in between migrants and local population

Table 4E Differences in the prevalence of at least one multidrug-resistant organism in between migrants and local population

Study

Country

N°

Migrants

N°
Population

Type resistance

Prevalence Migrants

Prevalence
Population

Difference

Costa et al (2017)

Italy

354

141

At least one MDRO at admission (MRSA; ESBL; CPE; VRE)

61.9%

24.8%

↑37.1%

Perniciaro et al (2018)

Germany

21

405

invasive pneumococcal disease isolates resistant to >3 classes of antimicrobial drugs

38%

2%

↑36%

Table 4F Differences in the prevalence of colonization by Ureaplasma urealyticum and Mycoplasma hominis

Study

Country

N°

Migrants

N°
Population

Type resistance

Prevalence Migrants

Prevalence
Population

Difference

Leli et al (2012)

Italy

121

312

Ureaplasma urealyticum

Mycoplasma hominis

39.6%

2.5%

33.3%

0.9%

↑6.3%

↑1.6%

Table 4 Prevalence of antibiotic resistance in migrants and host population by organism
MDRO=multidrug-resistant organisms. (CA)-MRSA=(community acquired) methicillin-resistant Staphylococcus aureus. PVL=Panton-Valentine leucocidin. ESBL=expended-spectrum beta-lactamase. MDRGN=multidrug-resistant Gram-negative bacteria; defined as Enterobacteriaceae with extended spectrum beta-lactamase (ESBL)–phenotype as well as Enterobacteriaceae, Pseudomonas aeruginosa, and Acinetobacter baumannii resistant against piperacillin, any 3rd/4thgeneration cephalosporin, and fluoroquinolones. CPE=carbapenemase-producing Enterobacteriaceae. MDRE=multidrug-resistant Enterobacteriaceae. VRE=vancomycin-resistant Enterococci

A systematic review on the relationship between international migration and antimicrobial resistance

Status:

Version 1

Abstract

Figures

Background

Methods

Study design

Types of studies

Types of participants

Types of AMR

Comparator

Identification of studies

Selection of studies

Data extraction and management

Assessment of methodological quality of included reviews

Data analysis

Ethical considerations

Results

Discussion

Summary of main findings

Quality of the evidence

Conclusions

Abbreviations

Declarations

References

Reference list

Tables

Supplementary Files

Status:

Version 1