Antimicrobial resistance among migrants in Europe: a systematic review and meta-analysis – update from 2017 to 2023

Summary Background Antimicrobial resistance (AMR) is a critical global health concern. A previous systematic review showed that migrants in Europe are at increased risk of AMR. Since the COVID-19 pandemic there have been rapid changes in patterns of antibiotic use, AMR, and migration. We aimed to present an updated evidence synthesis on the current distribution of AMR among migrants in Europe. Methods We carried out a systematic review and meta-analysis in accordance with PRISMA guidelines (PROSPERO ID: CRD42022343263). We searched databases (MEDLINE, Embase, PubMed and Scopus) from 18 January 2017 until 18 January 2023 to identify primary data from observational studies reporting any laboratory-confirmed AMR among migrants in the European Economic Area (EEA) and European Union-15 (EU-15) countries using over 7 key search terms for migrants and over 70 terms for AMR and countries in Europe. Outcomes were infection with, or colonisation of AMR bacteria. Methodological quality was assessed using Joanna Briggs Institute Critical Appraisal Checklist for Observational Studies. We meta-analysed the pooled-prevalence of infection and/or colonisation of AMR organisms. Findings Among 630 articles, 21 observational studies met the inclusion criteria and were included in this review. The pooled prevalence for any detected AMR was 28.0% (95% CI 18.0%–41.0%, I2 = 100%) compared to a 25.4% seen in the previous review; gram-negative bacteria 31.0% (95% CI 20.0%–44.0%, I2 = 100%), and methicillin-resistant staphylococcus aureus 10.0% (95% CI 5.0%–16.0%, I2 = 99%). Drug-resistant bacteria were more prevalent in community settings in large migrant populations (pooled prevalence: 41.0%, 95% CI 24.0%–60.0%, I2 = 99%) than in hospitals (21.0%, 95% CI 12.0%–32.0%, I2 = 99%). AMR estimates in ‘other’ migrants were 32.0%, (95% CI 12.0%–57.0%, I2 = 100%) and 28.0% (95% CI 18.0%–38.0%, I2 = 100%) in forced migrants. No firm evidence of AMR acquisition with arrival time or length of stay in the host country was found. Interpretation Studies investigating AMR in migrants are highly heterogenous. However, since the COVID-19 pandemic, migrants may be at higher risk of acquiring resistant bacteria, particularly gram-negative bacteria, within community settings such as refugee camps and detention centres in Europe. Our study highlights the importance of infrastructure and hygiene measures within these settings, to mitigate transmission of resistant pathogens. Policy-makers should screen for AMR in migrants prior to departure from countries of origin, where feasible, and upon arrival to a new country to ensure optimal health screening, infection control and effective treatment. Funding There was no funding source for this study.


Introduction
Antibiotics treat and prevent common infections in humans and animals. 1 Extensive use of antibiotics use contribute to antimicrobial resistance (AMR). 2 The most common bacteria linked to mortality from AMR are Streptococcus pneumoniae, Acinetobacter baumannii, Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, and Klebsiella pneumoniae. 3,4ften, mortality from AMR is exacerbated within settings caused by overcrowding, or poor water and sanitary conditions, which in turn often occurs in human migration. 5ddressing AMR is becoming increasingly challenging.Between 2001 and 2014, resistance to thirdgeneration cephalosporins in gram-negative bacteria increased by 13.3% in Europe. 6In England, E.coli resistance to piperacillin-tazobactam increased from 8⋅5% to 11⋅7%, while the resistance level of K. pneumoniae increased by 5.9% between 2011 and 2015. 7Meanwhile, in 2015, 63.5% of the 671,689 infections caused by AMR in the EU/EEA were linked to healthcare settings. 81][12] Furthermore, during the pandemic, antibiotics were frequently prescribed for patients with COVID-19, despite absence of evidence of a superadded bacterial infection. 8We therefore performed a systematic review and meta-analysis to investigate trends of AMR amongst migrants in Europe, following a period of mass changes in global antmicrobial prescribing following the COVID-19 pandemic.Our findings have public health implications for understanding the burden of AMR amongst migrants in Europe.

Methods
This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and

Research in context
Evidence before this study Antimicrobial resistance (AMR) is a global concern, especially within migrants.Previous work has shown that COVID-19 may have accelerated AMR, particularly for gram-negative organisms.Prior to the COVID-19 pandemic, a previous systematic review found the prevalence of AMR in Europe to be 25.4% within migrants.Given the changing patterns of migration and the influence of antibiotic use following the COVID-19 pandemic, there was a need for an update regarding the distribution of AMR among migrants to Europe.

Added value of this study
This systematic review and meta-analysis was conducted to identify and synthesise data on AMR, including colonisation or infection, in migrants to countries in Europe and the EU/ EEA up to January 2023.The pooled prevalence for any detected AMR was 28.0% (95% CI 18.0%-41.0%,I 2 = 100%) compared to a 25.4% seen in the previous review.Our findings show high rates of any AMR colonisation or infection among 'other' migrants and refugees and asylum seekers, and elevated rates in community settings compared to hospitals.We note a particularly high prevalence of gram-negative drugresistant organisms amongst migrants in Europe, which may reflect the types of congregant settings in which these organisms are transmitted.

Implications of all the available evidence
We show that within Europe, the prevalence of AMR in migrants, particularly within refugees and asylum seekers is increasing; particularly in community settings.These will often be refugee camps, transit hubs or detention facilities within receiving countries.Our results demonstrate the vulnerability of migrant communities to AMR exposure in Europe and the urgent need for interventions to better prevent, detect, and treat AMR infections in these settings, in line with better social, environmental and health conditions.

Data sources and searches
Search strategies and search terms were developed from similar research and previous systematic reviews in migrant health and AMR. 2,12We searched Embase, MEDLINE, PubMed, and Scopus for articles reporting primary findings from observational studies between January 18, 2017 and January 18, 2023.This start date was chosen since it is a follow-up study from a previous systematic review on AMR in migrants to Europe, reporting evidence up until January 18, 2017. 12A Boolean search strategy with search terms relating to migration, AMR, bacterial infections, EU-15 and EEA countries, and the appropriate MeSH headings was used for each database.Appendix I details the specific database search strategies and the number of studies found.Migrants were classified as persons born outside the country where the study was conducted, including forced migrants (e.g asylum seekers, refugees, migrant children) and 'other' migrant groups.Forced migrants were categorised as persons subjected to leaving their country of residence due to threats to life and livelihood, such as environmental disasters, political unrest, war, persecution, and famine. 14'Other' migrants were foreign-born and had migrated for different reasons, including work, education or reuniting with family.
Studies that examined drug resistance in tuberculosis were excluded. 12We also excluded articles in which migrant status was not defined or was determined by ethnicity, country of birth of participants' parents, and articles in which data were not separated or reported by migrant status.Studies that did not present original data or reported non-laboratory confirmed data on AMR, including editorials, comments, reviews, letters, and case reports, were also excluded.No language restrictions were placed on the searches or search results.

Outcomes
Our primary outcome was infection, or colonisation with laboratory confirmed antibiotic resistant organisms, such as methicillin-resistant Staphylococcus aureus (MRSA) and gram-negative bacteria, including extended-spectrum β-lactamase producing Enterobacteriaceae (ESBL-PE) and multidrug-resistant bacteria (combined resistance to three or more class of antibiotics).

Study selection and data collection
BC and LBN screened the bibliographies of included articles to identify additional eligible studies.Title and abstract screening, full-text screening, data extraction, and quality assessment were done independently.Any discrepancies were discussed until a mutual agreement was reached.

Data extraction and quality assessment
Mendeley V1.19.8 and Rayyan software were used to manage references, deduplication, and for screening.Data were extracted using a predesigned excel sheet and based on study design, study setting, type of migrant, country of study, and outcome reported.Methodological quality assessment of articles was done using the Joanna Briggs Institute Critical Appraisal (JBI) tools for observational studies. 15The tool consisted of an 11-point scale for cohort studies and an 8-point scale for crosssectional studies evaluating descriptions of the study population and setting, inclusion criteria, accounting for confounders and use of appropriate statistical methods.Articles were given a quality score percentage to reflect methodological rigour and clarity in reporting.Articles were not excluded based on their quality scores, although we did conduct sensitivity analysis to ascertain the robustness of our findings.

Statistics
Eligible studies that reported AMR prevalence were included in the meta-analysis.Data analysis was done in R V4.1.1 using the meta-packages to estimate the pooled AMR prevalence and 95% confidence intervals.Random-effects models were used to account for heterogeneity in the study, which was assessed through the I 2 statistic.DerSimonian and Laird estimator and Freeman-Tukey double arcsine transformations were used to account for variations in the true effect between and within studies. 16For all migrants, pooled estimates of the prevalence of AMR colonisation and infection were calculated, and stratified based on migrant type and settings.Sub-analyses were also performed for MRSA and drug-resistant gram-negative bacteria.Heterogeneity was graphically explored in forest plots to check potential sources which could be explained by study setting, migrant type, screening approaches and sample processing.Funnel plots and Eggers' test were used to check for asymmetry between included studies.on AMR either as colonisation or infections in 14,168 migrants were included in this meta-analysis.
31,[33][34][35] All studies reported prevalence rates of AMR in migrants identified during screening sessions intended for a particular Articles population, such as asylum seekers or refugees, and in specific settings, such as arrival facilities or at the time of hospital admission with an existing infection.In addition, four studies reported time taken to travel to the host country, ranging from 30 days to 350 days, depending on the travel route.17,31,32,35 Three studies reported AMR prevalence among refugees based on the length of stay in the host country.31,32,35 When described, clinical signs of infection were mostly skin and soft tissue infections or diarrhoea.Samples collected for laboratory testing included throat, nasal, and rectal samples, biopsies, wound swabs, and faecal samples.Different guidelines for determining the antibiotic susceptibility of clinical and screening samples used in various investigations and the rules for interpreting the antimicrobial sensitivity and minimum inhibitory concentrations were reported across studies.
All 21 studies recorded colonisation or infection, of one or multiple forms of resistance.Thirteen studies detected MRSA, of which four were community-associated, 19,21,32,35 ESBL-producing bacteria, 17,18,20,21,25,26,28,31,32,35,36 and vanco mycin-resistant enterococcus. 17,29Nine studies reported AMR distribution according to the region of origin (see Fig. 2).Using the JBI critical assessment checklist for observational studies, the studies received scores ranging from 60% to 100% on questions about their quality (Appendix II and III).Three studies 17,21,35 accounted for missing data by multiple imputations or by creating a separate group for categorical variables.1][22][23]26,29,[31][32][33][34][35][36] Furthermore, less than one-third of the included studies reported the travel duration to the host country. 17,31,32,35However, none addressed how travel time impacted the development of AMR.Instead, AMR rates were compared with the various durations of stay since migrants' arrival time.A detailed summary of the included studies is shown in Table 1.

AMR and travel time
Three studies investigated colonisation of AMR bacteria with regards to time taken to travel to the receiving country or the length of stay in the receiving country. 31,32,35Evidence from these studies suggested that the proportion of migrants who tested positive for MRSA or MDRE varied over time.However, there was no discernible trend of reduction or rise. 31No evidence was found between AMR acquisition and various migratory routes.

Sensitivity analysis
Sensitivity analysis, as shown in Appendix IV, was done to assess the influence of article quality on the predicted prevalence of AMR colonisation or infection.Studies with a quality level of 75% or below yielded a pooled prevalence of AMR (26.0%, 95% CI 11.0-46.0,I 2 = 99%) that did not vary significantly from when all studies were included (28.0%, 95% CI 18.0-41.0,I 2 = 100%).Heterogeneity changes were not significant in subgroups analyses.Funnel plots demonstrated absence of publication bias (Appendix V).

Discussion
Our study has three main findings.First, we found a high prevalence of AMR colonisation and infection among forced migrants and other migrant groups, particularly in communities with high geographical concentrations of migrants.Second, we found that resistant bacteria were more prevalent in community settings compared to hospitals.Finally, we found that there were low MRSA colonisation rates among migrants, with the majority of AMR attributable to gram-negative multi-drug resistant bacteria.

Articles
Our findings showed a higher prevalence of AMR among migrants compared to previous work, suggesting an increase in AMR since the COVID-19 pandemic in Europe. 12Many migrants travel from areas where there are minimal antimicrobial stewardship policies compared to the receiving country.Thus, these migrants often come from areas with a high prevalence of AMR organisms in the community, and may bring them to the receiving country.Migration to a receiving country in sub-optimal conditions can also contribute to resistance, depending on route and mode of travel.
When comparing other migrants with forced migrants, a slight difference in the rate of AMR was observed, in contrast to the previous review, but aligns with two studies that found a higher risk of resistance among family-reunited migrants than refugees. 31,38In addition, the prevalence of AMR was observed to differ across regions, with the Middle East/North Africa and Asia having the highest occurrence in migrants.These data needs to be interpreted cautiously, since most studies lacked information about the origin region and corresponding AMR data.
We found a higher prevalence of AMR in migrants within community settings compared to hospitals.Migrants are a highly diverse cohort; with certain subpopulations, such as refugee workers and undocumented migrants facing major difficulties to healthcare services and living within congregant settings, that may increase the risk of both AMR acquisition and transmission.Within healthcare settings, strict infection control procedures and testing limit the spread of AMR.These measures include good hygiene practices, and isolation precautions that are less strict within migrant communities outside the hospital.
Our findings emphasise the role immigrant dominated areas, camps, or transit, arrival, and detention  Articles centres might have in increasing the risk of acquiring AMR organisms for migrants.Pathogens which have AMR are more likely to spread in these environments due to poor socio-environmental factors such as overcrowding, improper environmental hygiene, and limited access to adequate health services, including medications or vaccines. 12,39,40These factors may have a more significant impact in determining AMR among migrants to Europe than the acquisition of resistant bacteria in their countries of origin. 12It is possible that a substantial proportion of migrants will have come from refugee routes within recent studies, due to the occurrence of lockdown during the COVID-19 pandemic and ban of travel in many countries within Europe.
Our findings also a greater prevalence of drugresistant gram-negative bacteria (GNB 31.0%,community 52.0%) than in the prior study (GNB 27.2%, community 32.1%). 12The high occurrence of multi-drug resistant gram-negative bacteria could translate to an increased prevalence of urinary tract and gastrointestinal tract infections (GIT) commonly linked with travelling and poor sanitary conditions.Our findings agree with other studies that reported a two to three-fold increase in the colonization of drug-resistant gramnegative bacteria among migrants compared to general community populations in receiving countries. 12,41,42In addition, a systematic review found that COVID-19 may have accelerated the emergence and transmission of AMR, particularly for gram-negative organisms in hospital settings globally. 43Many migrants to Europe travel from countries where high rates of ESBL-PE have been previously reported, such as North Africa and Asia. 42igrants in these areas may be at increased risk of exposure to AMR organisms. 44e found that there were low MRSA colonisation rates among migrants, a common cause of skin and soft tissue infections. 45However, in migrants who arrive with SSTI, MRSA and Panton-Valentine leucocidin positive (PVL) genes are frequently detected. 30,46The role of MRSA and PVL isolates in spreading AMR genes has been documented in previous research. 45,47,48However, compared with multidrug-resistant gram-negative bacteria, the risk of MRSA transmission among migrating individuals is substantially lower than gram-negative bacteria. 30One reason for why this may be is environmental stability; gram-negative persist longer in the environment due to a robust structural layer that slows down or inhibits the penetration of chemical agents. 49t remains unclear whether migrants bring resistant organisms from their country of origin, or whether they

Item Description
Improved Surveillance/Research Having reliable and efficient surveillance systems in place is crucial for monitoring the spread and patterns of antimicrobial resistance (AMR) within migrant communities.To achieve this, a unified system must be established to gather data on antimicrobial usage, resistance trends, and healthcare-associated infections.Additionally, it is critical to support research initiatives aimed at uncovering the underlying factors driving AMR among migrant populations, as well as developing innovative methods for prevention, diagnosis, and treatment.This can be achieved through conducting epidemiological studies, genomic surveillance, and clinical trials to inform evidence-based interventions.
Health Education/Stewadrship Tailored health education campaigns that meet the cultural and linguistic needs of migrant communities are vital for improving their overall health.These campaigns can increase awareness of the appropriate use of antibiotics, the risks of antimicrobial resistance (AMR), and the importance of completing antibiotic courses as prescribed.Additionally, it's crucial to implement antimicrobial stewardship programs in migrant healthcare facilities and provide healthcare providers with training on appropriate prescribing practices.Encouraging the use of diagnostic testing to guide antibiotic treatment decisions is also important.Providing health education and information about AMR, infectious diseases, and appropriate antibiotic use can empower migrants to take control of their health and make informed decisions.Ultimately, this can lead to improved health-seeking behaviors, better adherence to treatment regimens, and reduced reliance on antibiotics.

Restructuring migrant living conditions
Living in overcrowded conditions, such as migrant shelters or poor housing conditions, can cause the rapid spread of infectious diseases, some of which can be resistant to antibiotics.To prevent the transmission of these pathogens, it is crucial to ensure that migrants have access to adequate housing and are not forced to live in overcrowded spaces.Additionally, promoting good hygiene practices, such as hand washing and providing access to clean water and sanitation facilities, can help individuals protect themselves against infections and reduce their dependence on antibiotics.By taking these measures, we can also play a significant role in reducing the selective pressure for antimicrobial resistance.
Access to healthcare Enhancing the availability of healthcare services for migrant communities is of utmost importance.The language barriers, cultural disparities, and legal status are among the hurdles that they encounter which can impact their physical and mental health.Addressing these obstacles requires culturally knowledgeable healthcare providers, as well as promoting awareness about the existing healthcare resources within migrant communities.Prompt access to healthcare services is equally vital as it aids in the early detection and treatment of infections, thus lessening the likelihood of complications and the use of broad-spectrum antibiotics that contribute to antimicrobial resistance (AMR).

Community Engagement
The involvement of migrant communities is essential to developing and executing effective strategies to prevent antimicrobial resistance (AMR).Accordingly, there is need for robust partnerships with community leaders, advocacy groups, and healthcare providers.In doing so, we can enable these communities to become active participants in their healthcare and efforts to prevent AMR.A supportive environment must be created to facilitate the dissemination of health messages and encourage positive health behaviors, such as the responsible use of antibiotics.Articles acquired the organism in transit, or in refugee centres where living conditions may be limited.Evidence-based data on the prevalence of AMR colonisation in relation to travel time from country of origin or time since arrival in the host country remain limited.One study found a lower E. coli resistance (57.6%) in migrants with more than 10 years of stay compared to migrants with less than 5 years of stay (62.6%). 35Similarly, in a German study, the prevalence of gram-negative organisms was higher among refugees who recently arrived in Germany (72.4%), with a gradual decline seen after 18 months (14.3%). 32This shows that the duration of colonisation with resistant organisms may vary across strains.However, one study found no decrease in the colonisation rate of MDROs among asylum-seekers even after twelve months since arrival. 31Meanwhile, studies have shown that resistant organisms may be carried from country of origin to receiving country.For example, blaNDM-1 resistance gene in P. aeruginosa was first discovered in North America and Europe, from medical travellers arriving from Asia. 50It may be that the spread of resistant bacteria depends on the settings in which a migrant resides, within their host country; if they live in a refugee camp, it may be that they are constantly exposed to resistant pathogens from other refugees and detention centres compared to living with the locals, where they may be a lower prevalence of AMR organisms.More studies involving migrants should aim to record duration since leaving their country of origins to disentangle this issue.Additionally, analysing the genetic makeup of the strains and thoroughly examining their evolutionary relationships could identify information about transmission and clustering.
Our study had limitations.Due to the sampling sites and procedures across the included studies, the colonisation of some resistant organisms may not be detected.Sampling bias, introduced due to requiring a reason for testing (such as treatment failure) could lead to overestimations of AMR prevalence in this study.Notable lack of pre-migration and post-migration tests in many studies reduces the precision and meaningful inferences of the data.Variations in migrant type, settings, bacteria species, type of resistance reported, and standard of measurements utilised differed significantly, resulting in high heterogeneity between studies.Future efforts must focus on strengthening surveillance systems worldwide, ensuring unified reporting, and collecting comprehensive data on migrant patients.Our findings continue to be applicable in 2024; since conducting the study we performed an updated search since the last search date, from January 2023 to March 2024.Our search identified only two additional studies which would not have changed our main findings. 38,51n conclusion, we found an elevated rate of AMR among migrants in Europe since 2017. 12The prevalence of AMR in migrants were higher in community settings, especially those with high geographical concentrations of migrants compared to hospitals.The most common causative AMR organisms in migrants were gramnegative bacteria.Our findings emphasise the Articles importance of screening and treating AMR in migrants, especially those from refugee camps.Additionally, policy-makers must engage with migrant communities to ensure that any new health policies are feasible, acceptable and non-stigmatising.

Fig. 1 :
Fig. 1: PRISMA flow diagram, illustrating the flow of studies from identification to inclusion.

Fig. 2 :
Fig. 2: Distribution of AMR organisms according to migrants region of origin.Abbreviations used: AMR: antimicrobial resistance.

Fig. 4 :
Fig. 4: Forest plot showing pooled prevalence of AMR in community (A) and hospital (B) settings.Abbreviations used: AMR: antimicrobial resistance.

Table 1 :
Summary of included studies.

Table 3 :
Actionable strategies to tackle AMR among migrants.