Table 1.
Summary of main study characteristics and key findings by review outcomes of interest
| Volume of ED utilization | ||||||
|---|---|---|---|---|---|---|
| Reference and host country | Study design (sampling method) | Sample and number of migrants | Migrant definition | Overall quality assessment rating for internal validity (IV) and external validity (EV) | Key findings | |
| IV | EV | |||||
| Ballotari et al. (2013), Italy28 | Cohort (Record linkage of three databases) | Healthy singleton live births in the years 2008–09 followed for the first year of life (n = 8788) migrants (n = 2383) | Maternal citizenship. Mothers who were citizens of HMCs. | ++ | ++ | Higher use of ED in the first year of life by immigrant mothers.g |
| De Luca et al. (2013), Italy29 | Cross-sectional (Population survey: Italian health conditions survey 2004/05) | Nationally representative population sample (0–64 years) (n = 102 857) migrants (n = 5167). | Place of birth and citizenship. | ++ | + | Immigrants have a higher probability of using emergency services than natives. a,b,c,d,f |
| First generation migrants: born outside Italy without Italian citizenship. | ||||||
| Second generation: born in Italy without Italian citizenship. | Highest use in immigrants from Morocco, Africa and Albania. | |||||
| Naturalized Italians: (born outside Italy with Italian citizenship). | ||||||
| Zinelli et al. (2014), Italy16 | Cross-sectional (ED database) | Visits to the ED by Italian-native and foreign born patients during 2008–12 (n = 424 466 visits) (migrants 64 435 visits) | Country of birth. ‘Foreign-born’ persons born outside Italy, whose parents were either foreign citizens or born outside the national territory. (first generation) | + | + | Higher ED use in immigrants. |
| Clement et al. (2010), Switzerland25 | Cross-sectional (ED database) | Patients attending the ED with non-urgent problems (n = 11258) migrants (n = 2948) | Nationality. | + | + | Higher proportion of visits by non-Swiss nationals. |
| Diserens et al. (2015), Switzerland15 | Cross-sectional (Patient survey) | Patients (>16 years) presenting to ED with non-life-threatening condition (n = 1082) migrants (n = 465) | Nationality. | - | + | Higher proportion of visits by non-Swiss nationals. |
| Ruud et al. (2015), Norway27 | Cross-sectional (Patient survey) | Walk-in patients with non-urgent or semi-urgent health conditions attending A&E outpatient clinic (n = 3864). migrants (n = 1364) | Country of birth. | + | + | First and Second generation immigrants use ED more than Norwegians.a,b |
| First generation immigrants: patient and both parents born abroad. | Second generation: Norwegian born with immigrant parents. | |||||
| Higher use among Swedish, Pakistani and Somali. No difference in visits among Polish.a,b | ||||||
| Shah and Cook (2008), England31 | Cross-sectional (Population survey: British general household survey 2004–05) | Persons living in private households in Britain (n = 20421) migrants (n = 1728). | Country of birth. | + | + | No significant difference in use of casualty by immigrants vs. UK born person. a,b,d,f |
| Hargreaves et al. (2006), England26 | Cross-Sectional (Patient survey) | Walk-in patients attending the A&E (n = 1611). migrants (n = 720). | Country of birth and Nationality. | + | + | Overseas-born over-represented in A&E. |
| Nielsen et al. (2012), Denmark30 | Cross-sectional (Nationwide survey, data linked to healthcare registries) | Random sample of each immigrant group and Danes (≥18–66) from nationwide survey (n = 4952) migrants (n = 2866). | Country of birth and citizenship. | ++ | + | Higher among all immigrant groups (except Lebanese). Highest use among Pakistanis, former Yugoslavia and Iran.a,b,f |
| Higher use among second generation immigrants from Turkey.a,b,f No difference in service use for decedents from Pakistan. Increased use of emergency room with increased length of stay in host country, multiplicative effect 2.32-3.03 (except for Iraqi and Turkish immigrants)a,b,c,d,f | ||||||
| Norredam et al. (2004), Denmark22 | Cross-sectional (Data from Statistical office of the Municipality of Copenhagen) | Patients (≥20 years) attending ER (n = 152,253) migrants (n = 24 433). | Country of birth. | ++ | + | Higher ER utilization for persons born in Somali, Turkey and ex-Yugoslavia compared with Danish-born residents.a,b,c |
| Lower ER utilization for persons born in other Nordic countries, the European countries and North Americaa,b,c | ||||||
| No difference in utilization rates for persons born in Iraq, Pakistan and ‘other countries’.a,b,c | ||||||
| Carrasco-Garrido et al. (2009), Spain12 | Cross-sectional (Secondary analysis of survey data: Spanish National Health Survey 2006–07) | Sample of Non-institutionalized adults (≥16 years) resident in Spain (n = 29 478) migrants (n = 1436). | Nationality. Those not from EU, USA or Canada defined as ‘economic migrants’. | + | + | Higher use of emergency services by economic migrants.a,b,d |
| Carrasco-Garrido et al. (2007), Spain11 | Cross-sectional (Secondary analysis of survey data: Spanish National Health Survey 2003) | Sample of non-institutionalized adults (≥16 years) resident in Spain (n = 1506) migrants (n = 502). | Nationality. Non-EU migrants (not from EU, the USA or Canada). | − | + | No significant difference in emergency service use.a,b |
| Hernández-Quevedo and Jiménez-Rubio (2009), Spain13 | Cross-sectional (Secondary analysis of survey data: Spanish National Health Survey 2003 and 2006) | Sample of non-institutionalized adults (≥16 years), resident in Spain (n = 49 123) migrants (n = 2705). | Nationality. | ++ | + | Higher use among non-Spaniards.a,b,c,d,f |
| Highest probability of use among Latin-Americans and Africans. | ||||||
| Lower use among patients from EU and Europe. | ||||||
| No significant difference those from Asia, North America and Oceania.a,b,c,d | ||||||
| Antón and Muñoz de Bustillo (2010), Spain10 | Cross-sectional (Secondary analysis of survey data: Spanish National Health Survey 2006–07) | Sample of non-institutionalized adults (≥16 years), resident in Spain (n = 25 033) migrants (n = 3042). | Country of birth. | ++ | + | Higher use of ED among Non-EU15. EU-15 immigrants show lower rates of emergency service use.a,b,c,d,f |
| Sanz et al. (2011), Spain14 | Cross-sectional (Secondary analysis of survey data: Spanish National Health Survey 2006) | Sample of non-institutionalized adults (≥16–74), resident in Spain (n = 26 728) migrants (n = 3570). | Country of birth. | + | + | Higher, equal and less use by different sub groups. a,b,c,d,f |
| Higher use by men from Latin America; no difference those from Sub-Saharan Africa or North Africa and less use by those from: Western Countries, Eastern Europe and Asia/Oceania. a,c,d,f | ||||||
| Higher use by women from Sub-Saharan Africa.a,c,d,f | ||||||
| Buron et al. (2008), Spain19 | Cross-sectional (ED patient register) | All emergency care episodes for registered patients (≥15 years) living in study area (n = 29 451 visits). Visits by migrants (n = 10 224). | Country of birth. | ++ | ++ | Lower utilization of ED by foreign born.a,b,f |
| López Rillo and Epelde (2010), Spain23 | Cross-sectional (Medical records) | Patients attending the ED during a 2-week period (n = 5660) migrants (n = 792). | Country of origin. | − | + | No significant difference. |
| Rue et al. (2008), Spain24 | Cross-sectional (Hospital database) | Emergency visits in patients (15–64 years) during 2004 and 2005 (n = 96 916 visits) migrants (n = 20 663 visits). | Country of birth. | + | + | Higher use of emergency by immigrants.a |
| Women from Maghreb, Sub-Saharan Africa, Latin America, Eastern Europe and HIC had higher use than Spanish. | ||||||
| Men from Maghreb, HIC, Latin America and Eastern Europe. Rates were lower for other LIC and Sub-Saharan Africa.a | ||||||
| Patient’s presenting condition to the ED | ||||||
| IV | EV | |||||
| Grassino et al. (2009), Italy21 | Cross-sectional. | Patients (0-adolescent) admitted to the ED n = 4874 foreign (n = 2437). | Parents’ country of birth. One or both parents born outside Italy and the EU. | – | + | No difference in presenting pathologies.* |
| Survey of paediatric ED clinical notes. | ||||||
| Buja et al. (2014), Italy18 | Cross-sectional (not stated in paper) (Record linkage database) | Patients (18–65 years) attending A&E (n = 35 541) migrants (n = 5385) | ‘Citizenship’. Nationality assumed to be that of country of birth if not born in Italy. | ++ | + | Significant difference in presenting conditions. |
| Higher digestive disease in TPF males and those from HMPC. | ||||||
| Higher obstetric and gynaecology diagnoses in TPF women. | ||||||
| Clement et al. (2010), Switzerland25 | Cross-sectional (ED database) | Patients attending the ED with non-urgent problems (n = 11 258) migrants (n = 2948). | Nationality. | + | + | No significant difference in admission reason (trauma or other). |
| Ruud et al. (2015), Norway27 | Cross-sectional (Patient survey) | Walk-in patients with non-urgent or semi-urgent health conditions attending A&E outpatient clinic (n = 3864) migrants (n = 1364). | Country of birth. | + | + | Higher use of general emergency clinic (vs. trauma clinic) for migrants. |
| First generation immigrants: patient and both parents born abroad. | ||||||
| Second generation: Norwegian born with immigrant parents. | ||||||
| Buron et al. (2008), Spain19 | Cross-sectional (ED patient register) | All emergency care episodes for registered patients (≥15 years) living in study area (n = 29 451 visits). Visits by migrants (n = 10 224). | Country of birth. | ++ | ++ | Lower use of surgery, traumatology, medicine and psychiatry among foreign-born a,b,f |
| No significant difference In gynaecology, utilization among foreign-born women.a,f | ||||||
| López Rillo & Epelde (2010), Spain23 | Cross-sectional (Medical records) | Patients attending the ED during a 2-week period (n = 5660) migrants (n = 792). | Country of origin. | − | + | Higher rates of presentation with obstetric and gynaecological disease among migrant women. |
| Higher presentation with digestive tract disease among migrants. | ||||||
| Cots et al. (2007) Spain20 | Cross-sectional (Hospital database) | All emergency visits between 2002 and 2003 (n = 165 257 visits) (migrants = 32 822 visits | Country of origin. Neonates classified by parents’ country of origin. | + | ++ | Higher use of gynaecology and obstetric services among migrant women.* |
| Lower use of medicine and traumatology.* | ||||||
| Appropriateness of ED presentation by severity of presenting condition. | ||||||
| References and host country | Study design (sampling method) | Sample and number of migrants | Migrant definition | Quality assessment | Key findings | |
| IV | EV | |||||
| Ballotari et al. (2013), Italy28 | Cohort | Healthy singleton live births in the years 2008–09 followed for the first year of life (n = 8788) migrants (n = 2383) | Maternal citizenship. Mothers who were citizens of HMCs. | ++ | ++ | Immigrants more likely to visit the ER inappropriately.g |
| (Record linkage of three databases) | ||||||
| Grassino et al. (2009), Italy21 | Cross-sectional. | Patients (0-adolescent) admitted to the ED (n = 4874) foreign (n = 2437) | Parents’ country of birth. One or both parents born outside Italy and the EU. | − | + | Both immigrant and Italian patients access ED mostly for non-urgent or semi-urgent conditions. Higher proportion white triage codes among foreigners.* |
| Survey of paediatric ED clinical notes. | ||||||
| Brigidi et al. (2008), Italy17 | Cross-sectional (ED patient database) | Patients attending ED. 51 000 patients treated (Latin Americans n = 3832). | Country of origin: Latin America. | − | + | Latin American users of the ED use the ED for non-urgent rather than emergency medical treatment. Higher percentage of white triage codes among Latin Americans.* |
| Buja et al. (2014), Italy18 | Cross-sectional (not stated in paper) (Record linkage database) | Patients (18–65 years) attending A&E (n = 35 541) migrants (n = 5385). | ‘Citizenship’. Nationality assumed to be that of country of birth if not born in Italy. | ++ | + | Foreigners more likely to attend A&E with non-urgent clinical conditions. |
| Zinelli et al. (2014), Italy16 | Cross-sectional (ED database) | Visits to the ED by Italian-native and foreign born patients during 2008–12 (n = 424 466 visits. (migrants 64 435 visits) | Country of birth. ‘Foreign-born’ persons born outside Italy, whose parents were either foreign citizens or born outside the national territory. (first generation) | + | + | Higher rate of use of ED for non-urgent conditions among migrants. |
| Clement et al. (2010), Switzerland25 | Cross-sectional (ED database) | Patients attending the ED with non-urgent problems n = 11 258. Migrants (n = 2948) | Nationality. | + | + | Significantly higher attendance at ED with non-urgent conditions among foreigners. |
| Diserens et al. (2015), Switzerland15 | Cross-sectional (Patient survey) | Patients (≥16 years) presenting to ED with non-life-threatening condition. N = 1082 (Migrants n = 465) | Nationality. | − | + | Higher proportion of foreigners visits ED with non-urgent conditions.* |
| López Rillo and Epelde (2010), Spain23 | Cross-sectional (Medical records) | Patients attending the ED during a 2-week period (n = 5660) migrants (n = 792). | Country of origin. | − | + | No significant difference in severity of triage scores. |
| Cots et al. (2007), Spain20 | Cross-sectional (Hospital database) | All emergency visits between 2002 and 2003 (n = 165 257 visits) (migrants = 32 822 visits | Country of origin. Neonates classified by parents’ country of origin. | + | ++ | Lower cost of treating migrants in ED compared with Spanish patients reflects lower complexity of emergency care and workload.a,b,f |
| Patient’s arrival time at the ED | ||||||
| References and host country | Study design (sampling method) | Sample and number of migrants | Migrant definition | Quality assessment (IV) and external | Key findings | |
| IV | EV | |||||
| Zinelli et al. (2014), Italy16 | Cross-sectional (ED database) | Visits to the ED by Italian-native and foreign born patients during 2008–12 (n = 424 466 visits) (migrants 64 435 visits) | Country of birth. ‘Foreign-born’ persons born outside Italy, whose parents were either foreign citizens or born outside the national territory. (first generation) | + | + | No significant difference between the percentage of Italians and migrants seen during the day and night shifts. |
| Clement et al. (2010), Switzerland25 | Cross-sectional (ED database) | Patients attending the ED with non-urgent problems (n = 11 258) migrants (n = 2948) | Nationality. | + | + | Non-Swiss nationals significantly more likely to present to ED during unsocial hours. |
| López Rillo and Epelde (2010), Spain23 | Cross-sectional (Medical records) | Patients attending the ED during a 2-week period (n = 5660) migrants (n = 792). | Country of origin. | - | + | Immigrants significantly more likely to present during unsocial hours. |
| No differences in day of week patients attend. | ||||||
| Grassino et al. (2009), Italy21(paediatric) | Cross-sectional. | Patients (0-adolescent) admitted to the ED n = 4874. (Foreign n = 2437) | Parents’ country of birth. One or both parents born outside Italy and the EU. | − | + | No Difference* |
| Survey of paediatric ED clinical notes. | ||||||
| Buja et al. (2014), Italy18 | Cross-sectional (not stated in paper) (Record linkage database) | Patients (18–65 years) attending A&E (n = 35 541) migrants (n = 5385) | ‘Citizenship’. Nationality assumed to be that of country of birth if not born in Italy. | ++ | + | Patients arriving at weekends and bank holidays mainly Temporarily Present Foreigners and those from High Migratory Pressure Countries. |
| Most patients arrive at A& E between 08:00–16:00 h, patients arriving between 16:00 and 24:00 h mainly from HMPC group. | ||||||
Note: A study that reported more than one review outcomes of interest will appear more than once in the table. ++, all or most of the checklist criteria have been fulfilled; where they have not been fulfilled the conclusions are very unlikely to alter. +, some of the checklist criteria have been fulfilled, where they have not been fulfilled, or not adequately described, the conclusions are unlikely to alter. −, few or no checklist criteria have been fulfilled and the conclusions are likely or very unlikely to alter.9
a
Adjusted for age.
b
Adjusted for gender.
c
Adjusted for socio-economic status.
d
Adjusted for health status.
e
Adjusted for time in host country.
f
Adjusted for other factors (region, marital status, Attending speciality, Triage colour).
g
Adjusted for mother’s age at delivery, mother’s educational level, child gender, previous live births.
*
Not tested for significance.