1. Introduction
Given the continuous immigration flows to Europe and the increased numbers of migrants arriving to EU member states [
1], issues related to the more equitable and effective delivery of healthcare become a high priority for EU member states. Increasing access to basic health services for the migrant population in host countries is of utmost importance. The scant evidence that exists on healthcare access for migrants in Europe makes it difficult to make comparisons between systems and countries or to support public policy decision making [
2]. Although the lack of language services overall and the fear of being ethnically discriminated within the healthcare setting are highlighted in the literature as prevalent problems for migrants when accessing healthcare [
3], less is known about the experiences of the migrant population once they actually access the healthcare services in Europe.
The rights of refugees, asylum seekers, and migrants to healthcare access vary across European countries in terms of regulation and laws [
4]. Even when access to healthcare is granted by law, research suggests that migrant groups, in particular asylum seekers [
5] and undocumented migrants [
6,
7], face several obstacles when trying to access healthcare. In this regard, the difficulty to communicate effectively in a host country’s language can impede access to healthcare services [
8]. The provision of translation services can help to overcome these communication barriers; however, the availability of free and accessible interpretation services is highly variable across Europe [
9]. Studies indicate that the presence of professional interpreters can improve both the quality of care [
10] as well as patient satisfaction. Alternatively, inefficient communication between healthcare providers and patients increases the risk of misunderstanding and misdiagnosis [
11]. Moreover, patients who face communication difficulties visit their healthcare provider less often [
12] and are less compliant with medication and treatment advice [
13]. The lack of professional interpreters has been associated with unnecessary, extensive, and potentially harmful medical exams, treatments, and hospitalisations [
14,
15].
Furthermore, a large-scale study (QUALICOPC) conducted in 31 European countries showed that most differences on perceived healthcare discrimination were found between the native inhabitants of a country and first-generation migrants, reporting more discrimination within healthcare settings for the migrant population [
16]. In this regard, previous studies have identified that experiencing ethnic discrimination in the healthcare setting is associated with poor quality of care [
17], highest levels of distrust in healthcare providers as well as distrust in the healthcare system overall [
18,
19]. Moreover, different studies have also found an association between experiencing healthcare discrimination and decreased healthcare utilization and participation in preventive services. Studies have also indicated that discrimination within healthcare settings is related to increased emergency department visits and hospital admissions [
20]. Moreover, experiencing healthcare discrimination has a detrimental impact on health outcomes. As a result of racial discrimination, studies have shown that migrant patients with diabetes do not receive appropriate diabetes care [
21] and experience higher risk of diabetes comorbidities [
22]. Moreover, racial discrimination in healthcare is also associated with lower medication adherence in ethnic diverse patients taking hypertension medication [
23,
24].
Mig-HealthCare Background
The Mig-HealthCare project was a three-year project launched in May 2017 [
25]. The project’s main scope was to provide evidence-based information and practical guidance to primary healthcare professionals, primarily in the EU member states, on how to best address the health issues of refugee and migrant populations.
In order to support public policy decision making to reduce obstacles in healthcare access, and thus reduce health disparities for migrant populations in Europe, it is important to study the migrant patient experience as a result of them accessing healthcare services. Given the worse health outcomes, healthcare misuse and receiving lower quality of care associated with the perception of ethnic discrimination in healthcare and the lack of interpretation services, generating a greater understanding of the patient experience of the migrant groups in Europe is important to plan future tailored healthcare systems and services [
26]. This paper, has collected and analysed migrant patient experiences in 10 European countries (Austria, Bulgaria, Cyprus, France, Germany, Greece, Italy, Malta, Spain, and Sweden). The primary aim of this study is to assess the perception of ethnic discrimination within healthcare services in Europe while having a comparative framework to understand the commonalities and differences across European countries by country of origin and language proficiency of migrants. The secondary aim is to assess the need for translation services when accessing healthcare services and identify factors that predict the feeling of being ethnically discriminated when accessing healthcare in Europe.
3. Measures
3.1. Discrimination in Medical Settings (DMS)
Ethnic discrimination experienced in medical settings was measured through the Discrimination Scale in Medical Settings (DMS scale) [
28]. Participants were asked the following questions: “When getting healthcare of any kind, have you ever had any of the following things happen to you because of your race or ethnicity?”, with seven adapted items: (1) you are treated with less courtesy than other people, (2) you are treated with less respect than other people, (3) you receive poorer service than others, (4) a doctor or nurse acts as if they think you are not smart, (5) a doctor or nurse acts as if they are afraid of you, (6) a doctor or nurse acts as if they are better than you, and (7) you feel like a doctor or nurse is not listening to what you were saying. Response categories were 1 = never, 2 = rarely, 3 = sometimes, 4 = most of the time, and 5 = always. Then, all seven questions were summed and a mean score on the entire scale was computed, with higher scores indicating more perceived discrimination (range from 1–5 units). The DMS scale was evaluated for reliability. Pearson correlations between DMS score and its seven component items were positive and larger than 0.3 and the diagonal Cronbach’s α results scored excellent (more than 0.9), therefore confirming the reliability of the DMS scale (
Supplementary Tables S1 and S2). A mean score on the entire scale was computed, with higher scores indicating more perceived discrimination.
3.2. Mental and Physical Health Status
For this study, the relevant to mental health scale of the Short Form 36 (SF-36) [
29] was used to provide an assessment of mental health status. The following five questions were used to assess psychological distress and well-being, scoring from 0 (low) to 100 (high): “Have you been a very nervous person? Have you felt so down in the dumps that nothing could cheer you up? Have you felt calm and peaceful? Have you felt downhearted and blue? Have you been a happy person?”. The total mental health score was evaluated for reliability. Pearson correlations between the mental health score and its five component items were positive and larger than 0.3 and total Cronbach’s α was acceptable (>0.7), therefore confirming the reliability of the mental health score (
Supplementary Table S3).
To assess physical health status, participants were asked about chronic conditions. A list of 22 choices was provided (multiple options possible) following the question: “Do you suffer from any of the following chronic diseases or long-term conditions? (tick all that apply)”.
3.3. Accessibility to Healthcare Services and Translation Services
Accessibility to healthcare services was measured with the following question “Need to use healthcare services the last 6 months”, respondents were given the following response options: “Needed and did not have access”, “Needed and had access” and “Did not need”. Two more indicators of accessibility to healthcare were also asked. Participants were asked if they perceived having worse access to healthcare services compared with local people with the response categories being “yes” and “no”. Moreover, participants were asked about the need of translation services during their medical visits with the response scale being never, few times, most times, and always.
3.4. Sociodemographic Characteristics
The following variables were included to further elaborate on the study sample characteristics: age, sex, country of origin, country of interview, number of years in the education system, having children, speaking language of the country of interview, and legal situation in the country.
3.5. Statistical Analysis
A descriptive analysis was performed for all available data. We conducted linear regression analysis to investigate variations in DMS scale (dependent variable) by country of interview, country of origin, and socioeconomics. Confounding factors were examined by applying three different linear models, starting from the univariable model (forward procedure). To investigate relevant changes in DMS score by legal situation in the country and health status, we selected a negative binomial model due to poor fit of the linear (Poisson distribution of the residuals) and Poisson (overdispersion) models. Finally, a multivariable logistic model was used to compare the odds of having access between migrants with different health status, country of interview, and kind of permission to stay in the country. All variables in each model were initially conceptualized by the perspective of clinical interest. Following this, the final models with the best fit were selected using Collet’s method, based on Akaike information criterion (AIC). All variables in the models were tested for collinearity. The level of statistical significance was defined as alpha = 0.05.
4. Results
The general demographics of the study participants are presented in
Table 1. In
Table 2 we observed a high DMS score in Greece, Italy, Cyprus, and Austria. The lowest score was reported in Spain. Migrants from Afghanistan tended to score higher in the DMS scale.
In
Table 3, perceived discrimination in migrant women is presented by country of interview, country of origin, and age. We observed a high DMS score in migrant women in Cyprus and Greece. The lowest score was reported in Spain (same as
Table 2). Migrant women from Iran scored significantly higher in the DMS scale compared with other countries of origin. Higher age indicated lower DMS score among migrant women.
Migrants in Greece reported needing and not having access to healthcare services less frequently, compared with other countries (
Table 4). Almost all migrants in Bulgaria and the majority in Italy and Greece required a translator, while the lowest percentage of requiring a translator was observed in Spain.
Linear regression models are presented in
Table 5 to observe associations in DMS scale by country of interview, country of origin, and socioeconomic status. In Model 3, the natural logarithmic transformation of the DMS score as the dependent variable was used due to poor fit of the standard linear regression model. Migrants in Italy and Austria scored higher in the DMS scale compared with Spain (Models 1–3). The DMS score was lower for older participants and for those with more years of education (Models 2,3). Migrants from Nigeria, Syria, Iraq, and other countries scored lower in the DMS scale compared with migrants from Iran. Age was significantly associated with DMS scores, with younger migrants scoring higher DMS (−0.006; 95% CI [−0.012, −0.001]).
Speaking the language of the country of interview was negatively associated with the DMS score in the univariate model (−0.093; 95% CI [−0.164, −0.021]) (not shown in the Table) and it was not statistically significant in the multivariate model. This could be explained perhaps due to the potential confounding effect of country of interview and country of origin (
Supplementary Table S5).
DMS was transformed into an integer scale (e.g., from 1.257 to 1257) due to poor fit in the linear model and Poisson distribution of the residuals. We used negative binomial regression due to increased dispersion in the Poisson model (
Table 6). Older migrants reported better treatment experience (Models 1,2). Migrants with better self-perceived mental health scored lower in the DMS scale (0.994; 95% CI [0.993, 0.996]).
In Model 2, we added an interaction term for country of interview to include other kinds of legal permission. Migrants with no other kind of residence permission in Greece had higher DMS scores compared with migrants with some kind of permission (1.384; 95% CI [1.189, 1.611]). Migrants with no kind of permission in Austria had lower DMS score by 24% compared with migrants who had some kind of permission in Austria (0.763; 95% CI [0.632, 0.922]).
Logistic regression analysis was performed to investigate the likelihood of having access to healthcare services (
Table 7). Migrants in Greece were more likely to need and not have access to healthcare services compared with Spain (0.293; 95% CI [0.166, 0.516]). Female migrants had 60% higher odds of needing and not having access to healthcare services compared with males (1.613; 95% CI [1.183, 2.199]). Migrants with health problems (chronic problems from injury/accidents, gastrointestinal disease, diabetes, skin disease, headaches/migraines, and diseases related to bone and muscle) were more likely to needing and not having had access to healthcare services compared with healthy migrants. Migrants with chronic problems had the highest odds of needing and not having access to healthcare services compared with other health problems (3.292; 95% CI [1.585, 6.837]). Migrants with gastrointestinal disease or diabetes had higher odds ratios compared with migrants with skin diseases, headaches, or migraines, and diseases related to bone and muscle.
5. Discussion
To our knowledge, this is one of the first studies to assess the patient experience of the migrant population across 10 countries in Europe. The main finding from this study is that better self-reported mental health outcomes as measured by the SF-36 relevant mental health scale were associated with lower perceived discrimination in medical settings. Moreover, migrant women were more likely to not be able to access healthcare services when needed. Likewise, the same findings were reported for migrants suffering from chronic illnesses. Finally, older migrants reported higher feelings of health discrimination.
To our knowledge, this is the first study using the DMS questionnaire among a multi-cultural group of migrants in Europe. Two previous studies validated the DMS in two ethnic minorities samples in the United States [
21,
30]. Similar to these studies, we evaluated the reliability of the scale by calculating Cronbach’s α and Pearson correlations between the DMS score and its seven component items (r > 0.3). Although there is a large number of migrants arriving to Europe, no studies have been conducted assessing ethnic discrimination in healthcare services. For this reason, future studies should explore psychometric properties of this scale among specific migrant groups, different healthcare settings, and across Europe. In this regard, the results of this study contribute to healthcare systems efforts to assess and address healthcare discrimination in the migrant population [
31].
Moreover, findings indicate that the migrant population reported higher levels of healthcare discrimination in Greece, Italy, Cyprus, and Austria and lower levels in Spain (
p < 0.001). Additionally, migrants from Afghanistan tended to score higher in the DMS scale (
p < 0.05). The lower level of healthcare discrimination reported in Spain may be explained by some characteristics of the study population because 186 (92.08%) participants were from South America and were Spanish speakers. Therefore, our results demonstrate a high relationship between the need for translators and the feelings of being discriminated when accessing healthcare services, making the role of translation services highly relevant for the quality of care for migrant populations. Recent evidence has also suggested additional benefits at system and professional levels, such as cost-saving for the healthcare system, and reducing difficulties during the appointments [
8].
The results of our study also show significant differences of perceived ethnic discrimination in the healthcare system especially regarding age. Here, younger migrants reported worse treatment experience in our sample from 7 EU countries. A recent systematic literature review by Robards and colleagues [
32] found that for marginalized young people, the decision to access health services is affected by previous bad experiences during which they felt treated differently and with disrespect by healthcare professionals. In the same article, the study population highlighted different actions to be considered in the delivery of healthcare for migrant young groups such as culturally appropriate services, cultural sensitivity of staff, and the use of interpreters.
Our study also highlighted the vulnerability of migrant women regarding both the lack of access to healthcare services and the perception of higher discrimination within healthcare services. In addition to the common barriers of migrants in accessing healthcare [
33], the results of this study could be explained by differences in care seeking behaviour of this group [
34] or in health services-related factors [
35]. For this reason, health interventions aiming to mitigate gender-driven inequalities in accessing quality healthcare have to be put in practice in the European context [
36].
The management of chronic diseases in European migrants and refugees has been identified as a priority for health service provision [
27]. However, the results of this study show that migrant chronic patients reported the highest odds for needing but not having access to healthcare services. This could be explained by the higher contact frequency that chronic patients have with the healthcare systems, which may increase their health literacy and empowerment, making them more aware of their needs, their rights, and their expectations from the system. Hence, this group could be more perceptive to discrimination during their medical visits. Moreover, more frequent health visits could increase the likelihood to be exposed to experiences of healthcare discrimination towards them. Another reason may be that frequent visits expose general inefficiencies of the healthcare system and provision (such as lack of personnel or equipment) that are more attenuated in these groups.
Additionally, in our study differences were found concerning healthcare discrimination towards undocumented migrants depending on the host country. In this regard, undocumented migrants in Greece (1.384; 95% CI [1.189, 1.611]) had higher DMS scores compared with migrants with legal status. However, the opposite trend was reported in Austria, where undocumented migrants had a lower DMS score by 24% compared with migrants with legal status (0.763; 95% CI [0.632, 0.922]). In this regard, Robertshaw and colleagues [
37] found that immigration status and legislative policy are a challenge for the provision of healthcare by creating or reinforcing vulnerability of marginalized groups [
38]. However, the results of our study could be interpreted in the light of the results of Dauvrin and colleagues [
39] who also reported insufficiencies in the actual delivery of care for undocumented migrants despite the variations in healthcare entitlement related to the immigration status across Europe, suggesting that even in countries with “minimum rights”, health professionals may consider treating undocumented migrants more important than abiding by law (“pragmatic health professional”). For that reason, our results might be outlining a complex interplay of different factors that might be worsening the provision of healthcare for migrant patients in the host country beyond their legal status, as this is not a general result and differences appear between countries in our study.
Although specific migrant groups have reported experiencing discrimination in healthcare, ethnic discrimination, and translation services, these are still under-researched topics in Europe. In this regard, our results could be also interpreted from a structural and organizational point of view for healthcare delivery. Indeed, King and colleagues [
40] argue how (1) compulsory assigned residency, (2) resources (including language skills), and (3) freedom of movement (related to documented/undocumented status) could be consolidating heavy and stable forms of devaluation, reification, and stigma, denying access to healthcare for certain groups with negative consequences on the health of migrants.
This study shows the experiences of migrants and refugees between 2018 and 2019. The inequalities and vulnerabilities shown in this study may have been further amplified by the COVID-19 pandemic, which was not captured in our study but should be briefly discussed. Even though healthcare services face an unprecedented demand, groups such as undocumented migrants and migrant women are still facing barriers to access appropriate quality care that contributes to poorer health outcomes [
41,
42]. Moreover, epidemiological data shows that COVID-19 disproportionately affects patients who have chronic conditions and underlying comorbidities [
43]. This is in line with the report of Fiorini and colleagues [
38], indicating that due to limited access to appropriate care and quality of care, migrant and refugee populations, and specifically some groups such as undocumented female or chronic patients, might experience higher morbidity and mortality during the COVID-19 pandemic,
To the best of our knowledge, this is the first study assessing the migrant patient experience with a multicultural sampling in seven EU Member states regarding healthcare discrimination and the need for translation services. Another strength includes the range of data and populations investigated allowing for comparative and intersectional analyses across many dimensions of discrimination in healthcare services. On the other hand, limitations include relying on self-reported information which may introduce reporting bias. Furthermore, study participants varied greatly in country of origin, duration of stay in country of interview, and integration phase. The use of interpreters may have introduced additional information bias, and cultural barriers in female representation in the survey for some countries (such as Afghanistan) may have biased their responses. Moreover, considering the convenience sampling method used, respondents may have given socially desirable answers, leading to underestimation of healthcare discrimination. Finally, as this is a cross-sectional study which relies on a non-random sample, causal relationships cannot be established.