Access to cardiac rehabilitation and the role of language barriers in the provision of cardiac rehabilitation to migrants

The present study is a survey-based descriptive study. Data on the provision of CR services in hospitals and municipalities in Denmark are based on nationwide surveys conducted by the Danish Cardiac Rehabilitation Database (DHRD) [29] in 2013 and 2015. The Danish healthcare system is publicly financed through taxes and operates across three political and administrative levels, national, regional and local. The state has a regulatory and supervisory function. The state is divided into five regions that are responsible for hospitals, general practitioners and psychiatric care. The regions contain 98 municipalities that are responsible for providing primary healthcare services [30].

Information on core components of available cardiac rehabilitation programmes and organization of CR services were collected by way of parallel web-based questionnaires sent out to all hospitals offering CR services (n = 35) and all Danish municipalities (n = 98). Core components of CR included exercise training, patient education, psychosocial support, smoking cessation support and nutritional counselling.

The respondents were physicians, nurses, dietitians and physiotherapists, who are part of the multidisciplinary CR team. Each of them had to respond to a section of the questionnaire related to their profession.

The questionnaires were complied using the recommended national clinical guideline regarding the coverage of core CR components. A more detailed description of the survey can be found elsewhere [31]. Hospital-based questionnaires were derived from the DHRD, which collects programme-level CR data. The submission of data is required for hospitals according to Danish Law whereas the participation for the municipalities is optional. A combination of closed-ended and open-ended questions was applied to give respondents the opportunity to elaborate on their answers.

In the present study we used hospital data from 2013 and 2015 to examine provision of CR core components for migrants and data from 2015 to examine the distribution of provision across the five Danish regions, and whether hospital type and percentage of migrants in a hospital’s catchment area was related to provision.

The survey from 2015 contained additional questions about socially differentiated CR, socially vulnerable groups including migrants and special offers to these groups. Due to survey design, data on provision of interpreter services and multilingual information material was only obtained from hospitals and municipalities that offer socially differentiated CR.

Finally, we investigated factors that could potentially predict the provision of interpreter services and multilingual material. For hospitals, data variables applied were hospital type and percentage of migrants in a hospital’s catchment area. Municipality data variables were percentage of migrants and size of population in the municipality.

Response rates for hospitals reached 100 and 96% for municipalities, of which 100 and 93% responded that they offer at least one CR core component (Fig. 1).

The study found incomplete provision of CR core components to migrants in hospitals. These provision rates were virtually unchanged from 2013 to 2015 (Additional file 1: Table S1). Moreover, in 2015 the provision of each CR core component to migrants in the five Danish regions ranged from 44 to 100% and none of the regions had full provision of all core components in general (Table 1).

Socially differentiated CR was offered in 54% (n = 19/35) of the hospitals, and in 44% (n = 31/71) of the municipalities. Overall migrants were considered vulnerable in up to 84% of the hospitals, more specifically 74% were non-Danish speaking patients and 84% were patients with an ethnic minority background (see definition of migrants in this study in the methods section). The corresponding percentages for municipalities were respectively 58 and 61%. These proportions did not differ considerably in percentage terms from other potentially vulnerable groups such as patients with low education, cognitive disability and those living alone except from those with a long distance to provider. No statistical comparison measure was calculated between groups. There were no significant differences between hospitals and municipalities (Table 2).

Provision of interpreter services in hospitals reached full coverage, whereas it was lower in percentage terms in municipalities 84% (n = 26/31), the difference was not statistically significant. The availability of multilingual information material was low in hospitals 32% (n = 6/19) and in municipalities 3% (n = 1/31). The difference between hospitals and municipalities was significant (p = 0.009) (Table 2).

University hospitals had a higher provision of multilingual information material 44% (n = 4/9) than general hospitals 20% (n = 2/10), and hospitals with the highest proportions of migrants in their catchment area were more likely to have multilingual information material 67% (n = 6/10), although numbers were small. The association between provision of multilingual information material and proportion of migrants in a hospital’s catchment area was significant (p = 0.003).

Provision of interpreter services was not associated with the proportion of migrants. Municipalities with the largest populations were more likely to offer interpreter services 90% (n = 19/21) than the ones with smaller populations 70% (n = 7/10) but the difference was not significant (Table 3).

Some hospital respondents commented in the open-ended questions that they specifically referred migrants to the municipality where there are customized offers for non-Danish speaking patients.

Open ended responses also revealed that some respondents did not respond to consider migrants as vulnerable in the close-ended questions because they did not want to stereotype patients, but instead performed an individualized assessment of the patients’ needs.

According to our results, hospitals did not have full provision of CR core components to non-Danish speaking patients. Only minor changes in provision of CR services were observed from the years 2013 to 2015, e.g. the provision of exercise training increased from 91 to 92%; and the provision of patient education increased from 71 to 74%. These percentages must be interpreted with caution as a few respondents from hospitals explained that they did not exclude non-Danish speaking patients but would refer these patients to municipalities with whom they share responsibility for CR, and therefore had indicated not to provide CR services at the hospital. However, some of the negative responses were left without any comments, and the reasons behind this potential lack of available services needs further investigation. Where language barriers were the reasons given for some hospitals not to provide CR, it would contradict the recommendations of the national clinical guideline for CR [6]. The guideline recommends that CR should be offered to all eligible patients regardless of background and advises providers to be attentive to well-known barriers to participation, such as language. Even though the offer of interpretation is available in theory, this is often organized based on the principal of individual consultation. CR is group based by nature and therefore people with language barriers could be excluded [32]. The right to equal access to healthcare services is also established in Danish law and international human rights covenants [21, 33]. We found a variation in the proportion of hospitals that offered CR to migrants across the five Danish regions; and no region had full provision of core components. A recent Danish study on rehabilitation after brain injury showed regional variations in rehabilitation despite the existence of a national guideline [34]. Together with our findings, this could reflect that the implementation of guidelines can be challenging, and requires monitoring in order to ensure they improve quality in clinical practice [35]. The health system could reduce disparity in healthcare by standardized descriptions of disease management programmes, which are implemented by law, in order that all patients receive treatment of uniform high quality regardless of where they live. This has been proven to be effective in the field of cancer treatment’.

Socially differentiated CR that focused on vulnerable patients was offered in 54% of the hospitals, and in 44% of the municipalities that provided data but, the open-ended answers revealed that there was no uniform understanding of the term. Non-Danish speaking patients and patients with an ethnic minority background were considered vulnerable by 74 and 84% of responding hospitals respectively. The corresponding percentages were lower for municipalities, 58 and 61% respectively. These percentages must also be interpreted with caution as some respondents were reluctant to stereotype by certain vulnerability parameters; and would rather do an individual assessment of needs and provide individualized healthcare services. The provision of interpreter services in hospitals reached 100% whereas it was only 84% for municipalities. The full coverage in hospitals can be seen in the light of hospitals being legally required to provide interpretation [33], while municipalities are not subject to the section of the Danish Health Act about provision of interpreter services. Municipalities are solely obliged to provide secondary prevention care equal to that of hospitals. The section about interpretation in the Danish Health Act is the only law that explicitly states mandatory use of interpreters specifically aimed at the public hospitals, medical specialists and general practitioners, thereby leaving a gap in other health services outside of these health facilities regarding the provision of interpreter services.

A study in 2012 mapped the provision of interpreter services in 240 health services (48 emergency services, 48 mental health services, and 144 primary care services) across 16 different European countries and found that 42% of the services did not provide any form of interpretation. Countries that tended to have higher provision of interpreter services had policies and regulations concerning cost of interpretation [36]. It is therefore fair to conclude that legislation on interpretation plays a significant role in the provision. The importance of having a legal framework for language in access to healthcare services has been discussed in the US, and that it should not an impediment to health [37]. Although the Danish and US healthcare systems are not directly comparable, the essence of the message is applicable. The differentiation in the Danish law between hospitals and municipalities seems to be reflected in the provision of interpreter services, albeit it is the very same Danish Health Act that sets out the obligations for shared responsibilities for CR. This has the potential to increase inequality in access for migrants to CR between the two sectors.

An aim of the involvement of municipalities in CR was to achieve greater equality in health and increase the inclusion of socially vulnerable patients according to WHO’s vision of “Health in All Policies” [8]. However, it seems that the inclusion of migrants in CR is inadequate.

Another interesting finding of our study was the very limited availability of multilingual information material reported to be offered by 6 hospitals, mainly concentrated at university hospitals, and hospitals with catchment areas with higher densities of migrants, and only in one municipality. This runs contrary to the findings of a review by Coulter et al. [27] showing major benefits for patient information material.