How should health service organizations respond to diversity? A content analysis of six approaches

Background Health care organizations need to be responsive to the needs of increasingly diverse patient populations. We compared the contents of six publicly available approaches to organizational responsiveness to diversity. The central questions addressed in this paper are: what are the most consistently recommended issues for health care organizations to address in order to be responsive to the needs of diverse groups that differ from the majority population? How much consensus is there between various approaches? Methods We purposively sampled six approaches from the US, Australia and Europe and used qualitative textual analysis to categorize the content of each approach into domains (conceptually distinct topic areas) and, within each domain, into dimensions (operationalizations). The resulting classification framework was used for comparative analysis of the content of the six approaches. Results We identified seven domains that were represented in most or all approaches: organizational commitment, empirical evidence on inequalities and needs, a competent and diverse workforce, ensuring access for all users, ensuring responsiveness in care provision, fostering patient and community participation, and actively promoting responsiveness. Variations in the operationalization of these domains related to different scopes, contexts and types of diversity. For example, approaches that focus on ethnic diversity mostly provide recommendations to handle cultural and language differences; approaches that take an intersectional approach and broaden their target population to vulnerable groups in a more general sense also pay attention to factors such as socio-economic status and gender. Conclusions Despite differences in labeling, there is a broad consensus about what health care organizations need to do in order to be responsive to patient diversity. This opens the way to full scale implementation of organizational responsiveness in healthcare and structured evaluation of its effectiveness in improving patient outcomes.


Background
Health service users belonging to groups that differ from the majority population, such as migrants and ethnic minorities, often receive poorer care than majority users. Variously described as health care 'disparities' , 'inequalities' or 'inequities' , these problems in health care have been well documented in the United States [1,2] and are increasingly being recognized in other countries [3][4][5].
The evidence demonstrating health care inequalities implies that health services need to adapt in order to increase their accessibility and quality for minority service users. Promoting responsiveness to diversity requires interventions at individual, organizational and system level. Individual caregivers need specific knowledge, skills and attitudes [6,7]. For health care organizations, promoting diversity responsiveness involves putting into place specific service policies and practices [8]. Diversity responsiveness at the level of entire health systems includes measures at national or state level, for example by legislation guaranteeing financial access to health care.
In the US, responsiveness to diversity at organizational level has been promoted through the CLAS standards (Standards for Culturally and Linguistically Appropriate Services) [9], designed to address the needs of racial, ethnic, and linguistic population groups that experience disparities in health service provision [10,11]. In Europe, diversity responsiveness recommendations have been published under the heading of 'migrant-friendliness' [12], 'intercultural opening' [13], 'transcultural competence' [14] or 'difference sensitivity' [15], the latter targeting 'vulnerable groups' including but not confined to ethnic and cultural minority groups.
Responsiveness to ethnic and cultural diversity has for several decades been referred to as 'cultural competence' [16]. Regarding terminology, the term cultural competence has often been criticized in the literature [17,18], mainly because of limited operationalizations focusing too exclusively on expert knowledge about minority populations. Critics correctly point out that such knowledge may lead to stereotyping and a focus on the 'otherness' of patients. However, many cultural competence frameworks have a broader operationalization, emphasizing the triad of skills, attitudes and knowledge [6].
The development of uniform terminology and definitions to describe culturally competent or diversity responsive health care is considered as key for effective evidence-based development, implementation and evaluation of diversity responsive health care [17,19]. Therefore, we qualitatively analyzed and compared six approaches to organizational diversity responsiveness in health care. We explored whether the lack of agreement about terminology between such approaches reflects important differences at the level of content, or whether there is a reasonable degree of consensus about the measures that organizations should take to adapt health services to the needs of diverse patient populations. The paper addresses the following questions: 1. On which domains of health service provision do the approaches focus? 2. How much agreement is there about important domains? 3. What are the most consistently recommended elements for organizations to address in order to be responsive to diversity?

Method
Sampling strategy of approaches In recent years many guidelines or recommendations for increasing organizational responsiveness to diversity have been published internationally by both public and private bodies. We explicitly did not aim at conducting a comprehensive review of all available approaches. Instead, we purposively sampled six approaches to diversity responsiveness at organizational level using the following criteria: the approach was developed for wider than local use by a single health care organization; it was developed by an authoritative public body, such as a health ministry or a recognized group of experts; it was publicly available; the final selection included approaches from the US, Australia and Europe.
We started with the CLAS Standards [9] and continued adding approaches one by one, searching the published and grey public literature in an iterative process, until saturation was reached (i.e. adding a new approach did not provide new information). The sampling was conducted in 2012.
The following approaches were sampled: 1) CLAS Standards -National Standards for Culturally and Linguistically Appropriate Services in Health Care (further referred to as: CLAS) developed by the Office of Minority Health, part of the U.S. Department of Health and Human Services [9]; 2) Advancing Effective Communication, Cultural Competence, and Patient-and Family Centered Care: A Roadmap for Hospitals (further referred to as JCR), developed by the US Joint Commission [20]; 3) Cultural Responsiveness Framework. Guidelines for Victorian health services (further referred to as CRF) developed by the Victorian Government, Department of Health, Australia [21]; 4) Recommendation of the committee of ministers to member states on mobility, migration and access to health care (further referred to as COER) of the Council of Europe [22,23]; 5) The Equality Delivery System (further referred to as EDS) for the NHS, UK [24]; 6) Standards for Equity in Health Care for Migrants and other Vulnerable Groups (further referred to as EQS) developed by the WHO-HPH Task Force on Migrant-Friendly and Culturally Competent Healthcare [25]. Table 1 briefly describes each approach and clarifies our reasons for including the approach in our study.
In May 2013 the Enhanced CLAS Standards were published, entitled "National Standards for Culturally and Linguistically Appropriate Services in Health and Health Care: A Blueprint for Advancing and Sustaining CLAS Policy and Practice" [11]. A comparison of the new Standards with the original version showed that the underlying approach was virtually identical to that which informed the first edition. The aim of the revision was to increase the effectiveness of the Standards, by explaining them more clearly, ensuring that they reflected recent developments, and aligning them with other initiatives such as the Affordable Care Act and the work of the Joint Commission. Some differences of emphasis are described in Table 1. The version of CLAS used in this paper is the original one: we considered the original CLAS Standards as the oldest and most influential of all approaches and we did not feel the changes made in the enhanced version were fundamental enough to warrant a separate analysis.
Developing an analytical framework for the present analysis Encoding of content 1. An inductive approach was used. We first analyzed CLAS [9] by grouping the 14 standards into conceptually distinct topic areas (domains). We subsequently identified dimensions that showed the concrete operationalization within each domain. 2. A second approach was selected and its content was subsumed under the domains and dimensions identified in Step 1, new ones being created where necessary. 3. The remaining four approaches were treated in the same way. 4. The resulting framework of domains and dimensions was critically reviewed and discussed until consensus was reached by three of the authors in order to remove ambiguities and overlap.  [ 9 ]. These standards were developed by the Office of Minority Health, part of the U.S. Department of Health and Human Services. Some of the standards have the status of mandates, meaning that they are Federal requirements for all health care organizations that receive Federal funds; others are purely recommendations. We included the CLAS standards because they were the first one and probably the most comprehensive and influential approach currently in use. In May 2013 the Enhanced CLAS Standards were published [ 11 ]. Although largely similar, there are some differences of emphasis between the original and the Enhanced CLAS Standards: • The revised CLAS acknowledged that in order to address disparities in health care (for any target group), we need to go beyond cultural issues and deal with other (e.g., social, psychological) issues.
• In the vision on responsive care some slight changes of emphasis could be found, such as a shift from regarding diversity as a 'group' characteristic to 'appreciating the diversity of individuals'. The enhanced CLAS also places more emphasis on the importance of 'patientand family centred care', thus bringing it more into line with the JC Roadmap.
2. Advancing Effective Communication, Cultural Competence, and Patient-and Family Centered Care: A Roadmap for Hospitals (JCR) [ 20 ]. This 'Roadmap' has been developed by the Joint Commission (JC), an independent, not-for-profit organization which accredits and certifies health care organizations in the United States. The Roadmap was developed in addition to existing JC requirements "to inspire hospitals to integrate concepts from the fields of communication, cultural competence, and patient-and family-centered care into their organizations." We included the JC Roadmap because of the global influence of JC and the Joint Commission International (JCI) accreditation program on health care organizations through their accreditation programs (applied in over 50 countries). JC developed its own framework of recommendations in which cultural competence is embedded within effective communication and patient-and family centred care. Please note that 1) other existing JC requirements also include issues related to those issues discussed in the Roadmap, and 2) that the national Joint Commission Standards are different from the Standards of the Joint Commission International.
3. Cultural Responsiveness Framework. Guidelines for Victorian health services (CRF) by the Rural and Regional Health and Aged Care Services, Victorian Government, Department of Health (Australia) [ 21 ]. The CRF was developed to replace the Health Service Cultural Diversity Plans (HSCDPs) which since 2006 have required all Victorian health services to develop and implement policies for ethnic diversity in care. The intention of the CRF is to consolidate multiple requirements for reporting on cultural diversity initiatives within health services. All Victorian health services are required to submit plans and achievements according to the standards and measures in the CRF to the Statewide Quality Branch. We included the CRF because it has been disseminated and made compulsory in a large health care system in Australia.
4. Recommendation of the committee of ministers to member states on mobility, migration and access to health care (COER) of the Council of Europe [ 22 , 23 ]. The Council of Europe is an international organization set up "to achieve a greater unity between its members for the purpose of safeguarding and realizing the ideals and principles which are their common heritage and facilitating their economic and social progress" [ 37 ]. We included the COER because it has been endorsed by the Health Ministers of the 47 member states of the Council of Europe. The document is aimed at ministerial level, therefore it includes recommendations that have consequences for the whole health system. These recommendations focus on the diversity responsiveness in the context of migrants and their children. To make comparisons possible we have only included the recommendations at organizational level in our analysis.

Equality
Delivery System (EDS) for the NHS [ 24 ]. EDS originates from the Equality and Diversity Council within the British National Health Service (NHS). It is designed to help NHS organizations to comply with the 'Public Sector Equality Duty' (PSED) of the Equality Act. This act "requires public bodies to consider all individuals when carrying out their day to day workin shaping policy, in delivering services and in relation to their own employees" [ 38 ]. EDS is made available to the NHS as an optional tool. It was included because it is a European instrument which has been disseminated in a large health care system. 6. Equity Standards (EQS) of the Task Force on Migrant-Friendly and Culturally Competent Healthcare [ 25 ]. These Standards were developed by a group of mainly European experts set up within WHO's Health Promoting Hospitals network. The Equity Standards are a self-assessment instrument to enable health care organizations to carry out an 'equity evaluation' against a set of standards. The instrument was piloted in 10 European countries, as well as in two non-European ones. The Equity Standards were included because of the broad international context in which they were developed.

Comparison of content
Categorizing the content of the approaches in this analytical framework enabled us to see at a glance whether certain domains were unique to, or absent from, particular approaches. Within each domain, it revealed the differences in the ways in which approaches operationalized the domain. We listed the differences between them as well as their commonalities. MCS performed the analysis; results were independently reviewed by MLE-B and JDI. Disagreements were resolved by discussion.

Ethics statement
An ethics statement was not required for this work.

Results
This section starts with background information on the six approaches. We then provide an overview and analysis of their content, classified according to the domains and dimensions of the analytical framework. We then describe the similarities and differences between the six approaches. Table 2 provides background information on the six approaches, listed with their acronyms. As their aim, the approaches refer to reducing or eliminating existing inequalities in health and service provision between different populations (CLAS, COER, EQS), as well as improving outcomes for individual patients (JCR, CRF, EDS). Some approaches (CLAS, COER, EQS) start from human rights principles, regarding inequalities between groups as injustices which should be eliminated. JCR considers equality of patient outcomes as a performance indicator, while CRF and EDS combine both starting points. Two unique aims also emerge: CRF aims to enhance the cost-effectiveness of health service delivery, and EDS aims to create better working environments for staff.

Background information
In their vision on responsive care, three approaches directly invoke the concept of 'cultural competence'. CLAS refers to the classic definition of cultural competence (Cross et al., 1989) [16] and explicitly emphasizes language issues. For JCR, cultural competence is one of three fundamental concepts on which the Roadmap elaborates (the other concepts being 'effective communication' and 'patient and family centered care'). JCR and CLAS operationalize cultural competence similarly. CRF introduces another label: 'culturally responsive care' , but the vision implied is very similar to that of CLAS and JCR. A common characteristic of these three approaches is their emphasis on 'culture' in the labeling of their vision.
In the other three approaches the emphasis is not on 'culture' but on 'equity' or 'equality'. COER refers to "equitable access to health care of appropriate quality": in relation to service delivery, by referring to "improving the adaptation of health service provision to the needs, culture and social situation of migrants". EDS does not provide a definition of its concept of equality, but relates it to the pursuit of quality, which in turn is defined as recognizing the needs and circumstances of all (both patients and staff ) and ensuring accessibility, appropriateness, safety and effectiveness for patients. EQS explicitly distances itself from the concept of 'cultural competence' , instead highlighting Whitehead's definition of equity in health: "equal access to available care for equal need; equal utilization for equal need; equal quality of care for all" [26].
The target population of each approach refers to the user groups envisaged by the authors as beneficiaries. CLAS, JCR and CRF refer to the target population in terms of race, ethnicity, culture or language, while EDS and EQS also include gender, age, disability, religion, sexual orientation, transgender status (both EDS and EQS); marriage and civil partnership, pregnancy and maternity, nationality (only EDS); socio-economic status and minority status (EQS). COER focuses explicitly on migrants, a category that is not mentioned explicitly in CLAS, JCR, CRF or EDS. However, COER uses the term migrant "in a very broad sense, referring not only to those who change their country of residence voluntarily but also to asylum seekers, refugees and victims of human trafficking. Since the consequences of migration may also extend beyond the first generation, second and later generations are also discussed. In the case of Internally Displaced Persons, internal migrants are also included". There is thus considerable overlap between the category of 'migrants' as defined by COER and the term 'ethnic minorities' used in other approaches.

Organizational commitment
We classified elements of the various approaches in this domain if they mentioned commitment at management level to responsiveness to diversity. Two dimensions were found: policy and leadership and measurement of performance.

Policy and leadership
All six approaches maintain that organizations must make an explicit commitment to developing responsive care, rather than merely permitting individual initiatives that are not structurally embedded in the organization. In COER this requirement is implied by insistence on a 'whole organization approach'. Commitment can either take the form of an explicit plan (CLAS, EQS), which sets out how the organization intends to organize and guarantee responsiveness, or a policy of good leadership (JCR, EDS), or both (CRF). 'Good leadership' is explicitly committed to achieving responsive care and promotes this within the organization (JCR, EDS): leaders take responsibility for reaching this goal (CRF). All approaches emphasize that plans for change should be integrated in existing organizational policies and processes. EQS additionally promotes a 'proactive' approach: in all its plans, the organization should anticipate the effect the plans will have on accessibility and quality of care for vulnerable groups.

Measuring and improving performance
All approaches regard it as essential that organizations measure their performance in providing responsive services (e.g. outcomes of treatment for different groups), with the aim of identifying necessary improvements, taking action, and assessing the organization's progress in providing responsive care. CLAS, EDS and EQS further emphasize that performance measurement should be a continuous activity, incorporated in regular performance measurement systems. The approaches differ regarding the variables that should be measured: some focus on quality of care, some on accessibility and some on both (see also the domain 'collecting data' for the data sources to be used in measuring and improving performance). CRF is the only approach stipulating mandatory indicators for measuring organizational cultural responsiveness. These have to be submitted by affiliated organizations and are also used for benchmarking.

Collecting data
The second domain we identified concerns the collection of data, not as an end in itself but in order to inform the organization about equity of access to and quality of care delivered in their organization and to identify special needs or opportunities for improvement. Two types of data are distinguished, concerning the population at large and the organization's own users.

Data on the population at large
Five approaches (CLAS, JCR, CRF, EDS, EQS) recommend collecting data on the community or catchment area in order to adapt services to the needs thus identified. Organizations can use information that is already available, but CLAS and EDS also give organizations an active role in collecting these data themselves. Such data include demographic variables (e.g. age, gender, ethnicity), characteristics potentially affecting service use (e.g. language proficiency, health literacy), health status and exposure to health risks. In COER the importance of empirical evidence is strongly emphasized: governments are urged to collect it "in partnership with relevant organizations".

Data on the patient population
Patients' files can serve as a source of data on ethnicity, race, language and other characteristics considered relevant for quality of care. For CLAS, JCR, CRF, and EQS these data are considered important in order to identify and monitor health and health care inequalities. For CLAS, JCR and EQS, information in patients' files on individual characteristics associated with ethnic minority status (e.g. proficiency in the majority language) also enables adaptation of care to the needs of an individual patient. Additionally CLAS, JCR, CRF, EDS and EQS emphasize that outcome measures and patient feedback systems must be able to analyze results according to diversity characteristics. The approaches differ in the types of data they recommend organizations to collect in this respect.

Staff/Workforce
The third domain concerns the staff or workforce of the organization. Two dimensions can be distinguished: staff competencies and diversity in the workforce.

Staff competencies
Staff competencies in delivering responsive care, and the importance of education and training, are central themes in all approaches. CRF and EDS describe a comprehensive approach to improving confidence and competence among staff, for example through personal development programs (EDS) and adapted HRM policies (CRF). CLAS, CRF, COER and EQS emphasize that all staff should be trained (CLAS even includes affiliated and subcontracted staff ); CLAS, CRF and EQS also recommend monitoring the effects of training. CLAS recommends separate training in the provision of responsive care; JCR recommends the incorporation of such training in the existing curriculum, while COER and EQS support both. The approaches vary in the amount of information they provide on the content of training.

Diversity in the workforce
According to CLAS, JCR and COER, diversity among staff members is desirable for furthering responsiveness to patient diversity. Two arguments are given for the importance of staff diversity. The first is general: the workforce should be representative of the general population (CLAS, COER). The second is more specific: staff diversity is considered to further equity by making possible a higher degree of linguistic and ethnic concordance between patients and staff (CLAS, JCR). EDS and EQS discuss this issue from the perspective of equality among staff and include objectives for inclusive Human Resource policies relating to issues such as recruitment. CRF does not address the issue of staff diversity.

Ensuring access
Under this heading we classified all issues relating to barriers to entering the healthcare organization. The dimensions that emerged concerned entitlement to care, the provision of understandable information, and issues concerning geographical and other aspects of accessibility. Some issues discussed in this domain reappear in the domain of 'care provision' , because they are also relevant to the caregiving relationship.

Entitlement to care
Entitlement to care (i.e. whether patients are insured or are allowed to use national health services) is not mentioned in CLAS, JCR, CRF and EDS. This is understandable to the extent that entitlement is an issue covered by legislation and insurance rules, rather than by the policies of service providers. If service providers choose to give care outside the framework of formal entitlements, this is left to their own discretion. However, EQS goes a step further and charges organizations with a responsibility for patients who are not eligible for care: it urges that at the very least, steps should be taken to help them find appropriate care elsewhere. COER, which includes recommendations at health system level, makes a plea for "adequate entitlements to use of health services": concerning the role of service providers, it stresses that these must ensure that legislation is implemented properly and that all care providers are aware of existing rights.

'Understandable' information
Three approaches (CLAS, COER and EQS) stress that organizations should provide 'understandable' information in order to facilitate accessibility. This means providing information which is translated where necessary and is adapted to the health literacy level of the targeted populations.

Geographical accessibility
The importance of reducing geographical barriers to accessibility is briefly discussed in COER and EQS.

Other aspects of accessibility
In EDS and EQS two unique dimensions related to accessibility appeared. Firstly, EDS mentioned specific types of care (public health, vaccination and screening programs); secondly, EQS discussed the accessibility of organizations for specific 'disadvantaged' target groups such as HIV/ AIDS patients, disabled patients, and homeless people.

Care provision
Issues in this domain relate to the quality of care patients receive within an organization. Topics mentioned in the six approaches include: care that is responsive to the needs and wishes of patients, patient participation in the care process, overcoming communication barriers, providing 'understandable' patient information materials, trust, and patients' rights.

Care responsive to needs and wishes
All approaches underline the importance of this issue. The interpretation of this dimension is related to the different visions the approaches have concerning the nature of responsive care: CLAS and CRF focus on the cultural needs of patients, in accordance with their respective visions on responsive care ('culturally competent' and 'culturally responsive'). JCR refers to 'additional' and 'unique' needs that should be integrated in the clinical process: "it is important for hospitals to be prepared to identify and address not just the clinical aspects of care, but also the spectrum of each patient's demographic and personal characteristics". COER focuses on the needs of migrants (broadly defined), going beyond cultural factors to consider social position, migration history and legal situation. EDS and EQS focus on needs resulting from patients' individual characteristics.
Apart from identifying needs, JCR also discusses the points in the care continuum at which they should be taken into account. Although all approaches emphasize the importance of taking patients' needs and wishes into account, they leave it up to the professional to reconcile the demands that patients or their relatives may make with the dictates of their professional conscience.

Patient participation in the care process
Five approaches (JCR, CRF, COER, EDS) explicitly refer to the importance of patient participation or involvement in the individual care process, for example in shared decision making about treatment and care planning [27]. CLAS and EQS do not refer explicitly to patient participation in this context; however, the standards they provide show that they too consider patients as active participants in their treatment.

Overcoming communication barriers in patient-provider contact
All approaches except EDS emphasize that organizations should systematically tackle language barriers in the service delivery setting, placing the onus on the organization to provide patients with language assistance where necessary. Various types of interpreting are recommended such as professional interpreters, bilingual staff or intercultural mediators; approaches differ according to which type they prefer. CLAS, JCR and COER explicitly advise against using untrained, informal interpreters such as family members. CLAS, JCR, CRF and EQS assert that organizations are responsible for ensuring the quality and competence of language assistance that is offered. CLAS and JCR mention that patients should be informed about their right to language assistance. JCR and EQS also discuss support to patients with other communication barriers (e.g. hearing or speech impairments).

Understandable patient information materials
With the exception of EDS, all approaches stress that patient information materials should be understandable for all patients, in terms of both language and level of health literacy. When suitable materials are not available, CLAS asserts that patients have a right to orally translated information. These points concern not only patient folders providing information about specific medical problems or treatments, but also consent forms and labeling of medicines.

Trust
The approaches discuss several issues related to building trust between users and service providers. The first of these is related to the environment within the health care organization: CLAS, JCR and EQS stress the importance of making this welcoming and inclusive. Some approaches include statements underlining the security of patients, stressing that patients should not be exposed to any dangers and mistreatment that might arise from their vulnerability. Phrases used include "patients are free from abuse, harassment, bullying, violence" (EDS); and "the patient has the right to be free from neglect, exploitation, and verbal, mental, physical and sexual abuse" (JC Requirements (see Table 1), p.54 [20]).
A second issue related to trust concerns conflict and grievance procedures. CLAS and JCR recommend that these procedures should be capable of identifying issues that concern organizations' responsiveness to diversity, and that such conflicts should be dealt with in a respectful manner (CLAS, JCR, and EDS). The issue of access by minority patients to grievance procedures is also discussed (CLAS, JCR), including the need for personnel dealing with complaints and grievances to receive proper training (CLAS).

Patients' rights
CLAS and JCR discuss the importance of informing patients about their rights. This concerns (among other things) the right in the US to receive language assistance (CLAS, JCR) and not to experience discrimination (JCR). JCR and EQS also note the importance of adapting informed consent procedures to the patients' needs (e.g. health literacy level).

Patient and community participation at organizational level
The sixth dimension identified in the approaches concerns the involvement of users and communities in health care at the organizational level. In this domain one dimension appeared: involving patients in the development of services. Patient participation in the care process was subsumed under the previous domain.

Involving patients and communities in the development of services
The issue of participation at the organizational level is discussed by all approaches. The first argument put forward in favor of such participation is that it results in more responsive care (CLAS, JCR). Another advantage named is that patients and communities can contribute to the implementation of changes (EDS). The approaches explicitly (CLAS, CRF, COER, EDS, EQS) or implicitly (JCR) assume that their target populations often belong to disadvantaged groups that may normally be less likely to take part in participation processes. The approaches therefore pay attention to the challenge of creating inclusive participation processes.
Four approaches explicitly mention community as well as patient participation (CLAS, JCR, CRF, EDS; COER speaks of migrant participation). The important difference between patient (or user) and community participation is that the latter brings in the voice of people who did not get into treatment. However, only CLAS and EDS explicitly regard it as important to build partnerships (e.g. with community representatives or organizations) in the community served by the health care organization. Their argument amounts to the following: a health care organization serves a community; therefore the community has to be enabled to exert influence on what happens in the organization through a collaborative process. In the other approaches, patients and community members are regarded as complementing each other (usually in the same sentence), without making clear the additional value of community participation.

Promoting responsiveness
Issues were classified in this domain if they concerned the promotion of responsive health care in the wider society. We identified one dimension, 'sharing information on experiences' in improving care for ethnic minority patients.

Sharing information on experiences
All approaches except CRF mention the importance of sharing experiences in promoting responsiveness. This is proposed with different aims in mind: to increase support for responsive care from the general public (COER), to demonstrate an organization's commitment (CLAS, JCR), or to enable organizations to learn from each other (CLAS, EQS). CRF and EQS take this theme a step further, by proposing that organizations should enter into active partnerships with others that promote equity within the health care system (e.g. in research and other collaborative activities).

Unique issues
Two issues were unique to particular approaches. JCR repeatedly mentioned the identification and addressing of patients' mobility needs (e.g. using a cane, guide dogs). EDS emphasized "supporting the workforce to remain healthy", which is in line with its focus on equality in the workforce.

'Vertical' analysis (comparing approaches)
In the foregoing section we discussed findings in terms of the domains (the rows of the matrix in Table 3). In what follows we analyze differences between the approaches represented in the columns, to obtain insight into the specific nature of each approach.
The linguistic issues in CLAS have a legal basis in the Civil Rights Act of 1964, which requires "entities that receive Federal financial assistance to take steps to ensure that limited English proficient individuals have meaningful access to the health services" [9], p. 10). CLAS has a very explicit vision on responsive health care, for example in the details provided about the content of training or the types of data that should be collected. Although unequal access is mentioned as a problem which might be reduced by providing appropriate services, the CLAS standards discuss only linguistic barriers to access.
JCR follows the steps in the clinical care process, which is explained by its origin in the Joint Commission standards for hospitals. JCR was developed as a supplement to existing standards, so it may have left out some issues already covered by existing Joint Commission standards. The standards do not embody an elaborated vision on 'responsiveness' beyond cultural competence, communication, and patient-and family-centered care. Accessibility is not operationalized in JCR. Interestingly, the international branch of the Joint Commission does consider accessibility as an important aspect of health care, as demonstrated for example by the standard stating that "the organization seeks to reduce physical, language, cultural, and other barriers to access and delivery of services" (p.42 [28]).
CRF focuses, like CLAS, mainly on cultural and linguistic issues. CRF offers quantitative and qualitative indicators (standards and measures) for measuring organizational responsiveness to diversity.
COER states to address governments rather than individual health care organizations, by discussing issues concerning organizational responsiveness to diversity without specifying precisely the division of responsibilities between levels. In keeping with the Council of Europe's historical role, COER is primarily concerned with the ethical and human rights dimensions of social and health issues.
EDS addresses accessibility and quality of care, but its vision of responsive care for 'protected groups' remains rather implicit. The implementation strategy elaborates on steps such as 'engage with local interests' and 'analyze performance'. The content of responsive care is only briefly described in terms of goals such as 'better outcomes for all' and 'improved patient access and experience' (see Table 2). One of the main objectives of EDS was to provide a tool for NHS commissioners to comply with the UK's 'Public Sector Equality Duty' (PSED). This is reflected in the envisaged target groups of EDS and explains the unique focus of EDS on equality among staff, in line with the aim of the PSED to eliminate discrimination and enhance equal opportunities throughout the public sector.
EQS emphasizes the vulnerability of certain patients, which can result from many factors -'culture' being hardly mentioned as one of these. The focus is on patients' individual needs and characteristics, rather than their membership of specific ethnic, cultural or other groups. This approach seems to view 'patient centered care' as the best way to respond to diversity in care provision. Although EQS defines its target group as 'migrants and other vulnerable groups' , most of the standards focus on issues relevant to migrants, which may be explained by the Task Force's origin in the Migrant Friendly Hospital network [25].

Variations in the orientation of different approaches
Only the European approaches address issues of access to health care in the sense of entitlement. A common feature of the non-European approaches is their emphasis on 'culture'. On closer examination, this seems to be mainly a question of how factors are labeled: in CLAS, for example, differences in socioeconomic or legal status are regarded as 'cultural' ones. In the European approaches the focus is on individual characteristics, which brings EQS close to the approach known as 'patient centered care' [29] (Saha, Beach, and Cooper have discussed the relation between 'patient centered' and 'culturally competent' care [30]). A possible shortcoming of this individualistic perspective is that the social position that characterizes members of certain groups (e.g. asylum seekers, irregular migrants) may be overlooked. COER and EQS are the only approaches that explicitly refer to

Discussion
From this content analysis of six approaches to organizational responsiveness to diversity a broad consensus emerged on what health care organizations need to do to meet the differing health care needs of diverse patient populations. The following domains were almost universally regarded as important for creating responsive organizations: organizational commitment, collecting and analyzing data to provide empirical evidence on inequalities and needs, development of a competent and diverse workforce, ensuring access for all users, ensuring responsiveness in care provision, fostering patient and community participation, and actively promoting the ideal of responsiveness (see Tables 3 and 4 for the synthesis of the content of these domains). The described elements for organizations to address might be used as guidance for implementation of diversity responsive health care. The elements show that all stakeholders, including all patients and their communities, should be involved in developing improved services. This consensus opens the way to wider implementation of organizational responsiveness to diversity. Table 4 Description of classified domains and dimensions, and coverage of domains/dimensions by the six approaches (the orange cells visualize table 3's empty cells meaning that this dimension is not covered by that approach) There were, however, also some variations in the operationalization of diversity responsive health care, related to different scopes, contexts and types or definitions of diversity. The non-European approaches focus on improving quality of care and linguistic accessibility in the fight against health disparities, while the issue of 'entitlement to care' seems overlooked. This is remarkable since (lack of ) health insurance coverage explains substantial proportions of disparities in health care in the US [31]. The National Health Interview Study 2012 found that while 7.6 % of white respondents had been uninsured for more than one year, this percentage was 23.6 in the Hispanic population, 11.7 in the black population, and 11.3 in the Asian population [32]. Although entitlement to care is usually regulated by governments rather than service providers, approaches that aim to reduce disparities should at least acknowledge the importance of entitlement.
The consensus we found between these six approaches relates to some emerging debates in the area of diversity responsive care. First, there is the debate on the relative importance of individual patient characteristics versus cultural or group characteristics. Adapting routine health and preventive care services [33] using the concept of 'patient centered care' is a step in the right direction of acknowledging diversity among all patients. However, patient centeredness focuses on acknowledging the uniqueness of patients. This individualistic approach to diversity carries a risk, because the most serious inequities in health care are strongly associated with differences in group membership and social situation. This was illustrated in Philimore's study on maternity services in an era of superdiversity, which showed that immigration status was one of the most important factors in determining problems with maternity services [34]. Being an asylum seeker or undocumented migrant is neither a cultural characteristic nor a personal oneit is a social position, with important consequences for health and access to health care. An exclusive on 'culture' is misleading, and that diversity responsive health care means that cultural aspects, social position and individual needs all have to be taken into account.
A second debate touching the topic of this paper is the debate of discrimination versus diversity responsive health care. The latter explicitly requires care providers to differentiate between patients (if relevant for equal chances of optimal health care outcomes), whereas the non-discrimination principle prescribes not making a distinction between patients (under equal circumstances) [35]. Dovidio et al. report that overt racism among care providers has become rare, but that explicit egalitarian attitudes prevent care providers and organizations to diversify care provision when it is necessary [36]. The fear of prejudice and stereotyping may thus lead to suboptimal care. The topic of discrimination was only briefly touched upon in most of the six analyzed approaches.
We believe that future approaches should elaborate more on this topic so that health care organizations will have more guidance in dealing with this dilemma.
Our analysis had its limitations. The sample of approaches was confined to the US, Australia and Europe. We explicitly did not aim at a complete review by sampling and comparing all existing approaches, because too many have been developed to make that practical. It is possible that valuable approaches were missed. However, the six approaches included showed considerable consensus regarding the important elements of care that is responsive to diversity, and data were saturated. Because of variations in the level of detail provided by the approaches, classifying them may not fully do justice to the visions or ideas behind the approaches. For example, EDS contains only a rudimentary conceptualization of responsiveness, but it may be that within the NHS other documents elaborate this concept in more detail.

Conclusion
This study demonstrated that a considerable degree of consensus exists about the issues that health care organizations need to address in order to adequately respond to diversity. This creates the opportunity to move forward, to resolve purely terminological differences and to help health care organizations to respond to the diversity that is present in modern societies.