Basket | Login


Culturally sensitive assessment and diagnostic tools

Diagnosis, assessment and treatment.

Challenges surrounding the development of culturally sensitive assessment and diagnostic tools

Language, literacy and lack of culturally appropriate diagnostic tools

Although biomarkers are increasingly playing a role in the diagnostic work up of AD dementia, dementia is still very much a clinical diagnosis which relies on verbal interaction, observation of behaviour and cognitive assessments (Fields, Ferman, Boeve and Smith 2011).  Language difficulties and illiteracy may hinder the timely and accurate assessment and diagnosis of people from some minority ethnic groups. Although literacy is usually associated with schooling, the term is not dependent on a person having received formal education. A commonly cited definition is that provided by UNESCO in 1958 which was that “literate person is one who can, with understanding, both read and write a short simple statement on his or her everyday life.” In 2006, in an independent report commissioned by UNESCO, it was emphasised that literacy skills are independent of the context and background of an individual[1].

A high proportion of Arabic-speaking older people in the Eastern Mediterranean Region are illiterate (Chaaya et al. 2016), including many older people from Morocco, with people of Moroccan origin constituting the largest group of non-European migrants in Flanders (Berdai Chaouni and De Donder 2018). In the South Asian community in the UK, only 35% of older people over 65 years of age can speak English and only 21% can read and write English, often communicating in their first language, which for many is Urdu (Blakemore et al. 2018). A 2011 census of England and Wales further revealed that only 40% of people from traveller communities had any formal qualifications. In a study involving Turkish immigrants in Denmark, Nielsen and Jørgensen (2013) demonstrated how illiteracy can affect a person’s performance on cognitive tests. Considerable differences were found between people who were illiterate compared to those who were literature. Appendix 3 (p.101) provides examples of typical mistakes that people make when asked to copy four different line drawings of shapes and in filling in the numbers and hands on the drawing of a clock. When comparing the drawings, please note that none of the participants in the study had dementia. The drawings of the illiterate participants show how being illiterate might easily contribute towards the misdiagnosis of dementia by affecting scores on certain diagnostic tests.

In countries with a long history of migration (e.g. due to colonisation), people from minority ethnic groups tend to have fewer difficulties understanding the language and culture than in countries where migration is more recent (Nielsen et al. 2011a). Despite considerable linguistic and cultural diversity amongst members of minority ethnic groups in Europe, most countries in a cross-country comparison carried out by Nielsen et al. (2011a) used cognitive assessment instruments with people from minority ethnic groups which have only been validated in populations from the Western culture and relied on relatives for interpretation (as well as professional interpreters). The lack of access to culturally sensitive screening and diagnostic tools is a major challenge facing GPs and specialists, resulting not only in under-diagnosis of dementia in minority ethnic groups but also in over-diagnosis (Nielsen, Vogel, Phung, Gade and Waldemar 2011b).


Whereas some people from majority ethnic groups, especially of the older generations, may have had little or no formal education, this is more common amongst some minority ethnic groups. Standard screening tests used for screening and diagnosing dementia have repeatedly been found to be biased with regard to education (Goudsmit et al. 2018). Education often involves learning things and doing tasks which are not necessarily considered important or relevant at the time but people (usually children) have little choice and become accustomed to doing such tasks. Lack of formal education (such as schooling) may result in people finding seemingly simple tasks, such as drawing a clock face or a cube, difficult. They might never have drawn geometric shapes or done much drawing in general. People get used to using and holding pens and pencils at school and continue to use pens and pencils to write and draw in their everyday lives. Good performance on some drawing or copying tests, for example, requires fine motor finger movements as well as visuospatial skills, and tests which involve visually guided motor behaviour (e.g. moving a cursor on a computer screen) may be biased by literacy (Ardila et al. 2010). People who are illiterate lack this kind of familiarity with pencils and pens and this may affect their performance on tests (Soumaya 2018). In addition, people with limited or no schooling may feel under stress and be totally unfamiliar with the concept of cognitive testing.  According to Ardila et al. (2010), as little as 1 to 2 years’ schooling can make a significant difference in performance on some neuropsychological tests.

People with no or low levels of education also tend to perform worse on tests involving naming, comprehension, verbal abstraction, figure recognition and orientation (Goudsmit et al. 2018).  They also tend to be less ‘test-wise’, to be unfamiliar with the formal procedures associated with testing and sometimes to find line drawings difficult to interpret (Ardila et al. 2010, Goudsmit et al. 2018). It should also be noted that the content of formal education may be radically different in countries from which some people from minority ethnic groups migrated.

Education not only provides knowledge but also contributes towards the development of certain abilities and attitudes (Ardila 2003). According to Carreiras et al. (2009) and Castro-Caldas et al. (1998 – in Nielsen and Jørgensen 2013), lack of schooling as a child, resulting in people not learning to read and write, affects the functional and structural development of the brain and this is not limited to the language domain of the brain. Nielsen and Jørgensen (2013) also point out that being unable to read and write is often accompanied by a whole range of factors, especially linked to low socio-economic status of the person’s family, which may have an impact on cognitive function in later life.

Cultural norms

The concept of cognitive testing was developed in Western society and represents a culture-dependent activity affected by different cultural conditions and idiosyncrasies (Ardila 2003 and 2005). Ardila (2005) describes eight approaches or strategies typically used in cognitive testing which may be unfamiliar to people from some ethnic backgrounds and even violate their cultural norms. These are: 

  1. a one-to-one relationship (in communities where tasks are carried out in a collaborative way in a social context, a one-to-one, one-off relationship between two strangers involving one performing a task may run counter to cultural norms),
  2. background authority (the authority a person is perceived to have which allows him/her to make certain requests such as for the person to do various tasks. It might not be obvious why he or she has such authority and why the other person should comply with requests),
  3. best performance (doing one’s best may be well understood in competitive societies but in others restricted to important tasks. Some tests may seem absurd and pointless, resulting in the person not making a sufficient effort to ‘do well’),
  4. isolated environment (the person may be in a room, perhaps even locked, with a person he/she does not know, and this may be inappropriate in some cultures, especially for women),
  5. special type of communication (possible lack of familiarity with formal language use in the testing situation),
  6. speed (doing things quickly is not necessarily perceived by the people being tested as being better than taking longer and doing a better job),
  7. internal or subjective issues (some questions may be perceived as inappropriate and invasive),
  8. the use of testing elements or strategies (pictures, images and objects used in tests may be culture-specific).

People from minority ethnic groups may therefore sometimes find cognitive testing difficult and under-perform on tests that are not culturally sensitive. This may contribute towards misdiagnosis, unnecessary further examinations and inappropriate medical treatment.

Translation issues

Translating instruments for the assessment and diagnosis of dementia into the languages understood by members of minority ethnic groups may go some way towards more accurate and timely diagnosis of dementia amongst those groups. The points raised by Ardila (see above) suggest that this would not be sufficient on its own because there are also other cultural issues to consider. However, until more culturally appropriate instruments are developed, it is worthwhile looking at what the issues are linked to the translation of screening and diagnostic tools.

Translated versions of the MMSE exist in several languages from around the World. However, there is a shortage of translated version of the MMSE that have been validated on relevant minority ethnic groups in Europe. It may not be sufficient to translate an assessment or diagnostic instrument. It should also be validated on the group with whom it will eventually be used and there may be differences in the way a language is used by minority ethnic groups in different settings or countries and of different ages. There are hundreds of different languages spoken by people from different minority ethnic groups, in some cases, several within the same group. As highlighted by Parker and Philp (2004), some people cannot read or write in their mother tongue.

“Many Gujarati women from rural areas of the sub-continent are unable to read or write in their own language. Indian Punjabi speakers comprise Sikhs from Punjab state, and Hindus from Punjab or East Africa. Pakistanis from provincial districts generally speak a Punjabi dialect and may be unable to read or write, while those from the cities speak, read and write Urdu. Immigration from Bangladesh has been more recent: the most common language spoken is Bengali (also known as Bangla or Bangladeshi), with some speaking Urdu” (Parker and Philp 2004, p.447-448).

A major difficulty with some screening and even diagnostic tools for dementia, which might otherwise be suited to people from minority ethnic groups, is that they can only be used by clinicians who are bilingual or with the help of interpreters (Ahmed et al. 2017). Another issue which can be problematic is that healthcare professionals sometimes have difficulty obtaining consent for various investigations (e.g. blood tests and lumbar punctures) which may be needed for the diagnostic procedure (Nielsen et al. 2011a).

Interpretation issues

Assessment and diagnosis may be improved in some cases if good quality interpretation is provided. In some countries or healthcare authorities, people who are not fluent in the language of a country have the right to request an interpreter (Plejert et al. 2015, Majlesi and Plejert 2018). However, interpreters are not always available when needed (e.g. due to a shortage of suitably qualified interpreters in a particular language or dialect, practical issues linked to accessing one when needed and lack of funding for their services). Relatives are often asked or expected to fulfil the role of interpreter and in many cases, given the choice, some members of the family, would prefer a relative to interpret.

Relatives/informal interpreters

There are pros and cons to the use of relatives as interpreters. On the positive side, relatives are often familiar with the person’s situation and already know what the person is trying to communicate, they are in many cases trusted (i.e. they will not reveal what the person says outside of the family) and in some cases may be familiar with a particular language that is not spoken by many people and with a dialect that the person speaks. Moreover, at a recent meeting of the Race Equality Foundation (in Bristol in 2018), a healthcare professional from the African-Caribbean community drew attention to the changing nature of some languages whereby the language or dialect spoken by an older person having been born in a different country may be radically different (different words, different ways of using language to express a point etc.) to that currently used (e.g. by second or third generation members of the community). This issue is also relevant to the use of professional interpreters. However, there may be conflicts and family interests which impact on the type of information and details that the relative is willing to convey, resulting in censorship. Also, some interpreters might not translate a particular comment because they do not think that it is medically relevant (Bolden 2000). Acting as an interpreter may also have an emotional impact on relatives in that it may result in them putting their energy into providing interpretation when they actually need to focus on coping with their own emotions.

Involving relatives in interpretation also means overriding the general principles of respect for privacy and confidentiality. Younger, second or third generation members of the family are more likely to have a relatively good command of the national language and may therefore be expected to fulfil the role of interpreter. Depending on their age and role within the family, they may sometimes have little say in the matter. The use of informal/family interpreters may also mean that the person with possible dementia might be uneasy talking about certain issues which could call into question their standing within the family or lead to personal embarrassment (e.g. having a grandchild interpret information about intimate matters). 

Finally, interpretation is a complex skill which requires specialised training. Being fluent in more than one language does not make someone an interpreter. People who are bilingual but have not received interpreter-specific training are more likely to add or omit information in exchanges with the user and have a tendency to interject their own opinions and assumptions, resulting in the transmission of incomplete and/or inaccurate information (NHANES 2006). When using relatives for the purpose of translation, healthcare professionals have no guarantee with regard to the accuracy of the information the latter conveys to the person with possible dementia and to healthcare professionals. Relatives should not be used as interpreters simply to reduce costs or because it is quicker and easier to ask a relative than to find a professional interpreter. This does not mean that they should be totally excluded during the process of obtaining a diagnosis (unless the person seeking diagnosis is opposed to their presence) as they may be able to convey useful information about their relative’s difficulties and behaviour.

Professional/trained interpreters

One of the reasons for using relatives as ad hoc interpreters is that it is difficult to find professional, trained interpreters in the relevant languages, dialects and varieties[2], and at the exact time that they are needed. In some countries in Europe certain dialects, with limited numbers of older speakers, are gradually dying out, but in most European and non-European countries many dialects and varieties are spoken. In such cases, it is essential that interpreters speak and understand the actual dialect that a person uses. In addition, some people from minority ethnic groups would prefer to have an interpreter of their own age and background (Mohammed 2017). Due to a general shortage of suitable and qualified interpreters, and in some cases the limited number of interpreters who are fluent in a particular dialect, there is a risk of professional interpreters being members of the same minority ethnic community and of people in need of their services not wanting to have them as interpreters or to disclose certain details to them.

Plejert (2018) has drawn attention to the increasing practice in some countries of fast-track/crash courses for interpreters and of under-estimating the practical and professional skills needed for effective interpretation. Even with professional, well-qualified interpreters, their use in the context of the assessment and diagnosis of dementia raises certain challenges. In many cases, interpreters are not sufficiently knowledgeable about dementia and cognitive testing (Plejert et al. 2015). Similarly, many healthcare professionals lack experience working with interpreters. In the context of memory assessments of people from the Somali community in Finland, Jaakson (2018) points out that it has been necessary to remind interpreters that it is not their role to help the person do the test but just to translate the instructions and what was said. Sometimes, however, a certain degree of flexibility and ad hoc adaptation from interpreters may be helpful.  In clinical practice, when tests of cognitive functioning are not culturally appropriate, clinicians often find that they need to skip parts of the test or use interpreters to find a solution, sometimes replacing an element of the test or adapting itin situto make it more suitable for a person from a minority ethnic group (Plejert, Antelius, Yazdanpanah and Nielsen 2015). Such approaches are adopted on a daily basis because most of the tests being used are not optimal for practitioners, patients or interpreters.

Majlesi and Plejert (2018) provide a detailed case study of the interaction between a professional interpreter, a Kurdish-speaking elderly lady and a Swedish-speaking occupational therapist in the context of cognitive testing. They conclude that the completion of cognitive testing was the outcome of a collaborative approach in which the interpreter not only provided oral translation but co-facilitated the task by responding to posture and body language, prompting, rephrasing/reformulating the instructions to facilitate understanding, simplifying responses and directly answering requests from the person without going through the tester etc.[3] In this way, the interpreter played a role of monitoring and walking the person through the test, which could be seen as co-production between the interpreter, the person taking the test and the tester. Majlesi and Plejert (2018) emphasise the need to pay attention to such issues which may affect the reliability of interpreter-mediated tests and their implications for the diagnosis of dementia in people from minority ethnic groups. Plejert et al. (2015) describe a dialogical approach[4] whereby the interpreter is considered as an active agent, engaged in an interplay with and between those involved and the environment (a visible co-constructor of an interactional event), in contrast to a more traditional approach whereby the interpreter is considered as merely transmitting a message in a neutral manner (i.e. without influencing in any way its content or formulation). They point out that the perception of interpreters as co-participants in cognitive testing encounters is perhaps less widespread amongst clinicians and conclude that,

“when it comes to the training of clinicians in working with interpreters, it would be beneficial if the ‘myth’ of the interpreter as a neutral message transmitter could be broken. That would hopefully help clinicians understand and be prepared for practices that go beyond the interpreters’ pure renditions of what has been said only. It might also encourage clinicians to dare to interfere and ask about information that is being negotiated, but not rendered”(Plejert et al. 2015, p.183).

In sum, interpreter-mediated dementia assessments pose many challenges that may be related to the lack of appropriate tests for particular patients, and to interpreters frequently being asked to translate tests in situ, which may be necessary, but is in conflict with the interpreters’ professional codes of conduct. Clinicians may also sometimes put too much trust in the interpreters’ abilities, and therefore refrain from interfering, even in cases when the interpreter apparently does not report sequences of talk between the patient and themselves. Many interpreters are very skilled at what they do, but the term “professional” is sometimes complicated, since it may be associated simply with authorisation and medical specialisation, with no guarantee of the quality of the service provided. In addition, the interpreter may in some cases primarily have been trained to interpret medical interviews but not, specifically, formal tasks. Furthermore, they may not have been trained to work with people with cognitive or mental health conditions. Some fairly simple measures can, however, improve the situation for interpreters as well as for clinicians, and increase the security of interpreter-mediated medical encounters.

[1] For further information about literacy and illiteracy, including definitions from several countries, please see Chapter 6 of the 2006 report commissioned by UNESCO:

[2] There are numerous varieties of a given language. A dialect is often considered as a variety of a language connected to a specific geographical area, with features on all levels of language, including lexical, phonological, and grammatical ones. It is generally (but not necessarily) mutually intelligible by speakers of a language. Other varieties include style (level of formality) and register (adaptations to particular contexts and participants). What counts as a dialect is also heavily dependent upon cultural and political issues (Schilling-Estes 2006). 

[3] In another study reported by Plejert, Lindholm and Schrauf (2017), an interpreter changed an open question to a closed question with 4 responses to choose from (which changes the nature of the test and makes it somewhat easier).

[4] Please see Wadensjö’s (1998) seminal work on this perspective.

Key points

  • The lack of access to culturally sensitive assessment and diagnostic tools for dementia is a major challenge facing GPs and specialists in Europe.
  • Some tools have been translated into a range of languages used by minority ethnic groups but this alone does not make them culturally appropriate.
  • Most assessment and diagnostic tools for dementia, which are used for people from minority ethnic groups, have not been validated for use with those groups.
  • Language is not the only factor affecting the appropriate use of assessment and diagnostic tools. Other issues, such as cultural and educational bias, need to be considered as well as literacy levels.
  • Education is not just about acquiring knowledge and facts. It is also about learning attitudes, becoming ‘test-wise’, complying with requests to do tasks and becoming familiar with writing tools.
  • Lack of schooling as a child and illiteracy may have an impact on the development of the brain.
  • The concept of cognitive testing was developed in and reflects the values of Western society. This bias is reflected in many tests and puts people who were not brought up Western countries at a disadvantage.
  • It would not be sufficient to translate every test into every existing language. Apart from the sheer number of languages used in some countries by people from minority ethnic groups, as mentioned above, language is not the only obstacle.
  • Interpretation can be useful when conducting assessments with people who are not proficient in the language of the country in which they live. In some countries, people have a right to request an interpreter if they feel that they will not be sufficiently able to speak or understand the language used during the clinical encounter.
  • In practice, it is not always possible to find a suitable interpreter and sometimes there may be linguistic or cultural issues (e.g. finding an interpreter who speaks a specific dialect or an interpreter of the same sex etc.), which make this more difficult.
  • Relatives are frequently asked to act as interpreter. This is not ideal, either for the relatives concerned or for the people being assessed.
  • Healthcare professionals who work with interpreters for the purpose of assessment and diagnosis do not always fully understand the difficulties involved in interpretation, such as differences between languages and dialects, and the fact that words and expressions in one language are not easily translated into another language, or the goals of interpretation. This may affect their assessment of a person’s performance on tests.
  • Inadequately trained interpreters may inadvertently influence a person’s performance on tests.


  • Every person should have the right to be assessed and diagnosed in his/her best language.
  • Tools used for the assessment and diagnosis of people from minority ethnic groups should be culturally sensitive and have been validated for use with those groups.
  • More research is needed to develop culturally appropriate assessment and diagnostic tools for dementia.
  • To ensure effective assessment and diagnosis, people from minority ethnic groups should be informed of their right to have an interpreter and access to an interpreter should be free of charge.
  • Relatives should not be asked to act as interpreters. They should, however, be consulted during the assessment process if required and subject to the agreement of the person being assessed.
  • More interpreters are needed in different languages and dialects, who are authorised and have a medical specialisation, including training in person-centred language, in different activities and formal clinical tasks, and with people with different conditions such as dementia. Training in this respect varies greatly across Europe.
  • Clear standards for the training, accreditation and operation of professional interpreters should be developed and implemented across Europe.
  • Clinicians and interpreters should meet before an assessment in order to ensure that the interpreter is familiar with the instruments to be used and is informed about certain aspects or regulations with regard to why and how the test is to be carried out.
  • Skilled interpreters should, if possible, be re-appointed for the same patients over time.
  • Many clinicians are trained in working with interpreters but may sometimes rely too much on interpreters. Clinicians should be encouraged to make sure that they maintain the medical and overall responsibility for the task for which interpretation is needed. This would also facilitate the work of the interpreters.



Last Updated: Tuesday 09 April 2019


  • Acknowledgements

    This report received funding under an operating grant from the European Union’s Health Programme (2014-2020) and from the Robert Bosch Stiftung. The content of the report represents the views of the author only and is his/her sole responsibility; it cannot be considered to reflect the views of the European Commission and/or the Consumers, Health, Agriculture and Food Executive Agency or any other body of the European Union. The European Commission and the Agency do not accept any responsibility for use that may be made of the information it contains
  • European Union