Description of included studies

The initial search identified 6483 studies after duplicates were removed. Screening of the titles and abstracts left 122 studies which met the criteria for full-text review, 26 were then selected for inclusion.16–41 To update the review, the search was rerun in June 2017; one new paper was found to be eligible for inclusion, bringing the total number of included studies to 27.42 Figure 1 shows the flow of study selection and numbers of excluded studies with reasons.43

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Flow of included studies.

Table 1 shows a summary of the findings for all 27 included studies. For the purpose of narrative review, we grouped the studies into six categories: prevalence and characteristics; diagnosis validation and screening; knowledge, understanding and attitudes; help-seeking; experience of dementia; service organisation and delivery.

Prevalence and characteristics

We identified five articles that had looked at either the prevalence or characteristics of dementia in the UK South Asians.18 21 31 39 41

Bhatnagar and Frank looked at the prevalence of dementia in a community sample of people from the Indian subcontinent living in Bradford, UK.18 Diagnostic interviews were conducted with 100 participants and the prevalence of dementia was reported as 4%. However, when tested using a Hindi translation of the Geriatric Mental State (GMS-A), which found an initial prevalence of 7%, the reliability of the diagnosis was poor. The authors suggest that this is due to a lack of cultural adaptation of the measure. For example, some interviewees would have little understanding of western calendar months, a knowledge of which forms part of the screening measure.18

A cross-sectional study conducted in inner-city Liverpool also used the GMS interview to assess the prevalence of dementia in English-speaking versus non-English-speaking Asian participants.41 The authors do not use the term South Asian but rather Asian. However, we have included this paper as it is clear from the categorisation of other ethnic groups included in the study that Asian refers to participants from South Asia who make up 2% of the total study sample. The authors interviewed 12 Asian participants using the Geriatric Mental State Examination and found a prevalence of dementia of 9% among the English-speaking sample.

Two articles reported characteristics of ethnic elders from the Indian subcontinent living in west London and attending a psychogeriatric service.31 39 Both published reports were from the same sample population, the first consisting of cross-sectional data from a census conducted in August 1995,39 and the second a 2-year study of all new cases entering the service between August 1995 and July 1997.31 Sample sizes were small with South Asians making up only 12%31 and 17%39 of the total number of recruited participants. Dementia was diagnosed by physicians and eligible patients were identified by case note review. Redelinghuys and Shah39 found that only a small subsample of participants had dementia, with no significant difference found between those from the Indian subcontinent (n=6, 15%) and white British elders (n=43, 22%). However, Odutoye and Shah31 reported that over that time period there was a significant difference in the incidence of dementia with elders from the Indian subcontinent being less likely to have a dementia.

Haider and Shah conducted a study of the behavioural and psychological signs and symptoms of dementia using the Behavioural Pathology in Alzheimer’s Disease Rating Scale.21 44 They compared a sample of physician-diagnosed patients from the Indian subcontinent with a white British sample living in West London who had been referred to a day hospital for treatment. They did not find any major differences between the two groups.

In summary, studies have attempted to identify the prevalence of dementia in South Asians living in the UK. Most have reported that there is no difference in the prevalence between South Asian and white British participants. However, sample sizes in these studies have been unanimously small and the studies were conducted  over a period of 14 years, therefore it is difficult to draw conclusions about the current definitive prevalence. Several studies do not report sufficient detail on the methods and instruments by which dementia was diagnosed and if there was any translation or cultural adaptation.21 31 39 Bhatnagar and Frank noted that there is a lack of cultural adaptation of dementia measures for South Asians, which may result in high false-positive and false-negative diagnoses, which may result in unreliable prevalence rates.18

Diagnosis, validation and screening

Three papers were identified that discussed either the diagnosis of dementia in the UK South Asians or validation of screening tools. Two studies were identified that had looked to validate the Mini- Mental State Examination (MMSE) dementia screening instrument for use in the UK South Asian population. Lindesay et al investigated whether a Gujarati version of the MMSE could be used as a screening instrument in a Gujarati population in Leicester.28 The authors reported that they followed the precedent of Ganguli et al in developing this Gujarati version of the MMSE.45 They initially selected the items using consensus, translation and back translation, pretesting and validation against psychiatric diagnosis.28 Lindesay et al reported that the MMSE performed well in the identification of moderate-to-severe dementia but was less effective in detecting cases of mild dementia in the Gujarati group when compared with a group of white British participants. In a second study, using the same sample as Lindesay et al,28 Shah et al investigated the stability of dementia diagnosis over time.35 They reported that the diagnosis of dementia was stable at follow-up over a period of >2 years (26–32 months), which indicates that the measure Gujarati MMSE has test–retest reliability in this population.

Rait et al tested the MMSE in a population of South Asians, including Gujarati and Pakistani participants.33 The authors state that, as there were no definitive guidelines available for translating questionnaires such as the MMSE, they engaged both academics and members of the South Asian community to assess the cultural sensitivity of the items in three groups. The items were translated by a professional translation group and then back-translated with the South Asian community group. They reported that modifications were made to the MMSE, which centred around education and literacy and the modified MMSE was tested in a pilot study.46 They found high levels of sensitivity and specificity for the MMSE, but with a lower cut-off score for the identification of mild dementia in the South Asian group.

Findings indicate that the Gujarati version of the MMSE may be effective at identifying dementia, especially where the diagnosis is certain and therefore symptoms are moderate to severe. However, it may be less effective in identifying dementia where symptoms are mild or diagnosis is uncertain.

Knowledge, understanding and attitudes

We identified six studies that had set out to identify knowledge, understanding and attitudes to dementia in the UK South Asian community. Knowledge about dementia in Indian (n=91) and white UK/Irish/European (n=55) people was assessed using the Dementia Knowledge Questionnaire (DKQ) by Purandare et al.32 They found that both groups had poor knowledge of causes and symptoms, and Indian older people had significantly less ‘basic knowledge’ about dementia. Here, the authors define ‘basic knowledge' as including knowledge of epidemiology, aetiology and symptomatology as measured by the DKQ.47 A further series of focus groups investigated general population knowledge of dementia with 28 Sikh participants.37 Again they found poor knowledge of dementia and that Sikh participants placed greater emphasis on physical illnesses than disorders such as dementia. La Fontaine et al held focus groups with a population sample of 49 Punjabi Indian participants.25 One of their key findings was the failure of participants to directly refer to dementia or even to use lay terms to discuss it, which the authors conclude indicates a lack of recognition of dementia as a concept.

Turner et al explored knowledge and attitudes to dementia in a qualitative community study of South Asian (n=96) and white older people (n=96) and also found reduced knowledge of dementia in the South Asian group. Furthermore, their findings highlighted differences in attitudes to dementia care, with the South Asian group reporting that care should be provided by family and friends.38 In interviews with 10 South Asian family carers, Lawrence et al looked at traditional versus non-traditional caregiver ideology.26 The majority of South Asian carers possessed a traditional ideology in that they saw care giving as natural, expected and virtuous. However, where understanding of dementia is poor problems can occur within the family caring relationship due to beliefs about the causes of dementia in the South Asian community and the attribution of blame on to the person with dementia.16

Adamson interviewed 15 South Asian carers living in five cities across the UK.16 The study looked at the relationship between individual experiences, cultural factors and social structures within this minority population. Carers tended to interpret their caring roles as an expected part of their cultural heritage and a continuation of their family relationships. They also likened the experiences of this group to other chronic illness care and suggest that the understanding of the experiences of chronic care may be useful in understanding experiences in informal care for dementia.

In summary, several studies have explored the knowledge of dementia, and beliefs and attitudes to care in the UK South Asians. All report that knowledge is poor and this group is often more focused on physical illness rather than conditions such as dementia. This finding that this group does not identify with dementia as a concept further highlights the need for valid and reliable measures to be developed to identify and then diagnose dementia in this group.

Our findings show that care is preferred to be provided by family and within the community and therefore, the UK South Asians may not access formal healthcare for dementia.

Giebel et al have adapted the Barts Explanatory Model Inventory Checklist (BEMI-C) for use with South Asian ethnic minority groups (BEMI-Dementia (BEMI-D)).42 They conducted 25 qualitative interviews with South Asians and identified 123 new perceptions around their understanding of dementia. These were added to the BEMI-C to create a new checklist (BEMI-D) to better assess the barriers to dementia service uptake for this group in the future.

Help-seeking

Five papers were identified that focused on different aspects of help-seeking for dementia in the UK South Asian population.

Two studies had looked at barriers to help-seeking for dementia. First, in a community sample of English-speaking or Bengali-speaking UK South Asians who did not have a diagnosis of dementia, Mukadam et al identified the barriers and facilitators to help-seeking for memory problems.30 They identified four main categories of barriers, which interacted to prevent timely diagnosis of dementia: barriers to help-seeking for memory problems; the threshold for seeking help for memory problems; ways to overcome barriers to help-seeking; what features an educational resource should have. Second, Hailstone et al devised and validated a theory of planned behaviour (Attitudes of People from Ethnic Minorities for Help-seeking for Dementia) questionnaire to predict medical help-seeking for dementia in the UK South Asians.22 They found that attitudes to dementia predicted 77% of variance in help-seeking with the strongest predictor being perceived social pressure.

Furthermore, there were two studies which had specifically identified religious explanations as barriers to help-seeking for dementia in South Asians. Mackenzie interviewed 11 Pakistani and 5 Indian carers and found that stigma resulted from religious and magical beliefs around the causes of dementia and resulted in concealment from their own community and delays in help-seeking.29 Giebel et al looked at the differences in the perceptions of South Asians who do and do not consult a general practitioner (GP) about dementia.20 They found those who did not consult a GP were significantly more likely to consider memory problems as given by God, with the view that medical intervention was therefore inappropriate. In a case study of a Muslim, Pakistani patient accessing healthcare for dementia in the UK, Regan further highlighted the importance of understanding a person’s religious community and its role in providing both support and reducing stigma and isolation.40

The included studies highlight that attitudes and beliefs about dementia can serve as barriers to accessing healthcare, in particular understanding dementia within a religious context can delay help-seeking. However, religious communities can also play a vital role in supporting patients with dementia and in reducing stigma. This may be the case across other ethnic minority communities in the UK, not just South Asians. It is therefore important to learn more about the religious context in which people understand dementia and for healthcare providers to engage with local communities and religious leaders in order to work in partnership with them to increase support and reduce stigma.

Experiences of dementia

We identified three articles that looked at understanding the experience of dementia for the UK South Asians. Bowes and Wilkinson explored the experiences of South Asian patients with dementia in Scotland.19 Interviews were conducted with 11 health professionals, and 4 case studies were built up through multiple interviews with four people with dementia, as well as interviews with their family and carers. The authors reported overwhelmingly negative experiences of dementia and of health services from the case studies. From the interviews with health professionals, the authors concluded that there was a need to develop and promote culturally sensitive services. Furthermore, the article by Jutlla looked at the experiences of 12 Sikh carers, caring for a family member with dementia, in Wolverhampton.23 In particular, they looked at the influences of migration experiences and migration identities. They found that knowledge of a person’s background and experiences is important for understanding how they then experience health services and caring roles. This is presented as an advocate for culturally appropriate services and for person-centred dementia care.

Lawrence et al looked at experiences of dementia in three ethnic groups, 11 black Caribbean, 9 South Asian and 10 white British older people with dementia.27 The paper reported a comparison in the personal experiences of the condition in the three groups. They reported similar themes across groups with a key finding being that patients assessed their condition by the degree to which it interfered with ‘valued elements of life’. The authors concluded that development of culturally sensitive approaches to care should promote roles, relationships and activities that the patient values.

Service organisation and delivery

We found three articles that discussed service organisation and delivery for South Asian patients with dementia in the UK. First, in 1999 Shah, a Gujarati Psychogeriatrician, published a descriptive account of his own experiences working with 12 Gujarati patients in their language.36 Four of the patients had a diagnosis of dementia. He identified particular problems in eliciting cognitive symptoms due to problems with the translation and the lack of culturally appropriate assessments.36 Second, Kaur et al wrote about the role an Asian link nurse for Punjabi-speaking people of Asian origin in a dementia service in Wolverhampton.24 The authors reported success in providing appropriate and culturally sensitive help and information for healthcare professionals, voluntary services and South Asian patients with dementia. Third, Seabrooke and Milne discussed the Dementia Collaborative Project in North West Kent, which aimed to raise awareness of memory problems in South Asians and facilitate access to screening and diagnosis.34 The authors reported an increase in referrals to a specialist clinic for memory assessments, some of which were for South Asian patients, and an increase in health professional’s knowledge of memory problems in South Asians.