♥

♥

The Impact of Faith Identity on the health and healthcare of Pakistani Muslims

Ghazala Mir

Centre for Research in Primary Care, University of Leeds

http://www.surrey.ac.uk/Arts/CRONEM/Conference-2005-papers/Ghazala%20Mir.doc

Abstract

Two historical frameworks dominate European discourse about Muslim identity.  First, the Enlightenment notion that religion is a private matter to be disassociated from public life, particularly from the scientific enterprise.  Secondly, the Orientalist tradition of portraying Islam as inferior to Western culture and Muslims as people to be feared and controlled.  These discursive practices have consequences for the everyday lives of Pakistani Muslims in the UK and for their healthcare and health.

The incidence of chronic conditions such as diabetes and coronary heart disease is far higher in Pakistani communities than most other ethnic groups.  The higher prevalence of poor health, and disadvantage in areas affecting health, is mirrored in Muslim communities, of which Pakistani populations form a part. This paper reports the findings of a study to explore the significance of Muslim identity in relation to the healthcare and health of Pakistani people.

The study employed an ethnographic methodology to take account of context and other aspects of identity such as social class, ethnicity, gender and age. Findings show that dominant conceptualisations of religion and of Islam corrupt the communication process between Pakistani people and health practitioners;  discussion of religious influences on self-care is avoided by patients and practitioners alike and Pakistani people are exposed to stereotypical ideas about their beliefs and practices. Consequently, they receive inadequate support in decision-making about chronic illness management and are more likely to develop complications. This disadvantage is exacerbated by ethnicity and gender.

These dynamics of healthcare reflect discrimination that is mirrored in almost all contexts in the wider UK society, including education, employment and civic participation.  These areas affect health status, as does self-perception of social position and social relations.  The disadvantage to which Muslim identity appears to expose individuals and groups suggests a possible explanation for higher levels of mortality and morbidity within this community compared to other minority ethnic communities. 

The implications of these findings for practice, policy, research and activism are explored.  Developing shared understanding and common ground with Muslim perspectives is highlighted as a necessary focus for policy and practice development. Policy support for Muslims to organise on the basis of faith identity is also indicated if health inequalities within the Pakistani Muslim community are to be effectively addressed.

Introduction

The UK context

There is evidence of a longstanding tradition in Western models of healthcare of separating religion from the illness experience.  This has had the effect of marginalizing and stigmatising spiritual beliefs, so that patients may only reveal those parts of their experience that fit the normative expectations of clinicians (Walsh 1998).  At the same time, research into chronic illness conditions rarely focuses attention on such dynamics or explores how religious motivations affect an individual's attitudes or needs (Selway and Ashman 1998).

The practice of excluding religion from the realms of scientific inquiry has been linked historically to the ‘Enlightenment’ of seventeenth and eighteenth-century Europe as part of the process through which the power of the Church over the state was gradually eroded:

‘the constitution of the modern state required the forcible redefinition of religion as belief, and of religious belief, sentiment, and identity as personal matters that belong to the newly emerging space of private (as opposed to public) life ……it has become commonplace among historians of modern Europe to say that religion was gradually compelled to concede the domain of public power to the constitutional state, and of public truth to natural science…    This construction of religion ensures that it is part of what is inessential to our common politics, economy, science, and morality.’

(Asad 1993 pp 205, 207)

The development of scientific inquiry in Europe, of which sociological studies are a part, were thus shaped and developed by European culture, ideology and politics. Exploration of the impact of religion on the health of Pakistani Muslims has been limited by this context and the available literature is consequently sparse.

Studies on health and healthcare within Pakistani communities in the UK have most often adopted the theoretical framework of ethnicity as a basis for research, whilst excluding or marginalizing consideration of the religious identity of respondents.  Within qualitative studies of ethnicity and health, group identities based on shared physical features, ancestry, geographical origins and social and political heritage have been used to define ethnic communities (Bhopal 1997).  Such studies have been instrumental in highlighting the experiences of service provision and the healthcare needs that exist within minority ethnic groups (ibid).  

Despite the widespread use of this framework, however, the use of ethnic groupings has been criticised by a number of writers.  Specific ethnic categories may not correlate with terms used by social groups for self-identification (Ranking and Bhopal 1999). As a concept, ethnicity is recognised as underdeveloped in its ability to recognise diversity (Ahmad 1999) and studies present much data but little robust theory (Hillier 2001). Conceptions of ethnicity as fluid and ‘continually changed across contexts and over time’ (Karlsen and Nazroo 2002) may add to the confusion by attempting to consider fixed and changing aspects of identity, such as geographical origin and religion, within the same concept. 

Attempts have been made to include religion as a subset of ethnicity (Modood 1997; Nazroo 1997; Aspinall and Jackson 2004; Audit Commission 2004), however, the relationship between ethnicity and religion is not straightforward. Ethnic group categories (such as those used in the Census) focus on national heritage, which, in the case of Muslims and some other faith groups, can mask religious identity.  The latter has been shown to be a prime identity for Pakistani and Bangladeshi Muslims (Modood et al 1997). 

There is evidence that health inequalities in South Asian communities can be revealed more accurately using faith identity than ethnic groupings:  greater socio-economic and health inequalities exist between the different faith groups comprising the Indian population, for example, than between Indians and other ethnic groups.  This is likely to be because Indian Muslims have more in common with the lifestyles, beliefs, customs and historical perspectives of Pakistani Muslims than with those of Indian Sikhs and Hindus (Ahmad 1999).  This suggests that religion should be recognised as an aspect of identity that may at times overlap with ethnicity but cannot be subsumed by it. Both aspects of identity are important in terms of belonging and social relations and therefore both are likely to have a bearing on health and wellbeing.  Ignoring the specific influence of religion in research means that only a partial picture will emerge of the issues affecting a particular social group. 

Studies that focus on ethnicity may result in findings that relate to faith only because participants raise the significance of their religious practice, rather than as a result of deliberate exploration and inclusion in the design of a study (see, for example, Wheeler 1998 and Bhakta et al 2000).  Consequently findings of inequity are unlikely to differentiate between religious and ethnic discrimination.  Yet, this important distinction is applied in practice within UK society.  The concept of Islamophobia was coined in recent years to represent a particular form of racism experienced by Muslims and targeting their faith identity.  Indeed racism within the UK has historically focused on religious practices, particularly those related to Islam (Commission for Racial Equality 1998; Runnymede Trust 1997).

Islamophobia has historical roots dating back to the Crusades, however the strength of hostility which exists at present does not appear to be greatly diminished by the passage of time (Runnymede Trust 1997).   A survey of attitudes towards ethnic minorities by the Commission for Racial Equality found that all negative stereotypes documented referred to Muslim practices.   These appeared to be associated with a perception that Muslims were ‘invading’ the UK and the refusal to assimilate rather than colour was the main obstacle to acceptance (Commission for Racial Equality 1998).

Said’s (1995) study of Orientalism can be seen as a history of Islamophobia in Europe and the means by which this has been maintained in popular thought. His exploration of Western conceptions of ‘the Orient’ reveal that, despite posing as objective and impartial historical analysis, the majority of Orientalist literature adopts a profoundly anti-empirical approach, particularly towards Islam and Muslims.  Thus, for example, Muslims are represented as a monolithic community of orthodox believers rather than as diverse communities in which interpretations of Islam are debated and contested 'in a dialogue of equals' (p338).  Such literature repeatedly attempts to demonstrate the latent inferiority of Muslims and to turn Islam 'into the very epitome of an outsider' (p70).  This construction is made possible by denying credibility to both, using the notion that Islam and Muslims are incapable of self-analysis or of constructing their own formulations.

The discursive practices outlined in this introduction form part of the context in which Pakistani Muslims live in the UK and have consequences for their everyday lives.  The following section examines what is already known about health inequalities in Pakistani and Muslim communities.

Health inequalities in Pakistani and Muslim communities

Compared with the white population, Type 2 diabetes is up to six times more common in Pakistani people. The risk of death from diabetes is between three and six times higher, with this group also being particularly susceptible to the cardiovascular and renal complications of diabetes.  Death rates from heart disease are also two to three times higher in this group (Acheson 1998). The high level of health inequalities within the Pakistani community is noted in the NHS Plan:

‘children of women born in Pakistan are twice as likely to die in their first year than children of women born in the UK.’

(Department of Health 2000: paragraph 13.15)

Management of chronic conditions also appears to be poorer in this community than in the general population. Among men aged 50-64 and with a limiting long-term illness, the proportion of Pakistani men reporting their health as ‘not good’ is almost double the national average and higher than other minority ethnic communities(Office for National Statistics 2004b).

Census 2001 figures suggest that the health inequalities experienced by Pakistani communities are part of a bigger picture in which faith identity is significant.  Muslim communities suffer the highest rates of disability and self-reported ill-health of all religious groups.  Disadvantage is also apparent in areas affecting health, for example, Muslims experience the highest levels of unemployment, are least likely to have degrees and live in the most overcrowded households (ibid). 

Figure 1:  Socio-economic position of faith groups in Leeds

This national picture was mirrored in the fieldwork area.  In percentage terms, Muslim representation in higher socio-economic groups was lower than any other faith community and long term unemployment was at least five times higher (see Figure 1). Although the Pakistani and Muslim communities formed respectively 2.11% and 3% of the Leeds population, within Harehills, one of the most deprived wards in the UK (Office for National Statistics 2004), 25.7% of the population was Muslim and 18.7% Pakistani (ibid). 

The evidence presented suggests a relationship between Muslim identity and disadvantage which may help to explain why policies that have focused on ethnicity have not been as effective in overcoming disadvantage within the Pakistani Muslim community as in some other communities (Social Exclusion Unit 2004b para 27).

The context provided by this introduction constitutes an important background in which to understand findings from this study.  The following section outlines the methods used to collect and analyse data and the impact this had on the kind of data collected.

Methods

An ethnographic approach was adopted, combining observation, informal ‘interviews’ and triangulated case studies to investigate experiences, relationships and social processes (Hammersley and Atkinson 1995).  Fieldwork included diverse data sources to help build a detailed picture of the context in which chronic illness was managed and the influence of religious identity within this.  Other influences such as ethnicity, gender, age and socio-economic background were also explored within this framework to avoid essentialising religious influence.

Thirteen community organisations, selected because of a link to health or else to faith, were included as data sources in order to gain an understanding of how community networks operated in relation to these areas.  Semi-structured interviews were also conducted with 13 ‘key informants’,  such as GPs, interpreters and community-based professionals. Themes arising from these interviews were used along with themes from the literature review and community-based fieldwork to develop a topic guide for use in the first phase of patient interviews. 

Qualitative interviews were carried out with 31 Pakistani people who had been diagnosed with a chronic illness in the last year. Twenty-two of these individuals were interviewed a second time six months later and 20 a third time, a year to 16 months after their original interview.  The three separate interviews of Pakistani people with chronic illness offered opportunities to gather in-depth data at a group and individual level.  Interviews were conducted in English, Urdu, Punjabi and Mirpuri (a dialect spoken by many respondents who arrived from the Kashmir province of Pakistan).  Informal contact with people in the patient sample was also an important source of data for this respondent group.

The final sample was evenly balanced in terms of gender and ages ranged between 18 and 71 years old.  A mix of social class backgrounds was also achieved by including two affluent and two inner city wards in the fieldwork area, all served by the same network of Pakistani community groups.  All except two individuals were born abroad and had lived in the United Kingdom for between 12 and 43 years.  All people in this sample lived within a two-mile radius, which constituted the fieldwork area.

Ten people with chronic illness nominated a carer and professional person who was influential in their decision-making about health. These interviews enabled exploration of the kind and quality of support available and triangulation of carer, professional and patient perspectives. 

A cross section of primary and secondary care health professionals from various health services related to chronic illness management were included.  Detailed observations of interactions were conducted in a diabetes clinic, a cardiac rehabilitation clinic and two community health clinics  Semi structured interviews with health professionals included a range of diabetes practitioners (consultants, specialist nurses, dietitians and podiatrist) along with consultants in cardiology, oncology and mental health as well as GPs, interpreters and community health visitors. 

Local, national and international events relating to Pakistani and Muslim populations that were   reported in the media formed an important backdrop to the study.  Over the three year period in which the research was conducted, disturbances in Harehills and the terrorist attack in New York in 2001 had a particular impact on the Pakistani community in the fieldwork area.  The significance of these events and the way in which they were interpreted was discussed both within community settings but also through e-lists organised by Muslim groups at local, national and international levels. 

These groups, which required a high level of literacy in English, were made up of people who were, in some ways, significantly different to the majority of people within the patient sample. However, comparison of the data from these lists with that from community-based ethnography and semi-structured interviews highlighted similarities in the way Muslims interpreted such issues.  At the same time, use of these different data sources also allowed exploration of social class and ethnic diversity in the way people responded to these events. 

Findings  

Findings from the study are presented in the context of family and community settings, patient and practitioner interactions and the wider context of UK society. Understanding how religious identity influenced interaction in all these contexts helped to develop a sophisticated  understanding of this aspect of identity and its significance for the healthcare and health of Pakistani individuals and communities.

Family and community contexts

In family and community-based contexts, religious expressions were often used naturally within conversations about health, indicating the close link respondents often felt between their physical and spiritual life. Religious teachings provided meaning to the experience of illness and could encourage ways of responding that decreased anxiety and depression.  The use of faith beliefs to achieve emotional distance and an objective perspective was encouraged as a response to diagnosis of a chronic condition:

‘She at least reassures me, like she might say “Never mind, Allah is in control, you will get better, don’t worry, these things happen”.’

  (OB - female with undiagnosed lump)

Religion was an important value framework and influenced the health decisions of over half the final sample, with a prime or secondary level of influence for a third of patients. The data suggests that for respondents who accorded religion a prime position, these beliefs and teachings came prior to and determined their own judgement in all areas. In such instances, religious teachings had more authority than any other influence on decision-making.  For others, religion had a more advisory role and was often grouped with the influence of family members, health professionals and friends.  

An individual’s ability to draw on faith beliefs was affected not only by personal choice but also by the family and community context.  Lack of support to incorporate ritual aspects of religious practice, such as prayer, into everyday life weakened the ability of people to draw on these as a resource in chronic illness management.  Visible manifestations of Islam, such as purpose built mosques and community fundraising events, were a source of self-esteem for the community, however these were financed and driven by Pakistani community members on an almost entirely voluntary basis without external support.  Community-based organizations set up to meet health and social care needs within this community packaged themselves in terms of ethnic identity because of known hostility to faith-based activity, particularly by Muslims.

These community organisations were often run by Pakistani Muslim staff and did sometimes make use of faith as a resource in health activities. Using religious teachings that conveyed health messages was an important means of adapting health promotion material to the cognitive framework of those who attended these activities and increasing trust in the workers.  Notions of the body as a trust from God were used to promote the idea of personal responsibility, preventive healthcare and an active approach to seeking out advice for health problems. Specialist services to meet the needs of South Asian communities had developed expertise in their knowledge of these needs yet no mechanism existed by which this knowledge could be transferred to mainstream services, where most patients were seen.  Furthermore, the lack of secure infrastructure in the voluntary sector led to high turnover of experienced staff so that Pakistani patients did not receive a satisfactory service in either area of provision.

Healthcare contexts

Professionals often thought that poor health outcomes for Pakistani people were the result of patients "not treating themselves or their health seriously”, "not engaging" with their health, "not wanting advice" or feeling the advice was not significant to them.  Being "cut off" through language or social isolation was also seen as a possible reason why some individuals might find it difficult to follow the advice given by professionals.   

Apart from linguistic barriers to communication, findings suggest that none of these reasons hold true for Pakistani people with chronic illness and that the influence of physicians is most often decreased because of difficulties in concordance - the process through which chronic illness management is negotiated (Stevenson et al 2000).  This process was hampered in interactions by three barriers, two of which bore a significant relationship to religious beliefs:

a)    linguistic differences

b)    professionals’ assumptions about the culture of Pakistani patients, and

c)    unwillingness on the part of both patients and professionals to discuss cultural influences on decision-making if they had little common ground in this area. 

Reference to religious beliefs in clinical contexts was rare and, when this did occur, was initiated by people with chronic illness in relation to specific acts that required adjustments to treatment or self-care, such as fasting or going on Hajj[1].  Patients often expected health professionals to consider acts such as fasting to be unimportant and consequently did not consult them about how to manage treatment.  These respondents felt that professional advice in relation to religious obligations would be subjective rather than impartial and inclined towards undermining rather than supporting their faith.  A minority of health professionals did engage with this aspect of patients’ lives to support their decision-making about self-care, however most tended to treat such decisions as a personal matter to be left to the patient’s discretion.  Patients often contributed to this lack of dialogue through their own assumptions about health professionals who did not share their cultural background. 

‘Fasting and prayer do not concern the doctors… They don’t even know what it is.’

(NC - male patient with coronary heart disease)

Health practitioners often denied that religion or ethnicity had any effect on their communication with Pakistani patients and felt differences were related to ‘class, education and age, rather than race’.  Analysis of data from the study showed, however, that ethnic and religious identity did inform the assumptions of both patients and professionals, not only in relation to their own views but also in terms of their responses to each other.  Stereotypical ideas about fatalism and about the position of Pakistani women in their families were particularly damaging to the achievement of shared understanding.  For example, some professionals ascribed non-compliance with medication to religious beliefs rather than seeing this response as generic.  Challenging such interpretations could lead to recognition of cross-cultural similarities in patients’ responses (see Extract 1).

Extract 1 illustrates a particular area of misinterpretation by health professionals. Most respondents accepted their illness as part of their destiny and interpreted this experience as a test of their patience, or (more rarely) a punishment for past misdeeds.  Individuals sometimes expressed their belief in ways that could be interpreted as fatalism; when asked about his expectations for the future, NC, for example said that this was in God’s hands, ‘A person can’t do anything’.  Mrs N similarly fasted in Ramadan saying ‘If my health is meant to become worse then let it, it doesn’t matter’.  At the same time, both respondents were taking regular medication, paid attention to any deterioration in their symptoms and respected the advice they had been given by health professionals.   Such statements could however be taken at face-value and out of context by health professionals.  Religious justifications that were used to support an individual’s decision about treatment could be understood in isolation and could prevent health practitioners from considering other reasons why a Pakistani person might resist certain forms of intervention. 

Stereotypical views about Muslim women’s position in the family were also applied to Pakistani women, who were often seen as oppressed by male members of their family  though there was no evidence of this.  This use of stereotypes to fill the gap in practitioners’ knowledge about Pakistani patients’ lives could lead to misdiagnosis and prevent accurate assessment of the issues affecting patients.  AB, for example had experienced a very traumatic labour in which she was subjected to physical abuse and racism by midwives. This had left her feeling extremely distressed and she went to the GP seeking help for depression.  Her GP’s response (Extract 2) indicates that she was obliged to negotiate a professional judgement, not only of the extent of her mental distress, but also of its causes, before practitioner support was made available:

There was evidence that widespread health beliefs about medication in the Pakistani community were also not being addressed by practitioners who were responsible for chronic illness management.  These included an association between lifelong medication and addiction and the notion that medication was an alternative to changes in lifestyle rather than one part of an intervention to manage chronic illness.  Evidence from other studies suggests that these health beliefs may be common to patients in the general population and that  communication barriers may relate more to differences in lay and professional health beliefs than to ethnicity or religion (Stevenson et al 2000; Lambert and Sevak 1996).  However, findings from this study suggest that the ability of professionals to engage with Pakistani individuals in order to address differences in perspective is far less well developed than with white patients (see Extract 3)

Dynamics relating to an undercurrent of unspoken racism were also described by some respondents.  These could result in hostility from professionals and resistance to using resources, especially for those who did not speak English. Religious and ethnic identity were thus combined in the discrimination to which Pakistani people were exposed and both could affect access to healthcare and health information.

The study provided evidence that these dynamics in healthcare mirrored discrimination in a range of contexts within the wider UK society.  Pakistani individuals who had tried to maintain social contact with white friends or social groups often experienced insensitivity and undermining of their values by those with whom they tried to interact, and lack of support to challenge racism within group settings.  Even within their own homes respondents experienced hostility via the media, which often promoted attitudes and values that were hostile to Islam and Muslims.  The common disregard for beliefs held by Muslims and the danger of 'losing' children to other value frameworks was very distressing to a number of people and posed a barrier to developing relationships with people outside the community.  Some respondents had learnt to avoid situations in which they experienced this stress and felt that problems with integration arose from the dominance of European cultural values. Until the values of Pakistani people were given equal status to those of the dominant culture, these respondents felt Pakistani people would continue to be perceived as 'inferior' and 'outsiders'. 

Discussion

Implications for policy and practice

This study details the mechanisms of social exclusion by which health inequalities are produced for Pakistani people.  These mechanisms operate both within healthcare contexts and within arenas which have been shown to determine health, such as employment, education, housing and civic participation (Scambler 2002; Marmot 2004). The findings reveal a dichotomy between the significant personal resource that faith can provide to Pakistani people with chronic illness and the exposure to discrimination that Muslim identity triggers in the wider UK society.

Lack of engagement with alternative beliefs and values has been shown to result in a disparity in expectations and an emotional distance between patients and professionals (Bhopal 1997). It has also been suggested that inability by health practitioners to form a holistic relationship lies at the root of suboptimal care,  poor take-up of services and other types of discrimination such as stereotyping and inadequate response to expressed needs (Auluck and Iles 1991; Bowler 1993; Vangen et al 1996). 

Findings from this study support the argument that practitioners who are equipped to engage with patients’ cultural context are more likely to develop relationships in which concordance can be developed and better health outcomes achieved.  For many professionals, however, engagement with Pakistani individuals is considered almost solely in terms of language and even at this preliminary level is not adequately addressed.  Shared language is clearly a first step to negotiating with non-English speakers; however, many of the barriers highlighted by this study affected English-speaking patients in the sample also.   Steps to improve management of chronic illness within the Pakistani community therefore need to address the range of impediments identified and not focus solely on language.

 The study also provides evidence that expertise need to be shared between the voluntary and statutory sectors of service provision – to develop more appropriate services for patients in mainstream services and to retain experienced staff in voluntary sector organisations.      

Findings show that strong links exist between ethnicity and religion in relation to the prejudice to which people may be subjected.  This indicates that in some areas these aspects of identity must be simultaneously addressed, rather than considered separately. It does not mean, however, that each aspect of identity does not deserve independent consideration and this analysis demonstrates that faith identity has a distinct and significant influence on the lived experience of Pakistani people with chronic illness. 

There is evidence that awareness of the discrimination they experience is as significant to the health of Pakistani people as the discrimination itself.  Marmot (2004) has demonstrated that self-perceptions of social position and social relations are markers of psychological wellbeing, which is an important indicator of health status (ibid).  Although stress factors have previously been identified as relevant to addressing health inequalities in minority ethnic communities, these have been related to socio-economic position and membership of ‘deprived income communities’ (Horn and Beal 2004).  Findings from this study indicate that stress factors for Pakistani communities are likely to include Islamophobia, and that this may equally affect those from higher socio-economic groups.  There is further evidence that this kind of discrimination may be more intense for those who choose the ‘visual identifiers’ of Muslim identity, such as the hijab[2] (Allen 2002).

Religious identity is rarely addressed in policies aiming to tackle health inequalities, where ethnicity is the key concept for addressing the needs of Pakistani communities (Department of Health 2000; Social Exclusion Unit 2004). Evidence from this study demonstrates that, although religion cannot replace ethnicity (or for that matter social class, age and gender), as a way of defining Pakistani communities, it can contribute to a more complete picture of their experience.  As such, religion provides an alternative means of identification that may at times be more appropriate than other frameworks to understanding health inequalities. 

Supporting, rather than restricting, faith-based activity within Pakistani communities is indicated as a way of addressing the problems outlined above. Findings suggest that the visible social presence of Islam in communities increases self-esteem and thus psychosocial wellbeing and supports the ability of individual Muslims to sustain faith activity at a personal level and draw on their faith as a resource for health.  Furthermore, the message of social inclusion that policy support would give has the potential to act as a powerful counter to discrimination that currently operates in areas that determine health status.

REFERENCES

Acheson Sir Donald (Chair) 1998, Independent Inquiry into Inequalities in Health The Stationery Office, London. http://www.archive.official-documents.co.uk/document/doh/ih/part2h.htm (Accessed June 2004)

Ahmad W 1999, “Ethnic Statistics:  Better than Nothing or Worse than Nothing,” in Statistics in Society, Dorling D and Simpson S, eds., Arnold. pp 124 - 131

Asad T 1993, Genealogies of religion : discipline and reasons of power in Christianity and Islam  Johns Hopkins University Press, Baltimore.

Aspinall P and Jackson B 2004, Ethnic Disparities In Health and Health Care: A focused review and selected examples of good practice Department of Health/London Health Observatory, London.

Audit Commission 2004, Journey to Race Equality:  Delivering improved services to local communities, Audit Commission, London.

Auluck R and Iles P 1991, "The Referral Process - A Study of Working Relationships Between Antenatal Clinic Nursing Staff and Hospital Social-Workers and their Impact on Asian Women", British Journal of Social Work, vol. 21,1, pp. 41-61.

Bhopal R 1997, "Is research in ethnicity and health racist, unsound, or important science?", BMJ, vol. 314:1751, no. 14 June.

Bowler I 1993, "Stereotypes of women of Asian descent in midwifery:  some evidence", Midwifery, vol. 9, pp. 7-16. 

Cinnirella M and Loewenthal K M 1999, "Religious and ethnic group influences on beliefs about mental illness: A qualitative interview study", British Journal of medical Psychology, vol. Vol 72, no. 4, pp. 505-524.

Commission for Racial Equality 1998, Stereotyping and Racism:  Findings from two attitude surveys, Commission for Racial Equality, London.

Department of Health. The NHS Plan.  A Plan for Investment.  A Plan for Reform. (Cmnd 4818-1)   2000. London, The Stationery Office.

Hammersley M and Atkinson P 1995, Ethnography Routledge, London.

Hillier S 2001, "Book Review: Ethnicity, Disability and Chronic Illness by Ahmad W I U (ed)", Sociology of Health and Illness, vol. 23:4, pp. 541-542.

Karlsen S and Nazroo J 2002, "Agency and Structure:  The impact of ethnic identity and racism on the health of ethnic minority people", Sociology of Health and Illness, vol. 24, no. 1 pp1-20.

Marmot M G 2004, Status Syndrome - How Our Position on the Social Gradient Affects Longevity and Health Bloomsbury.

Modood T, Berthoud R., Lakey J, Nazroo J, Smith P, Virdee S, and Beishon S 1997, Ethnic Minorities in Britain:  Diversity and Disadvantage Policy Studies Institute, London.

Nazroo J 1997, The Health of Britain's Ethnic Minorities Policy Studies Institute, London.

Office for National Statistics 2004a. Census data  http://www.national-statistics.gov.uk (Accessed June 2004)

Office for National Statistics 2004b. Census data. http:/www.statistics.gov.uk/CCI/nugget.asp?ID=348andPos=andColRank=2andRank=320 (Accessed June 2004)

Runnymede Trust (Commission on British Muslims and Islamophobia). 1997, Islamophobia : a challenge for us all : report of the Runnymede Trust Commission on British Muslims and Islamophobia
Runnymede Trust, London.

Said E 1995, Orientalism Penguin Books, London.

Scambler G 2002, "The New Inequality and Health," in Health and Social Change, Open University Press, Buckingham; Philadelphia, pp. 86-109.

Selway D and Ashman A 1998, "Disability, Religion and Health:  a literature review in search of the spiritual dimensions of disability", Disability and Society, vol. 13, pp. 429-439.

Social Exclusion Unit 2004a  http://www.socialexclusionunit.gov.uk . (Accessed June 2004)

Social Exclusion Unit 2004b. Jobs and enterprise in deprived areas: Scoping Note. http://www.socialexclusionunit.gov.uk/downloaddoc.asp?id=77 (Accessed April 2005)

Stevenson F A, Barry C A, Britten N, Barber N, and Bradley C P 2004, "Doctor-patient communication about drugs: the evidence for shared decision making", Social Science and Medicine, vol. 50, no. 6, pp. 829-840.

Vangen, S., Stoltenberg, C., and Schei, B. 1996, "Ethnicity and use of obstetrical analgesia: Do Pakistani women receive inadequate pain relief in labour?", Ethnicity and Heatlh., vol. Vol 1, no. 2, pp. 161-167.

Walsh F 1998, Strengthening Family resilience The Guildford Press, New York.