(6) Researching "race", ethnicity and health: a critical review

Sociology and Pharmacy Practice IJPP 2003, 11: 183–197 ß 2003 Int J Pharm Pract DOI 10.1211/0022357022151 ISSN 0961-7671 (6) Researching ‘‘race’’, ethnicity and health: a critical review Paul Bissell, Janine Morgall Traulsen and Lotte Stig Haugbølle Abstract In a previous article in this series, we posed some questions about the visibility of gender and the relevance of feminism for pharmacy practice research.1 Gendered distinctions are just one of the ways in which sociologists have attempted to categorise and explain the contours of the social world. Another obvious feature of modern societies is their ‘‘racialised’’ or multi-ethnic composition. Recognition of this has focused attention — if somewhat belatedly — on how to account for the health and illness experiences of ethnic or ‘‘racial’’ groups. In this paper, we describe and critique some of the health services and social science research which has explored these issues. Our focus has mainly been the UK literature. Our aim is to raise the profile of questions around ‘‘race’’ and ethnicity within pharmacy practice research, which have hitherto received scant attention. Some basic problems Centre for Pharmacy, Health and Society, School of Pharmaceutical Sciences, University of Nottingham, England Paul Bissell, lecturer in social pharmacy and pharmacy practice Department of Social Pharmacy, The Danish University of Pharmaceutical Sciences, Copenhagen Janine Morgall Traulsen, associate professor Lotte Stig Haugbølle, associate professor (head of department) Correspondence: Dr Bissell, Centre for Pharmacy, Health and Society, School of Pharmaceutical Sciences, University of Nottingham, England NG7 2RD E-mail: [email protected] A quarter of a century ago, ``race’’ and ethnicity did not figure prominently in the work of health services researchers, medical sociologists or sociologists more generally. A general sociology text from the 1970s would have chapters discussing the social significance of class, the family, religion, urban sociology, health and health services and perhaps a chapter discussing gender. However, the experiences of ethnic minority groups were generally either absent, or confined to specialist texts. The health and illness experiences of minority groups received similarly scant attention within health services research and medical sociology texts of the time. In the final quarter of the 20th century, the relationship between ``race’’, ethnicity and health began to figure more prominently in mainstream medical and social science journals. However, as recently as 1997, a professor of social epidemiology in Britain felt compelled to make the following comments about the research base in the British Medical Journal: ``race and ethnicity in health research have seldom given fundamental new understanding of disease. Most such research remains `black box epidemiology’. Researchers have not overcome the many conceptual and technical problems of research into ethnicity and health. By emphasising the negative aspects of the health of ethnic minority groups, research may have damaged their social standing and deflected attention from their health priorities. Unless researchers recognise the difficulties with research into ethnicity and health and correct its weaknesses, 20th century research in this subject may suffer the same ignominious fate as that of race science in the 19th century.’’2 In a similar vein, sociologists have argued that much of the social science research exploring the health status and health experiences of ethnic minority groups has been marred by a focus on ``exotic’’ illnesses and a tendency to construct such groups as the ``Other’’. 3 One author claims that: ``a lack of sensitivity to the meaning of categories such as `race’, ethnicity and the concept of racism, alongside the propensity of sociologists of health and illness to keep their heads `buried in the sand of white health concerns’ has all but inhibited the development of an effective body of research.’’3 This makes for depressing reading. However, although the relationship between ``race’’, ethnicity and health remains contested terrain, there are grounds for optimism 183 184 The International Journal of Pharmacy Practice, September 2003 with the development of a more critical and theoretically informed literature. 4±10 We examine some of the arguments around the use of the terms ``race’’ and ethnicity in relation to health in the following sections. Brief history of ‘‘race’’ and health The ``science of race’’ has an entirely dishonourable history. Although reference to it first appeared around the 17th century, it was not until the 18th and early 19th century that it began to be used in Europe and North America to explain differences between humans. 11 During this latter period, it was common to distinguish between several defined ``races’’, which could be identified by physical traits such as skin colour, facial characteristics etc Ð with the crucial distinction that members of each ``race’’ shared common biological attributes, which differentiated them from other ``races’’. These physical markers of difference Ð phenotypes Ð were assumed to arise as a result of the group’s genotype (heritable genetic makeup). While phenotypical difference are self evident across human populations, modern genetics has shown that classifying people into ``races’’ is neither useful nor accurate. Most genes occur identically in all human beings and most biologists have either abandoned racial classifications or adopted more sophisticated multivariate techniques for analysing genetic differences between populations. In short, the scientific basis for the existence of ``races’’ has been shown to be erroneous and the category is now known to have almost no scientific worth, 2,3,11 although, of course, it continues to possess considerable currency within political discourse. The 19th century belief that distinct ``races’’ had a ``biologically determined capacity for cultural development’’ 11 was strongly held by many powerful groups. Colonial expansion played a key part, drawing attention to the phenotypical variation among humans and prompting the need for explanation. Coupled with the rise of modern science and medicine during this period, a ``race science’’ took hold, which explained cultural development as a product of biological difference. As we saw in our illustration of the history of gender, 1 the ``facts’’ of gender difference were often articulated through and underpinned by science and medicine: so it has been in relation to ``race’’ and ethnicity. This can be most clearly seen through some historical examples. Kreiger,12 quoted in Annandale, 3 reminds us that physicians played a prominent role in the slavery debate in America. The idea of ``biology as destiny’’ found expression in a number of medical studies around the mid 19th century. Kreiger12 refers to a book produced by a Dr Pendelton of Georgia, who argued that the reason why twice as many blacks as whites died of tuberculosis was because they were not suited to the cold of Georgia’s climate. The fact that most blacks lived in overcrowded, poor quality housing without adequate clothing and food was ignored. Commenting on the observation that ``coarse muscular Negresses’’ experienced more miscarriages than white women, Dr Pendelton noted that this was a product of the ``unnatural tendency of the African female to destroy her offspring’’. Dr Samuel Cartwright, a noted advocate of slavery, argued that ``black inferiority’’ was a product of an inability to consume as much oxygen as whites. In his view, this predisposed slaves to lethargy, the only antidote being forced labour. The ``race science’’ of the time provided a justification not only for slavery, but also for imperialism and antiimmigration policies in many countries. 5,13±15 Bhopal 2 points out that ``diseases’’ such as drapetomania (the irrational and pathological desire of slaves to run away) and dysaethesia Aethiopica (rascality) were invented as a means to underpin the social and economic status quo. To quote a medical textbook from the early 20th century: ``the pelves becomes increasingly lower and broader the more civilised the race from which it is obtained’’ and ``coloured children weigh considerably less than white, a fact which, in large cities at least, is indicative of the physical degeneration which characterise the race’’. 16 In other words, inequalities between different ``races’’ were considered self-evident. The 1939±45 War provides another appalling example of the application of scientific racism. Aided and abetted by anthropologists and psychiatrists, the Nazi state produced a ``scientific’’ basis for assumptions about racial and genetic inferiority. Health professionals were heavily implicated in the forced sterilisation of ``mental’’ patients; those considered to be from inferior ``races’’ were employed in labour camps, experimented on and sent to the gas chambers. The list of ``inferior’’ groups included homosexuals, communists, Romany Gypsies and of course Jews. Gilman17 has noted that ``there is no space more highly impacted with the sense of difference about the body of the Jew than the public space of medicine’’. Gilman17 reminds us how Jews were marked out as unfit for military service (and therefore full citizenship) on the grounds of having inferior feet. The notion of the ``limping Jew’’ gained credence within much 19th century and early 20th century neurological medicine. What these examples remind us of is the way in which the body was marked out by racial discourse, with the complicity of medical science. 3 The ``science’’ underpinning ``race’’ was largely repudiated by the end of the 20th century and most of the complex classificatory schemes that were developed have been forgotten. Instead, ethnicity is now often used instead of ``race’’ to denote certain characteristics of social groups in both sociological and everyday discourse. We now turn to consider the term ethnicity and its application to health research. Ethnicity: some comments Ethnicity is derived from the Greek word ``ethnos’’ Ð meaning nation. As it is currently used, ethnicity alerts us to the exclusionary connotations implicit in the term given that in reality it is often a conflation of religious, cultural and national identities. In many countries, current ethnic distinctions tend to be strongly identified around an ideology of ``race’’ which focuses on physical appearance (and particularly skin colour) as the fundamental marker of September 2003, The International Journal of Pharmacy Practice difference. However, other ethnic distinctions can and have been made around language, religion, historical or territorial identity as well as notions such as culture. In Britain, people of Jewish, Irish, Lithuanian, Caribbean and Pakistani extraction can all be considered to constitute ethnic groups. Wallman 18 refers to ``symbolic identifications’’ Ð diet, dress, kinship systems and so on. We are left with the notion of ethnicity as a quite elastic concept, potentially encompassing numerous markers of ``difference’’ which are both claimed by a group, and which are ``placed on them’’ by significant or powerful groups. To put it slightly differently, the ``ethnic’’ markers which people choose to discriminate between themselves and others, are in fact essentially arbitrary. While factors such as physical difference, language and religion often do demarcate social differences between individuals, there is no a priori natural or logical reason why this should be so. Perhaps not surprisingly, some have objected to using ethnicity in this way. Firstly, objections have been made on the grounds that while ethnicity may be a preferable term to ``race’’, explanations for ethnic difference have been and still are couched largely in genetic terms.19 Others have pointed out that defining ethnicity in terms of arbitrary ethnic markers ignores the political and historical factors by which these markers are given concrete meaning. Some writers prefer to maintain a role for ``race’’. 20 The point these writers are making is that while the markers which define ethnicity may be arbitrary, the circumstances in which they are used are not. Typically, ethnic distinctions are most forcefully made when issues to do with economic and political power are at stake. To take a simple example, the meaning of identifying oneself as a Catholic or Protestant as a marker of ethnicity is likely to be quite different in Belfast as compared with Birmingham. Smaje 21 suggests that if ``race’’ and ethnicity are social constructions, what meaning can there be in talking about ``ethnic differences’’ in health status? This debate has been important within the ethnicity and health research literature. Sheldon and Parker 4 point out: ``If health workers continue to use the term `race’ because people act as though race exists, they are guilty of conferring analytical status on what is nothing more than an ideological construction.’’ On the other hand, Marmot 22 has suggested that: ``the vagueness of the term `ethnic’ does not invalidate this area of study. If two groups, however defined, have different rates of disease, productive aetiological investigations may follow.’’ What these two quotes highlight is the lack of consensus about the status of ``race’’ and ethnicity as explanatory variables within health research, a theme which permeates this paper. The point is that neither position disputes that differences in health status are to do with factors associated with ethnicity rather than ethnicity per se. However, a focus on ethnicity, rather than other associated factors (social class, poor housing, education, racism etc), can lead to 185 the inference that it is ethnicity itself which causes poor health. It is not difficult to see that such an approach links with victim blaming: that minority groups are themselves responsible for their ``poor’’ health by virtue of their adherence to deviant ``cultures’’ or practices which are held to shape health behaviours and health outcomes.5 These arguments help to illustrate some of the problems encountered when concepts such as ethnicity are applied to health. At a political level, this dispute is mirrored in the ongoing debate about the merits of ``multiculturalism’’ versus ``anti-racism’’. Although it does not have an exact meaning, we can define multiculturalism as an attempt to emphasise the existence and the validity of differing cultural traditions and to promote tolerance and understanding of these traditions. Anti-racism on the other hand, although broadly embracing the multiculturalist perspective, places rather more emphasis on identifying and changing the forces that structure relationships and determine access to power in society. In other words, for anti-racists, dominant ideologies are inherently racialised and racism is seen as a pervasive instrument of social control which marginalises people from ethnic minority populations. As Smaje 21 notes, this debate is endemic within social research and there is no simple solution to it. However, it forms an important backdrop to the debate about ethnicity and health in this paper. Moving on from this debate, what we have also witnessed over the last 25 years, as ``race’’ and ethnicity have become more culturally visible, is the construction of a body of statistical data reflecting various ways of operationalising the concept of ethnicity. We now look briefly at the debate over ethnic statistics, using an example from the British experience. Operationalising ethnicity: ethnic statistics The most authoritative source of demographic data in Britain comes from the 10-yearly census. It is only recently that questions relating to ethnicity have been asked in the British census. For example, the 1971 census asked only for country of birth and mother and father’s country of birth. This question picked up on many, but by no means all, ethnic minority populations resident in Britain. However, recognition of the increasing number of respondents from such populations born in Britain made the need for additional questions increasingly apparent. To rectify this, researchers could either continue asking for country of birth of parents/grandparents/great-grandparents, or abandon this focus on migration and devise an alternative classificatory scheme from which individuals could choose their preferred identity. What transpired was that measures of self-reported ethnicity gained the upper hand in official British statistics. The arguments for focusing on ethnicity rather than migration centred on the notion that systematic social forces exist within the ``host’’ country, which lead to the reproduction of social differences, persisting beyond the migrant generation into its indigenous offspring. 21 These differences, it has been argued, lead to inequalities in access to resources which may affect health and other experiences. Thus, this 186 The International Journal of Pharmacy Practice, September 2003 argument states, it is not the fact of migration per se which is important, but the continuing emphasis on social, economic and other factors which should be the focus of academic inquiry. This argument provides a more explicitly political rationale for research to grasp the social realities of ethnic pluralism, notwithstanding the debate about how that plurality can best be described by empirical research.4 Whatever the merits of these debates, by the time of the 1991 British census, the question on ethnic group shown in Table 1 had been devised. This question was subsequently revised and extended in the 2001 census. The inclusion of questions on ethnicity has undoubtedly helped to move ethnic statistics from the realm of migration to broader issues. In relation to health, researchers have used other techniques to focus on the health of ethnic minority groups. Perhaps of particular note is the use of ``name’’ searches or electoral registers, death certificates and so on in order to identify particular ethnic populations. Obviously, this approach requires populations with names distinctively different from the majority population and with a high degree of endogamy (marriage within the same ethnic group). The technique has been used with some success among South Asian populations and has been shown to yield quite accurate results, 23 although its accuracy will decline with decreasing endogamy. Although the method could extend beyond South Asian populations, its accuracy in groups with names less distinctively different from the majority populations is clearly questionable. 24 Perhaps not surprisingly in what is a heavily contested field, although there are some obvious advantages of ethnic statistics for researchers, others have criticised them.25,26 The main arguments raised include the fact that ethnic statistics deflect attention from racism, they ignore the fact that people do not always classify themselves in the same way, they are inflexible to changes in ethnic identity over time, they conflate conceptually distinct categories in their construction and they do not provide data refined enough to be useful (see Smaje 21 for a review of these arguments). Many of the points raised remind us of the problem of focusing on ethnicity itself, instead of racism or some other variable such as social class as the factor producing disadvantage and exclusion, and point to the ``victim-blaming’’ that can accompany a focus on ethnicity. On the other hand, Leech27 argues: ``It was ethnic statistics that revealed unemployment rates among black teenagers to be twice the national average. It was census data that was used to show how racial minorities had been concentrated in the most derelict, most overcrowded and least secure sectors of inner city housing. . . it was census data that helped demonstrate that far from racial minorities making disproportionate demands on social services, the opposite was in fact the case.’’ Others have been less convinced of the case for ethnic statistics, in themselves, to significantly help combat ethnic disadvantage: ``The lack of political will on the part of the government has long been accepted as the framework . . . within which data are produced . . . the real need is not for better definitive data but for a new framework of political will aimed at reducing and eradicating racial disadvantage and discrimination.’’28 These arguments are an important contextual backdrop to the use of ethnic statistics in health research. From a research point of view, although there is a strong case for collecting ethnic statistics, one needs to be aware of the assumptions underlying them and the uses to which they will be put. 29 We now turn to how ethnicity has been used to explore health status among minority groups. Ethnic minority groups and health status In this section, we provide a brief overview of data on the health status of ethnic minority groups in Britain. The data from the 1991 census demonstrate that standardised rates of limiting longstanding illness are lowest for those who identify themselves as Chinese. Those identifying themselves as ``Other Asian’’, along with white men and women and Black African men, also have lower rates of limiting longstanding illness. Rates are significantly higher for Pakistani men and women and for Bangladeshi men.3 The 2001 census confirms that Pakistani, Bangladeshi and Table 1 1991 Census question on ``ethnic’’ group. Ethnic group Ð please tick the appropriate box White Black Caribbean Black ± African Black ± Other (please describe) Indian Pakistani Bangladeshi Chinese Any other Ethnic Group (please describe) ``If the person is descended from more than one ethnic or racial group, please tick the group to which the person considers he/she belongs, or tick the `Any other ethnic group’ box and describe the person’s ancestry in the space provided.’’ 187 September 2003, The International Journal of Pharmacy Practice now also ``Other Asian’’ people have well above average rates of limiting longstanding illness when analysed by age group. For example, among men aged 50±64 with a limiting longstanding illness, the average proportion reporting their health as ``not good’’ is 13.7 per cent. For Bangladeshi men this figure is 30.9 per cent and for Pakistani men 26.3 per cent.30 Given the new category for Irish within the 2001 census, it is interesting to note that Irish and Black Caribbean groups also report poorer health than the average. For England and Wales as a whole, 68.6 per cent say their health is good, 22.2 per cent are in ``fairly good health’’ and 9.2 per cent ``not in good health’’. However, the proportion of the Irish group reporting health as ``not good’’ is 14.8 per cent overall and for the Black Caribbean group it is 10.7 per cent. The Black African group appears to have better health than average, with 81.2 per cent saying their health is good and just 4.4 per cent saying their health is ``not good’’. There are more data that allow us to compare mortality rates across different ethnic groups. 31 As can be seen from Table 2, persons from the Indian subcontinent, the Caribbean Commonwealth (women) and the African Commonwealth all experienced excess mortality when compared with the host population of England and Wales. Looking at cause-specific mortality by country of birth, it is apparent that despite some evident differences, the major causes of death are similar to those of the indigenous population, namely, circulatory disorders, respiratory disorders and cancers.32 However, as shown in Table 3, all three of the ethnic minority groups experienced a markedly higher burden of death from diabetes and from stroke. Perhaps most notable are the lower levels of death from cancers for all minority groups when compared with the rates for England and Wales. We refer those interested in finding out more about this area to Smaje 21 and Nazroo.33 Of course, these data tell us nothing about the reasons for the higher death rates and rates of limiting longstanding illness among the various ethnic minority groups. We now look at some of the explanations that have been forward to explain these observed inequalities. ‘‘Race’’, ethnicity, social class and health Many explanations have been put forward to explain patterns of health status experienced by ethnic minority groups (see Smaje 21 for a discussion of these). In this paper, we concentrate on the debate over the relationship between social class, ethnicity and health. As discussed in an earlier Table 2 Standardised mortality ratio (SMR) at ages 20±69 from all causes by sex and country of birth, 1979±83 (England and Wales as base). Country of birth Males Females Scotland All Ireland Indian Subcontinent Caribbean Commonwealth Mediterranean Commonwealth African Commonwealth Australia Canada New Zealand France Germany Italy Poland England and Wales 118 128 106 79 87 109 85 96 84 80 87 65 99 100 118 120 105 105 78 114 83 85 96 75 93 78 97 100 Adapted from Smaje21 Table 3 Standardised mortality ratio (SMR) from selected diseases by place of birth, age 20‡ , England and Wales, 1979±83. Birth place Sex Coronary heart disease Cerebrovascular disease Cancers Diabetes Indian sub continent Caribbean Commonwealth African Commonwealth England and Wales M F M F M F 136 146 45 76 113 97 100 153 125 176 210 163 139 100 59 69 65 71 71 83 100 297 103 292 424 219 161 100 Adapted from Smaje.21 188 The International Journal of Pharmacy Practice, September 2003 paper in this series, sociologists have long been interested in the impact of social class and socio-economic factors in determining health. 34 However, as some authors have argued, ethnic minority groups endure ``working class health inequalities and then some’’.35 Why is this? At the heart of the debate among epidemiologists and sociologists about the relative significance of ethnicity and socio-economic factors for health inequalities is the recognition that although ethnicity tends to be strongly associated with material disadvantage, the correlation is not clear cut and, consequently, the health issues under investigation cannot always be explained by reference to socioeconomic factors alone. In other words, there may be an independent effect of ethnicity/culture (or some other related factor) upon health. As Smaje has put it, ethnicity cannot be ``simply emptied into class disadvantage’’. 7 The arguments are complex, but we will explore them in depth, given their importance to explaining inequalities in health among ethnic minority groups. In particular, we draw on the recent work of Nazroo 9,33 in the fourth national survey of the health of Britain’s ethnic minorities. As discussed previously, 34 many researchers have identified a clear social gradient in relation to morbidity and mortality in the general population. In Britain, both the Black report36 and the Acheson report37 favoured materialist or socio-economic explanations for the observed inequalities in health. Where does ethnicity fit in? Unfortunately, there is rather less research exploring the relationship between socio-economic status, ethnicity and health. One early publication in this area38 explored rates of mortality of immigrants and came to quite opposite conclusions to Black and his colleagues. Marmot et al38 argued that social class (and consequently, material explanations for poor health) were unrelated to the higher mortality rates experienced by most migrant groups in Britain. Smaje’s21 review calls for ``more refined approaches to the dynamic interactions between culture, socio-economic status and health experience’’ and argues that evidence from the more extensive US literature is inconclusive about whether socio-economic status alone can account for racial disparities in health. In contrast to this, more recent work in the US39±41 and work from the UK33 suggest that material factors are relevant to the health of ethnic minority groups and make a key contribution to differences in health between different ethnic groups. Part of this debate revolves around how ethnicity is operationalised. For example, much epidemiological work has taken a very crude approach to this, using country of birth as recorded in death certificates and in the census. As Nazroo9 points out, this has clear limitations since the ethnic groupings identified use inappropriate boundaries (Asian, or South Asian, Black, etc). The data collected are then interpreted as though the individuals within them are ethnically homogenous, whereas in fact, they are very clearly heterogeneous Ð conflating groups with differing migration histories, religions, cultures and geographical and socio-economic locations. In short, they are insufficiently sensitive to the realities of ethnic identity and experience. One aspect of Nazroo’s9,33 work demonstrates this neatly. He details the relative risk (compared with whites) for reporting fair or poor general health for various ethnic groups, operationalised firstly as ``South Asian’’ and then as Indian, Pakistani and Bangladeshi. He shows that using the category South Asian obscures important differences in health within that group. In particular, the relative risk of reporting fair or poor health is much higher for Pakistanis and Bangladeshis, but lower for Indian people. Using the category South Asian obscures this detail. Nazroo goes on to explore the relationship between health, ethnicity and socio-economic position in more detail. In terms of economic position itself, he suggests there are few differences between the white and Indian groups; however, Caribbeans, Pakistanis and Bangladeshis are, to varying degrees, worse off than the other groups and these socioeconomic differences are mirrored in the health of these groups. What Nazroo9 shows is that class effects are similar for ethnic minority and white people. In other words, there is a clear and strong relationship between class and reporting fair or poor health for all the ethnic groups studied (see Figure 1). However, as he points out,9 differences in social class distribution do not explain ethnic differences in health. Within each class group, Pakistanis and Bangladeshis (who have the poorest health) were more likely than the equivalent whites to report poor or fair health, and the same is true for non-manual and manual Caribbeans. The implication of these data Ð and other data sets Ð is that there remains some unidentified component of ethnicity that increases certain ethnic minorities groups’ risk of poor health. This unexplained variance, is often assumed to be a result of cultural and/or genetic factors. However, as Nazroo9 shows, drawing on various data sources and arguments, this may not be the case. He suggests that rather than reverting to arguments about culture or genetics, it may be more important in explaining this variance to explore the cumulative effects of a lifetime of disadvantage, living in a racist society, and ecological effects (the geographical location of ethnic groups in areas that are quite different to those populated by the white majority). Indeed, 45 Non-manual Manual No full time work 40 35 30 25 20 15 10 5 0 Caribbeans Indians and Pakistanis and African Asians Bangladeshis Whites Figure 1 Percentage reporting fair or poor health by ethnic group and class (age and gender standardised) (reproduced with permission from reference 9). September 2003, The International Journal of Pharmacy Practice in subsequent work,10 it was demonstrated that ethnicity, in terms of racialisation and class experience, is strongly associated with health for ethnic minority people living in Britain. This debate represents an important context for the work reported on here. ‘‘Race’’, ethnicity and health inequalities It is worth returning to Nazroo’s 9,33 work before we move on to look at other research which has critically explored some of the arguments put forward for the health of ethnic minority groups. He makes the key point that although research needs to be sufficiently sensitive to encapsulate ethnicity in a meaningful way, identifying ethnic groups is never a theoretically neutral process. For example, taking the work cited earlier by Marmot et al38 the authors have a clear notion of the significance of the country of birth variable: ``Comparisons of disease rates between immigrants and non-immigrants in the `old’ country, between immigrants and residents of the `new’ country and between different immigrant groups in the new country have helped elucidate the relative importance of genetic and environmental factors in many diseases.’’ However, the problem is that these authors interpret their findings on the basis of assumptions about genetic or cultural characteristics of individuals within the ethnic group at greater risk. Consequently, explanations tend to fall to unmeasured genetic or cultural factors based on stereotypes, because such meanings can be easily imposed on ethnic categorisations. As Nazroo 9 explains: ``Theory is brought in surreptitiously ± ethnicity, however measured, equals genetic or cultural heritage. This then leads to a form of victim blaming, where the inherent characteristics of the ethnic (minority) group are seen to be a fault and in need of rectifying.’’ Another example of this can be found in the widely quoted paper by Gupta et al42 in the British Medical Journal which explored the greater risk of ``South Asians’’ for coronary heart disease. They used their research findings to attribute this problem to a combination of genetic (``race’’) and cultural (ie, ethnicity) factors that are apparently associated with ``being South Asian’’. Concerning the genetic factors supposedly at play, the authors argue that ``South Asians’’ have a shared evolutionary history that involved adaptation to ``survive under conditions of periodic famine and low energy uptake’’. This resulted in the development of an insulin resistance syndrome which (apparently) underlines ``South Asians’’’ greater risk of developing coronary heart disease (CHD). As Nazroo9 points out, this is tantamount to viewing ``South Asians’’ as a genetically distinct group (a ``race’’) with a common history, which is highly suspect. Gupta et al42 go on to speculate about the ``cultural’’ factors involved. They suggest that use of ghee in cooking, a lack of physical exercise and a reluctance to use health services are all important factors. However, these are all 189 questionable assumptions. Ghee is not always used by all the groups that comprise ``South Asians’’; there is evidence that South Asians do understand the significance of exercise 43 and, as discussed later, South Asians do use medical services.21,44 The policy implications that flow from this are significant. The authors state that community leaders and ``survivors of heart attacks’’ should spread the message among their communities and that ``South Asians’’ should be encouraged to undertake healthier lifestyles. In other words, the problem is apparently something to do with being South Asian, nothing to do with the context of the lives of South Asians (eg, racism, material disadvantage) and only rectifiable if South Asians are encouraged to modify their behaviour and to address their genetic and cultural weaknesses. What this extended example reminds us is that although scientific racism may have been swept away and viewed as erroneous, it can be argued that it has been superseded by a new ``cultural’’ racism, which, although it has swapped the language of ``race’’ for culture and national belonging, is no less repressive in its effects. As various sociologists have pointed out, although this new racism is more fluid and unstable, it still attempts to naturalise the difference between ``belongingness and otherness’’. 45 Other sociologists argue similarly that the new racism does not rely so much on biological difference, rather, in response to global economic and social change, it is concerned with demarcating inclusion and exclusion: who is one of ``us’’ and who is the ``other’’. 20,46 It is argued that the current focus on ethnic or cultural difference is part of this new, more covert racism. Another example comes from mental health where higher rates of schizophrenia have been diagnosed among AfroCaribbean men. Harrison et al47 found an incidence around 12±13 per cent higher than for the general population in a study based in Nottingham. To what extent these figures are the result of variations in presentation of symptoms, the result of racism or the propensity of white professionals to label them as ``mad without a cause’’48 is not known. The possible implication of racialisation is made more apparent by the fact that black people are far more likely than others to reach the mental health system via the courts, the police and prisons and to experience more harsh and invasive forms of treatment (such as ECT).3 Pilgrim and Rogers48 argue that the debate on ``race’’ and madness is central to the workings of the new racism which attempts to banish and exclude undesirable ``others’’. In this context, ``racism becomes the modality through which mental illness is represented, and this is something which is reinforced by psychiatric discourses and practices’’. 49 It has been suggested by some sociologists that psychiatry then becomes a part of the ``postcolonial, Europeanised alternative to repatriation’’. 48 Many sociologists have also noted that much health education/promotion for ethnic minority groups has overt and sometimes latent racist or stereotypical assumptions. 3,5 Rickets among South Asians living in the UK is a well-known example. Caused by vitamin D deficiency, as a disease of poverty, rickets was dealt with in the white 190 The International Journal of Pharmacy Practice, September 2003 UK population in the 1940s through fortified margarine. When Asian school children were experiencing rickets in the 1970s, it was explained in terms of ``un-British’’ eating habits and the solution was to adopt a ``British’’ diet and lifestyle. 5 Solomos and Back50 have described this process of cultural inscription using a variety of coded signifiers as ``producing a racist effect while denying that this effect is the result of racism’’. In a similar vein, Bowler’s study of midwives’ representations of women of ``Asian’’ descent showed how professionals stereotyped ``Asian’’ women’s behaviour and negatively contrasted this behaviour with white women.51 Midwives rarely differentiated between different South Asian groupings, tending to subsume them all under the rubric ``Asian’’. The point about such a discourse is that it renders other groups as ``different’’ and ``explains’’ behaviour on the basis of such putative claims. As Bowler51 points out, while none of the midwives actually made racist remarks along the lines of not ``liking’’ black or Asian women, the negative stereotyping was both profound, ingrained and discriminatory. Annandale 3 has pointed out that health education, medicine and the social sciences have colluded in constructing an ``intellectual apartheid’’ in their conceptualisation of ethnic minority groups as carriers of problematic ``cultures’’ and ``exotic’’ illnesses. In a similar vein, Ahmad contends that medicine has presented the world of ethnic minority groups as one where cultures are ``lifeless, limp, cellophane-wrapped . . . rather than cultures that are empowering, changing, challenging and flexible’’. 52 It is the invisibility of ``white’’ culture as an assumed but unstated standard in so much health services research which is the problem. Whiteness is ``unmarked and unnamed as only Others are seen to be carriers of `culture’’’. 3 Frankenberg 53 noted that in studying AfroCaribbean (or equally South Asian) women, researchers typically ``address race and culture; a study of white women . . . probably will not’’. Smaje 7 contends that it is this silence about the way in which white culture is constructed that is the biggest omission in research exploring race and health. The problem that follows from the use of ``white’’ culture as an unarticulated standard by which to judge ethnic minority culture, ``is the tendency to subsume people under the oppressive and deterministic cloud of cultural difference’’. 3 The tendency to focus on culture in research findings is also accentuated by the fact that much research is funded on the assumption that there are fixed distinctions in beliefs and behaviour between different ethnic groups. 54 The task of research predicated on such assumptions is to construct a set of findings that facilitate the provision of more appropriate or culturally sensitive health education messages and health promotion materials. 52 This creates a tendency to interpret findings with reference to such beliefs and practices, thereby accentuating or reifying these. For example, such an approach has been criticised for assuming that the problem under consideration rests with ``faulty’’ health beliefs which are a function of monolithic ``cultural’’ values. By addressing ``culture’’ it permits health professionals and policy makers to believe that they are attending to the problem in question without having to consider the more politically sensitive problems of poverty, deprivation, discrimination and racism, about which they may not be able to respond effectively. Such a focus blames individuals, pathologises ``culture’’ and fails to address what may be the more significant causes of health differences: ``It is not that in considering health, illness and health care of minority ethnic groups, culture is not important. It is that stripped of its dynamic social, economic, gender and historical context, culture becomes a rigid and constraining concept which is seen somehow to mechanistically determine people’s behaviours and actions rather than providing a flexible resource for living . . . the rigid conception of culture, which all too often is apparent in health research . . . provides a description of people which emphasises their `cultural’ difference and helps to obscure the similarities between broadly defined cultural groups and the diversity within a cultural group . . . What is pursued through an adherence to simple, un-reflexive and essentialist notions of `culture’ is a politics of victim blaming which constructs ethnic minority communities as dangerous to their own health.’’52 As Ahmad makes clear, it is not that culture is unimportant or insignificant in shaping health behaviours. However, the point is that it is crucial that researchers question their potentially ethnocentric assumptions about ``culture’’ and explore other avenues before assigning ``culture’’ to a predominant position in their explanations for behaviours. Kelleher55 makes a sensitive and thought provoking case for maintaining a focus on the role of culture and ethnicity as analytical categories within health research and we have considerable sympathy for his argument. He argues: ``Researchers who want to use the concepts of culture and ethnicity do not believe . . . that in constructing ethnicity and culture people are simply taking over a fully formed monolithic set of rules for living; . . . the process of culturally based research is not one of digging up a past culture in order to be able to predict precisely in a positivistic way what people believe and how they will behave . . . it is guided by an awareness of some of the problems people in particular ethnic minority groups currently face and a desire to know more about how they perceive a situation and how they are using their cultural resources to address their current problems.’’55 This seems a sensible and sensitive approach to conducting research and it is clearly important to retain some role for culture and the complexity of culture in analyses of health-related behaviour. However, our emphasis in this paper has been to provide a critical, sociologically informed perspective. Use and access to health services by ethnic minority groups Having reviewed some of the more critical and sociologically informed research around explanations for health inequalities among ethnic minority groups, in this section we describe the major findings detailing use of and access September 2003, The International Journal of Pharmacy Practice to health services. Much of this data is based on a relatively small body of research and we focus mainly on Britain. One of the major findings from this body of work is that adults and children from ethnic minority groups in Britain, particularly from the South Asian population, appear to have higher levels of use of general practitioner services. 56±58 Lower hospital use among ethnic minority groups relative to the white population has also been a consistent finding. 8,59,60 For example, Balarajan et al58 examined ethnic minority patters of GP consultations for people under the age of 65, using a general household study of 64,000 people which included 2,200 people from ethnic minority groups. The investigators computed odds ratios to indicate the probability that people from each ethnic minority group consulted a GP in the previous two weeks relative to the probability of white people consulting. What they found was that in the 16±64 age group, utilisation was raised for males and females of all ethnic minority groups, and these values differed significantly for males from all ethnic minority groups and for Pakistani females. McCormick and Rosenbaum, 61 reporting on the 3rd national GP morbidity survey, found that after adjusting for age, there were significantly higher consultation rates for Caribbean-born men and women for endocrine, nutritional and metabolic disorders, diseases of the nervous system and for ``symptoms, signs and ill-defined conditions’’, compared with the white British population. Caribbean-born women also had significantly higher consultation rates for neoplasms and skin disease, whereas Caribbean-born men consulted more often with mental disorders. They also showed that men and women from the Indian subcontinent had higher rates of consultations for endocrine, nutritional and metabolic disorders, respiratory diseases, skin diseases, diseases of the digestive system and for ``symptoms, signs and ill-defined conditions’’. These higher rates of consultations for certain ethnic groups are confirmed by work by Benzeval and Judge 62 which showed that after adjusting for age, gender, socioeconomic status and morbidity, South Asian people were significantly more likely to have had contact with a GP Ð consulting 1.5 times more often than whites. No other significant ethnic differences were found in this study. More recent work by Cooper et al, 63 looking at use of health services by young people according to ethnicity and social class, suggested that South Asian children in Britain used GP services more than any other ethnic group after controlling for socio-economic background and perceived health status, but the use of hospital outpatient and inpatient services was significantly lower for children and young people from all ethnic minority groups compared with the white population. Their analysis showed that Indian children in particular were most likely to consult their GP. There is interesting research looking at the relationship between ethnicity, equity and use of the health services in Britain. Drawing on a review of the literature and data from the General Household surveys of 1984±1991, Smaje and Le Grand 8 argue that although the general trend of high GP utilisation is confirmed, particularly among 191 South Asian groups (and in some studies, Caribbean groups and the Chinese population living in the UK), there do not appear to be gross patterns of inequity in the use of health services by different ethnic groups in Britain. They suggest that the major exception to this is the Chinese population, which in their study displayed a striking pattern of lower utilisation of health services. As they point out, little research has been conducted looking at the health experiences of Chinese populations living in Britain. More generally, they point out that there is less use, relative to need, by certain ethnic minority groups for inpatient and, particularly, outpatient care. For example, particularly for younger age groups, respondents from all South Asian groups, as well as Caribbean and Chinese groups, report low use of outpatient services (after controlling for confounding variables). Coupled with the consistent finding of broadly equivalent or higher GP use, the authors suggest that if inequity is present it is not at the point of the GP consultation, but at the referral stage. They contend that it may be that people from ethnic minority groups tend to be referred on to secondary care less than their white counterparts with similar needs. This work also highlights some other areas of concern. Pakistani females from younger age groups appear to utilise GP services to a rather lesser degree than white counterparts. Given that the Pakistani population in Britain is characterised by high levels of deprivation and reported ill health, and high perinatal mortality rates, and there is evidence of obstacles in accessing care, this is a particular area of concern. The evidence suggests that levels of GP registration is high among ethnic minority groups in Britain, at over 90 per cent.21,64 In other areas of primary service provision, it has been suggested that lower levels of utilisation are mirrored by less knowledge of these services. For example, Donaldson 65 found a low level of knowledge about meals on wheels services and home helps amongst the Asian elderly in Leicester. Atkin et al66 found similar results although McFarland 67 suggested that ethnic minority elders had low expectations about such services. Hek68 found little evidence of knowledge of district nursing services among South Asian elders in Bristol. Firdous and Bhopal 69 found less knowledge about health topics such as cervical smears and breast self-examination among South Asian than white maternity patients in Glasgow, although other studies have questioned whether actual uptake of such services is lower among such groups. 70 Studies have also explored the choice of GP by ethnic minority groups, focusing on language and communication difficulties and whether problems exist for females in consulting a white GP. The research has produced widely differing results. Studies in the 1980s suggested that the key determinant of GP choice was proximity, and that the ethnicity of a GP was less important, 71 although there is clearly some coincidence between areas of South Asian residence and the availability of South Asian GPs.72 Some studies have pointed to a lack of communication difficulties for ethnic minority patients. For example, Rashid and Jagger 73 found that only 6 per cent of a sample of South Asian patients reported difficulties in communicating 192 The International Journal of Pharmacy Practice, September 2003 their symptoms in English. In Bristol, Pilgrim et al24 found that although people who did not speak English fluently reported additional problems with GP consultations, these problems were not always as severe as might be imagined: less than half of respondents who said their English was ``slight or none at all’’ found it impossible to contact the GP on the phone. However, other research paints a different picture. Considerable problems in English fluency have been reported amongst older South Asians in Leicester65 and South Asians, particularly Pakistanis, in Bradford. 74 Both studies also found poorer fluency among women as compared with men. Another study found that South Asian patients tended to consult more often with a South Asian doctor,75 a finding consistent with Donovan’s study.76 Of course, as others have pointed out, the proximity or availability of GPs from the same ethnic background as their patients does not mean that such groups do not experience discrimination in accessing care.70 Turning to the gender of the GP, it is often assumed that a generally greater sense of modesty among South Asian (particularly Moslem) women makes them more reluctant to consult with a male doctor when physical or vaginal examination may be involved. This assumption is not entirely supported by the evidence which suggests that most South Asian women consult with male doctors and relatively few express a preference for a woman. 71,73 On the other hand, there is some evidence of GPs reporting difficulties examining female ``Asian’’ patients. 77 Smaje 21 suggests that the GP is, if anything, an even more crucial conduit for health services for people from ethnic minority backgrounds than for the majority white population. He concludes that ethnic patterns of health care utilisation are complex and it cannot be assumed on the basis of crude utilisation rates that appropriate access exists. Knowledge and beliefs about health and health services among ethnic minority groups It is often assumed that beliefs and knowledge about health, disease and health care vary between ethnic groups, and there is of course a voluminous anthropological literature exploring the health and healing activities of cultural groups. Our focus here is on exploring the beliefs and behaviours of ethnic minority groups living in Britain and other Western countries. One of the most striking conclusions to be drawn is the degree of congruence with the Western medical model of disease exhibited by ethnic minority groups, 76,78±81 coupled with a general faith and confidence in Western medicine. Of course, respondents might under-report beliefs which they may feel will be judged as superstitious or quaint, but there appears to be little evidence for this, or that this dissuades people from using health services. Donovan 76 conducted in-depth interviews and life histories with 30 Caribbean and South Asian Londoners who had been born abroad. She found that some had extensive personal theories about health and illness which could be viewed as ``incorrect’’ by the standards of the medical model. For example, some South Asian people talked about bodily balance in terms of hot and cold, a humoral lay theory common in a number of societies.82 Some Caribbean people expressed doubt that smoking was bad for health and some South Asian people had doubts about the ill effects of obesity. Interestingly, many of her respondents had experienced racism and believed this to be relevant to their health. There was also a degree of fatalism about illness. One problem is that in the absence of comparison with white groups it is difficult to know if this fatalism is confined to these groups or may also be a feature of the white population. Howlett et al81 attempted this comparison in an analysis of the Health and Lifestyles Survey. Although there are problems with the survey’s ethnicity variable, they suggest that Afro-Caribbeans and whites were more likely to describe health in terms of strength and fitness, whereas Asians were more likely to think of it in functional terms. More Afro-Caribbeans and Asians than whites regarded health as a matter of luck, whereas more whites regarded poor health as the fault of the individual. They also found that fewer Afro-Caribbeans and Asians than whites regarded stress as a cause of disease and fewer viewed smoking and alcohol as a cause of disease. In a study of largely Punjabi people living in Glasgow, 78 Bhopal found that informants’ knowledge about health was rooted strongly in the Western medical model, and most were highly scornful of many traditional ``Asian’’ beliefs and practices. However, he also found evidence that they were poorly informed on health issues ± particularly older people ± with poor knowledge about CHD, the dangers of alcohol and lead-based surma cosmetics. One exception to the finding of a general congruence with the western medical model is shown in a study of Traveller Gypsies.83 The author found strong evidence of beliefs about contamination which led them to reject many health interventions. Smaje 21 points out that different cultural norms regarding the definition and nature of illness may also affect use of services. For example, a Canadian study84 showed how responses to chronic childhood sickness or disability are mediated by an ``ideology of normalisation’’ among white families which seek to re-define the child as normal and healthy. By contrast, they found that migrant Chinese families were less likely to do this, partly through lack of material resources, but also because of greater cultural acceptance of the deviance of ill health. The cultural context in which illness is viewed and the availability of an appropriate sick role may be of some significance in the way individuals interact with services. In a study of Bangladeshi people with type 2 diabetes living in London, Greenhalgh et al85 found some evidence of fatalistic attitudes towards the management of the disease which they attribute to ``cultural’’ factors, although they also suggest that problems with complying had as much to do with material and structural factors. In another study of Bangladeshi people with diabetes, September 2003, The International Journal of Pharmacy Practice Kelleher and Islam 86 found similar evidence of fatalistic attitudes and problems associated with complying with the type 2 diabetic regimen. Another paper explored influences on smoking among Bangaldeshi and Pakistani adults. 87 The authors found that smoking was a widely accepted practice among Pakistani and particularly Bangladeshi men and was associated with socialising, sharing and male identity. On the other hand, smoking among women was often associated with stigma and shame and often hidden from family members. Interestingly, the authors also explored the role of Islam in relation to smoking and found that opinions differed on whether the Muslim religion allows smoking. The authors concluded that while there are some culturally specific contexts for smoking behaviour among Bangladeshi and Pakistani adults (around religion and gender), there are also similarities with white people’s smoking practices. Once again, this latter point reminds us of the importance of thinking critically and laterally about behaviours across all cultural and ethnic groups in the population. Relevance to pharmacy practice research Published journal articles discussing aspects of ``race’’ or ethnicity in relation to what we define as pharmacy practice research Ð use of medicines and the practices of pharmacists Ð are somewhat thin on the ground. What we therefore set out here is a brief review of the published research, before moving on to suggest some areas that pharmacy practice researchers might seek to address, on the basis of the themes discussed above. Perhaps the most obvious omission, given the growing research base that has explored the use of general health services by ethnic minority groups, concerns such groups’ experiences of using community pharmacy services. We could identify only one published study that addressed this area Ð that by Jesson et al, 88 published in 1995 from the UK. The authors surveyed ethnic minorities regarding their use of community pharmacy services in the UK, drawing on a sample of 202 people (73 Black-Caribbeans and 129 ``Asians’’). They found that over half of the respondents typically used one pharmacy, usually an independent pharmacy, and that most (77 per cent) were aware that pharmacies provided advice about minor ailments, although only 39 per cent had used this service. Thirtyfour per cent were taking some prescribed medication and most were able to identify these. The authors concluded that use of, and access to, pharmacy services by this minority population appeared to be lower than by the majority population, although their health needs may be higher. Another study set out to explore GPs’ perspectives on the role of community pharmacists in meeting the health needs of ethnic minority groups for a local population. 89 A paper from the US explores the ``cultural sensitivity’’ of pharmacy interns in New York City90 and another discusses cultural considerations around the provision of pharmaceutical services.91 From New Zealand, Norris et al92 have explored levels of understanding of pharmacyrelated terms by three ethnic groups (Pakeha, Maori and 193 Tokelauan). They found a low level of understanding of key pharmacy and medicine-related terms by these groups. There is clearly a paucity of studies exploring the needs, views and experiences of ethnic minority clients of community and hospital pharmacy, and whether the experiences of these groups are in any way different to those of the majority population. We would suggest that basic research is urgently needed in this area to provide some insights into how pharmacy services are actually used in practice by ethnic minority groups. Given that certain ethnic minority groups appear to have higher rates of consultations with GPs (at least in Britain) as compared with the majority population, researchers might also wish to explore what impact this has Ð if any Ð on rates of consultations with community pharmacists. Another fruitful area for research might be to explore and compare what ailments ethnic minority clients bring to the pharmacist’s attention and whether these differ from those brought by the white majority population. We would also suggest that it might be useful to explore issues around the ethnicity of the pharmacist and their clients in order to ascertain what impact this has on care and whether this is a pertinent factor in the presentations of symptoms. In addition, researchers might also wish to explore whether clients experience any barriers to access on the grounds of language, or whether they typically seek out and consult with pharmacists who speak the same language. Again, basic health services research is urgently needed in this area. The self-care practices of ethnic minority groups would also appear to be an area ripe for exploration. There is some research exploring beliefs about medicine and medicines among certain ethnic minority groups in the UK.93,94 Others have explored issues around health care and compliance during particular religious festivals for certain ethnic minority groups 95 and the use of leadcontaining preparations 96 in the ``Asian’’ community. However, there is clearly also a need to explore the general self-care practices of ethnic minority groups, in order to ascertain whether or not they are in fact any different from the white majority. Do cultural traditions have an influence on how minor (and more major) illnesses are managed across different ethnic groups, and if so, in what way and why? How does this differ according to socioeconomic factors and factors such as gender and age? Do the higher rates of consultations with GPs for certain minority groups have any bearing on self-care practices and what might be the implications of this? One small study conducted by Rashid et al97 in the UK suggests that differences exist between Asians and non-Asians in terms of the range of products they feel should be available over the counter and the amount of information they wish to receive. Another study by Charupatanapong et al98 from the US examined low income older Asian people’s ability to practise self-medication with three non-prescription drugs (laxative, topical antibiotic and headache medication). The ability to practise self-medication appeared to be very low. Given the varying findings on levels of literacy about health, disease and treatment outlined above, researchers might wish to extend such 194 The International Journal of Pharmacy Practice, September 2003 studies to other contexts in exploring self-medication practices among ethnic minority groups. On a slightly different note, in an interesting study conducted in Britain, Platts et al99 suggested that the ethnic minority pharmacy fulfils a role for patients who feel less confident in a pharmacy owned or managed by white British pharmacists and that this, at present, enables the small pharmacy business to remain viable in a competitive environment. Compliance with treatment among ethnic minority groups is a key area of relevance for pharmacy practice and there has been some work in this area.100±103 Again, issues discussed in this paper around explaining and interpreting compliance need to be considered by pharmacy practice researchers. In relation to concordance, there is a need to establish whether the needs of ethnic minority groups are being met by all health professionals. It is sometimes assumed that ethnic minority groups do not expect or express a wish for a concordant approach to consultations with health professionals.101,102 However, there is evidence that this may not be the case and the complex web of factors that shape ethnic minority groups’ experiences of consultations with health professionals needs to be appropriately explored. 102 One area this paper has not touched on concerns the experiences of health professionals from ethnic minority groups. This of course is a major research area in its own right. It is one area where pharmacy practice researchers have been active Ð at least in Britain. 104±106 For example, Platts and Tann 106 explored whether or not there were significant differences in the career profiles of UK pharmacists from different ethnic (and gender) sub-populations. They found that both male and female community pharmacists from ethnic minorities (which they describe as mainly Asian) favoured ownership roles relative to white pharmacists among all the age groups they measured. They suggest that family involvement may have enhanced the availability of human and financial capital for business start-up. In relation to hospital pharmacy, there were indications of under-achievement and career anxiety among pharmacists from ethnic minorities, with the proportion of hospital pharmacy managers from ethnic minorities being particularly low. Hassell et al105 have conducted one of the most extensive comparative surveys of the experiences of white and ethnic minority pharmacists working in community pharmacy in Britain. Drawing on three cohorts of pharmacy graduates (1975, 1985 and 1991), they found that over the three years studied, ethnic minority groups accounted for 21 per cent of the total number of qualifying pharmacists. In 1975, ethnic minority groups accounted for 15 per cent of all qualifiers, whereas by 1991 this had risen to 23 per cent. This is particularly interesting, given that the total ethnic minority population at the time was only 5.5 per cent. Of all the pharmacists in the three years studied, 9 per cent were East African Asian, 6 per cent Indian, 2 per cent Pakistani and 2 per cent Chinese. Bangladeshi and Black Caribbean pharmacists had a very small representation in this study. The study also confirms that ethnic minority pharmacists are more likely than white pharmacists to work in community pharmacy and are dispropor- tionately represented as self-employed contractors. In accounting for this latter finding, the authors argue that preference plays a key role in the decision to opt for selfemployed status. In particular, although racism plays a small part, personal choice, available resources and opportunities for business development within the profession largely account for the different work patterns found among white and ethnic minority pharmacists. It is interesting, and somewhat ironic, to note that on the issue of ethnicity and ``race’’ the profession itself appears to have generated more interest from pharmacy practice researchers than its client group. While such research is of obvious importance, the dearth of research exploring the needs, concerns and experiences of ethnic minority clients is a cause for major concern. Conclusion There appears to be a huge potential for further research in all of the above areas for pharmacy practice researchers. In particular, there is an urgent need to identify and research the experiences of ethnic minority client groups in relation to their use of community and hospital pharmacy, and there is considerable scope for adding to the research on ethnic minority groups’ use of medicines and their self-care activities. In framing explanations about patterns of care for ethnic minority users of services, we would argue that researchers need to bear in mind a number of points raised above. Drawing on Bhopal’s 1997 comments in this area2 , we would suggest there is a need for researchers to: 1 2 3 4 5 6 Appropriately operationalise ethnicity in studies, for example, not simply assume heterogeneity between potentially differing ethnic groups. An obvious example from the UK involves not conflating Pakistani, Bangladeshi and Indian groups under the catch all term South Asian Reflect on assumptions when interpreting research findings. As we have seen, it is far too easy to suggest that culture or ethnicity are ``factors’’ explaining differences or deficits in a particular area of study. Appropriate comparison groups or reflections on the literature seem highly pertinent here Explore socio-economic factors when considering or interpreting differences in health and health practices between ethnic groups Recognise the potential influence of their personal values, including ethnocentricity Recognise and reflect on the methodological and practical limitations of classifying ethnic groups Exert great caution in the generalisation of results, given the fluid and dynamic nature of ethnicity Ethnicity and health is an increasingly important component of health services and social science research, and represents a major public health concern. As yet, this is more or less virgin territory for pharmacy practice research and we strongly suggest that more sociologically informed research is needed in this area. September 2003, The International Journal of Pharmacy Practice References 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Traulsen J, Bissell P, Haugbolle LS. Feminist theory and pharmacy practice. Int J Pharm Pract 2003;11:55±68. Bhopal RS. Is research into ethnicity and health racist, unsound, or important science? BMJ 1997;314:1751± 6. Annandale E. The sociology of health and medicine: A critical introduction. Cambridge: Polity; 1998. Sheldon T, Parker H. Race and ethnicity in health research. J Public Health Med 1992;14:104± 10. Ahmad WIU. Making black people sick: ``race’’, ideology and health research. In: Ahmad W, editor. Race and health in contemporary Britain. Buckingham: Open University Press; 1993. Hillier S, Kelleher D. Researching cultural differences in health. London: Routledge; 1996. Smaje C. The ethnic patterning of health: new directions in theory and research. Sociol Health Illness 1996;18:139± 71. Smaje C, Le Grand J. Ethnicity, equity and use of health services in the British NHS. Soc Sci Med 1997;45:485± 96. Nazroo J. Genetic, cultural or socio-economic vulnerability? Explaining ethnic inequalities in health. Sociol Health Illness 1998;20:710± 30. Karlsen S, Nazroo J. Agency and structure: the impact of ethnic identity and racism on the health of ethnic minority people. Sociol Health Illness 2002;24:1±20. Miles R. Racism after `race relations’. London: Routledge; 1994. Kreiger N. Shades of difference: theoretical underpinnings of the medical controversy on black/white differences in the US, 1830±1870. Int J Health Serv 1987;17:259± 78. Stepan N. The idea of race in science. London: MacMillian Press; 1982. Gould SJ. The mismeasure of man. London: Pelican; 1984. Barkan E. The retreat of scientific racism. Cambridge: Cambridge University Press; 1992. Whitridge-Williams J. A text-book for the use of student and practitioners. New York: Appleton; 1926. Gilman S. The Jew’s body. London: Routledge; 1991. Wallman S. Ethnicity and the boundary process in context. In: Rex J, Mason D, editors. Theories of race and ethnic relations. Cambridge: Cambridge University Press; 1986. Senior P, Bhopal R. Ethnicity as a variable in epidemiological research. BMJ 1994;309:327± 30. Gilroy P. One nation under a groove: the cultural politics of race and racism in Britain. In: Goldberg D, editor. Anatomy of racism. Minneapolis: University of Minnesota Press; 1990. Smaje C. Health `race’ and ethnicity Ð making sense of the evidence. London: Kings Fund Institute; 1996. Marmot M. General approaches to migrant studies: the relation between disease, social class and ethnic origin. In: Cruikshank J, Beevers D, editors. Ethnic factors in health and disease. Sevenoaks: Wright; 1989. Ecob R, Williams R. Sampling Asian minoritiesto assess health and welfare. J Epidemiol Comm Health 1991;45:93±101. Pilgrim S, Fenton S, Hughes T, Hine C, Tibbs N. The Bristol black and ethnic minorities health survey report. Bristol: University of Bristol; 1993. Ahmad WIU, Sheldon T. Race and statistics. Radical Statistics Newsletter. 1991;48:27± 33. Nanton P. Official statistics and problems of inappropriate ethnic categorisation. Policy Politic 1992;20:277±85. Leech K. A question in dispute: the debate about an ethnic question in the Census. London: Runnymede Trust; 1989. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 195 Booth H. Identifying ethnic origin: the past, present and future of official data production. In: Bhat A, Carr-Hill R, Ohri S, editors. Britain’s black population. Aldershot: Gower; 1988. Mason D. Race and ethnicity in modern Britain. Oxford: Oxford University Press; 1995. Census 2001: Ethnicity and religion in England and Wales. London: ONS. www.statistics.gov.uk/census2001/profiles/ commentaries/ethnicity. Britton M. Mortality and geography: a review in the mid 1980s. OPCS Series DS No 9. OPCS: London; 1990. Hillier S. The health and health care of ethnic minority groups. In: Scambler G, editor. Sociology as applied to medicine. London: Balliere Tindall; 1991. Nazroo J. The health of Britain’s ethnic minorities: Findings from a national survey. London: Policy Studies Institute; 1997. Bissell P, Morgall-Traulsen J, Haugbolle LS. How relevant is Marxist sociology to pharmacy practice? Int J Pharm Pract 2002;10:127± 40. Grimsley M, Bhat A. Health. In: Bhat A, Carr-Hill R, Ohri S, editors. Britain’s black population. Aldershot: Gower; 1988. Department of Health and Social Security. Inequalities in health: report of a research working group (the Black report). London: Stationery Office; 1980. Independent inquiry into inequalities in health (the Acheson report). London: Stationery Office; 1998. Marmot MG, Adelstein AM, Belusu L, and OPCS. Immigrant mortality in England and Wales 1970±78: Causes of death by country of birth. London: Stationery Office; 1984. Rogers RG. Living and dying in the USA: socio-demographic determinants of death among blacks and whites. Demography 1992;29:287±303. Sterling T, Rosenbaum W, Weinkam J. Income, race and mortality. J Natl Med Assoc 1993;85:906± 11. Davey-Smith G, Blane D, Bartley M. Explanations for socio-economic differentials in mortality: evidence from Britain and elsewhere. Eur J Public Health 1994;4:131±44. Gupta S, de Belder A, O’Hughes L. Avoiding premature coronary heart deaths in Britain: spend now on prevention or pay later for treatment. BMJ 1995;311:1035± 6. Beishon S, Nazroo J. Coronary heart disease: Contrasting the health beliefs and behaviours of South Asian communities in the UK. London: Health Education Authority; 1997. Nazroo J. Mental health and ethnicity: findings from a national community survey. London: Policy Studies Institute; 1997. Hall S. New ethnicities. In: Donald J, Rattansi A, editors. Race, culture and difference. London: Sage; 1992. Rattansi A, Westwood S. Modern racisms, racialised identities. In: Rattansi A, Westwood S, editors. Racism, modernity and identity. Cambridge: Polity Press; 1994. Harrison G, Owens D, Holton A, Neilson D, Boot D. A prospective study of severe mental disorders in AfroCaribbean patients. Psychol Med 1988;18:643± 57. Pilgrim D, Rogers A. A sociology of mental health and illness. Buckingham: Open University Press; 1993. Westwood S. Racism, mental illness and the politics of identity. In: Rattansi A, Westwood S, editors. Racism, modernity and identity. Cambridge: Polity Press; 1994. Solomos J, Back L. Conceptualising racisms: social theory, politics and research. Sociology 1994;28:143±61. Bowler I. They’re not the same as us: midwives’ stereotypes of South Asian descent maternity patients. Sociol Health Illness 1993;15:127± 78. 196 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 The International Journal of Pharmacy Practice, September 2003 Ahmad WIU. The trouble with culture. In: Kelleher D, Hillier S, editors. Researching cultural differences in health. London: Routledge; 1996. Frankenberg R. White women, race matters: The social construction of whiteness. London: Routledge; 1993. Lambert H, Sevak L. Is cultural difference a useful concept? In: Kelleher D, Hillier S, editors. Researching cultural differences in health. London: Routledge; 1996. Kelleher D. A defence of the use of the terms `ethnicity and culture.’ In: Kelleher D, Hillier S, editors. Researching cultural differences in health, London: Routledge; 1996. Johnson MR, Cross M, Cardew SA. Inner city residents, ethnic minorities and health care. Postgrad Med J 1983;59:664± 7. Modood T. Ethnic minorities in Britain: diversity and disadvantage. London: Policy Studies Institute; 1997. Balarajan R, Yuen P, Raleigh VS. Ethnic differences in general practitioner consultations. BMJ 1989;299:958± 60. Wolinsky F, Aguirre B, Fann L, Keith V, Arnold C, Niederhauer J, et al. Ethnic differences in the demand for physician and hospital utilisation among older adults in major US cities: conspicuous evidence of considerable inequalities. Millbank Q 1989;67:412±49. Rudat K. Black and minority ethnic groups in England: Health and lifestyles. London: Health Education Authority; 1994. McCormick A, Rosenbaum M. 1981±1982 morbidity statistics from general practice. Third National Study: Socio-economic Analysis, Series M85, No 2. London: Stationery Office; 1990. Benzeval M, Judge K. The development of populationbased need indicators from self-reported health care utilisation data. London: King’s Fund Institute; 1993. Cooper H, Smaje C, Arber S. Use of health services by children and young people according to ethnicity and social class: secondary analysis of a national survey. BMJ 1998;317:1047± 51. Watt I, Howel D, Lo L. The health care experience and health care of the Chinese: a survey based in Hull. J Public Health Med 1993;15:129±36. Donaldson L. Health and health status of elderly Asians: a community survey. BMJ 1986;293:1079± 82. Atkin K, Cameron E, Badger F, Evers H. Asian elders health knowledge and future use of community social and health services. New Community 1989;15:656± 8. McFarland E, Dalton M, Walsh D. Ethnic minority needs and service delivery: the barriers to access in a Glasgow inner-city area. New Community 1989;15:405±415. Hek G. Contact with Asian elders. J District Nurs 1991;December:13±5. Firdous R, Bhopal R. Reproductive health care of Asian women: a comparative study with hospital and community perspectives. Public Health 1989;103:307± 15. Bradley S, Friedman E. Cervical cytology screening: a comparison of uptake amongst `Asian’ and `non-Asian’ women in Oldham. J Public Health Med 1993;15:3079± 84. Jain C, Naryan N, Naryan K, Pike L, Clarkson M, Chaterjee J. Attitudes of Asian patients in Birmingham to general practitioner services. J R Coll Gen Pract 1985;35:416± 8. Atkin K, Rollings J. Community care in a multi-Racial Britain: a critical review of the literature, London: Stationery Office; 1998. Rashid A, Jagger C. Attitudes to perceived use of health care services among Asian and non-Asian patients in Leciester. Br J Gen Pract 1992;42:197± 201. 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 Ahmad WIU, Kernohan E, Baker M. Patients choice of general practitioner: influence of patients’ fluency in English and the ethnicity and sex of the of the doctor. J R Coll Gen Pract 1989;39:153±5. Ahmad WIU, Kernohan E, Baker M. Patients’ choice of general practitioners: importance of patients’ and doctors’ sex and ethnicity. Br J Gen Pract 1991;41:330±1. Donovan J. We don’t buy sickness: it just comes. Aldershot: Gower; 1986. Wright C. Language and communication problems in an Asian community. J R Coll Gen Pract 1983;33:101± 4. Bhopal R. Asians’ knowledge and behaviour on preventative health issues: smoking, alcohol, heart disease, pregnancy, rickets, malaria prophylaxis and surma. Community Med 1986;8:315± 21. Bowes A, Domokos T. South Asian women and health services: a study in Glasgow. New Community 1993;19:611± 26. Currer C. Concepts of mental well and ill-being: the case of Pathan mothers in Britain. In: Currer C, Stacy M, editors. Concepts of health, illness and disease. Oxford: Berg; 1986. Howlett B, Ahmad WAI, Murray R. An exploration of white, Asian and Afro-Caribbean peoples’ concept of health and illness causation. New Community 1992;18:281±92. Manderson L. Hot-cold food and medical theories: overview and introduction. Soc Sci Med 1987;25:329± 40. Taylor J. Health behaviour and beliefs. Health Visitor 1991;64:223± 4. Anderson J, Elfert H, Lai M. Ideology in the clinical context: ethnicity, ideology and re-structuring of life. Sociol Health Illness 1989;33:253± 78. Greenhalgh T, Helman C, Chowdury M. Health beliefs and folk models of diabetes in British Bangladeshis: a qualitative study. BMJ 1998;316:978± 83. Kelleher D, Islam S. How should I live? Bangaldeshi Asians with NIDDM. In: Kelleher D, Hillier S, editors. Researching cultural differences in health, London: Routledge; 1996. Bush J, White M, Kai J, Rankin J, Bhopal R. Understanding influences of smoking amongst Pakistani adults: community based qualitative study. BMJ 1993;326:962± 8. Jesson J, Sadler S, Pocock R, Jepson M. Ethnic minority consumers of community pharmaceutical services. Int J Pharm Pract 1995;3:129±32. Lakhani N, Johnson M, Cannaby AM, Farooqi A. Role of the community pharmacist in health care needs for members of the south Asian minority ethnic population. Int J Pharm Pract (Suppl) (accepted for publication). Agrawal M, Sause R. Serving multi-ethnic populations: measuring the cultural sensitivity of pharmacy interns in New York City. Int J Pharm Pract 1998;6:83±90. Cannon-Bobb ML. Cultural considerations in the provision of pharmaceutical services. Acta-Farmaceutica-Bonaerense 2001;1:33±49. Norris P, Simpson T, Bird K, Kirifi J. Understanding of pharmacy-related terms among three ethnic groups in New Zealand. Int J Pharm Pract 2001;9:269±74. Morgan M. Perceptions and use of anti-hypertensive drugs among cultural groups. In: Williams SJ, Calnan M, editors. Modern medicine : Lay perspectives and experiences. London: UCL Press; 1996. Thorogood N. Caribbean home remedies and their importance for black women’s health care in Britain. In: Beattie A, Gott M, Jones L, Sidell M, editors. Health and well-being: A reader. Basingstoke: MacMillan; 1993. September 2003, The International Journal of Pharmacy Practice 95 96 97 98 99 100 101 102 Aslam M, Healy M. Compliance and drug therapy in fasting Moslem patients. J Clin Hosp Pharm 1986; 11:312±25. Healy M, Aslam M. Lead containing preparations in the Asian community Ð a retrospective survey. Public Health 1986;100:149± 51. Rashid A, Jagger C, Goodyer L. Views of Asians and nonAsians on sources of drug information and the desirability for medication to be made available over-the-counter. Br J Gen Pract 1996;46:609± 10. Charupatanapong N, Cheung HC, Hunyhn JH, Nguyen KV. Self-medication practices of the low income Asian elderly. Am Assoc Coll Pharm Annual Meeting (abstracts) 1996;143:53. Platts A, Tann J, Chishti Z. Ethnic minority pharmacy practice. Int J Pharm Pract 1997;5:72±80. Joshi MS. Adherence amongst ethnic minorities: the case of South Asians in Britain. In: Myers L, Midence K, editors. Adherence to treatment in medical conditions. London: Harwood Medical Publications; 1998. Bissell P, May CR, Noyce P. From compliance to concordance: meeting the needs of patients? Int J Pharm Pract 2001; 9 Suppl:R7. Bissell P, May C, Noyce P. From compliance to concordance: barriers to accomplishing a re-framed model of health care interactions. Soc Sci Med (in press). 103 104 105 106 197 Bashir A, Asif M, Lacey FM, Langley CA, Marriott JF, Wilson KA. Concordance in Mulsim patients in primary care. Int J Pharm Pract 2001;9 Suppl:R78. Hassell K. White and ethnic minority pharmacists’ professional practice patterns and reasons for choosing pharmacy. Int J Pharm Pract 1996;4:43±51. Hassell K, Noyce P, Jesson J. White and ethnic minority self-employment in retail pharmacy in Britain: an historical and comparative analysis. Work, Employment Society 1998;12:245± 71. Platts A, Tann J. A changing professional profile: ethnicity and gender issues in pharmacy employment in the UK. Int J Pharm Pract 1999;7:29±39. Further reading Davey Smith G, Charlsey K, Lambert H, Paul S, Fenton S, Ahmad WIU. Ethnicity, health and the meaning of socio-economic position. In: Graham H, editor. Understanding health inequalities. Buckingham: Open University Press; 2000. Navarro V. Race or class or race and class. Int J Health Services 1989;19:311±4.