Introduction
National data suggest that Bangladeshis in Britain are three to four times more likely to suffer from ill health than the general population (2001 Census), five times more likely to get Type 2 diabetes mellitus and 50% more likely to develop heart disease (Balarajan, Reference Balarajan1995; Nazroo, 1997). Its status as one of Britain’s most socio-economically deprived communities (Atri et al., Reference Atri, Falshaw, Livingstone and Robson1996; Modood and Berthoud, 1997) is thought to be an important factor in this picture of ill health (Roper et al., Reference Roper, Bilous and Kelly2001) (Box 1). Within the age group of 50–74 years, Bangladeshis are the most likely to describe their health as poor and visit general practitioners (GPs) twice as frequently as the average UK citizen (Gillam et al., Reference Gillam, Jarman, White and Law1989; Rashid and Jagger, Reference Rashid and Jagger1992; Nazroo, 1997; 2001 Census).
Box 1 Background to the Bangladeshi community in Britain
Young men from Bangladesh (then East Pakistan) first started settling in Britain in the mid-1950s, mainly from the poor rural district of Sylhet. They were often seamen, who took jobs as kitchen workers in London hotels. The temporary nature of their stay, together with difficulties in competing with indigenous workers who had the benefits of education and language, resulted in poorly paid jobs and a low standard of living. Wives and children joined in the early 1980s, once it was recognised that it would not be practicable to move back to Bangladesh permanently. These families came with no knowledge of English. Basic schooling and integration into mainstream British life has been a tremendous struggle for many.
The Bangladeshi community makes up 0.5% of the total population of the UK. There are approximately 8000 people in the Bangladeshi community in Cardiff, mainly employed in the restaurant trade (2001 Census data). The practice used to approach patients for this study had 311 adult patients registered from the Bangladeshi community, from a total list size of 5000 patients.
Bangladeshis report dissatisfaction with GPs, and there is evidence that the health promotion advice given to them should be delivered in ways that are more accessible and culturally appropriate (Hoare et al., Reference Hoare, Johnson, Gorton and Alberg1992; Naish et al., Reference Naish, Brown and Denton1994; National Surveys of NHS patients, 1998; Leedham, 2000). Professionals cite communication and cultural barriers, lack of understanding of the system and unrealistic demands by patients, resulting in inappropriate use (Hawthorne et al., Reference Hawthorne, Rahman and Pill2003). In planning cross-cultural health education, the need for prior systematic cultural assessment has been stressed in order to ensure its relevance and acceptability, and increase the likelihood of positive outcomes (Nolde and Smillie, Reference Nolde and Smillie1987). Even where culturally tailored health education programmes are introduced, if there has not been evaluation of the target community, the outcomes can be mixed, with poor attendance and marginal effects on health status (Griffiths et al., Reference Griffiths, Motlib, Azad, Ramsay, Eldridge, Feder, Khanam, Munni, Garrett, Turner and Barlow2005).
This study aimed to explore the barriers to engaging with the outside world from the point of view of the Bangladeshi community in Cardiff, and to identify the important decision makers of health in these families. It was part of the scope for a larger project to design culturally appropriate diabetes health education and advocacy for the community, and so particular interest was taken in the meal patterns and contents of meals taken by families. The approach taken here, although informed by social anthropology and qualitative methods, is a practical and pragmatic example of what is possible within a primary care setting and by a practice-based team.
Participants and methods
Design of the methodology was informed by recommendations from seminal papers on qualitative research within minority ethnic communities (Currer and Stacey, 1986; Hennings et al., Reference Hennings, Williams and Haque1996; Hillier and Rahman, Reference Hillier and Rahman1996; Greenhalgh et al., Reference Greenhalgh, Helman and Chowdhury1998), and by the grounding of the study in a primary care setting, with practical implications for resources and approach. It was based on qualitative one-to-one interviews with participants, and needed to include aspects that valued the culture and beliefs of the community. As part of this approach, community leaders from the local Bangladeshi community were invited to give their advice and opinions on the acceptability of the study. They stressed the importance of finding a credible interviewer who would be culturally acceptable (for example, a female interviewer for female participants) and be able to speak Sylheti fluently (the dialect spoken by most living in this community). They also helped in the recruitment and appointment of a graduate Sylheti-speaking researcher (JC).
An age-stratified sample (n = 20) from the Cardiff Bangladeshi community, registered with a group practice, was invited to participate in semi-structured one-to-one interviews. Consent to participate in the study and allow the interviews to be audiotaped was obtained by our researcher (JC) with initial approaches being made by the practice health visitor (HV) or one of the GPs who spoke Bengali. The sampling followed a sequence, beginning with eight mothers with young children, identified by the HV. Young mothers were visited at home by the HV and JC together, to explain the purpose of the study. They were also given an information letter in Bengali and English, so that they could discuss it with their families (Sylheti has no written form). Further interviews were conducted with four young fathers, four ‘older’ women and four ‘older’ men. We defined ‘older’ as meaning men and women with married children (and young grandchildren) living in Cardiff. The practice did not follow a personal list system, so participating patients could have previously seen any of the four GPs in the practice. Six of the participants had Type 2 diabetes, but they were not specifically chosen for this.
Not everyone was literate enough to give signed consent, but in these cases JC made verbally sure the participant understood the purposes of the study and was willing to take part (Hennings et al., Reference Hennings, Williams and Haque1996). Once informed consent had been given, an appointment time at the patient’s home or a nearby health centre was arranged.
Interviews were semi-structured and based on a discussion proforma, lasting for about 40 min and were in the language of the subject’s choice (Appendix 1). If subjects refused to be audiotaped, notes were kept during the discussion and compared with additional observations of the conversation as soon as possible afterwards (Currer and Stacey, 1986).
Audiotapes were translated independently by JC and a Sylheti-speaking linkworker. Any disagreements about the translated meaning of a phrase were settled by calling up the subject for clarification, although this was only necessary on two occasions (the respondents’ consent having been obtained at interview for this). Translations were made as literally as possible, presenting participants in their own terms and speaking patterns. Transcripts were entered into the N*DIST programme and coded according to emerging themes as identified by KH and JC. Early interview transcripts were read for these themes and new ideas raised by participants were introduced into subsequent interviews for confirmation or otherwise (iterative process). They were also read by LP and RP to look for overarching themes and connections. Initial meetings of the researchers discussed the coding strategy, the detailed description of the data and further ways the data could be explored to bring out more themes. Later meetings covered the connections between themes, and discussions of explanations why certain findings predicted difficulties in using the current primary care services as they stood.
The study was submitted to Bro Taf Local Research Ethics Committee (LREC) and approved. The researchers were very aware of the cultural aspects of the ethical issues involved and took every care to follow them (see informed consent and translation issues mentioned previously) (Prideaux and Rogers, Reference Prideaux and Rogers2006).
Findings
Two younger women, who initially showed interest in the study at one of the HV visits, later withdrew by telephone, after discussing it with their husbands. Our understanding was that they decided the study would be too personal and they did not want to reveal details of their family lives. They did not sign consent forms or get to the stage of being included in the study. Overall, 20 people were entered into the study (Table 1 shows the demographic details of participants).
Table 1 Demography of participants
The data fell into two main themes: family structure and behaviour, and health beliefs, experiences and opinions of the local health care services. A main theme identified was the role of women in this community, and the barriers they faced in accessing primary care.
Data sources have been coded according to the category of the person being interviewed (L = young lady, OL = older lady, M = young man, OM = older man).
Family structures
Individual family circumstances dictated family structure (Box 2), but living in extended family situations was not necessarily the norm. Three of the eight younger women interviewed lived in nuclear families, four lived with relatives in Cardiff and one had a mother-in-law who came to stay often. The splitting of families may account for the close contact maintained with the parent country, as the welfare of children and property left behind was often mentioned and closely monitored.
Box 2 Family structure influenced by migration
OL1 had waited to come to Britain for 10 years. Her brother-in-law in Bangladesh was responsible for her family and at one point he was looking after three related families while their menfolk were setting up in Britain: ‘my brother-in-law in Bangladesh is a fine person’ she said.
OL3 had come alone to Britain, living with a cousin for four years until her husband and five of their children were allowed to join her. Two children were still in Bangladesh, and their father had just returned to arrange the marriage of his oldest daughter.
The older generation had a dominant role in families, were heavily involved in choosing spouses for their children and described a family life of mutual respect and duty (OL1). Mothers-in-law had a powerful role, especially if living in an extended family situation.
Daughters were more likely to get married and leave the parental home, often taking on the responsibility of looking after their in-laws (L5). A number of young women had moved out of the in-laws’ house into a nuclear family situation, and although this caused some discord within the family, they held out for separate housing (Box 3). Some challenged advice given by the older generation (Box 4).
Box 3 Attitudes towards living in an extended family situation
L4 ‘Like my sister, she got married somewhere else so she’s gone. They like me being in the family. Because I got married I could go, but I stayed there even though I was married so yes they were pleased.’
L2 ‘It’s nice to have a house of your own with your own children, rather than living with someone. You want your own privacy don’t you.’
L8: ‘I know some of my friends are doing what they want with their husbands but, some of them are like stuck at home with in-laws. Even if she has to make something she has to think about it before she makes it and there’s lots of them…’ ‘If I had a mother-in-law, she wouldn’t like me going out’. ‘We want to go out and have a nice time with our husbands just me and my husband no one else but, they think: sit together, everyone together, but, we want privacy of our own’. ‘Yes, they think: look at him doing everything for his wife and he’s forgetting about his mother and he’s not, he’s doing his bit.’
Box 4 Generational differences
L4: ‘We eat these things in hospital and nothing happened.’
L1: ‘Now the things we know about the older generation didn’t have the opportunities to know about. The way we understand now, people in the past didn’t. …There is a lot of difference between them and us.’
L8: ‘They used to say you shouldn’t do this, you shouldn’t do that, nowadays we don’t follow all those traditions. They’ve got a different mind. What we like now, they wouldn’t like. What we think is right they’ll think no that’s wrong.’
M4: ‘Young people here become more Westernised while the parents are still far behind and have different opinions on a lot of matters. For example, the parents take their children back to get married in Bangladesh, and we now have a lot of marriages breaking up in the community.’
Young married women tended to stay at home and did not go out unaccompanied. Shopping was largely done by men (noted also in East London by Hillier and Rahman (Reference Hillier and Rahman1996)), and women did not go to the mosque, so their life outside the home consisted of visiting in-laws, friends nearby, shopping with their husbands and occasional prolonged visits to Bangladesh. The Imam in the sample (M3) gave his reasons why women were not expected to go to the mosque (Box 5). (An Imam is an educated religious cleric who, among other duties, conducts prayers and the affairs of the mosque.)
Box 5 Imam transcript detailing reasons why women don’t go to mosque
‘Our religion doesn’t allow men and women to pray together, it used to be allowed during the time of the prophet Mohammed now it’s been forbidden. It was stopped at the time of the second Khalifa of Uamr Rhadiallah. Three out of the five prayers are in the night. When we pray we can’t leave spaces between us so if a woman is standing next to a man they are bound to touch each other. A boy has to pray from the age of 12 he could be praying next to a woman. If a young man touches a young woman they could get wrong ideas and bad influences, that’s why it’s not allowed. If all the women all get together and want to pray separately then that’s acceptable. It is not compulsory for the women to pray in the ‘Jamath’ in mosque like it is for men. They could read at home there is not really any need for them to go to the mosque.’
In fact, there are different schools of thought and cultural practices attached to the practice of different Imams. It is common for Bangladeshi women not to go to a mosque at all, whilst Pakistani women often pray at the mosque and can be involved in mosque activities (for women and children).
As a result, some young women led very sheltered lives. They felt it was not appropriate to discuss personal issues, and conversations with friends and relatives centred around their children and films (L1). L5 said: ‘my father was getting me married so I had to be happy’. When her father died in Bangladesh some time later she felt very unhappy, but was not able to talk about her feelings to her in-laws and had no friends she could talk to so intimately – ‘if you talk to someone you feel lighter’. She found there was no safe place to unburden her feelings or talk about intimate matters.
Housework and childrearing and the domestic division of labour
Nearly all the housework and cooking for the family was done by the younger women in the household, who cooked from basic ingredients every day. Men did not get involved, and the older ones consistently said that issues such as healthy nutrition for babies and pregnant women had nothing to do with them. ‘You should talk to the mother of the children about these matters. I just buy them things when they need something, take them out now and then, that’s it’ (M4, OM2). However, as the main shoppers, they often chose the food for the family, sometimes being given a list, ‘sometimes she would tell me what to get but I know from the freezer being empty the things I need’ (OM4).
It was not considered seemly for a young woman to go shopping in the local Bengali shops: ‘some ladies don’t go because there are a lot of Bengali men, they think it is not respectful, so the man goes … I think in a way it’s wrong and then again I have to see what he says … as you go in some people look at you, you know…I can go shopping in Tesco’s’ (L8). However, this seemed to be a cultural rather than a religious rule, as explained by M3: ‘I think women should go because they can buy what they need as long as they are covered. You can show your face and hands there isn’t a problem with that. I say to her ‘don’t say I didn’t give you anything after’, I would pay for what she needs’.
The key players in day-to-day decision making (for example, what to buy, where to go with the children, if they should be taken to Baby Clinic for vaccinations) depended on family circumstances and who else was living in the house. In three-generation families, the older members had a greater say than the younger ones (L4, L5). Five of the eight young women interviewed said decisions were made by the male head of the household. In some cases (L2, L8) there was room for discussion, but husbands were still in control (Box 6).
Box 6 Decision making in the family
L8: ‘See I was living with my in-laws, I did live with them a year, I wouldn’t make a decision, they would, but, because I haven’t got no one here, just me and my husband I can’t just make all the decisions he’s got to make them as well. So, say I wanted to do something with the children I can’t do it myself I’ve got to ask him first, shall I do this or shall I do that. I’ve got to ask him and he if he says yes you can go, I go, otherwise I don’t. But, if I was living with my in-laws I wouldn’t make the decisions they would have to do it, everything.’
L7: ‘He will sit down and say I am telling you we need this for our children.’
Although young fathers tried to be involved in the day-to-day activities of the family, late nights and work sometimes made this difficult (Box 7). It also meant that the young mothers at home were unable to interact with outside events as their decision-maker was not there.
Box 7 Bangladeshi men in Britain
M4: ‘in Bangladesh the husband is still in control of the baby’s growth and health. Here it’s the mother because the timing of our profession which is mostly restaurant work, hardly leaves us with any time to take care of them. Fathers hardly see their children. It’s a hard life here but then its all just show in Bangladesh. We never work so hard like this in Bangladesh. When we go back from here we drive around like a king. Men that are coming to this country after they get married are struggling because they expect something different.’
L7: ‘he doesn’t know much about kids actually. Most of the time he’s at work.’
Meal patterns
Most families followed meal patterns dictated by the ‘restaurant hours’ of the breadwinners. Breakfast was a choice of cereals, bread, biscuits and tea. The main meal of the day, consisting of rice with two or three curries, was at 3–4pm, after men had woken following their night shifts. The timing of this meal was also affected by school finishing and prayer times. While a rice-based diet seemed universal among the people we interviewed, many ate it from habit and convenience (easier and quicker to make than chapatti, said L7). Some people felt that they had eaten so much rice in their lifetimes that cutting down would cause withdrawal symptoms (Box 8) (Greenhalgh et al., Reference Greenhalgh, Helman and Chowdhury1998).
Box 8 Comments about importance of rice in the diet
OM1: ‘If I don’t eat rice it feels as though I haven’t eaten anything. Being Asian you have to have it from childhood.’
OM2: ‘The body gets used to the strength from rice everyday.’ ‘We eat rice from birth therefore our body gets used to diet with rice.’
M3: ‘You could become weak without rice. A Bengali is made of rice and fish.’ (This is apparently a Bangladeshi saying.)
Husbands left for work after this meal, and the children had a smaller meal at around 7pm – pizza, fishfingers and chips, spaghetti, pakoras, eggs and biscuits were mentioned. Mothers were aware these foods had a high fat content, and were sparing with their use. Desserts were not eaten, but people enjoyed mangoes when in season, and apples, bananas and grapes. They drank tap water, tea and fizzy orange.
None of the young women ate a full meal in the evenings. Men ate in the restaurant at about 9pm and again in the early hours of the morning. Older people were much less likely to eat Western foods, and most people interviewed would never consider trying a Chinese or other ‘exotic’ meal. Paan (made from betel nut and betel leaf) was eaten by all sections of the sample, especially the older generation. It was mostly eaten to round off a meal – ‘the whole inside of the mouth feels lighter after eating paan’, ‘…it’s the taste. The English they have a ‘Tune’ after food.’ (Tunes are a sweet-based product widely advertised in the Western press for unblocking sinuses) (L7). The women and younger men in the sample did not smoke, although the older men did. No one drank alcohol.
Six of the eight young mothers and all of the older women interviewed felt that breast milk was best for the baby, although none had fed beyond three months and most had stopped within a few days. L2 feared her children might acquire her diabetes through breast milk. L5 was the only one to have a full understanding of the health benefits, including protection from infection. Babies were weaned at 4–6 months of age, starting with prepared baby foods and baby rice. Some babies were first given ‘jao’, a traditional salty rice pudding. These foods were supplemented with home cooking, soft, well-cooked lentils, boiled vegetables and boiled egg. Small children ate Bengali food; mothers either washed pieces of cooked meat to reduce the spiciness (L7) or would cook separately for them with less spices and oil (L2).
Traditional practices and health beliefs
People generally held views that were very similar to ‘Western’ health beliefs and were happy to accept explanations and practice, with regard to the aetiology of viral respiratory infections, and safety of prescribing during pregnancy, echoing previous findings in the Pakistani community of Glasgow (Bhopal, Reference Bhopal1986). They often cited advice from their HV, practice nurse and GP (L9). Many people held views that were both Western and traditional; too much oil and sugar in the diet is unhealthy, and also that eating bananas if you had flu was bad. Younger women did not follow traditional practices, but recounted advice from older female relatives. For example, L1 had been told that children and laundry should be brought in by early prayers (magrib, or sunset), and that Saturdays were inauspicious for children to go out. L10 was told to pray regularly in pregnancy, move slowly, not wear high-heeled shoes and avoid violent or sad movies on television. She was told to eat chicken, fruit and drink milk, and to avoid sour or bitter foods and eggs.
Two of the older men and one of the older women, who had diabetes, were taking traditional remedies in addition to their tablets. These were based on dried leaves and herbs easily available in Bangladesh, usually very bitter or sour in taste. OM4 grew his herbal remedies at his home in Bangladesh and brought them over. He believed that they burnt fat and cleaned the blood. Some people would purchase a ‘taveez’ from the Imam – a religious tract from the Koran with special significance for the person that would protect them and heal them from illness. OL2 mentioned a practitioner called a ‘kobiraj’, who would give special oil to rub onto an affected part and water to drink that had been prepared by reading a verse of the Koran, and breathing over it. However, she said that the kobiraj could not help with diabetes, it was mainly useful for aches and pains, and she did not use one in Cardiff.
The Imam (M3), whose main role was in conducting the prayers at the local Bengali mosque, teaching Islamic studies to children both at the mosque and at their homes, also found himself being called out for health problems: ‘they feel brave if we go to see them. If we go we pray for them as well as giving them more courage. We say you should stay under control, go to the health centre and see your doctor. If you have asthma, don’t go out in the cold, try not to catch a cold. People believe us, they have faith in us.’
Sources of advice on health issues
Most people said that they would go to the doctor if they had any queries about their health, ‘if you talk to anyone, they say ‘see the Dr I know there’s nothing wrong but, just go and have a check up or something’ (L8). They had little knowledge on how to access other services, and dependence on the doctor’s advice appeared to be due to a need for reassurance that nothing was about to happen, or going to get worse. Some younger women found it less threatening to speak to the HV first. Others would consult their older female relatives.
Experiences of healthcare
People were generally happy with the healthcare they had received, although many commented on the difficulties in getting appointments. These difficulties centred around access to a doctor for acute problems as well as co-ordinating appointments with the availability of men to take their wives and children to the surgery. Older respondents feared getting into a situation of being very ill, with no one to translate for them or to call for help. However, there were many comments that the primary care clinical staff treated them well and gave personal service.
Routine care, as opposed to emergency care, was not well attended – L7 said she had not been told about ante-natal clinics and another mother had just had her third baby but did not know what the post-natal checkup was for. The HV came to her at home, and she was happy with that. (In fact, the HVs in this practice had taken on a considerably heavier home-visiting workload since they had recognised that this was the most productive way to build relationships with mothers and to get babies regularly checked.) This mother said she had no problems with appointments and wanted her children to have their immunisations, but the baby clinic time interfered with her housework and her daughter’s nap time.
People were aware that administrative staff did not really see them as individuals and Bangladeshis, although they were keen to tell them who they were. M3 said ‘I tell them and introduce myself as a Bengali. I like that.’ OM4 said ‘I feel proud being a Bengali, people should ask about origins and background, otherwise they wouldn’t know me’. L8 said ‘I’d rather them asking than ignoring. I’d rather them to know who I am’. The effect of having to take family members along to translate was mentioned as a major problem: several felt there ought to be a Sylheti-speaking receptionist available. OM4 recounted how a conversation with a Pakistani Urdu-speaking doctor in secondary care had persuaded him to stop smoking – ‘he is our own people’.
Discussion
This work was a qualitative exploratory study, providing a snapshot of family life in the Bangladeshi community in Cardiff. It should be of interest and relevance to British GPs and primary care services working with Bangladeshi communities, as it provides vivid portrayals of the range of family dynamics and the ways they affect peoples’ interaction with the outside world. It also has relevance to GPs working with other minority communities as an example of what is possible in a non-research environment by a ‘journeyman’ GP.
There were a number of difficulties encountered during the study period, the main one being the reluctance of various sections of the community to talk about themselves. Reasons for this included the unfamiliarity of the situation, reluctance to talk about personal matters and, for women, the unusual situation of taking a lead role in conversations with professionals or outsiders. Box 9 outlines the main learning points from the methodology we developed for working with this community. Experienced researchers with the necessary language skills can be difficult to find outside the main conurbations in Britain, as communities are smaller in number and the choices become less.
Box 9 Important learning points from the methodology that aid working with minority communities
• Community leaders give valuable insights into acceptable ways of approaching community members and provide support to researchers. They can lend validity to a study in the eyes of the community.
• Take time to find the right research worker. In our case, a fluent Sylheti speaker with academic credentials was vital, and her gender and marital status made her acceptable to participants of both genders.
• Approaching potential participants through valued, known professionals increased uptake for the study.
• Provide information about the study in the languages read by the participants and give them a choice.
• Be flexible about venues and times for conducting interviews. Allow participants to bring relatives and/or friends if they wish, even though the study is designed for 1:1 interviewing.
The limitations of the study are the relative inexperience in qualitative interviewing of the Sylheti-speaking researcher ((JC), the reasons and justification for which has been discussed already), and the use of a single practice for sampling patients. This was a pragmatic decision, as the wider research programme was centred on this practice, which was one of the few in Cardiff at the time with both a large proportion of Bangladeshi patients and an ethnic profile of its patients. Patients had a variety of ethnic and gender choices in deciding which doctor they went to, and experienced communication and cultural barriers first hand.
It was interesting to discover what people felt appropriate to say to a stranger, in what was a novel and strange situation for most, if not all, of them. No one withdrew from the study after giving formal consent, although one woman rang later to get assurance of anonymity, which implies that she felt some concern about the personal nature of her conversations and worried they might be traced back to her.
The consensus on health and illness issues within the sample and compared with the local community was striking. The studied community is dynamic and changing, with the younger generation trying to keep in with their elders and a traditional lifestyle, while adapting to a Western environment (Modood and Berthoud, 1997). As individuals are each at a different stage in this process, it becomes difficult to generalise about the community, but the data help us to understand how people live and relate to each other, and the context of their presentations in primary care. Once the outward differences are accounted for (language and communication, the reduced exposure of women to the outside world, and the timetable of mealtimes imposed by prayer times and working restaurant hours), the family aspirations, intergenerational communications and health beliefs of this Bangladeshi/Sylheti community in a circumscribed area of Cardiff are not dissimilar to those of the local community living around them. Most aspire to be home owners, to have regular employment, and to educate and bring up their children in a stable family environment, mindful of their cultural heritage. Nazroo (Reference Nazroo2003) has pointed out the unreliability of focussing purely on cultural or ethnic origins as explanations for why and how people live, giving evidence that the socio-economic position of families and their experiences of discrimination are also important in determining their life-course choices and lifestyles.
Once the barrier of language is removed, much of the reasons for apparent compliance problems (such as not being able to make appointments during surgery times, and not attending health promotion clinics because husbands are still at work) become understandable, as also pointed out by Kelleher and Islam (Reference Kelleher and Islam1996). We found that the similarities with the local community with regard to their relationship with primary care services outweighed their differences, a finding previously reported for a Pakistani community and reinforced here (Bhopal, Reference Bhopal1986). For example, the Bangladeshi understanding of food values was very similar, they were willing to accept Western-style health promotion advice, and placed great value on the roles of GPs, HVs, midwives and practice nurses. The use of traditional remedies, while it took place, was secondary to reliance on Western medicine.
Several distinct themes cut across this main theme
a) Firstly, the ‘protected’ role of young married women in the community and how this influences access to primary care. Their relative restriction to the home, low profile in family decision making (see also Talbani and Hasanali, Reference Talbani and Hasanali2000) and problems related to speaking in English if born abroad result in significant difficulties in receiving the same quality of attention available to other sections of the public. (Only 14% of Bengalis in 1996 were reported to have a survival level of competence in the use of English, with women, the elderly and those born outside the UK being particularly disadvantaged (Modood and Berthoud, 1997).) They are also more likely to be seen if accompanied by someone else. Not only does this result in physical constraints in accessing health care other than as ‘emergencies’, but also has implications for confidentiality and autonomy.
b) The older generation is likely to speak limited English, and value traditional customs such as deferring to elders, spending time together as extended families and strict observance of religious practice. They are anxious to be healthy, and worry how they would cope if they became ill. Despite difficulties of access, they place heavy reliance on the opinion and support of doctors (also reported by Greenhalgh et al., Reference Greenhalgh, Helman and Chowdhury1998), and are keen to be seen as Bangladeshis. The majority of younger people interviewed were either living in nuclear family structures or aspiring to do so, which belies the general expectation that South Asians live in extended family units. This is a finding that could be important to those who might otherwise assume that the extended family would support someone with special health needs (Badger et al., Reference Badger, Cameron, Evers, Atkin and Griffiths1989; Nazroo, 1997; Merrell et al., Reference Merrell, Kinsella, Murphy, Philpin and Ali2005).
c) Meal timings are governed by ‘restaurant hours’, which, together with prayer schedules and school pick-up times, result in a regimented day and difficulties making appointments during ‘normal’ surgery hours. Primary care teams need to consider the responses they can make to this situation, otherwise the current difficulties in access will continue.
d) The pastoral advice offered by Imams to individuals in their communities appears to be both timely and helpful. This is perhaps a health promotion resource that could be explored constructively at a local level, but it will be up to Primary Health Care Teams (PHCTs) to find and reach out to influential members of local communities.
Applicability to other Bangladeshi communities in Britain
A great deal has been written about the Bangladeshi community, but it should be noted that it is a community that is not homogeneous and is developing all the time. However, there are resonances with studies of Bangladeshi communities carried out elsewhere in Britain (Bhopal, Reference Bhopal1986; Hillier and Rahman, Reference Hillier and Rahman1996; Greenhalgh et al., Reference Greenhalgh, Helman and Chowdhury1998). Testing this knowledge at the grass-roots level is vital for putting research into practice. We suggest that such an approach is relevant not only to those in primary care working with Bangladeshi families but also to all those working within our increasingly multi-cultural, ethnically mixed communities.
Applicability to other primary care situations
When there is a mismatch between primary care and a community’s expectations of service delivery, a number of questions inevitably arise. Which ‘party’ should alter their behaviour? And how exactly is this to be achieved? Time pressures in busy clinical environments make it difficult to find out about communities and start the internal debate that is needed to enable modification of services to improve access and communication. Experience is often based on anecdote and opportunistic observation, which can lead to misleading conclusions about ‘norms’ if mainly ‘problem’ families are remembered (Hardy, Reference Hardy1998). It can be argued that primary care offers an ‘equal’ service to all, but our data reveal disparities in access to services as ‘inequalities in quality’ or to a form of ‘colour blindness’ (Szczepura, Reference Szczepura2005).
By avoiding issues of nationality, language needs and socio-economic background determinants of health status, primary care staff can continue to remain disengaged from the patients, who are seen as inaccessible foreigners rather than as individuals. It has been suggested that to be patient-centred, General Practice consultations should be based on the ‘lifeworld’ of the patient, rather than that of the health professional (Barry et al., Reference Barry, Stevenson, Britten, Barber and Bradley2001). In order to do this, we need an understanding of the everyday routines and practices that shape and structure the patients’ world. PHCTs are aware of some of the constraints on patients that affects the use of their services, but continue to call for patient education and changed behaviour (Hawthorne et al., Reference Hawthorne, Rahman and Pill2003). However, PHCTs have a responsibility to examine their own systems, learn about their patients and adapt accordingly. The data collected from this study have been fed back to some of the PHCTs in the area (as part of the Diversity Training offered by Cardiff Local Health Board within their ‘HeartLink’ Project). The challenge for PHCTs is to gather and use such information constructively, be prepared to change or adapt their systems and enter into a real partnership with patients.
Acknowledgements
Mrs Maya Ali (linkworker) for double translations, Mr M. Haque (Bangladeshi Association, Cardiff) and the four GPs and two Health Visitors at Riverside Health Centre, Cardiff, for support with the project. The project was funded by Bro Taf Health Authority.
Authorship: KH, RP and LP conceived and designed the work, KH and JC conducted the interviews and all authors were involved in the analysis and interpretation of the data. KH drafted the article and JC, RP and LP all contributed to its revision.
Appendix 1. Topic areas for discussion during interviews
Introduction and explanation of the work
Activities centred on food and eating – who does the shopping, planning, cooking and eating.
Attitudes towards maternal and child nutrition – eating in pregnancy, breast feeding versus bottle feeding, weaning practices, feeding school-age children.
Experiences of health care and health promotion in Britain, compared with Bangladesh.
Attitudes towards traditional beliefs and remedies, generational differences.
