Primary care

Understanding influences on smoking in Bangladeshi and Pakistani adults: community based, qualitative study

Judith Bush, ESRC/ODPM postdoctoral research fellow ^a, Martin White, senior lecturer in public health ^a, Joe Kai, professor ^b, Judith Rankin, senior research associate ^a, Raj Bhopal, Bruce and John Usher chair of public health ^c. 

^a School of Population and Health Sciences, University of Newcastle, Newcastle upon Tyne NE2 4HH, ^b Division of Primary Care, School of Community Health Sciences, University of Nottingham Medical School, Nottingham NG7 2UH, ^c Division of Community Health Sciences, Public Health Sciences Section, University of Edinburgh Medical School, Edinburgh EH8 9AG

Correspondence to: J Bush Judith.bush{at}ncl.ac.uk

	Abstract

Top Abstract Introduction Methods Results Discussion References

Objective: To gain detailed understanding of influences on smoking behaviour in Bangladeshi and Pakistani communities in the United Kingdom to inform the development of effective and culturally acceptable smoking cessation interventions.
Design: Qualitative study using community participatory methods, purposeful sampling, one to one interviews, focus groups, and a grounded approach to data generation and analysis.
Setting: Newcastle upon Tyne, during 2000-2.
Participants: 87 men and 54 women aged 18-80 years, smokers and non-smokers, from the Bangladeshi and Pakistani communities.
Results: Four dominant, highly inter-related themes had an important influence on smoking attitudes and behaviour: gender, age, religion, and tradition. Smoking was a widely accepted practice in Pakistani, and particularly Bangladeshi, men and was associated with socialising, sharing, and male identity. Among women, smoking was associated with stigma and shame. Smoking in women is often hidden from family members. Peer pressure was an important influence on smoking behaviour in younger people, who tended to hide their smoking from elders. There were varied and conflicting interpretations of how acceptable smoking is within the Muslim religion. Tradition, culture, and the family played an important role in nurturing and cultivating norms and values around smoking.
Conclusion: Although there are some culturally specific contexts for smoking behaviour in Bangladeshi and Pakistani adultsnotably the influence of gender and religionthere are also strong similarities with white people, particularly among younger adults. Themes identified should help to inform the development and appropriate targeting of smoking cessation interventions.

What is already known on this topic Smoking is common among Bangladeshi and Pakistani men in Britain but rare among the women Smoking is particularly common in Bangladeshi men Socioeconomic status is thought to influence smoking uptake in Bangladeshi men Influences on smoking in South Asians in Britain are poorly understood What this study adds Smoking among Pakistani and Bangladeshi men is strongly seen as socially acceptablea "normal" part of being a man Smoking in Bangladeshi men is more deeply socially ingrained than in Pakistanis, contributing to group cohesion and identity Smoking in Bangladeshi and Pakistani women is associated with a strong sense of cultural taboo, stigma, and non-acceptance Islam forbids addiction and intoxicants, but opinions differ on whether the Muslim religion allows smoking Culturally sensitive smoking cessation interventions for Bangladeshis and Pakistanis are needed

	Introduction

Top Abstract Introduction Methods Results Discussion References

Studies in the 1980s of combined heterogeneous South Asian populations in Britain suggested that smoking rates were similar to or lower than rates in the white population, ^{1 2} but recent surveys in the United Kingdom have shown that smoking is much more common among Bangladeshi men (49%) than among white (29%), Pakistani (28%) or Indian men (19%).^3-5 The rate is particularly high (56%) in Bangladeshi men aged 50-74.⁴ Cancer of the trachea, lung, and bronchus is the commonest cause of death from cancer in South Asian men, and the second commonest in South Asian women.⁶ Smoking is the principal risk factor for these cancers.

Sex and age differences in smoking rates in South Asian populations are marked. In South Asian women, smoking rates are reportedly low (Bangladeshi women 4%, Indian women 1%, Pakistani women 2%)⁴ though possibly underestimated.⁷ National smoking prevalence is strongly associated with socioeconomic status in Bangladeshi people living in the United Kingdom, although the association is less clear in the Pakistani population.⁵

Detailed understanding of attitudes, beliefs, values, and behaviours in relation to smoking in minority ethnic groups is lacking. Such understanding is necessary to inform development of smoking cessation strategies that are culturally appropriate for these communities.^8-10 We report community based, qualitative research (March 2000 to March 2002) that aimed to gain such insights in Bangladeshi and Pakistani communities.

	Methods

Top Abstract Introduction Methods Results Discussion References

Participatory approach
We used a community participatory approach previously developed successfully in these communities,¹¹ in which members of the Bangladeshi and Pakistani communities in Newcastle (box 1) participated in study development, implementation, and analysis. After community publicity, application, and interview, 13 bilingual "community researchers" (six men and seven women) were recruited from the local South Asian population and attended a 14 week, accredited training programme in qualitative research.

These researchers were responsible for organising, recruiting, undertaking, and translating in-depth interviews and focus groups, facilitated by JB. In discussion with the research team, the community researchers also developed interview topic guides, publicity for the study, and strategies for recruiting participants and contributed to data analysis.

Research methods
The community researchers held semistructured, in-depth interviews with 37 participants and 24 focus groups (with 104 participants). Interviews and focus groups were based on topic guides, translated into relevant South Asian languages by the community researchers. Twenty pilot interviews and focus groups took place to give the community researchers confidence, test the feasibility of recruitment techniques, and refine the topic guides. Topics discussed included smoking behaviour, views on what influences smoking, and understanding of how smoking affects health.

Research participants were sampled purposively from the local Bangladeshi and Pakistani communities on the basis of ethnic group (that is, Bangladeshi or Pakistani), sex, age, smoking status, and occupation. Both male and female smokers were recruited. Participants were recruited informally through existing community based religious and non-religious organisations, groups, and social networks by using a "snowballing" techniquewhereby a small number of informants put the researcher in touch with others, who then nominate friends, colleagues, and other contacts, and so on.¹⁴

The community researchers were "matched" as closely as possible to the participants in terms of language spoken, sex, and age.¹⁵ Single sex and ethnic focus groups were held to respect and increase cultural acceptability, and, as far as possible, focus groups contained participants of similar ages.

A total of 141 people (87 Bangladeshi and 54 Pakistani) aged 18-80 years participated. The table shows characteristics of the participants. Participants were broadly typical of the Bangladeshi and Pakistani populations of Newcastle, with slight under-representation of people aged over 50 and slight over-representation of people aged 18-29. 

Translation
Sixty per cent of focus groups and interviews were conducted in English, the rest in Punjabi or Urdu (for Pakistanis) or Bengali or Sylheti (for Bangladeshis). All interviews and focus groups were audiotaped and transcribed verbatim with participants' permission. The community researchers translated the tapes in Punjabi, Urdu, Bengali, and Sylheti into English. About a fifth of these translations were also sent to an independent translation agency. The two sets of translations were compared for consistency; no substantial differences in meaning were identified.

Analysis
We analysed the transcripts by identifying recurring, emergent themes using constant comparison of the interview transcripts. ^{16 17} Data generation and analysis continued until no new themes or ideas were emerging. JB led the analysis, with the community researchers and members of the research team reading a proportion of transcripts to agree a thematic framework to be used for coding, thus improving the reliability of the analysis. We used the NUD*IST 4 textual analysis software to help us to do this. To refine our interpretations, we discussed the analysis at a meeting with local community workers and organisations.

	Results

Top Abstract Introduction Methods Results Discussion References

Four highly inter-related themes were found to influence views on smoking: gender, age, religion, and tradition.

Gender
Smoking in men was viewed with a strong sense of social acceptance, social bonding, and tradition and was seen as a "normal" part of "being a man." This view was particularly strong among Bangladeshi participants. It was intimately bound up with notions of male identity. Macho and fashionable images were associated with smoking and reinforced by Indian films and popular media (box 2).

Many of the men described how they were more likely to smoke, and to smoke more cigarettes, when they were in the company of other smokers (box 2). A minority thought that the trend was now changing, however, that it was becoming more sociable not to smoke. For a minority of young men, smoking in peer groups was also linked with drinking alcohol.

Stress was also thought to influence smoking in men. Both men and women viewed Bangladeshi and Pakistani men as having stressful lives owing to pressures associated with being separated from family and to poorly paid work. Participants thought that Bangladeshi men working in the catering industry suffered particularly severe stress as a result of unsocial and long working hours in restaurants.

In contrast, participants often regarded smoking in Bangladeshi and Pakistani women with a sense of taboo, stigma, and non-acceptance, using such words as "bad," "labelled," "tainted," "shamed," and "disrespectful." Smoking was also perceived to affect the chances of a woman marrying. Women were regarded as having fewer opportunities to smoke, both culturally and economically (box 3).

However, some men and women (and especially the younger participants) held the view that the prevalence of smoking in young Bangladeshi and Pakistani women is increasing as they become more westernised, influenced by white women and peer pressure from white children at school. Most of the female smokers who participated in the study were under 30. Motivation for young women to smoke often centred on "rebellion" or expressing independence from family members (box 3). Although smoking in women was usually viewed as a covert activity, some younger Bangladeshi and Pakistani women seemed to smoke in peer groups (box 3).

Age and generation
Age seemed to influence the cultural acceptability of smoking. Because of elders' respected status in South Asian society the participants viewed it as more acceptable for older men and, to a lesser degree, older women to smoke openly. In contrast, smoking in young people tended to be regarded as "disrespectful," particularly in front of elders. Thus, smoking in young people tended to be hidden from older members of the community (box 4).

Elders who smoked were perceived to lack knowledge of the health effects of smoking and have a more fatalistic approach to life. Younger people were viewed as being more likely to smoke because of influence of peer pressure, image, and rebellion (box 4). However, if a participant's peer group comprised predominantly non-smokers, this sometimes resulted in that person being less likely to start smoking.

Religion
Unlike alcohol, tobacco is not specifically banned or prohibited in the Islamic faith. The Koran does prohibit intoxicants and addictions, however. Most participants agreed that it was religiously unacceptable to smoke in a mosque and that the potential for women to smoke was reduced as they have a protected status in the Muslim religion.

Participants held conflicting perspectives on how religiously acceptable it is for men to smoke and to what degree smoking is permitted in the Muslim religion. Many believed that, although smoking is not banned or prohibited, it does not fit comfortably within the Islamic religion, that it "isn't right" or is "makroo." This argument was used in two contradictory ways. Some used it to justify the need to ban smoking; others used it to justify the acceptability of smoking in the Muslim religion as long as the smoker was not "addicted to" or "intoxicated by" cigarettes (box 5). Non-smokers (particularly women) thought that many smokers who are addicted to cigarettes do not realise or admit to this.

Only a small number of participants felt that smoking was unacceptable in the Islamic religion because it damaged health (although the Koran states that Muslims should not allow their hands to contribute to their own destruction) (box 5).

Tradition, culture, and family
Tradition and culture also seemed to play a role in creating, perpetuating, and regulating cultural norms, acceptance, and "fashion" around smoking (box 6).

Views on why smoking levels in Bangladeshi and Pakistani men differed focused on socioeconomic differences and social disadvantage; the more recent arrival of Bangladeshis into the United Kingdom; stressful jobs and working long unsocial hours in unregulated environments (particularly restaurants); and the fact that smoking has traditionally been viewed as macho among Bangladeshi men.

The family was felt to be an important medium though which cultural norms and values associated with smoking were shaped and negotiated. Young boys often learnt to smoke by observing male elders smoking (box 6), whereas opportunities for women were limited owing to the cultural restrictions imposed on a Muslim woman by her parents. There were strong rules and standards surrounding what was expected of a "good family."

Our analysis also suggests generational changes in acceptability of smoking. Several participants described how their childrenwho had learnt at school about the risks to health from smokingcriticised their fathers for smoking (box 6). This had sometimes led to changes in behaviour, such as starting to smoke outside the house or in a different room from their children or trying to give up smoking altogether. In some families, however, criticism of elders was not acceptable.

Other traditional practices involving tobacco (smoking hookahs, Pakistanis; and chewing "paan" (betel leaf and areca nut), Bangladeshis) were viewed as being less common today among younger people. Cigarette smoking was often viewed as becoming the modern equivalent of smoking hookah for young people.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Although our results show some similarities with those of studies of smoking behaviour in predominantly white populations, ^{18 19} they also highlight important differences, particularly the influence of a person's gender and religion. The findings must be interpreted with regard to the characteristics of our sample and the participatory nature of the research. We had a broad range of participants in terms of age, sex, occupation, socioeconomic status, educational level, and smoking status. By working with members of communities at all stages of the research, we used our participatory approach to increase the validity of our findings.¹¹ The socioeconomic characteristics of Bangladeshi and Pakistani people living in Newcastle upon Tyne are broadly typical of Bangladeshi and Pakistani people nationally.¹³ Thus, although we must be guarded, our findings are likely to be generalisable to other Pakistani and Bangladeshi communities in the United Kingdom, although local variations reflecting differing regional origins within Bangladesh and Pakistan may exist.

Comparing Bangladeshi and Pakistani populations
Despite smoking being less common in Pakistanis than in Bangladeshis^3-5 we found few differences in beliefs or attitudes between these two groups. Differences in smoking levels between Pakistani and Bangladeshi men may be explained largely by socioeconomic factors and social disadvantage, ^{3 5 13} which, although acknowledged by several participants, are difficult to confirm in qualitative research. However, our data suggested that compared with Pakistani men, smoking in Bangladeshi men may be more deeply socially ingrained, contributing to group cohesion and identity.

Gender and age
Our analysis shows that in Pakistani, and particularly Bangladeshi, men smoking is central to socialising and identity and is an antidote to stress. In women, smoking has been traditionally regarded as disreputable, and cultural restrictions have reduced opportunities for women to smoke. However, both male and female participants said that smoking, which was often private and covert in women, seemed to be increasing among young women.⁷ This increase was attributed to westernisation, peer pressure, and rebellion.

Smoking was viewed as more acceptable in male elders and, to a lesser degree, in female elders. Young people smoking in the presence of elders was deemed disrespectful. Peer pressure and image were viewed as powerful influences on young people starting to smoke.

Islam, tradition, and family
As others have found in relation to diabetes and heart disease, religion plays an important but contested role in influencing attitudes and behaviour. ^{20 21} Islam forbids addictive substances, intoxicants, and those that harm health but does not expressly forbid tobacco (which was unknown in the Old World when the Koran was written). There were conflicting interpretations of how religiously acceptable it is for Muslim men to smoke. These conflicting interpretations were linked with participants' understanding of whether smoking is an addiction or intoxication.

The family was viewed as an important medium through which traditional norms, rules, and values associated with smoking were shaped and negotiated.

Implications for smoking cessation
Currently ethnic minority groups are not given special mention in national policies on smoking cessation, ^{22 23} although the Department of Health has recently launched the NHS Asian tobacco education campaign, and local smoking cessation services are increasing. Our findings and those from other studies ^{4 10} suggest that substantial effort and investment is needed in culturally sensitive smoking cessation interventions for South Asian people, involving the government and national and local health agencies (in particular, primary care trusts). The Race Relations (Amendment) Act 2000, which obliges public authorities, including the NHS, to promote racial equality in access to services will underline and add urgency to this requirement.²⁴

Appropriate targeting and involvement of ethnic minority groups and respect for cultural norms is essential.¹⁰ Work with Muslim religious leaders should clarify the religious acceptability of tobacco use in the Islamic faith and support the dissemination of an agreed policy nationally. Further work should develop and evaluate culturally sensitive smoking cessation interventions with South Asian communities.

	Acknowledgments

We thank the following people: all the research participants; Jean King from Cancer research UK for her guidance through the project; Terry Lisle for secretarial and administrative support; all members of the project's steering group, particularly Zakia Chowdhury, Iain Miller, Lucy Hall, and Ana Kulkarni; Azad Kashmir Women's Association (Leeds) and the Newcastle City Council Civic Centre Language Department for translations; North Tyneside Open College Network for help with the training programme; South Tyneside College Languages Department for holding and facilitating language tests for the community researchers; Veena Bahl from the Department of Health for instigating the programme and organising support; the Pakistani Muslim School, Roshni, and the Bangladeshi Workers Association for their time and allowing us to use their facilities for running focus groups; and Sue Zeibland for helpful referees' comments on an earlier draft.

Contributors: JB contributed to the supervision, management, and training of the community researchers; design of research materials; data collection; and data validation. She also took the lead in data analysis, report writing, and drafting of this paper. RB, MW, JK, and JR contributed to the study hypothesis, research design, data analysis, research materials, and data validation; commented on drafts of the text; and gained funding for the research. JK and JB designed the training programme. Jane Harland contributed to the research design and obtaining funding. All authors are the study guarantors. Thirteen community researchers organised (including recruitment), held, and translated in-depth interviews and focus groups facilitated by JB. The community researchers also contributed to developing interview topic guides, publicity for the study, participant recruitment strategies, and data analysis. The community researchers were Masuk Ahmed, Asif Shariff, Shubh Ghai, Khalid Mohammed, Akla Rahman, Anita Sarkar, Neelam Varma, Rushna Ahmed, Afzal Choudry, Afroz Qureshi, Rurkinder Kaur, Momotaj Rahman, and Jamal Sarwar.

	Footnotes

Funding: Cancer Research UK and the Department of Health. The guarantors accept full responsibility for the conduct of the study, had access to the data, and controlled the decision to publish.

Competing interests: None declared.

Ethical approval: Ethical approval was obtained from Newcastle and North Tyneside local research ethics committee.

	References

Top Abstract Introduction Methods Results Discussion References

1.	Balarajan R, Yuen P. British smoking and drinking habits: variations by country of birth. Community Med 1986; 8: 237-239[ISI][Medline].
2.	Kohli HS. A comparison of smoking and drinking habits among Asian and white schoolchildren in Glasgow. Public Health 1989; 103: 433-439[CrossRef][ISI][Medline].
3.	Modood T, Berthoud R, Lakey J, Smith P, Virdee S, Beishon S, eds. Ethnic minorities in Britain: diversity and disadvantage. London: Policy Studies Institute, 1997.
4.	Health Education Authority. Black and minority ethnic groups in England. London: HEA, 2000.
5.	Department of Health. Health survey for England. London: Stationery Office, 1999.
6.	Bhopal RS, Rankin JM. Cancer in minority ethnic populations: priorities from epidemiological data. Br J Cancer 1996; 74(XXIX): S22-S32.
7.	Pooransingh S, Ramaiah S. Smoking amongst South Asians in the United Kingdom. [electronic response to Barratt H. UK government launches anti-tobacco campaign for Asians]. BMJ 2001. www.bmj.com/cgi/eletters/323/7309/359/b#16178
8.	Bhopal R, White M. Health promotion for ethnic minorities: past, present and future. In: Ahmad WIU, ed. "Race" and health in contemporary britain. Buckingham: Open University Press, 1993:137-66.
9.	Bahl V. Cancer and ethnic minorities: the Department of Health's perspective. Br J Cancer 1996; 74(suppl 29): S2-10.
10.	Farren C, Naidoo J. Smoking cessation programmes targeted at black and minority ethnic communities. Br J Cancer 1996; 74(suppl 29): S78-S80.
11.	Kai J, Hedges C. Minority ethnic community participation in needs assessment and service development in primary care: perceptions of Pakistani and Bangladeshi people about psyhcological distress. Health Expectations 1999; 2: 7-20[CrossRef][Medline].
12.	White M, Bush J, Kai J, Bhopal R, Rankin J. A study to explore the feasibility and acceptability of smoking cessation interventions among Bangladeshi and Pakistani men and women. Newcastle upon Tyne: University of Newcastle, 2002.
13.	Bhopal R, Unwin N, White M, Yallop J, Walker L, Alberti KGMM, et al. Heterogeneity of coronary heart disease risk factors in Indian, Pakistani, Bangladeshi, and European origin populations: cross sectional study. BMJ 1999; 319: 215-220[Abstract/Free Full Text].
14.	Burgess RG. In the field: an introduction to field research. London: Routledge, 1991.
15.	Douglas J. Developing appropriate research methodologies with black and minority ethnic communities. Part 1: reflections on the research process. Health Educ J 1998; 57: 329-338[Abstract/Free Full Text].
16.	Glaser B, Strauss AL. The discovery of grounded theory. New York: Aldine de Gruyer, 1967.
17.	Strauss A, Corbin J. Basics of qualitative research (grounded theory procedures and techniques). Thousand Oaks, CA: Sage, 1990.
18.	Butler CC, Pill R, Stot NCH. Qualitative study of patients' perceptions of doctors' advice to quit smoking: implications for opportunistic health promotion. BMJ 1998; 316: 1878-1881[Abstract/Free Full Text].
19.	Cook PA, Bellis MA. Knowing the risk: relationships between risk behaviour and health knowledge. Public Health 2001; 115(1): 54-61[Medline].
20.	Health Development Agency. Tobacco and England's ethnic minorities: a research report. London: HDA, 2000.
21.	Kelleher D, Islam S. "How should I live?" Bangladeshi people and non-insulin-dependent diabetes. In: Kelleher D, Hillier S, eds. Researching cultural differences in health. London: Routledge, 1996:220-237.
22.	Raw M, McNeill A, West R. Smoking cessation guidelines for health professionals: a guide to effective smoking cessation interventions for the health care system. Thorax 1998;53(suppl 5):part 2.
23.	West R, McNeill A, Raw M. Smoking cessation guidelines for health professionals: an update. Thorax 2000; 55: 987-999[Abstract/Free Full Text].
24.	Race Relations (Amendment) Act. London: Stationery Office, 2000.

(Accepted 14 March 2003)