Article Text

Download PDFPDF

Evaluating social and behavioural impacts of English smoke-free legislation in different ethnic and age groups: implications for reducing smoking-related health inequalities
  1. K Lock1,
  2. E Adams2,
  3. P Pilkington3,
  4. K Duckett1,
  5. A Gilmore4,
  6. C Marston1
  1. 1Faculty of Public Health and Policy, London School of Hygiene and Tropical Medicine, London, UK
  2. 2Department of Health and Social Care, Oxford Brookes University, Oxford, UK
  3. 3School of Health and Social Care Research, University of the West of England, Bristol, UK
  4. 4School for Health, University of Bath, Bath, UK
  1. Correspondence to Dr Karen Lock, Faculty of Public Health and Policy, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK; karen.lock{at}


Objective To explore social and behavioural impacts of English smoke-free legislation (SFL) in different ethnic groups.

Design A longitudinal, qualitative panel study of smokers using in-depth interviews conducted before and after introduction of SFL.

Participants A purposive sample of 32 smokers selected from three ethnic groups in deprived London neighbourhoods with approximately equal numbers of younger and older, male and female respondents.

Results SFL has had positive impacts with half smoking less and three quitting. Although there were no apparent differences in smoking and quitting behaviours between groups, there were notable differences in the social impacts of SFL. The greatest negative impacts were in smokers over 60 years, potentially increasing their social isolation, and on young Somali women whose smoking was driven more underground. In contrast, most other young adult smokers felt relatively unaffected by SFL, describing unexpected social benefits. Although there was high compliance, reports of illegal smoking were more frequent among young, ethnic minority smokers, with descriptions of venues involved suggesting they are ethnically distinct and well hidden. Half of respondents reported stopping smoking in their own homes after SFL, but almost all were Somali or Turkish. White respondents tended to report increases in home smoking.

Discussion Although our study suggests that SFL can lead to reductions in tobacco consumption, it also shows that impacts vary by ethnicity, age and sex. This study highlights the importance of understanding the meaning of smoking in different social contexts so future tobacco control interventions can be developed to reduce health and social inequalities.

  • Smoke-free legislation
  • England
  • evaluation
  • ethnic groups
  • environment
  • public policy
  • qualitative study
  • tobacco products
  • young adults

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Comprehensive smoke-free legislation (SFL) is an important public health policy that has now been introduced in over 15 countries and many other jurisdictions including 12 of 13 Canadian provinces and 25 of 51 US states. Yet, SFL evaluations have focussed on impacts on the population as a whole, including impacts on air quality, cigarette consumption, smoking prevalence rates or health (notably admissions to hospital for acute coronary events).1–4 Evaluations have not examined differences in compliance with, or social and behavioural impacts, of legislation between different ethnic groups, other than in New Zealand which explored impacts in the Maori. This lack of research may be partly due to limitations in the coding of ethnicity in routine data and because many jurisdictions that have previously gone smoke-free have small ethnic minority populations. However, given that smoking is a major cause of health inequalities, this is an important omission.

SFL covering all enclosed public places was introduced across England in July 2007. To our knowledge, England is the most ethnically diverse jurisdiction to go smoke-free. Nine per cent of the English population (over 4.5 million) and 29% (over 2.1 million) of Londoners are from Black, Asian or other minority ethnic groups.5 Consideration of ethnicity in development of tobacco control policies is important because ethnic groups vary widely in their attitudes to and beliefs about tobacco use and in the type and extent of tobacco used.6 7–9 For example, among UK men, smoking prevalence ranges from 20% (Indian) to 40% (Bangladeshi), compared with the then national average of 24%.10 Additionally, although most surveys focus on measuring cigarette smoking, many people use tobacco in other ways. Hand-rolled unfiltered cigarettes, bidis, and chewing tobacco are common in South Asian communities,11 12 with use of the latter unaffected by SFL. Hookah or shisha, a waterpipe used to smoke flavoured tobaccos, is traditional in North Africa but increasingly used in Europe and North America.13 Both bidis and hookah are wrongly perceived to be safer than cigarette smoking.11 14

In England, the Department of Health conducted a race equality assessment before introducing new SFL and concluded that “as action will affect all groups equally, we do not think there are serious race equality issues”.15 Nevertheless, equal impact of SFL on different ethnic groups is uncertain, given the documented cultural differences in attitudes, beliefs and behaviours, which in turn may affect the success of interventions to reduce smoking prevalence and thus health inequalities in different communities.7–9 With an increasing number of culturally and ethnically diverse countries proposing or introducing SFL, it is important to understand whether SFL has different impacts in different ethnic groups, and if so, why. Such evidence should help guide the implementation of SFL in other ethnically diverse jurisdictions and help maximise the effectiveness of existing legislation.


The study was a longitudinal, qualitative panel study of smokers living or working in an area of North London where non-White ethnic groups comprise 25% of the population. We used purposive sampling to recruit people of Turkish, Somali and White (British or Irish) backgrounds, primarily from lower socio-economic neighbourhoods. North London has the largest UK Somali and Turkish populations. These communities have contrasting histories. While the Turkish community has been established since the 1950s, there has been rapid increase in Somali immigration since 1991 due to conflict.

Ethnicity was self-defined by the interviewees. We recruited 32 smokers 1 month before the introduction of SFL, during June 2007 (‘pre-legislation’), either directly, by approaching smokers in ‘natural smoking settings’ (including bars and cafes) or indirectly via community contacts. As SFL was implemented immediately in all locations, we conducted follow-up interviews between 6 and 12 months post-legislation. Twenty-three of the original interviewees were successfully followed up. Ten replacement smokers were recruited for the post-legislation interviews, matched as closely as possible with respect to age, sex and ethnicity to those lost to follow-up. In total, 65 interviews were conducted with approximately equal numbers of younger (18–30 years) and older (>30 years), male and female participants in all three ethnic groups interviewed at both stages.16 Characteristics of study participants, including replacements, are given in table 1.

Table 1

Characteristics of study participants

The face-to-face, semi-structured interviews were conducted in English. The pre-legislation interview schedule was based on a literature review of ethnicity and smoking, previous SFL evaluations and developed further by discussion with a range of community stakeholders.17 Pre-legislation interviews aimed to explore interviewees' perspectives on SFL, their smoking and associated behaviours. Post-legislation interviews examined whether and how their smoking behaviours had changed due to SFL and the perceived impact on social networks and community norms.

All interviews were audio-recorded digitally with consent and transcribed verbatim. We used inductive, thematic analysis techniques,18 using a reflexive, iterative process of constant comparison involving team discussions and coding the data systematically using Nvivo7 software. In this respect, the process followed basic principles from grounded theory.19 Although we sampled using two age groups, during the analysis, it became apparent that the responses of the oldest adults (over 60 years) differed from those of other adults over 30 years. At the analysis stage, we reported the results for those older participants >60 years old separately from those 30–59 years (whom we subsequently term middle-aged; table 1).


Changes in smoking behaviour

There were two striking similarities in smoking behaviour before and after legislation across the different sex, age and ethnic groups in the study. First, half the participants described signs of heavy nicotine dependency (including “having to have a cigarette as soon as they wake”),20 with men from all ethnic groups heavier smokers than women, smoking 20–40+ cigarettes per day. Second, interviewees highlighted the importance of smoking as both an individual and social behaviour linked to other social activities (eg, drinking with friends).

After the legislation, respondents described themselves in a range of states of smoking, from successful and failed quitters to those who did not intend to stop smoking. Only three people claimed to have quit successfully after SFL, with only one suggesting that this was directly attributable to the legislation. Of the others, approximately half were still smoking the same amount after SFL, and half were smoking less, with no apparent patterns by ethnicity or sex. Reasons given for cutting down on the number of cigarettes smoked included actively making an effort to cut down after SFL and the legislation reducing the opportunities to smoke while out.

Although all groups displayed good knowledge about the wide range of smoking cessation services and products available, many did not recognise or acknowledge the potential health and other benefits of stopping smoking. There was also mistrust of smoking cessation products on offer including a misperception among several people that nicotine, rather than the other components of cigarette smoke, is the most harmful aspect of smoking (box 1).

Box 1 Perceptions of negative effects of nicotine replacement therapy

“I did go to a smoking thing, and they asked me if I would try and stop…And she said, ‘You can have some tablets, if you want.’ But I never—she said, ‘Come back next week,’ but I never went back…Because, I mean, I'm on a lot of medication…And I didn't know if it would counteract.” (White woman aged >60 years)

“I don't think that smoking nicotine patches are really good…cause you know…, it's just…… putting nicotine in your blood…it's the same thing, innit.” (Young Somali man)

There were some sex and ethnic differences in attitudes to quitting. Some Somali respondents felt that smoking cessation services would not help as they focussed on cigarette use and did not address shisha smoking.

Men often reported relying on willpower as a main factor in quitting. Female participants referred to the need for willpower, but rather than stopping them from seeking help, it seemed to lower their expectations of the effectiveness of help. Men from Somali and Turkish backgrounds reported a strong sense of having to solve the problem on their own.“[Be]cause it's just what you tell yourself I think…, psychology.” (Young Somali man)“Patches and the chewing gum, but its rubbish…it depends on your brain, it depends on you want to stop, you don't need all this…you need the willpower.” (Middle-aged Turkish man)

Changes in the geographical location of smoking and its social impacts

The introduction of SFL resulted in major changes in the geographical location of smoking in our study population, with smoking increasingly becoming an outdoor activity, or shifting from public to private places (particularly homes). Among our interviewees, these changes had greater negative social impacts on older compared with younger smokers and on Somali smokers in particular.

Differential impacts on older versus younger smokers

Although smokers of all ages discussed the inconvenience of smoking outdoors, it was mostly the oldest participants (>60 years old) who said that they now smoked less in public because they did not want to go outside to smoke. This seems to have arisen for a range of reasons (box 2)

Box 2 Reasons for older people over 60 years smoking less in public places

“But we can't sit down, you know……, and I can't stand very long. So I have to make sure that I come back [home].” (White woman >60 years)“…people standing outside in this freezing weather, they're more liable to be with their doctor with chest complaints and all that.”(White man >60 years)“Smokers became like social outcasts…I mean, it feels that way…when they are sent out and said… ‘You cannot smoke here’.” (Somali man >60 years)
including it being physically difficult, that they thought it was socially inappropriate to smoke outside or being concerned about catching infections as a result of being outside in bad weather.

Before SFL, the older study respondents had described smoking as an important part of their social lives, taking place, for instance, when men met friends in pubs or traditional cafes and when women met at social clubs. This changed significantly after SFL with all but one of the oldest respondents over 60 years from all ethnic groups mentioning they now choose to stay at home more and smoke rather than go out to social venues (box 3), thus reducing their social contact with others.

Box 3 Reasons for staying home after SFL in older people over 60 years

“I used to go out to restaurants three times a week, and to other places certain times a week. Over the last 4 months, 5 months, I just can't be bothered. I think to myself, I go in there, I'm not going to enjoy myself, because I like to have a smoke, and if you're in a restaurant, you've got to go outside…” (White man over 60 years)

“Well I'm not going to go to the bookie, I'm not going to go to restaurants, that's what I do…I don't go to pub, cause I like to sit down [to smoke] and I don't wanna take my glass and get out. Just stay [at home]…read my papers, watch my television.” (Turkish man >60 years)

“At the…pub, you've got to go out and stand in the street. There's no comfort, no relaxation so really, my way of thinking, I'm better off sitting here [at home].” (White woman >60 years)

By contrast, younger people, under 30 years, generally said they felt relatively unaffected and appeared comfortable navigating the impact of SFL by simply smoking outside. Many young respondents across all groups (except Somali women) reported that they were surprised that the legislation had little impact on their smoking or social lives.“I can still smoke when I want to, all I just have to do is just get my jacket and go outside and stand.” (Young Turkish woman)

In fact, all but one of the younger respondents from white and Turkish backgrounds said the legislation had unexpected positive benefits by facilitating new social opportunities (see box 4). Respondents discuss the new outside spaces where smokers are able to meet people, which some young smokers perceive as making it “even more sociable to smoke” than before SFL.

Box 4 New social opportunities for young smokers after SFL

“You find that you talk to people… when you go outside the clubs, you talk to people that you'd never talk to if you was inside.” (Young white woman)

“I thought I will miss out most of the social life, but I don't. I see more people standing outside.” (Young Turkish woman)

“Everyone just goes out, so it's become more of a social thing, to be honest, because you get a whole gathering of smokers so, you know, it's become even more social to smoke.” (Young white man)

Young respondents only cited two negative social impacts, both concerning safety fears mentioned only by women. These were increased risk of fights occurring outside pubs and clubs and the potential danger of drinks left unattended inside being tampered with.

Differential social impacts by ethnic groups

There were several findings that suggest differential impacts of SFL in the different ethnic groups.

Changes to smoking in the home

After the legislation, approximately half the sample stated that they no longer smoked in their home (ie, had imposed voluntary home smoking bans), while the other half stated that they now smoked more in the home.

Because smoking outside in public was sometimes seen as socially unacceptable, many women, mainly Somali and Turkish, said they increasingly met in private houses to socialise.“For those who smoke Shisha they have to be home…… the gathering that used to take place in a restaurant takes place home a lot now.” (Middle-aged Somali woman)

However, of those who said that they had personally stopped smoking in their own homes after the legislation, all but one were Turkish or Somali (both men and women), implying a complicated picture with some Turkish and Somali women smoking more in private houses (of friends), while some had stopped smoking in their own homes. ‘Family’ was most commonly given as the explanation for not smoking at home. Younger Somali and Turkish participants explicitly stated respect for their parents as motivation. Older participants gave their own health, or that of their children, as a reason for not smoking indoors at home.“I used to go out and smoke outside the house…even before the prohibition came into force. I did that because I did realise at a very early stage that smoking, among your family members….and particularly the children…it's very, very, you know, dangerous to their health …because they are young.” (Middle-aged Somali man)

Seven participants said they no longer smoked at home explicitly due to SFL. Many respondents used the term ‘closed space’ to explain this, although only one person incorrectly believed smoking was actually banned at home.“Yeah, it has to be outside, even in my house, it has to be in the garden…or in the street. Not any closed places…because they are [his family] not smoking—I don't feel to disturb them.” (Middle-aged Turkish man)

Women and shisha smoking

Somali women appeared to experience the greatest social impact of SFL. All Somali respondents discussed the traditional importance of shisha,21 with all but one of the Somali women currently or previously smoking shisha, while few smoked cigarettes. Despite an estimated 17% of Somali women in this community who admit to smoke,22 it is considered culturally unacceptable for Somali women to smoke, especially in public. Respondents said this was the custom rather than because of specific religious beliefs.

“It's more a taboo…and that's one of the things that's…relevant in Somalia. Yes, woman smokers are looked down on…you know, more than men”. (Middle-aged Somali man)

Both Somali men and women agreed that the legislation has had a greater impact on women because of increased social restrictions. Before, SFL Somali women smokers could hire separate indoor smoking rooms in public shisha venues where they could socialise in private with friends. Women who continue to smoke shisha say they feel that they now can only smoke in private homes or, if continuing to smoke publicly, by taking measures to conceal themselves, travelling away from their local community or smoking in illegal venues (box 5).

Box 5 Illegal smoking in pubs, clubs, bars and cafes

“In one of the clubs we go to, we smoke in the toilet at the back…and no bouncers come in there. But everyone does, you can smell it as you walk past, so they must know that people do it.” (Young white woman)

“It's a different situation, I mean, it's a Turkish coffee shop and in there it says that, you know, that smoking is banned…., so it's illegal to smoke blah, blah, but everybody smoke in there, but it's kind of, not an open coffee shop…it's something under cover anyway.”(Young Turkish woman)

“I think I told you, that these people will go underground…and, yes, they did. There were restaurants that had a lower floor and I think they will let only their regulars in….I sat there and I could easily say that the space occupied about 50 to 60 people and I wouldn't be able to see the person at the far corner.” (Young Somali woman about shisha smoking)

“…one place…it was a normal restaurant upstairs, which used to be a normal Shisha bar…and you'd have to go downstairs and there was a room in the basement, that was for Shisha smokers. And you can still find places that are the same as before, like, inside, but it's just the fact that you have to pay more for them just cause it's inside and that's not allowed.” (Young Somali woman)

“… them snooker halls where the landlord just lets you do what you want. That's the only place we can go [to smoke]. And them places, sort of, dangerous, …and trouble starts.” (Young Somali man)

“For girls, they cannot sit outside. They feel a bit embarrassed. A friend—a family friend, like, someone might see them and tell the family. So what happened was, they put hoods on, a bit clothing on. Now they face on the wall, and they're just smoking…but they cannot sit there for a long time. And when they're, like, smoking the Shisha, they're not feeling comfortable…. I did it a couple of times, but it was at night time anyway. So I'm sure that my family aren't around. It was far away from where I live and I went out with a couple of friends, and even though it does matter, the way you dress up I just put my hood on like this, and …nobody's gonna see your face. ….I was not feeling comfortable…you know, before it was really okay, not anymore…It's the shame.” (Young Somali woman)

Smoking illegally

We also found age and ethnic differences between those who said they smoke in places where it is no longer legal.

At least half of study participants (from all groups) mentioned bus stops and railway stations as places where either they had smoked illegally or seen others smoke since SFL. Along with entrances to buildings, these appear to be ambiguous public spaces which many wrongly believed were “outside spaces” not covered by SFL.“I'll not be rude, half the bus shelters are not enclosed, all they are is a bit of a pole with a lid on it. Where is the commonsense in that?” (Older white man over 60 years)

In contrast, there was a range of experiences and opinions about what was clearly illegal smoking inside public hospitality venues. Many older respondents recalled situations which they largely perceived as accidental flouting of the law, either people forgetting initially or people not understanding which places were included such as private parties in public venues.“[I] went to a Turkish Kurdish wedding…and especially the young women, they were smoking in the toilets. They thought it wasn't illegal.” (Middle-aged Turkish woman)

Younger people were far more likely than older people to report smoking in illegal places and some talked about deliberately flouting the law. Although most respondents did not regularly do this, many young participants said that they knew of, or participated in, smoking inside venues that was tolerated or endorsed by the establishments' managers. There was a clear difference between places people smoked inside knowing they would not be penalised (toilets in clubs) and formally organised illegal smoking venues (box 5). Young Somali and Turkish participants were most forthcoming in discussing indoor public venues where some of them still smoked. These places seemed to be frequented mainly by members of the same ethnic group. Some young Somali men discussed visiting places they perceived as threatening or potentially dangerous and which they would have avoided before SFL. Interviewees described these basements and closed rooms as having little ventilation, which would result in higher exposure to second-hand smoke. From their reports, it is highly unlikely that these venues met other health and safety (including fire) regulations. SFL might even play into the hands of a few unscrupulous bar owners, who charged more for hosting illegal activity.


This study was designed to explore the range of smoking in different ethnic groups before and after SFL. The methodological limitations of the approach need considering. First, most of the participants lost to follow-up were Somali men. We recruited replacements of similar ages from the same areas in the follow-up phase. We consider this replacement approach to be legitimate because new respondents can respond retrospectively to the impact of SFL on them. Although this may have limited the ability to draw conclusions over time, it does not restrict the study conclusions on the impact of SFL. This also highlights the difficulty in conducting research in harder to reach communities, especially where people may be more mobile. Second, the sub-group analyses were based on small numbers. As it is an exploratory study, the sample was chosen to provide a wide range of views and experiences rather than to be generalisable to all people within any categories included. Small sample size or use of a single site does not itself limit the validity or generalisability of a qualitative study.23 Qualitative approaches have been previously used to gain important insight into issues of tobacco control including home smoking,24 but this study shows the relevance of such study designs to future SFL evaluations.

Preliminary evidence suggests the English smoke-free legislation has, in the short term, reduced smoking prevalence25 and cigarette consumption. This is reflected in this study, with about half the respondents reporting reductions in cigarette consumption or quitting. Although there are clear public health benefits associated with any reduction in active and passive smoking, our research highlights some potentially important differences in the impact of SFL by ethnicity, age and sex which have implications for tobacco control policy and smoking-related health inequalities.

These differences were most notable in relation to the social impacts caused by changes in the geographical location of smoking. Based on our respondents' experiences, SFL appears to have had a more negative impact on older smokers >60 years. While the reluctance of these oldest smokers to smoke outside may reflect their poorer health, potential increased social isolation since the ban may exacerbate other problems of older people such as loneliness and depression.26 This could have the effect of making an already isolated population even more marginalised. In turn, this could reduce their motivation and ability to quit smoking while also making the older, already poor users of smoking cessation services, an even less accessible group for health promotion efforts.27–29

In contrast, most young adult smokers in our study felt relatively unaffected by the new law. Despite stated resistance before SFL, the reality since is a high degree of compliance perhaps because most say they have found the legislation easier to live with than they anticipated. Indeed, many perceive positive benefits such as increased social opportunities for smokers. It was clear that peer and social pressure, which can contribute to social participation in smoking,30 are still active in the new smoking locations created. Previous social boundaries can be discarded and new associations formed while smoking outside.31 This raises concerns that in some population sub-groups, particularly those with high smoking rates, SFL may continue to normalise, rather than ‘de-normalise’ smoking.

It is important to understand the differences found between ethnic groups after SFL. Overall, the social impacts appeared most restrictive for young Somali women who, due to cultural sensitivity around female smoking, were often now unable to smoke in public where they might be seen and were thus taking measures to hide their smoking (including visiting illegal venues). Somali respondents also perceived that smoking cessation services were not culturally sensitive, focussing on cigarette, and this may have contributed to some of the ethnic differences seen in the lack of willingness to use cessation services.14 32 A further issue highlighted, as in previous studies, is poor understanding of the health impacts of quitting and misunderstandings about cessation products and the role of nicotine.33 This is an issue that needs to be addressed particularly given the growing policy focus on harm reduction.

Changes in home smoking since SFL appear complex. Despite concerns that SFL might increase second-hand smoke exposure in the home, there is little evidence to support this claim; indeed, the opposite appears to occur.24 34–38 Yet, most of these evaluations have been based on population level data, and our study suggests that effects may differ by population sub-groups. Social tobacco use among female Somali and Turkish respondents moved to private venues including friends' houses, despite personal tobacco consumption being less likely to occur in their family home settings (by imposing home smoking bans). A recent study measuring children's exposure to second-hand smoke showed that Black and Asian children are less exposed than white children in England.39 In line with these findings, white British respondents in this study reported allowing smoking in their homes, which was rare among the other ethnic groups. Reducing home exposure will become increasingly important post-SFL. Previous qualitative work has provided insights into the barriers to and reasons for restricting smoking in the home particularly among disadvantaged parents, but to our knowledge, no research has yet examined these issues between ethnic groups.40

Reports of illegal smoking were more common among young smokers with the descriptions of the venues suggesting that they are ethnically distinct, often well hidden, with other health and safety concerns. This raises the importance of implementing both culturally sensitive communication strategies about how and where SFL applies, as well as monitoring and enforcement strategies that recognise the variety of smoking venues used by different groups.

Overall, this study highlights the importance of exploring the complexity of the meaning of smoking in different social contexts. It draws attention to how policy impacts can differ in unexpected ways in different groups. The increased isolation of older smokers and the perceived stigma for some women associated with smoking outside in public since SFL may make already disadvantaged groups even more difficult to target or engage in future smoking cessation strategies.37 41 Further research is needed to explore what impact it may have on quitting in these population sub-groups. The study also illustrates the ongoing difficulty for public health policy in reducing smoking prevalence in young people highlighting the need to take a broad approach to youth smoking policy beyond SFL.


Although preliminary evidence suggests SFL may have helped smoking prevalence rates to fall in the UK,4 previous research has shown that falling prevalence rates can disguise widening relative inequalities in smoking and smoking-related health outcomes among disadvantaged and ethnic groups.42 Our study shows that legislation may well have different impacts for people of different ages, ethnicity and sex, with corresponding effects on health and social inequalities. Further research is needed to examine these equity effects in more detail to ensure that future and current smoking control legislation can be enacted while minimising any unintended negative effects.43

What is already known

Smoke-free legislation contributes to reductions in overall population rates of smoking prevalence and tobacco consumption.

What this paper adds

Social and behavioural impacts of smoke-free legislation may vary by age, ethnicity and sex, potentially contributing to tobacco-related health inequalities.


We wish to thank the following people for their contribution: Nicola Henderson, Natashca Mueller Hirth, Marianna Thomas, Caroline Coen, George Gallagher, Dawn Jenkin, Fatou N'Jie, Alexsis Prarrer, Verena Thompson, Laura Wu, Ufuk Genc, Keren Wiltshire, Helen James, Esther Sharma, Abdi Elmi, Shamis Elmi, Habiba Hersi, Mohamud Gure, Ilyaas, Somali On Air Radio, Somali Eye Media, Somali Speakers Association, IMECE Centre, Anadolu Halk Kültür Merkezi, Anatolian People's Cultural Centre.



  • Funding Islington Primary Care Trust, Goswell Road, London, N1, UK Other Funders: NHS.

  • Competing interests None.

  • Ethics approval The research protocol was approved by the LSHTM ethics committee and informed written consent was obtained from all study participants.

  • Provenance and peer review Not commissioned; externally peer reviewed.