There is a strong relationship between cigarette smoking and social disadvantage. Smoking prevalence and levels of addiction to tobacco are highest in the most disadvantaged social groups [1, 2] as are difficulties in quitting due to perceived low self-efficacy [3, 4] and triggers for habitual smoking in terms of stressors [5, 6]. Despite the fact that deprived smokers are as likely or more likely than advantaged smokers to want to quit smoking [7], smoking prevalence has decreased markedly in the most socially advantaged groups over the past two decades but remained unchanged in the most disadvantaged smokers [1], many of whom could be categorised as 'hard core' smokers [8]. As a result, smoking remains the largest single identified cause of the inequalities in health between rich and poor in the UK.

In 1998, the UK government Tobacco White Paper [9] outlined a strategy to develop smoking cessation services in the National Health Service, based on effective [10] and cost effective [11, 12] models, and with particular emphasis on helping disadvantaged smokers. These services have been successful in reaching smokers from disadvantaged communities [13], largely through locating them within areas of greatest deprivation, but otherwise there is little evidence to develop strategies to attract this target group [14]. It is therefore important to explore how services are perceived by those in the target groups to provide insights into how access to and uptake of services could be improved. Nottingham's smoking cessation service, New Leaf, provides a standard range of evidence-based smoking cessation interventions typical of those available nationally[15]. In this study, we have attempted, using qualitative methods, to determine whether disadvantaged smokers in Nottingham are aware of existing local smoking cessation services, to explore how they view the services on offer, and to identify specific barriers and motivators to improve access to smoking cessation services in this target population.

We used Manchester Information and Associated Services[16] to identify 5000 households in the 5% most deprived (highest Townsend Score) enumeration districts in Greater Nottingham, and posted a short self-completion questionnaire to these addresses. A second questionnaire and reminder letter was sent to non-respondents six weeks after the first questionnaire. The questionnaire asked how many smokers there were in the home over age 16 and whether each smoker wanted to quit smoking and had made an attempt to quit smoking within the previous 12 months. The questionnaire also asked for contact details of any smokers in the home who would be willing to help further by attending a discussion group. We offered to provide a language translation or help by telephone in completing the questionnaire as appropriate. We used the responses to select purposively participants who were current smokers, living in one of the 5% most deprived enumeration districts in Greater Nottingham, and who had made an attempt to quit smoking within the last year without using local smoking cessation services. Assuming that up to 35% of adults in manual occupations smoke, with an average of 2 adults per household, our questionnaire had the potential to reach around 3,500 smokers, of whom about one third (1167) would have made a quit attempt in the past year [7]. We chose to do this using qualitative methods which, although they have a background in marketing and social research, are increasingly being used in medicine. They aim to provide a holistic perspective which preserves the complexities of human behaviour [17] and explore the ideas and concerns which the subjects themselves come up with [18] rather than imposing external research themes. The study was approved by the Nottingham City Hospital research ethics committee.

Focus groups were held in local premises in the early evening, inviting 10 participants to each session. Each participant was sent a study information sheet in advance of the group session, and at the beginning of each session we also gave a verbal explanation of the aims of the study, along with the assurance that there was no obligation to participate in the research, and obtained signed consent to proceed. The discussions were semi-structured, and facilitators (2 per group) used a discussion guide which covered the following broad topics: participants' smoking behaviour, experience of quitting, knowledge and perceptions of existing smoking cessation services, and characteristics of services that would be more attractive to the smoker. We used a grounded theory approach [19], interweaving data collection and analysis from the start, and refining the topic guide to reflect the emerging themes. We used a variety of visual stimuli to prompt discussion of smoking cessation methods including leaflets, pictures of nicotine replacement therapy, flip chart listing of unprompted cessation methods, and pile-sorting exercises – sorting pictures representing various aids to smoking cessation into 'good' and 'bad' piles. Sessions lasted between 90 and 120 minutes, with breaks as necessary. Focus group sessions were tape recorded and transcribed in full.

We used a group approach to analyse the data in a systematic and rigorous manner, and to minimise bias in interpretation of findings from the text of the transcripts. Initially each researcher (ER, AM & SL) worked independently reading a subsample of three transcripts to identify themes (main issues relevant to smoking cessation). We met to agree these themes and to develop provisional definitions of the main themes, then independently reread the same transcripts to identify key issues or categories within themes, and met once again to agree working definitions of categories within themes. Two researchers (ER & AM) read the remaining transcripts to further develop themes and categories, in an iterative process of refining and developing categories where data did not fit existing definitions [19]. One researcher (ER) then coded all transcripts using agreed definitions, and using NUD*IST 6 software (QSR International, 2000) to facilitate a systematic inspection of text coded under each category. We aimed to describe the themes and the principal variations in opinions and attitudes within them, rather than imposing any predetermined categories. Anonymous quotes from the transcripts have been used to illustrate the themes raised.

A total of 358 questionnaires were returned with a median Townsend score of 8.1 (interquartile range 7.9 – 8.3). Of those returned, 186 were from smokers, of whom 92 declined or were unsuitable to participate in focus groups, 94 agreed to take part, and met the eligibility criteria, and 39 (median age 45 years, age range 27 – 77 years, 23 males) actually participated. Nine focus groups of sizes 2, 2, 3, 4, 4, 5, 5, 7, 7 participants were carried out.

The participants in this study were smokers who had previously attempted to quit smoking without formal support, and who lived in extremely deprived areas. Although motivated to quit smoking, they felt that their smoking was intractable and were torn between thinking that only intensive measures would help them to stop but also that all that was really required was willpower. They felt increasingly marginalised by society and government and felt that their addiction was not taken as seriously as addiction to heroin or alcohol. They knew little about the services available to help them, but perceived them to be ineffective and expensive despite evidence to the contrary. Participants stated the need for a wide variety of cessation group timings and locations without being aware that these services already existed.

This study addresses access and perception of services by deprived smokers, an important public health issue which has been under explored in previous research. To our knowledge, this is the first study that directly addresses barriers and motivators to accessing smoking cessation services amongst the social groups with the greatest need although studies have been carried out with pregnant women [20–22] and smokers also dependent on alcohol [23]. We expected a low response to our questionnaire, and indeed of the estimated 1167 potentially eligible smokers expected among the 5000 households included in our initial questionnaire, we recruited just 39 to take part in focus groups. Those who took part are unlikely to be representative of all eligible smokers in this social group. In particular, they were recruited to the focus groups by means of letters and phone calls, and these may be individuals who were more responsive to a personalised approach. Moreover, focus groups always provide a platform for the more opinionated in each group and despite the best efforts of the facilitators may not be representative of the true views of each individual within a group, and some of the groups were very small. However, the participants were, by definition, motivated smokers whose needs were not being met by current services, and who therefore provided an appropriate context for the current study. That our study was based in an area served by a single smoking cessation service, New Leaf, potentially limits the external validity of our findings. However, New Leaf provides a standard range of evidence-based smoking cessation interventions, with group or individual support at flexible times and locations, typical of the services offered nationally. Moreover, since the New Leaf service is based in one of the original Health Action Zones, offers predominantly individual support and has a strong relationship with local Primary Care Trusts, the reach of the service to disadvantaged smokers is likely to be at least as great as that for services elsewhere in the UK [24]. We suggest therefore that the unmet needs identified for deprived smokers in Nottingham are likely to be applicable more widely.

The main themes of our findings are however supported by evidence from other studies, showing that smokers from areas of disadvantage feel unable to cope without cigarettes [25, 26], guilty about their continued smoking [27], are sceptical about the effectiveness of nicotine replacement therapy [5] and less likely to be able to overcome barriers to cessation despite high levels of motivation to quit smoking [26, 27]. The perception that smokers from deprived areas are in some way discriminated against [26] and discouraged from making healthy choices is also reported elsewhere [28], as is the easy availability of contraband cigarettes in areas of deprivation [29], which may act as a barrier to cessation. A recent cross sectional survey of adult UK smokers identifies the need for targeted interventions, particularly towards smokers in areas of socio-economic deprivation [8]. The English smoking cessation services were instructed to target this group, as well as young and pregnant smokers, and there is evidence showing that these services have been successful in reaching smokers from disadvantaged communities [13]. However, these services may have been hindered in the need to meet throughput targets and by lack of evidence to develop strategies to attract this priority group [14]. Our findings go some way to identify the needs of these smokers, and specifically to identify factors which act as barriers or motivators to them gaining access to these services.

Smoking cessation services are amongst the most effective and cost effective interventions available in medicine [10–12], but if they are to deliver their enormous public health potential, particularly in relation to smokers in deprived areas, they need to be more widely and appropriately promoted in an attempt to dispel some of the myths surrounding their availability and effectiveness held by these smokers. A more personalised approach to promoting services that are non-judgemental, and with free pharmacotherapy and flexible support may encourage more deprived smokers to quit smoking.