Developing the evidence base for addressing inequalities and smoking in the United Kingdom
Correspondence to: Ann McNeill, UK Centre for Tobacco Control Studies, National Addiction Centre, Institute of Psychiatry, King's College London, Addiction Sciences Building, 4 Windsor Walk, Denmark Hill, London SE5 8AF, UK. E-mail: email@example.com
Smoking is an increasing cause of health inequalities in high-income countries. This supplement describes pilot projects set up in England to develop and test pathways to ensure that disadvantaged groups, where smoking is frequently the norm, are reached, encouraged and supported to stop their tobacco use. Target groups were: smokers attending centres set up for highly deprived parents; smokers with serious and enduring mental illness; pregnant smokers; prisoners/other offenders who smoked; South Asian tobacco chewers; and recent quitters from ‘routine and manual’ occupational groups.
Commonalities observed across the six projects are summarized, alongside recommendations for implementation.
A significant barrier to implementation was the lack of mandatory identification of tobacco users across primary, secondary and community health-care settings and routine use of expired air carbon monoxide monitoring, particularly for high-risk groups. Appropriate use of financial incentives and national guidance is probably necessary to achieve both this and the adoption of ‘joined-up’ tobacco dependence treatment pathways for these target groups. Further research is needed on the impact of ‘opt out’ pathways: while resulting in increased referral rates, success rates were lower. In general, smoking cessation service targets were a barrier to implementation. Flexibility and tailoring of interventions were required and most projects trained those already working in relevant settings, given their greater understanding of target groups. Mandatory training of all frontline health-care staff was deemed desirable.
Implementing the findings of these projects will require resources, for training, incentivizing health-care workers and further research. However, continuing with the status quo may result in sustained tobacco use health inequalities for the foreseeable future.
The association between tobacco use and disadvantage
Smoking is the most important cause of premature death and an increasing cause of health inequalities in high-income countries . Indeed, in countries with the longest history of smoking, such as the United Kingdom, smoking is now the most important cause of health inequalities . Furthermore, a recent review concluded that, internationally, smoking prevalence is higher among disadvantaged groups and that these inequalities are likely to grow as smoking spreads to low-income countries . Globally, disadvantaged adults are much more likely to die from smoking .
The relationship between smoking and health inequalities reflects the strong and increasing patterning of smoking by social disadvantage [1, 5, 6]. Disadvantaged groups include those with low socio-economic status (SES) (reflecting low occupational, educational and/or income level), the unemployed, lone parents, the homeless, those with mental health problems, prisoners and some ethnic minorities. A recent study which scored people in England according to number of personal indicators of low SES found that 15% of those with no indicators of low SES smoked compared with 61% of those with the most (six to seven) indicators of low SES . Other studies have shown that 76% of prisoners in Scotland smoke, an estimated 90% of homeless people in England are smokers and people with mental health disorders are more than twice as likely to be smokers as those without [8-10]. Smokers from more deprived groups also have higher levels of cigarette consumption and are less likely to be successful when trying to quit [11, 12]. In addition, some ethnic groups have high rates of other forms of tobacco use; in the United Kingdom, groups originating from the South Asian diaspora sometimes chew tobacco, often alongside smoking .
Addressing tobacco use inequalities
Addressing inequalities in tobacco use has been identified by British governments as a national public health priority. The current Coalition Government's tobacco control plan for England  identified tackling tobacco use as central to achieving its commitment to ‘improve the health of the poorest, fastest’. Over recent years, increased resources have been directed at national, regional and local levels to tackling inequalities and tobacco use . Since 1998, a suite of policies has been introduced in England, including a ban on tobacco advertising and promotion, regular price increases, national smoke-free legislation, a raising of the minimum purchase age, graphic health warnings and a national network of smoking cessation services [National Health Service (NHS) Stop Smoking Services (SSS)] and similar initiatives were introduced across the United Kingdom. Considerable progress has been made in reducing cigarette smoking among UK adults since this time, but there has been less success in reducing socio-economic inequalities in smoking [1, 15]. In England, between 2001–03 and 2006–08 smoking in the more affluent declined from 22.8 to 19.4%, but there was no significant decline in smoking in the more disadvantaged groups (42.6 versus 42.4%) . It is not clear whether chewing tobacco use has decreased in the United Kingdom .
Developing effective strategies and programmes to target disadvantaged tobacco users has been hampered by the lack of evidence on what works. Much is known about how to reduce adult smoking, but few reviews and studies have looked at the equity impact of tobacco control interventions. A recently published systematic review of the international evidence on the effectiveness of tobacco control interventions to reduce socio-economic inequalities in adult smoking identified only 90 papers (nine reviews and 81 primary studies) published on this issue between 2006 and 2010 . This review concluded that for population-level policies there was strong evidence that price (tax) increases reduce socio-economic inequalities in smoking, although for individual smokers who do not quit as a result of price rises the economic burden of smoking increases. There was also evidence that certain types of mass media campaigns, when tailored to low SES smokers, could have a positive equity impact. However, the evidence on the equity impact of other types of population-level interventions was either insufficient or unavailable. For individual-level interventions the evidence showed that combined behavioural and pharmacological cessation support had lower quit rates among low SES smokers but can reduce inequalities if targeted effectively at these groups. Other types of cessation support had a negative equity impact or lacked sufficient evidence to draw conclusions.
English pilot projects
There was therefore an urgent need to develop and strengthen the evidence base on more effective interventions with disadvantaged groups. The English Department of Health thus commissioned work to assess the effectiveness of pilot projects aimed at reducing smoking among six hard-to-reach and deprived groups in England, and these are the subject of the papers in this supplement. The target groups were: smokers attending ‘children's centres’ (centres set up in areas of high deprivation); smokers with serious and enduring mental illness; pregnant smokers; prisoners and other offenders who smoked; people of South Asian groups who chewed tobacco; and a project aimed at reducing relapse among recent quitters from ‘routine and manual’ occupational groups. For many of these groups, tobacco use is the norm in their social environments and cessation support either non-existent, not accessed or delivered inconsistently. The main aim of the pilots was to develop pathways and systems to ensure that tobacco users in these most disadvantaged groups were reached, encouraged and supported to stop their tobacco use and this included, where appropriate, referrals to the NHS SSS.
The NHS SSS were set up to offer universal support with a particular focus on disadvantaged smokers . The provision of locally provided, evidence-based clinical smoking cessation services has been a major success of public health policy and practice in the United Kingdom over the last 14 years. Demand for these services has grown over that time; people utilizing the services have high levels of satisfaction with them, whether or not they quit , and these services provide the best likelihood of long-term quitting against any other currently available method [18, 19].
At the time that the work reported here was commissioned, the services were also reaching disadvantaged smokers, thereby having the potential to reduce health inequalities , one of very few public health interventions to do so. However, concerns remained, as disadvantaged smokers had lower success rates than smokers from other socio-economic groups , and there were also concerns that services were not being accessed by the most disadvantaged groups in society. Several factors militated against such a focus. These included: targets were set for the NHS SSS for throughput (number of quit dates set with the services) and success rates (4-week quit rates) which, while raising the profile and importance of smoking cessation among commissioners, also reduced the likelihood of disadvantaged smokers with their higher dependence being sought out and/or supported appropriately for long enough . Support was standardized through guidance and subsequently training, and therefore tailoring of treatment to individual circumstances and needs was not actively encouraged or widely employed; and support was provided by advisers who typically did not necessarily have the skills and experience to deal with potentially challenging user groups. The NHS SSS were not set up to treat tobacco chewers.
The pilot projects aimed to redress these imbalances. Building on previous research and available ‘best practice’, each project developed a pathway and intervention deemed to be effective and maximize long-term success, while being feasible and acceptable to providers and users. The focus of the pilots varied, as will be seen in subsequent papers. However, a number of common issues emerged during implementation, and these are described here, alongside recommendations for overcoming barriers if the projects are to be taken forward and, where appropriate, rolled out nationally.
Identifying tobacco users
One of the first issues that arose across all projects was how to identify smokers or other tobacco users, as support could be offered only to those identified appropriately. Each pilot project began with a mapping exercise of existing practice, and this indicated that there was a paucity of data on tobacco use status for some of the target groups and hence little historical data available. In the United Kingdom, the Quality and Outcomes Framework (QOF) system within primary health care helps to ensure that patients are asked their smoking status at least every 27 months and that patients with some severe illnesses, including diagnoses of schizophrenia, bipolar and other psychotic disorders, are asked about their smoking and offered advice to stop every 15 months.
However, it became clear that significant proportions of people, particularly those from deprived or hard-to-reach groups, were not registered with primary health-care teams. For example, in the prisons pilot , it was noted that a third of those attending probation services were not registered with a general practitioner (GP), and in the mental health pilot  we found that, for people with mental health problems living in the community, it was rare to have a record of their tobacco use status in their case-notes. In the South Asian communities involved in the chewing tobacco pilot , community outreach (e.g. via community and faith centres as well as primary health care) was utilized, as it was perceived to be the optimum way of identifying and reaching tobacco chewers, particularly women.
Perceived barriers to tackling tobacco use among deprived groups have been identified previously as fear of being judged, fear of failure and lack of correct knowledge about cessation services and of the medication available . Given that some groups, such as pregnant smokers and carers of young children, may try to hide their tobacco use status, using biomarkers may be a useful way of detecting tobacco use instead of, or in addition to, self-report. Collecting biomarkers brings complexity to routine appointments, but they were nevertheless used successfully across this programme of projects to either help identify tobacco users or to validate quitting. The most commonly used biomarker within NHS SSS is expired air carbon monoxide (CO), but we found few other health professionals using CO routinely outside the services. Measuring CO is relatively inexpensive and is quick and easy to carry out. If CO measurement was carried out as routinely as measuring blood pressure, the importance of smoking would be underscored and identification of smoking would become routine practice. A comparison of CO with urinary cotinine (a metabolite of nicotine) to identify pregnant smokers  found that cotinine was superior and that the CO test was not picking up some women who were smokers. However, given that cotinine has to be sent away to a laboratory for analysis, a routine CO test, using the optimum cut-off of 4 parts per million (p.p.m.), would be more cost-effective and immediate as an indicator of smoking with this target group. For validation of abstinence from tobacco chewing, CO monitoring is not appropriate and an alternative such as cotinine would need to be measured instead, although this would be elevated among users of nicotine replacement therapy (NRT).
In many of the projects, concerns were aired that raising the issue of tobacco use with a ‘client’ could reduce interest in other services being offered, or that it could affect the therapeutic relationship negatively. Tobacco can be seen as a stigmatized behaviour, and therefore professionals involved in the projects frequently had to find imaginative ways of raising the issue of tobacco use sensitively and in a non-judgemental manner. For example, for parents who smoked and attending the children's centres, staff found it easier to raise the issue of smoking through a more generic discussion around promoting child health rather than parental health . Indeed, throughout the projects, the most successful strategies for raising the issue of tobacco use appeared to be those that embedded tobacco among other issues about wellbeing that were being discussed routinely. There was no evidence that raising the issue of tobacco use, particularly if conducted sensitively, had negative effects.
A key recommendation, therefore, is to ensure that in both community as well as health-care settings, asking and recording tobacco use status should be mandatory, wherever possible and feasible, and CO measured routinely. If, as seems likely, incentives are needed before this will happen in practice, then similar mechanisms to QOF, such as Commissioning for Quality and Innovation (CQUIN) targets, should be employed.
Pathways for treatment and ‘opt out’ mechanisms
In order to ensure that tobacco users receive appropriate and professional support, robust pathways need to be developed to ensure that the tobacco user has access to optimum support. The input and agreement of frontline staff implementing the pathway and affected by the processes involved was essential. The pathway also needed to take into account the movement of people within systems. For example, the offender pathway can involve several potential stages including police custody, prisons and probation and into the community. For mental health, passing back and forth between in-patient and community living can be common. It is therefore important that the identification and treatment of tobacco use is ‘joined up’ across the full pathway of someone's care and any inconsistencies in content (e.g. the accessibility of effective medications) and delivery of support across different establishments are corrected.
In England, the National Institute for Health and Clinical Excellence (NICE) produces evidence-based national guidance which sets the standards for high-quality health care and also public health. The only pilot that had an existing national guidance policy was pregnancy (NICE guidance requires the NHS to implement ‘opt out’ referral pathways and routine CO testing for all pregnant women), but this was not mandatory and unsurprisingly, therefore, the project found that the pathway was not followed comprehensively. An ‘opt out’ mechanism was tested in three of the pilot projects. In the relapse prevention pilot , all those offered the intervention on an ‘opt out’ basis took it up, but only 43% of those offered it on an ‘opt in’ basis did so. In the children's centres and pregnancy pilots, tobacco users once identified were informed that it was routine practice for them to be referred for further support unless they ‘opted out’. This worked better in practice for pregnancy (where a number of different issues are being discussed routinely with the patient) than in the children's centre pilot, where there was some reluctance (on both sides) to put a strong emphasis on referring all patients to stop smoking treatment. In the latter project, it became apparent that making referrals to smoke-free home support as the primary intervention was more acceptable to clients and could act as a gateway to the smoking cessation service. A concern that emerged from both these pilots was that the quality of the referrals decreased when opt out processes were used—i.e. they were less likely to convert to a successful quit. Nevertheless, benefits could still accrue, such as increased knowledge about the dangers of tobacco use, the importance of stopping and where to obtain help in the future. However, such referrals would, in the short term, reduce stop smoking services' local quit rates, thus affecting targets and incentive payments. This is a fundamental issue which needs to be addressed urgently if services are to be proactive in these areas in the future.
NICE guidance is being developed to support pathways to treatment in mental health as well as for smokeless tobacco users who are currently excluded from NHS SSS, where their treatment is a discretionary activity, and any outcomes are excluded specifically from target monitoring. Without national policy, services do not deliver and the topic falls down the list of competing priorities or does not feature at all. It is therefore important that national practice should be established, through NICE or other mechanisms, for pathways of support for those target groups not covered currently by NICE guidance—this would include prisoners, children's centres and relapse prevention. This would also help to identify who are the key professional groups involved in the pathways.
The delivery of cessation support and referrals to NHS SSS
Owing to the complex array of mitigating factors involved in tobacco use in these settings, flexibility in terms of who delivers support, and how it is delivered, is warranted. Most of the projects utilized people working within the settings (e.g. mental health professionals, maternity and community outreach workers) who were then trained in tobacco cessation advice rather than placing stop smoking advisers into these settings. The rationale for this stemmed from previous research and experience of the applicants and project workers, that staff working in the settings would be more familiar with their clients, know how best to approach them, and perhaps be a more trusted source of advice and support. These placements all resulted in increased referrals.
Linked to improved methods to identify tobacco users, databases needed to be developed further to ensure quality control at data entry (e.g. to ensure contact details are accurate) and enable error-proof rapid referrals to be made, where appropriate. In order for people to be followed-up quickly, a system needed to be found to ensure that the relevant professionals were given appropriate information in a timely fashion, otherwise the window of opportunity to intervene may be lost. This was particularly evident in the children's centre and relapse prevention pilots; in the latter pilot, e-mail alerts were set up to enable timely follow-up to any text message responses sent by clients. If data collection is to be integrated into other routine data collection systems, adding the relevant fields to existing NHS SSS databases would minimize impact on work-loads. Robust electronic referrals would help to avoid ‘lost’ referrals from fax or paper referral systems.
Flexibility was also needed in how cessation support was delivered once referrals were received, as these smokers had particular needs. There may be a need for rapid response, persistent follow-up of referrals and out-of-hours working in order, for example, to have a presence at relevant community events. Most of the projects involved in providing interventions tailored support to the clients. For example, in prisons, patch exchange schemes (‘used’ for ‘new’) were commonly developed to reduce the likelihood of misuse of patches as prison currency.
Levels of knowledge about the risks of tobacco use, what tobacco cessation support involves and the range of pharmacotherapies available seemed particularly poor among some health professionals. The lack of knowledge among psychiatric professionals was of particular concern, given the relationship between smoking and certain neuroleptic medications.
Standardization of training for smoking cessation is being led in England by the National Centre for Smoking Cessation and Training (NCSCT). Currently, NCSCT training programmes are for standard smoking cessation interventions, although other specialist modules such as mental health and pregnancy are now online. It became apparent during the pilots that further specialist training was required, such as chewing tobacco use cessation training (where the pilot provided further support for the use of NRT to support tobacco chewers when they stop), and training for other frontline community and health-care staff. Training was a low priority for some of the professionals involved in the pilot projects, and many staff were not given time to attend training sessions. Training gives confidence to staff to raise the issue of tobacco use and answer queries from clients. We believe that brief intervention training for smoking and for second-hand smoke should be mandatory for all frontline health-care staff and recommended for frontline community workers (www.ncsct.co.uk/vba and www.ncsct.co.uk/shs).
Involvement of stakeholders and other support
Local champions, for example an enthusiastic and committed manager of a relevant organization, such as a children's centre, were thought to be critical to the success of the pilots. A manager putting the spotlight on smoking and/or cessation helped to prioritize tobacco use and its prevention. The prisons pilot found that the support of governors and other senior officers appeared crucial to promoting the importance of tobacco control and smoking cessation within prisons. We believe that so much more could be achieved if other relevant staff, such as psychiatrists within mental health settings, were more supportive of the need to reduce tobacco use. It is also helpful when appropriate frontline staff set a good example, such as by not smoking or chewing tobacco in sight of clients.
Good communication was important, most especially in the prisons pilot, and this required resources not only in terms of time and money, but the skills to make best use of multi-media such as websites, e-mail newsletters, text messaging, etc. The prisons pilot demonstrated that effective communication and networking helped to bring together professionals from both similar and different occupational backgrounds, enabling shared learning and more productive partnerships and service developments.
Tobacco use cessation in the areas chosen for these pilot projects was often seen as very low down the list of competing priorities, and health commissioners therefore need to find ways to ensure that the services engage proactively with these groups of disadvantaged people.
Each project also involved an economic analysis undertaken from the NHS perspective and hence covering adaptations to service provision, staff time and training. In all cases, the cost was less than £500 per service user (data not shown ). While still highly cost-effective, the cost estimates indicated that each quit in these disadvantaged groups would require more time and resources and immediate short-term benefits were not guaranteed. Start-up costs were also associated with establishing new services, but over time it is likely that by meeting previously unmet need, new services would start to generate demand. With regard to the prisons pilot, which did not seek to establish smoking cessation services as part of the project, a great deal was achieved through awareness-raising and raising the profile of tobacco use which, in turn, stimulated service developments, often within existing resources, and affected smoking policies across the criminal justice system.
Several of the pilot results have identified important predictors of effectiveness and acceptability which can be added to the evidence base. Some pilots have already affected long-term change by seeking local funding to continue and, through use of the evidence gained, influenced national policy and practice. In terms of future sustainability, ensuring that these target groups are on the agenda of local tobacco alliances might help to ensure continued work in this area.
A series of specific recommendations based on these pilot projects is available . While the projects were implemented in England, given the commonalities of tobacco use in these populations internationally we believe that the findings are widely generalizable, and may contribute to a renewed emphasis on reducing the health inequalities caused by tobacco use world-wide. Implementing these recommendations may face some opposition, particularly in times of austerity, as they will require investment in time and resources. However, failure to do so may result in sustained tobacco-caused inequalities for many years to come.
Declarations of interest
This research was carried out as part of a programme of inequalities projects implemented by the UK Centre for Tobacco Control Studies and funded by the Department of Health. Funding for the UKCTCS from the British Heart Foundation, Cancer Research UK, the Economic and Social Research Council, the Medical Research Council and the National Institute of Health Research, under the auspices of the UK Clinical Research Collaboration, is gratefully acknowledged. Andy McEwen receives a personal income from Cancer Research UK via University College London. He has received travel funding, honorariums and consultancy payments from manufacturers of smoking cessation products (Pfizer Ltd, Novartis UK and GSK Consumer Healthcare Ltd) and hospitality from North51 who provide online and database services. He also receives payment for providing training to smoking cessation specialists; receives royalties from books on smoking cessation and has a share in a patent of a nicotine delivery device. EC previously worked at the English Department of Health as the delivery lead for tobacco control policy. She has received travel funding, honoraria and consultancy payments from manufacturers of smoking cessation products (Pfizer, J&J, McNeil, GSK, Novartis and Sanofi-Aventis) and she also receives royalties from a book on smoking cessation and a book on health promotion.