Yorkshire Enhanced Stop Smoking Study

Lung cancer has the highest mortality of all cancers in the UK and is the single biggest cause of premature death in Leeds. It disproportionately affects those from lower socioeconomically deprived communities where rates of smoking are highest. In 2014 in the Leeds South and East and the Leeds West Clinical Commissioning Group (CCG) areas, lung cancer standardized incidence rates were 131.4 and 103.3 respectively, compared to an average of 78.3 across England. Lung cancer standardized under-75 mortality rates were 52.4 and 39.3 respectively, compared to 33.26 across England. Poor outcomes from lung cancer treatment are typically the result of advanced, incurable disease at the time of symptomatic presentation. In 2011, the US National Lung Screening Trial (NLST) demonstrated a 20% reduction in lung cancer specific mortality, and a 6.7% reduction in all-cause mortality by annual low-dose CT (LDCT) screening, and lung cancer screening is now being adopted across North America having been recommended by the US Preventative Services Task Force (USPSTF) in 2013. A number of European nations are deliberating introducing screening programmes for lung cancer, and await the results of the NELSON study, the only European study powered to demonstrate a mortality reduction, which remains in follow-up. Despite widespread adoption in North America, several important questions remain regarding the best way to implement lung cancer screening. These include the optimum selection criteria to identify high-risk individuals to maximise clinical efficacy and cost-effectiveness, how to increase participation among 'hard to reach' populations, and how most effectively to embed smoking cessation support into these programmes. The Yorkshire Lung Screening Trial, funded by Yorkshire Cancer Research ( ISRCTN reference number 42704678) , seeks to address the first two issues identified above, and is due to commence screening in 2018. However, there are no currently funded UK studies specifically investigating optimal smoking cessation provision within a lung cancer screening programme.

More than 85% of cases of lung cancer are caused by tobacco smoking, and stopping smoking, at any age, significantly reduces lung cancer risk. It is therefore essential that any lung cancer screening programme provides smoking cessation support for participants, and doing so significantly increases the overall success of lung cancer screening programmes. One concern that has been raised around lung cancer screening for current smokers is the potential 'moral hazard' arising from a negative (i.e. reassuring) screening result, which may reduce motivation to quit. Conversely, attendance at a lung cancer screening programme offers a 'teachable moment' for smoking cessation, occurring at a time when participating smokers may be particularly receptive to offers of help to quit and indeed a negative screen result has been reported as being perceived as a 'clean slate' as a motivator to stop smoking. This observation fits with unpublished findings from a screening initiative in Manchester that despite a high prevalence of emphysema (68%) and coronary artery calcification (74%), most attendees were fit and relatively asymptomatic (85-90% performance status 0-1 and MRC dyspnoea scale 1-2), and hence potentially more susceptible to a message that lung damage had already occurred but clinical impact could be reduced by stopping smoking. One report from the NELSON study suggested a lower quit rate in screened persons compared to the unscreened control group, although a statistically significant effect did not persist following intention to treat analysis. No significant difference was seen between screened and control populations in the Danish Lung Cancer Screening Study, and the quit rate was significantly higher in screened versus control populations in the UK Lung Screening (UKLS) Pilot. Evidence suggests that smoking cessation and LDCT screening have additive effects on survival. An analysis of participants in the NLST reported a 38% reduction in lung cancer mortality with the combination of smoking abstinence at 15 years with LDCT screening; in addition, detection of an abnormality on the CT scan was associated with increased likelihood of smoking cessation. Further analysis of the NLST showed that arranging the smoking cessation support at the time of the screening increased quit rates by 46% (although this intervention was delivered to only 10% of study participants). In addition, research has indicated that adding smoking cessation interventions to LDCT screening improves the cost effectiveness of such programmes.

However, while smoking cessation is clearly an effective preventive intervention there is little evidence on the design and optimization of the smoking cessation interventions and support used in the screening setting. Evidence from a systematic review published in 2014 showed a benefit of materials tailored to the characteristics of individual smokers, although the included studies were conducted predominantly in the general population, rather than screening participants. Less research has been conducted investigating approaches to cessation in smokers attending for lung cancer screening, although a recent study by Marshall and colleagues reported that it was feasible to deliver a single tailored session of motivational interviewing counselling on the day of screening. For historical reasons NHS Stop Smoking Services (SSS) have evolved across the UK as opt-in facilities, whereby smokers who decide or agree to try to quit smoking then seek and are delivered help to do so. During the past five years however, evidence has grown that provision of smoking cessation support as an opt-out default generates far more quit attempts. We have demonstrated this to be the case in hospital inpatients, in which the opt-out model approximately doubled service uptake. Similar findings have been reported in relation to pregnant women setting quit dates in one NHS Foundation Trust, with uptake more than doubling after 'opt out' referrals were implemented. A study from the US found that 83% of eligible smokers did not opt out of a bedside consultation. Pro-active and systematic approaches to smoking cessation provision were recommended by NICE for pregnant women in 2010 and across all acute, maternity and mental health settings in 2013.

A recent UK study has also demonstrated the efficacy of including personalised risk information when inviting participants to smoking cessation services. A randomised controlled trial compared an individually tailored risk letter with an invitation to attend an introductory smoking cessation session to a standard generic letter advertising smoking cessation services. The personalised information was presented as a "Personal Health Risk Report", and included information on the participant's general health derived from Quality and Outcomes Framework (QOF) registered conditions on the GP record. In addition disease-specific health risks were included in the risk report, with the aim of making the individual aware of the personal health consequences of continuing to smoke, and their own individual risk of serious illness. The aim of providing this information was to change the individual's balance of perceived 'benefits' against their understanding of the harm caused by smoking. The group receiving personalised risk information had significantly higher attendance at stop smoking service (17.4% vs. 9.0%, adjusted OR 2.2, 95% CI 1.8- 2.7), higher completion of a 6-week NHS course (14.5% vs. 7.0%, adjusted OR 2.3, 95% CI 1.8-2.9) and higher 6- month validated abstinence (9.0% vs. 5.5%, adjusted OR 1.67, 95% CI 1.29-2.14).

To maximize smoking cessation among lung cancer screening participants it is therefore clearly important that all smokers who attend are provided with cessation support as a routine opt-out component of the screening visit, and that this support provides cessation pharmacotherapy and behavioural support in accordance with evidence-based NHS SSS practice. Furthermore the screening process, including risk assessment for lung cancer and imaging of the lung parenchyma and coronary arteries, allows an unprecedented opportunity for personalising the risks and harms of continued smoking, or conversely, the health gains to be made by successful smoking cessation. However there is currently no evidence on how best to design services to maximize the opportunity provided by screening participation, and how best to use the personalised information forthcoming from the screening process.