Through consecutive cycles of collaboration and reflection, the intervention progressed from a separate physical activity component delivered by researchers to a comprehensive and integrated strategy to promote physical activity as a cessation aid. It incorporated a comprehensive self-help guide and pedometers, which involved promoting physical activity in the 2 weeks prior to quitting and 4 weeks after quitting. The smoking cessation advisors were enthusiastic about implementing the refined version of the intervention, quickly increased their confidence to promote physical activity (based on survey responses and field notes), and reported that clients had started using physical activity as a way of dealing with situational cues, cravings and withdrawal symptoms. The pedometers were particularly popular, both as a motivational tool and as a form of distraction.
Our finding that the reaction to the pedometers was generally positive is consistent with previous research in a variety of populations. For example, sedentary individuals with type 2 diabetes engaged in a pedometer-based intervention liked the pedometer as a self-monitoring tool , ethnic minority populations reported enthusiasm for pedometer use , and midlife women found the pedometer functioned as a motivational tool . Similarly, our finding that some clients were disheartened by the 10,000 steps goal was consistent with previous research that has found that the goal may be unrealistically high for sedentary people or those with chronic diseases and can lead to attrition from research and reduced adherence to pedometer-based programmes . This goal was also found to be unpopular among middle-aged men in Australia , although a meta-analysis suggested that a step goal such as 10,000 steps was an important predictor in increasing physical activity levels . While the pedometers in the present study were popular, it was a challenge to get clients to self-monitor their daily steps using the recording sheets in the self-help guide. Prochaska and colleagues  also reported that less than 15% of their sample completed 6 weeks of data sheets, recording pedometer steps, to facilitate smoking cessation. Finally, our finding that the advisors found it possible to promote physical activity right from the start of the clinic is consistent with previous research that found the simultaneous promotion of multiple health behaviours changes to be more effective than the sequential promotion of multiple changes  despite a common contrary belief .
The training and delivery of the intervention did not significantly impact on advisors' stage of readiness for promoting physical activity as a smoking cessation aid, outcome efficacy beliefs or pro and con beliefs. This is perhaps not surprising, given the small sample involved. Another explanation may be that advisors already had favourable scores on these measures towards physical activity, and comparison with data from a national survey suggests that this was the case . Such advisor selection bias was unavoidable as we wished to collaborate with a Stop Smoking Service that was interested in promoting physical activity at this stage in our development, implementation and evaluation of 'Walk to Quit'. However, time spent promoting physical activity over a typical 6/7-week clinic and self-efficacy for promoting physical activity in smoking cessation clinics increased from before the training to the end of the delivery of the first clinic. Thus, the process of conducting the intervention (and undergoing associated training) appears to have raised the profile of physical activity promotion in smoking cessation clinics, indicating feasibility of this strategy, albeit without a parallel control group. Further research, with adequate power, is needed to explore if training can impact on the professional practice of advisors who have less favourable beliefs about physical activity as a smoking cessation aid.
Ussher and colleagues  reported that about 5 mins of counselling, in addition to usual care, increased quitters' physical activity. In our study, advisors reported an increase from about 6 to almost 10 mins per week in physical activity counselling, which could lead to even greater increases in quitters' physical activity, with hypothetically greater effects on smoking cessation. Ussher had a separate block of time to promote physical activity to individual quitters who had been recruited for a study on the effects of exercise on smoking cessation. In contrast 'Walk-2-Quit' is optimally fully integrated into a whole clinic. This may create problems in recalling when and how much physical activity counselling took place, and more objective observational research is needed to code advisor - client interactions to identify physical activity promoting behaviours and client discussion on physical activity to overcome this limitation.
The intervention impacted upon some client behaviour and beliefs in the anticipated direction, namely stage of readiness to use physical activity as a cessation aid, self-efficacy for smoking cessation and self-efficacy for dealing with stress during the quit attempt. However, it is possible that the latter two outcome effects are a function of the smoking cessation programme rather than 'Walk-2-Quit' per se, and these findings should be interpreted with caution in the absence of a control condition. Future parallel controlled studies with adequate power would ascertain the extent to which such an effect could be attributed to physical activity promotion.
The intervention did not appear to statistically change our measure of clients' mean self-reported levels of physical activity (though clearly there were increases in the median score). This may have been due to the small sample size. It could also be due to the nature of the physical activity being promoted. Sporadic lifestyle physical activity, perhaps useful for temporary management of controlling smoking urges and withdrawal symptoms is less easy to recall and report on a survey than sessions of more intense longer-duration structured physical activity designed for fitness gain. Thus it may be that either the quitters did not consider such sporadic bouts as being 'physical activity' (i.e. a measurement error), or the total amount of activity taken for controlling urges to smoke was not sufficient to raise the total amount of physical activity recorded using the 7-day physical activity recall survey. Further research, controlled and adequately powered, is needed to assess the impact on physical activity using accelerometers (not pedometers) to detect more subtle changes in physical activity. The intervention also did not significantly impact upon clients' self-efficacy for physical activity or outcome expectancy beliefs regarding physical activity as a cessation aid. A longer-duration or more intense/more belief-focused physical activity promotion strategy may be required to impact on these variables, if indeed they are important mediating variables. When the baseline data was compared with a larger national sample from a cross-sectional survey , the clients in the current study had stronger self-efficacy beliefs for being regularly active, weaker self-efficacy beliefs for quitting smoking, did less moderate physical activity in the previous week, and were slightly older. The extent to which the characteristics of the present sample impacted on the findings is unclear, but again, a ceiling effect could explain the lack of change in self-efficacy for physical activity.
Whereas a structured exercise programme may require greater planning and inconvenience, the present study and previous work  suggests that short bouts of physical activity could be promoted at the same time as a quit attempt, with less propensity for 'cognitive overload' for quitters. Indeed, brief bouts of physical activity were seen as a positive coping behaviour that could naturally diminish the urge to smoke and snack [28–30], and could be promoted by advisors for craving self-regulation, particularly during the first four weeks after quitting. This fits well with a recently developed multiple affective behaviour change approach (e.g. [31, 32]), in which the promotion of regular brief bouts of physical activity may regulate mood, which in turn reduces desire to smoke and sugar snack, and engage in other mood regulating behaviours.
During the time of the intervention implementation and evaluation (January to August 2008), 2464 clients accessed the South Birmingham Stop Smoking Service, 46% of whom were verified, with standard CO monitoring, as being quit at 4 weeks. We had very mixed success in recruiting clients through advisors, rather than a designated researcher as is common in research trials, despite frequent contact with the advisors. Comments from the advisors suggested that they were aware of the additional burden of clients completing surveys at the beginning and end of each clinic, and that this may have limited participation. In addition, some advisors reported that the length of the surveys could be off-putting to some clients. However, examining the feasibility of delivering and evaluating the intervention among smokers being treated (rather than recruits into a study) was one of the strengths of the present study.
Clients in the study joined respective Stop Smoking Service clinics unaware of the 'Walk-2-Quit' intervention and evaluation, until asked to give consent and complete baseline evaluation forms. While self-selection bias may have taken place at this point we are confident that the intervention was delivered in a contextually generalisable setting, at least within NHS Stop Smoking Services. In comparison with our published national survey of clients attending a Stop Smoking Service  and advisors , clients in the present study were comparable on age, sex, BMI, cigarettes per day before current quit attempt and times quit smoking in the last year, but did more vigorous physical activity, less moderate physical activity and were less ethnically diverse. Advisors were comparable on BMI, sex and smoking history, but did less vigorous and more moderate physical activity, were slightly older, spent more time promoting physical activity and were less ethnically diverse.