A Comparison of the Effectiveness of Two Approaches in Achieving Smoking Abstinence

Introduction:

A brief intervention with a small to moderate effect size can potentially benefit a large number of smokers and increase smoking cessation within the community if it is carried out routinely in clinical practices by all or most healthcare professionals or people with minimal training. It is also the most cost-effective smoking cessation programme, because no extra or minimal funding is needed to provide the venue, manpower, and other expenses (but incentives or payments to healthcare professionals and follow-up support would be needed). There is no evidence that longer interventions are more effective than shorter interventions. Brief cessation interventions have been shown to be effective with strong evidence from the investigators' randomised controlled trials and in systematic reviews.

The negotiating self-determination to reduce cigarette consumption is guided by social cognitive and self-determination theories. According to social cognitive theory, self-efficacy is an important personal determinant of human behaviour and has been defined as the belief in one's capability to engage in behaviour to solve difficult tasks. This belief influences decisions on whether a certain form of behaviour will be adopted and maintained. Because self-efficacy is built on a successful experience of overcoming challenging tasks, smokers who have more successful experiences in reducing cigarette consumption tend to have higher levels of self-efficacy. Some evidence has shown that a reduction in smoking may lead to greater self-efficacy to resist smoking, which could increase subsequent quitting. According to self-determination theory, autonomy is another influential determinant of behaviours which is emphasized by freedom of choice. Studies have shown that patients having an opportunity to decide on their own treatment may feel more eager to comply with instructions. The subjects in this study will be allowed to select their own schedules of smoking reduction after the negotiation with the counsellor, such as the percentage of smoking reduction over an acceptable period of time. It is anticipated that the subjects will show more willingness to adhere to their own schedule as a result of an increase in autonomy. Moreover, some evidence has shown that autonomy is positively associated with competence; that is, people have greater autonomy demonstrate higher competence in achieving behavioural change. Consequently, autonomy will facilitate their gradual reduction or cessation of smoking. Most importantly, as a result of reducing the cigarette consumption and lowering down the nicotine dependence, it would be much easier for such smokers to further reduce or quit.

Methods:

Chinese patients attending the general out-patient clinics in Hong Kong for routine follow-up visits who fulfil the following inclusion criteria will be invited to participate in the study.

All smoking patients will be approached by a nurse counselors.

Analysis:

Data analysis will be performed using the Statistical Package for Social Science. In particular, we will use inferential statistics (independent-samples t-test and χ2 test) to examine the homogeneity between the QI and CDTQ groups. We will conduct logistic regression analyses to calculate the odds ratio (OR) and 95% confidence interval (CI) for smoking outcomes, i.e. self-reported and biochemical validated quit rates, self-reported smoking reduction rates and number of quit attempts. We will apply mixed between-within subjects analysis of variance (ANOVA) to determine whether CDTQ is more effective than QI in enhancing self-efficacy (perceived importance, difficulty and confidence) for quitting smoking. We shall adopt the principle of intention to treat. Subjects who are lost to follow-up, refuse to participate in biochemical validations, or withdraw from the study are considered unable to quit or reduce smoking. Depending on the amount and distribution of missing data, other methods, e.g. sensitivity analysis and multiple imputation technique will be used if necessary.