Background
Like many difficult-to-change health behaviours, the process of quitting smoking is complex and often unsuccessful. Although wanting to quit is a necessary condition for attempting to quit, it is not in itself sufficient to ensure success. We know the vast majority of smokers express regret at ever having started to smoke [
1], that most smokers want to quit, and that every year about half attempt to quit [
2], yet annually only 3–5% of smokers successfully quit for at least 12 months [
3,
4].
Clinical practice guidelines and telephone quit-lines in countries such as Australia generally advise smokers that their chances of quitting successfully will improve if they plan their quit attempt in advance. Smokers are advised to set a quit date, address perceived barriers to quitting, seek social support, use pharmacological or behavioural support, and practice strategies to deal with cravings to smoke [
5‐
7]. However, in 2005 the widely held belief that planning is a necessary prerequisite for quitting was challenged when a Canadian GP reported more than half of the smokers and ex-smokers she had interviewed had quit or attempted to quit without any pre-planning [
8]. This finding was subsequently supported by studies in the UK [
9‐
11], USA [
12], and Sweden [
13]. Several of these studies also reported that spontaneous quit attempts were more successful than planned quit attempts [
8‐
10,
12,
13]. In contrast, several International Tobacco Control (ITC) studies reported that neither prior consideration nor delay between decision to quit and implementation was clearly related to quitting success and that there was no clear benefit of planning on short-term (1 month) cessation outcomes [
14,
15]. Interestingly, a recent qualitative study has highlighted the difficulties involved in measuring concepts such as planning and spontaneity in relation to quitting, and the limitations of questionnaire-based surveys when assessing the prevalence and impact of planning on quitting success [
16].
Aim and scope of the grounded theory study
Qualitative research has the potential to make a significant contribution to our understanding of the process of quitting by offering deep insights into the experiences of smokers when they quit. Grounded theory is a qualitative methodology that has already been used to better understand processes involved in difficult-to-change health behaviors [
17]. For example, grounded theory studies in the UK have provided valuable insights into why clients seek professional treatment for drinking problems, which lead to the development of a model of the behavior-change process while utilizing these services [
18,
19].
In this paper we report on a grounded theory study using in-depth, one-on-one interviews with recent ex-smokers (quit >6 months but <24 months). The current study is part of a larger qualitative study exploring how and why many smokers in Australia quit without using assistance despite pharmaceutical and professional smoking cessation assistance being affordable and widely available. It is anticipated the results of this study could provide rich information about the complex and highly variable process of quitting. It is hoped this information could inform a more nuanced response to the challenge of smoking cessation perhaps, for example, by providing campaign developers with insights that might allow them to develop more targeted quit campaigns tailored to the needs of specific audiences. Our purposive sampling strategy initially focused on ex-smokers who had quit without pharmacological or professional assistance as this was our primary area of interest and is an understudied area of research [
20]. We subsequently expanded our sampling to include smokers who had used assistance to quit to allow us to make analytical comparisons across cases and conditions. Our initial analysis indicated that there were more similarities than differences between the two methods of quitting and that using assistance appeared to be only one of many parts of a complicated process. In the initial analysis we also noticed very few participants appeared to have quit spontaneously (i.e. without any planning or preparation). This was noteworthy as this contradicts what many quantitative, survey-based studies into spontaneity and quitting have reported. Based on our initial findings, our subsequent analysis examined: (1) the process of successful quitting from the recent ex-smokers’ perspective; and (2) the concepts of spontaneity and planning in the participants’ accounts of quitting.
Discussion
We have created a typology that accounts for the experience of quitting as reported by all 37 participants. The typology is based on a number of characteristics seen across the different accounts of quitting. These characteristics interact to create a typology of quitting experiences: measured, opportunistic, unexpected or naïve. Three of these typologies were directly observed in participants’ accounts of quitting; the fourth (naïve) remains speculative. We hypothesise that naïve quitters are likely to have been light, social, intermittent, phantom or defensive smokers who may not have self-identified as smokers and therefore may not self-identify, once quit, as being an ex-smoker [
35,
36]. It is possible such ex-smokers did not come forward in response to our recruitment strategies as they may not have considered the study relevant to them or their experience of quitting.
This typology of quitting experiences may help smoking cessation researchers better understand what spontaneity and planning mean in relation to successful quitting, concepts that have been acknowledged by some to be more complex than the way in which they are currently conceptualized [
16,
37,
38]. The typology provides a new conceptual framework for understanding the process of successful quitting that accounts for: (1) how quit attempts and quitting success can be driven by rational plans
and impulsive behavior, and (2) how the concept of planning should not necessarily be limited to the period immediately prior to the quit attempt but could be expanded to include planning learnt, left-over or carried forward from an earlier quit attempt.
In this study we found very few participants quit completely out-of-the-blue with little or no preparation or planning. For most participants quitting involved some form of pre-planning or preparation, making them measured or opportunistic quitters rather than unexpected quitters. In contrast, many other studies on quitting report that a significant proportion of smokers and ex-smokers quit without planning (37–52%) [
8‐
13]. Several of these studies also report that spontaneous quit attempts are more successful than planned quit attempts [
8‐
10,
12,
13]. Our findings are in line with those of a recent prospective US study of quit attempts in real-world settings which reported that although unplanned attempts were more prevalent (defining “planned” quit attempts as “attempts preceded by an intention not to smoke the next day”), planned attempts were more likely to succeed [
38].
We suggest that some of the reported differences in prevalence and effectiveness of spontaneous versus planned quitting might be explained by two factors. The first is the lack of clarity surrounding what spontaneity and planning mean and the consequent difficulties inherent in measuring these concepts, an issue others have raised when attempting to understand the different results from studies into spontaneous quitting [
11,
14]. We note that several studies [
10,
12,
13,
39] reporting on the prevalence of planned versus unplanned quitting relied on a single question from the 2005 British Marketing Research Bureau household omnibus survey [
9]. The question asked: ‘Which of these statements best describes how your most recent quit attempt started?’ to which the first response was ‘I didn’t plan the quit attempt in advance; I just did it’. It is possible the emotive Nike
® slogan-like phrase (‘I just did it’) may have influenced how participants responded. Smokers, like others seeking to change health-related behaviors, often see themselves or wish to be perceived as central to their success even when they have used some form of assistance [
19,
40]. Furthermore, its position as the first response of eight may have resulted in a response-order effect [
41]. These factors may in part explain Murray’s 2010 finding that on in-depth questioning, many of their participants who had originally responded ‘I didn’t plan the quit attempt in advance; I just did it’ had been misclassified as spontaneous quitters. Murray’s in-depth interviews revealed that these participants had either delayed their quitting or had used some form of assistance when they quit and therefore had not actually quit spontaneously [
16].
The second explanation for the difference between studies is that previously researchers have tended to assume that spontaneity and planning are mutually exclusive: our findings challenge this assumption [
8,
9,
12]. At first glance, a substantial proportion of our participants did indeed appear to have quit spontaneously, often in response to what was essentially a minor trigger. However on examining their smoking and quitting history it became clear that for many of these participants quitting had not come out-of-the-blue. This is in keeping with what Cooper and colleagues report, that most quit attempts were not made on the spur of the moment but were preceded by a period of serious consideration [
14]. Many of the participants in the current study had invested time and effort into thinking about quitting, and some had made plans to quit. In these participants it was the exact timing or initiation of the quit attempt that was spontaneous or unplanned, not the quitting
per se. Thus, these opportunistic quitters demonstrate that quitting can include elements of both spontaneity and planning. The presence of spontaneity and planning in the process of quitting reflects current theorizing about how people think, how they make decisions, and how their motivational system generates action. The presence of spontaneity and planning is reminiscent of Haidt’s elephant and rider metaphor [
42], and Kahneman’s explanation in
Thinking, Fast and Slow of why human beings depart in systematic ways from standard economic approaches to rationality [
25].
Our analysis suggests the process of quitting involves both sudden (impulsive) and gradual (reflective) components. The existence of impulsive and reflective components lends further support to claims that behaviourist theories [
26,
27] and rationality-based cognitive theories (e.g. the transtheorectical model of behaviour change, also known as stages of change or SOC) [
28] only go so far in explaining hard-to-maintain behavior change such as quitting smoking [
30]. For example, the SOC model assumes individuals make rational, coherent and stable plans that gradually move them closer to achieving a permanent change in their behaviour. This would mean smokers make a clear decision to quit, set a date to quit, and then act on this intention (i.e. decide, plan, implement). In the current study, the SOC model would be able to account for the behavior of measured quitters, but would not be able to account for opportunistic or unexpected quitters.
A number of researchers have already challenged the relevance of rationality-based cognitive theories such as SOC to smoking cessation [
8,
9,
13,
43]. Our analysis supports parts of West’s 2005 critique of the SOC model, notably the suggestion that transition through pre-action stages is not always the norm or even necessary for successful change, that the change process is much more dynamic, heterogeneous and stimulus-driven than is implied by the model, and that the SOC model places too much emphasis on conscious decision-making [
43]. In addition, the SOC model fails to take into account the strong situational determinants of behavior, and the fact that behavior change can arise from a response to a trigger even in apparently unmotivated individuals.
A number of alternatives have been proposed that take into account the unpredictable and dynamic nature of quitting and in particular the role of spontaneity in quitting. The catastrophe theory, based on chaos theory, proposes that tensions develop in systems in such a way that even small triggers can lead to sudden catastrophic changes [
9]. According to catastrophe theory, quitting can take place unexpectedly without the smoker going through the slow process of cognitive shifts, quitting plans and intentions, and finally action. Instead, the catastrophe theory proposes smokers experience tension, or dissonance, about their smoking over a period of time but don’t act until a precipitating event triggers action. Although compelling, the catastrophe theory’s premise that many if not the majority of quit attempts are sudden and spontaneous and largely devoid of anticipatory planning does not fit with our typology of quitting experiences: as mentioned earlier, many participants who at first appeared to have quit spontaneously had actually invested time and effort into thinking about quitting.
Our typology of quitting experiences is perhaps more consistent with comprehensive theories of addiction such as West’s PRIME theory of motivation [
29] and theories of hard-to-maintain behavior change such as Borland’s CEOS dual process theory [
30]. These theories integrate both spontaneity and planning into the process of smoking cessation. Our typology of quitting experiences demonstrates explicitly what Borland has proposed that PRIME theory implicitly assumes: ‘spontaneity relates to peaks in fluctuating levels of longer term concern; that is, that “spontaneous” quit attempts are typically preceded by periods of deliberation that are not strong enough to trigger action rather than occurring completely out of the blue’ [
37].
Successful quitting, like other behaviour changes, appears to be a struggle between our rational, reflective selves and our impulsive natures [
25]. Most smokers know smoking is harmful, and most smokers want to quit. Yet their behavior is often at odds with what they know they should do. The current study indicates that for many of the participants quitting was characterized by a slow movement towards achieving that goal, with only a few of the participants taking an accelerated pathway triggered suddenly and unexpectedly by significant external events such as a diagnosis of a smoking-related illness. Many of the participants were instead influenced by a multitude of environmental and social factors and gradually come round to accepting that what they were doing (smoking) was at odds with what they valued or believed in (being in control, staying healthy, being a good role model). For some, this was a slow slog with multiple attempts to quit before success was achieved, others managed to opportunistically leverage their success off a timely trigger, while relatively few quit suddenly and unexpectedly when faced with an existential or identity threat.
Strengths and limitations
We spoke directly and in-depth to successful recent ex-smokers. By allowing participants to talk freely and at length about their quitting experiences the data collection focused on what smokers perceived to be important. Data collection and analysis were not restricted to variables predetermined by the researchers or to a pre-existing theoretical framework. By recruiting ex-smokers who had quit in the previous 6 months to 2 years we balanced risk of relapse to smoking [
31,
32] against potential for recall bias [
33]. Approximately two-thirds of participants had quit on their own, reflecting recently reported Australian rates of smoking cessation assistance use [
34].
We did not observe any naïve quitters among participants. We believe naïve quitters are likely to have been former light, social, intermittent, phantom or defensive smokers [
35,
36], and may potentially have self-identified as non-smokers rather than serious or regular smokers. On quitting such smokers may not self-identify as being a former smoker, making our study irrelevant to them. In contrast, our study is likely to have appealed to former smokers who had smoked heavily or regularly and for whom quitting had been a far more significant event in their lives. Future research could explore the hypothesised category of naïve quitters to establish whether this quitting experience and its proposed characteristics exist.
Acknowledgements
We wish to acknowledge the participants who generously shared their time and their experiences of quitting with us.