Does Self-Reported Smoking Cessation Fatigue Predict Making Quit Attempts and Sustained Abstinence Among Adults who Smoke Regularly?

Abstract

Background:

Quitting smoking is difficult and many people who smoke experience cessation fatigue (CF) as a result of multiple failed attempts. This study examined the association of CF with making and sustaining a smoking quit attempt.

Methods:

Data analysed were 4,139 adults (aged 18 years or older) who smoked daily or weekly and participated in the 2018 and 2020 International Tobacco Control Four Country Smoking and Vaping Surveys (ITC 4CV) conducted in Australia, Canada, England, and the US. CF was assessed at baseline using a single question: “To what extent are you tired of trying to quit smoking?” with response options: “Not at all tired”; “Slightly tired”; “Moderately tired”; “Very tired”; or “Extremely tired”. We used binary logistic regression models to test the hypothesis that baseline CF would predict lower odds of both making a quit attempt and sustaining abstinence for a month or longer at follow-up adjusted for socio-demographic and smoking/vaping-related covariates.

Results:

Persons who currently smoked and reported at least some CF were more likely to make a quit attempt, but less likely to sustain abstinence for at least one month, than those who reported no CF. These associations were independent of socio-demographic variables, and they did not differ by country.

Conclusion:

Contrary to expectation, CF was positively associated with making a quit attempt and non-linearly associated with lower rates of sustained abstinence at follow-up. While these findings should be replicated, they suggest that people with CF may benefit from targeted support to remain abstinent after a quit attempt.

1. INTRODUCTION

Commercial tobacco use remains a leading cause of preventable death and disease globally, including cancer, cardiovascular, and respiratory diseases (ABS, 2022; Centers for Disease Control and Prevention, 2020; World Health Organisation, 2021). Most people who smoke tobacco want to quit and about half make a quit attempt every year, however 90% of those who attempt relapse within 6–12 months (Hughes et al., 2004). Evidence suggests that many people who smoke expend significant effort attempting to quit smoking, making on average 30 attempts to quit before maintaining long-term abstinence (Borland et al., 2011; Chaiton et al., 2016).

The effort expended to engage in so many attempts to quit has been hypothesized to contribute to cessation fatigue (CF). CF consists of a loss of motivation to quit, reduced utilization of coping skills, loss of hope in cessation success, and decreased self-efficacy (Heckman et al., 2018; Piasecki et al., 2002). CF has been hypothesized to undermine motivation to quit, resulting in fewer quit attempts and fewer attempts that result in successful long-term abstinence (Piasecki et al., 2002). To date, research that examines the influence of CF on future attempts to quit and the outcomes of these attempts is limited. Understanding CF’s role in cessation outcomes may help identify people who are less likely to succeed in stopping smoking and who may benefit from additional support to help to quit.

Smoking dependence is difficult to overcome for many people who smoke. Smoking cessation often involves repeatedly choosing between the known, immediate, and reinforcing experience of smoking and a variety of known (e.g., withdrawals) and unknown consequences associated with quitting smoking. These choices happen in the context of continually fluctuating neurobiological, environmental, and cognitive influences, all while managing the demands of daily life. The cognitive and behavioural adaptations made throughout the day to support cessation include self-monitoring, planning, goal setting, and finding alternative strategies for managing stress and negative feelings without smoking (Liu et al., 2013; Piasecki et al., 2002). For instance, eliminating the first cigarette of the day with coffee entails planning for and carrying out alternative plans, all while possibly experiencing significant withdrawal symptoms. Actively managing a series of such events throughout the day, for weeks or months, can feel overwhelming. Such experiences of the toll on oneself from smoking dependence and smoking cessation attempts may ultimately contribute to the feeling of ‘tiredness’ of trying to quit smoking, in other words, contributing to the experience of CF during smoking cessation attempts.

Several factors might potentially influence the relation between CF and quit attempts and abstinence from smoking. Sociodemographic factors such as sex, gender, age, ethnicity, education, and income are all associated with the likelihood of achieving abstinence from smoking (Girvalaki et al., 2020; Heckman et al., 2018). Other factors associated with abstinence include indicators of nicotine dependence and withdrawal, such as the number of cigarettes smoked per day, strength of urges and cravings, and time to first cigarette (Liu et al., 2013; Mathew et al., 2017). Increased CF has already been linked with reduced self-efficacy and motivation, two factors which are important to achieving short and long-term abstinence (Hopkins et al., 2022; Piasecki et al., 2002). The use of smoking cessation aids such as nicotine replacement therapy or electronic cigarettes may also influence CF by suppressing withdrawal-related cravings and dependence on tobacco for nicotine (McDermott et al., 2021). Finally, the number of past quit attempts might influence the development and influence of CF on abstinence. A greater number of quit attempts might increase the level of CF and the relation between CF and long-term abstinence. An examination of the relation between CF and quit attempts and abstinence should account for these factors.

To date there are a limited number of studies on CF and smoking cessation outcomes (Liu et al 2013; Heckman et al 2018; Mathew et al. 2017). In a small study of people who smoke and who were highly motivated to quit followed up over a two-month period, Heckman et al. (2018) found that CF was associated with a delay in smoking cessation initiation, and a lower likelihood of successful abstinence providing a strong suggestion that CF may undermine smoking cessation. However, it is unclear if the findings would generalise to the real world given the study sample was based on individuals who were highly motivated to quit. Furthermore, the focus of this study was on the first 12 weeks of quitting which limits the ability to identify any long-term effects of CF on smoking cessation initiation and successful maintenance of smoking cessation. Using a cross-sectional design, Mathew et al. (2017) recruited a small sample of people who currently smoke and had recently relapsed for their study and found that respondents with higher CF scores reported greater severity of nicotine withdrawal and difficulty quitting. However, this study was based on cross-sectional data and does not provide insights into the directionality of effects. Liu et al (2013) studied clinical trial data from a sample of people who smoke daily who were motivated to quit smoking and also found that CF was associated with a lower likelihood of successful smoking cessation. Like Heckman et al.’s study, the findings of this study require replication as they may not generalise to the general smoking population. Additionally, all three studies were limited to participants in the United States. It would be beneficial to determine if such findings can be replicated in Australia and other countries outside of the US.

The current study sought to gain a better understanding of the role of CF experienced in the real world by adults who smoke in population surveys conducted in Australia, England, Canada, and the United States. This study addresses the following research questions:

1. Are people who currently smoke and experience CF less likely to make a quit attempt as compared to their counterparts who do not experience CF?

2. Are people who currently smoke and experience CF less likely to sustain smoking abstinence for a month or more if they make a quit attempt compared to their counterparts who do not experience CF?

3. Does any of the above differ by country?

   H1: People who experience higher levels of CF are:

   1. less likely to make quit attempts, and
   2. less likely to achieve abstinence from cigarette smoking for one month.

Being exploratory, no specific hypothesis was tested for research question 3.

2. METHOD

2.1. Participants

The data analysed came from the International Tobacco Control Four Country Smoking and Vaping Survey (ITC 4CV; ITC Project 2020). The ITC 4CV is a cohort study consisting of people who currently smoke cigarettes and people who have recently quit (within two years) that were aged 18 years or older. The study involves parallel online surveys conducted in Australia, Canada, England, and the United States. Two successive waves of the ITC 4CV were utilised in our study, these being the 2018 Wave (our baseline, February-July 2018) and the 2020 follow-up Wave (February-June 2020). The survey took participants approximately 40 minutes to complete. Participants were offered an incentive for completing the survey. Further description of the survey methodology has been documented elsewhere (ITC Project, 2020; ITC Project, 2021; Thompson et al., 2019) and the full surveys are available online (https://itcproject.org).

The current study is a prospective cohort study which utilised data from participants who currently smoked daily/weekly regardless of vaping status at baseline (n=9,815), of whom 4,139 were successfully recontacted at follow-up wave. Figure 1 presents the study sample flow diagram.

Figure 1.

Flow diagram showing the inclusion/exclusion criteria for the study sample.

2.2. Measures

2.2.1. Baseline Predictor Measure

Cessation Fatigue.

CF was determined by asking participants: “To what extent are you tired of trying to quit smoking?” with response options “Not at all tired”; “Slightly tired”; “Moderately tired”; “Very tired”; or “Extremely tired”. “Don’t know” responses were allowed. This single item measure has been shown to predict smoking relapse 6 months after a quit attempt (Liu et al., 2013). For sample distribution, see Supplemental Table S1.

2.2.2. Outcome Measures at 2-year Follow-up

Quit attempt.

Quit attempt data was collected from participants with the question “Since you completed the last survey in 2018, have you tried to stop smoking?” (“yes” vs. “no”). “Don’t know” responses were treated as not having made any quit attempt and coded as “No”.

One-month sustained abstinence from smoking (1MSA).

1MSA was derived based on smoking status of participants who reported making any quit attempts between Baseline and Follow-up waves with ‘Yes’ being those who reported not smoking at Follow-up and had stopped smoking for at least one month, and ‘No’ otherwise. Those who were still quit at Follow-up and had stopped for less than one month were excluded from these analyses since their outcome was indeterminate (n=15).

2.2.3. Covariates

Sociodemographic variables assessed at Baseline were used as control variables to account for their potential influence on quit attempts and abstinence including gender, age, ethnicity, education, income, and country. Other baseline smoking and vaping related variables known to predict quitting behaviours (Borland et al., 2011; Le Grande et al., 2021; Partos et al., 2013) were also included as covariates such as e-cigarette use, smoking frequency, quitting self-efficacy, motivation and intention to quit, number of quit attempts in past 2 years, strength of urges to smoke in the past 24 hours, and time to first cigarette after waking. Quitting self-efficacy was assessed using: “If you decided to give up smoking in the next 6 months, how sure are you that you would succeed (not at all/slightly/moderately/very/or extremely sure)?”. Motivation to quit was assessed using: “How much do you want to quit smoking (not at all/a little/somewhat/or a lot)?” and intention to quit was assessed using: “Are you planning to quit smoking (within the next month, within the next 6 months, some time in the future beyond 6 months, or not planning to quit)?”. Strength of urges to smoke and time to first cigarette after waking were used as indicators of nicotine dependence.

2.3. Data Analyses

Descriptive analyses involved frequency and crosstabulations for categorical variables of interest to examine the sample characteristics and percentage distribution of CF levels, utilising the chi-square test of differences to check for significance.

To examine the association between CF and two outcome variables, ‘making a quit attempt’ and 1MSA, we utilised binary logistic regression models performed in a stepwise manner. In the first model, a binary logistic regression was conducted to examine the unadjusted relationship between CF and each outcome variable. In the second model, sociodemographic variables were added to the model including sex, gender, ethnicity, age, income, education, and country to control for any confounding effect. In the third model, other potential confounders such as e-cigarette use, smoking frequency, self-efficacy, number of past quit attempts, strength of urges to smoke in the past 24-hours, and time to first cigarette were added to the model. In the fourth model, motivational variables (wanting to quit and plan to quit) theorised to be strongly linked to CF (Hopkins et al., 2022; Piasecki et al., 2002) were added to the fourth model to assess whether CF effect was over and above these variables. Lastly, to test for any moderating effect of country, ‘CF x country’ interaction terms were added to the model. Odds ratios and 95% confidence intervals were computed to aid in the interpretation of results. All data analyses were conducted on unweighted data using SPSS Statistics Version 29 and a p-value of < 0.05 was considered statistically significant.

3. RESULTS

3.1. Sample characteristics

Table 1 presents the study sample baseline characteristics along with those lost to attrition. Of note, 43.8% of the sample were aged 55 or older and 87.7% reported being part of the majority ethnic group. All socio-demographic variables and CF levels were associated with differential attrition rates (Table 1).

Table 1.

Baseline Sample Characteristics

Variables	Study Sample (n=4139)	Participants lost to attrition (n=5676)	Test of group differences, p-value
	%	%
Gender
Male	47.4	50.4	0.004
Female	52.6	49.6
Age Group (in years)
18–24	6.1	30.3	<.001
25–39	17.9	23.3
40–54	32.3	22.0
55+	43.8	24.4
Ethnicity
Majority group	87.7	83.4	<.001
Minority group	12.3	16.6
Education
Low	33.6	32.0	0.014
Moderate	40.0	42.9
High	26.4	25.0
Income
Low	33.0	32.3	0.005
Moderate	32.1	34.8
High	29.7	27.1
No answer	5.2	5.8
Country
Canada	28.9	25.1
US	23.6	17.7	<.001
England	30.3	48.1
Australia	17.2	9.0
Smoking cessation fatigue
Not at all tired	20.2	15.6	<.001
Slightly tired	17.6	16.6
Moderately tired	19.5	22.0
Very tired	13.2	12.6
Extremely tired	6.2	7.3
Don’t know	23.3	25.9

Note. Percentages were based on unweighted data.

3.2. The association between cessation fatigue and smoking quit attempts

Out of the 4,139 participants included in the analysis, 42.3% (n = 1,751) made a quit attempt between Waves. Table 2 shows the quit attempt rate by levels of CF. The results of the binary logistic regression analyses are presented in Table 3. Contrary to our hypotheses, those with higher levels of CF were more likely to make a quit attempt than those without CF. For example, in Model 1, those who reported being extremely tired had the highest odds of attempting to quit smoking compared to those who were not at all tired (unadjusted odds ratio [OR] = 2.75, 95 % confidence interval [CI] = 2.05–3.68, p <.001). The association between being extremely tired and quit attempts became slightly weaker after controlling for sociodemographic variables (adjusted odds ratio [AOR] = 2.55, 95 % CI = 1.89–3.43, p <.001) but remained significant after adjusting for tobacco smoking and vaping related variables (smoking frequency, strength of urges to smoke in past 24 hours, minute to first cigarette after waking, how sure will succeed at quitting, past quit attempts, e-cigarette use status) in model 3 (AOR = 1.66, 95 % CI = 1.19–2.31, p = .003). Only in model 4 did it become non-significant, after including motivational variables (wanting to quit and plan to quit) into the model (p = .188). The same pattern of results was observed for the other levels of CF except for moderately tired level which remained significant in model 4 (AOR = 1.30, 95 % CI = 1.02–1.64, p = .032). Interestingly, those who answered ‘Don’t know’ showed the opposite pattern, being less likely to make attempts (AOR = 0.56, 95% CI = 0.45–0.70, p<.001), and this became non-significant from model 3 onwards. There was no significant ‘CF x country’ interaction (p = .504).

Table 2.

Proportion who made any quit attempt and achieved 1-month sustained abstinence at follow-up by levels of cessation fatigue at baseline.

Baseline cessation fatigue level	Cessation outcomes at follow-up
	Made any quit attempt (N=4139)		Achieved 1-m sustained abstinence (N=1736)
	Yes - % (n=1,751)	No - % (n=2,388)	Yes - % (n=510)	No - % (n=1,226)
Not at all tired	34.4 (288)	65.6 (548)	32.9 (94)	67.1 (192)
Slightly tired	45.9 (334)	5.1 (393)	24.5 (81)	75.5 (250)
Moderately tired	51.9 (419)	48.1 (389)	27.4 (113)	72.6 (300)
Very tired	58.7 (320)	41.3 (225)	25.5 (81)	74.5 (237)
Extremely tired	59.7 (154)	40.3 (104)	27.9 (43)	72.1 (111)
Don’t know	24.5 (236)	75.5 (729)	41.9 (98)	58.1 (136)

Note. Percentages were based on unweighted data.

Table 3.

Logistic Regression Results for the Associations of Baseline Levels of Smoking Cessation Fatigue with Making Quit Attempts and 1-Month Sustained Abstinence at Follow-up.

		Model 1			Model 2			Model 3			Model 4
Variables		Unadjusted OR	95% CI	P	Adjusted OR	95% CI	P	Adjusted OR	95% CI	P	Adjusted OR	95% CI	P
Made Quit Attempt (N = 4,139)	% made attempt
How tired of trying to quit				<.001			<.001			<.001			.023
Not at all tired	34.4	ref			ref			ref			ref
Slightly tired	45.9	1.58	1.29–1.95	<.001	1.50	1.21–1.85	<.001	1.31	1.04–1.65	.024	1.19	0.94–1.52	.153
Moderately tired	51.9	2.01	1.64–2.46	<.001	1.89	1.54–2.32	<.001	1.48	1.18–1.86	<.001	1.30	1.02–1.64	.032
Very tired	58.7	2.61	2.08–3.28	<.001	2.47	1.96–3.11	<.001	1.54	1.19–2.00	.001	1.25	0.95–1.64	.107
Extremely tired	59.7	2.75	2.05–3.68	<.001	2.55	1.89–3.43	<.001	1.66	1.19–2.31	.003	1.26	0.89–1.79	.188
Don’t know	24.5	0.60	0.49–0.74	<.001	0.56	0.45–0.70	<.001	0.85	0.68–1.07	.174	0.87	0.68–1.10	.230
1-Month Sustained Abstinence (N = 1,736)	% 1MSA
How tired of trying to quit				<.001			<.001			.202			.366
Not at all tired	32.9	ref			ref			ref			ref
Slightly tired	24.5	0.67	0.47–0.96	.027	0.63	0.44–0.90	.011	0.77	0.53–1.12	.170	0.81	0.55 – 1.19	.278
Moderately tired	27.4	0.77	0.55–1.07	.124	0.74	0.53–1.04	.082	0.97	0.68–1.38	.858	1.01	0.71 – 1.45	.951
Very tired	25.5	0.71	0.50–1.01	.057	0.71	0.49–1.01	.059	1.05	0.71–1.56	.797	1.10	0.74 – 1.64	.647
Extremely tired	27.9	0.79	0.51–1.21	.278	0.73	0.47–1.14	.170	1.08	0.67–1.75	.739	1.08	0.67 – 1.75	.756
Don’t know	41.9	1.53	1.06–2.21	.023	1.47	1.01–2.14	.044	1.31	0.88–1.96	.187	1.29	0.86 – 1.93	.222

Note. OR = odds ratio; CI = confidence interval; Ref = reference group; P = p-value; 1MSA = 1-month sustained smoking abstinence;

Model 1 = unadjusted OR;

Model 2 = OR adjusted for potential socio-demographic confounders (country, gender, ethnicity, age, income, education);

Model 3 = OR also adjusted for smoking and vaping-related variables known to be associated with outcomes (smoking frequency, strength of urges to smoke in past 24 hours, minute to first cigarette after waking, how sure will succeed at quitting, past quit attempts, e-cigarette use status);

Model 4 = OR also adjusted for motivational variables (how much one wants to quit and plan to quit smoking cigarettes).

3.3. The association between cessation fatigue and 1-month sustained abstinence (1MSA)

Of the 42.3% of participants who made a quit attempt between Waves, 29.4% (n = 510) were able to maintain abstinence from smoking for one month or more. As presented in Table 3, there was an overall significant association between levels of CF and 1MSA (p<.001). However, this association seemed to follow a non-linear pattern. When compared to the ‘not at all tired’ group, the group least likely to achieve 1MSA were those who reported being slightly tired, rather than those who reported being extremely tired (unadjusted OR=0.67, 95% CI= 0.47–0.96, p=.027 vs OR=0.79, 95% CI=0.51–1.21, p=.278, respectively). After controlling for sociodemographic variables, the slightly tired group remained less likely to achieve 1MSA (AOR = 0.63, 95% CI = 0.44–0.90, p =.011) but became non-significant in models 3 and 4. Those who responded ‘don’t know’ were more likely to achieve 1MSA compared to those who were not at all tired (unadjusted odds ratio [OR] = 1.53, 95% CI = 1.06–2.21, p =.023) and remained so after controlling for sociodemographic variables (AOR = 1.47, 95 % CI = 1.01–2.14, p = .044), but became non-significant in both models 3 (p = .187) and 4 (p = 0.222). There was no significant ‘CF x country’ interaction (p = .722).

4. DISCUSSION

Contrary to our hypothesis, we found that any level of CF was significantly associated with increased odds of making a smoking quit attempt, but the association was greatly reduced and of marginal significance when controlling for other motivational measures. Generally consistent with our hypothesis, among those who made a quit attempt, CF reduced odds of sustaining abstinence for at least one month, but this association appeared non-linear and was not independent of dependence and motivational measures. The influence of CF appeared to have similar effects among people in the four countries studied.

The finding that CF is associated with an increased likelihood of making quit attempts is inconsistent with the findings of prior research showing the opposite result (Heckman et al., 2018; Mathew et al., 2017). One plausible reason for the discrepant finding may be the time interval between baseline and follow-up surveys in our study. Consistent with the notion that fatigue dissipates over time (Piasecki et al., 2002), as our study interval spanned 24 months, this might have provided an adequate length of time for those with some CF to recover from said fatigue, enabling them to make another quit attempt. By contrast, the study by Heckman et al. (2018) was conducted over a 12-week period, which might not be a satisfactory length of time for those experiencing CF to recuperate and try again, hence the different findings to this study. Further research into the recovery timeframe required for those who experience high level of CF before making another quit attempt may be beneficial to understand this further.

Our finding suggests that CF may also serve as an indication of an individual’s motivation to take action. Individuals who experience more CF may be those who have greater motivation to quit in the first place but the experienced difficulty to successfully quit is more likely to be seen as fatigue if the underlying motivation to quit remains strong. Given the small residual association once motivational measures were controlled for, it seems that CF might also temporarily reduced the reported strength of motivational measures such as planning and wanting to quit which are strong predictors (Borland et al., 2011; Le Grande et al., 2021; Partos et al., 2013). Thus, CF, when present, is a useful addition to understanding how quitting motivation changes over time.

The finding for sustained abstinence was generally consistent with findings from prior research showing a negative association between CF and abstinence (Heckman et al. 2018; Liu et al., 2013; Mathew et al., 2017) suggesting that there is something about CF that may negatively influence people’s ability to achieve abstinence for a sustained period. This may be because CF interferes with their ability to self-monitor and cope with cravings engendered from nicotine deprivation (Liu et al., 2013; Piper, 2015). Therefore, it is unsurprising that individuals who experience more CF will be at a greater risk of relapse.

There are two noteworthy aspects of our findings. Firstly, we found a non-linear association between CF and 1MSA. Surprisingly, those who were slightly fatigued were the least likely to achieve 1MSA rather than those who were extremely fatigued, and this might be due to those experiencing higher CF being more likely to take mitigating measures such as use of quit aids to improve their likelihood of success. This explanation appears credible based on additional analysis indicating a positive linear association between CF and use of nicotine replacement therapy for last quit attempt (see Supplementary Table S2).

Secondly, the effect of CF on 1MSA was reduced once other vaping and smoking related variables were controlled for in our model, suggesting that one or more of these variables may act as a mediating or confounding factor in the relationship between CF and 1MSA. Future research is needed to explore both the direct and indirect effects of CF on smoking abstinence in order to acquire further insights.

Also of note is the substantial number (56%) who expressed experiencing CF in our sample which is consistent with the difficulty of quitting and repeated failures experienced by most smokers trying to quit (Partos et al. 2013). The subgroup of almost a quarter (23%) who expressed “Don’t know” to the survey question on CF warrants further research to gain insights into why this subgroup were less likely to initiate a quit attempt but surprisingly were more likely to succeed if they tried.

4.1. Strengths and limitations

Study strengths include the prospective cohort study design which enabled us to assess the impact of CF over a period of 18 months, a large sample size and multi-country data to enhance the generalizability of our findings, and ability to control for a large range of potential confounders and possible alternative influences. Several limitations warrant discussion. First, the high rate of attrition between Wave 2 and Wave 3 may bias our findings. To mitigate this bias, socio-demographic variables associated with differential attrition were controlled for in our analysis, but due to the fact that CF was also associated with attrition, future research is needed to confirm our findings. Second, reliance on self-report data which are subject to memory recall bias particularly given the long follow-up interval and social desirability biases, which may have led to short duration quit attempts undertaken early in the interwave period being forgotten. Third, in the item measuring CF the word ‘tired’ is not operationalized, therefore, participants may have had different interpretations of the meaning of this item. Fourth, the use of a single-item measurement of CF may not adequately capture all aspects of CF. While our single-item measure has demonstrated predictive utility for smoking cessation behaviour, future research could investigate the degree to which CF is a multidimensional construct that could be assessed with a multi-item measure which would have greater predictive sensitivity.

4.2. Clinical implications

Our findings are consistent with CF having a temporary effect, but one where there is recovery over time. Those reporting CF might be allowed a bit of time to recover, but then should be supported to try again when they are ready. Where CF occurs, it likely indexes a high underlying motivation to quit, but limited capacity, so urging the use of an optimal mix of support tools should be encouraged.

Our single-item CF measure can be a useful tool that can be utilized during triage in clinical practice to detect and address CF in individuals, through tailored support from the clinician (Heckman et al., 2018). Such management of CF may be a critical aspect of effective care (Mathew et al., 2017). Being easy to administer, clinicians can use this measure over time to track any changes in CF levels, which may therefore aid in monitoring interventions and adapting relapse prevention plans as needed. Further research may be conducted to determine how existing smoking cessation evidence-based treatments affect fatigue, and if there is need for CF specific interventions (Heckman et al., 2018). An additional focus of future research may be a measure which incorporates the experience of CF for those who have already quit smoking and are making efforts to sustain abstinence from smoking.

4.3. Conclusion

This study found that contrary to expectation, adults who smoke regularly and experience any CF were not discouraged from making another smoking cessation attempt but were in fact more likely to do so over a 2-year period compared to those not experiencing CF. This unexpected finding may be due to the extended time between surveys, allowing sufficient recovery time from CF for people to make another quit attempt and CF being associated with a high underlying level of motivation to quit. The association between CF and sustained abstinence from smoking needs further replication. If confirmed, it would suggest that those prone to CF may be more likely to fail again in their next quit attempt as CF would set in quicker or be more intense during quitting, underscoring the need to identify and provide extra support to those experiencing CF to improve their cessation outcomes.

Data Availability Statement:

In each country participating in the international Tobacco Control Policy Evaluation (ITC) Project, the data are jointly owned by the lead researcher(s) in that country and the ITC Project at the University of Waterloo. Data from the ITC Project are available to approved researchers 2 years after the date of issuance of cleaned data sets by the ITC Data Management Centre. Researchers interested in using ITC data are required to apply for approval by submitting an International Tobacco Control Data Repository (ITCDR) request application and subsequently to sign an ITCDR Data Usage Agreement. The criteria for data usage approval and the contents of the Data Usage Agreement are described online (http://www.itcproject.org).