The Dynamics of Smoking Quit Motivation in Daily Life: Associations with Momentary Self-Regulation and Nightly Quit Intentions

Abstract

Due to prior work suggesting dynamic fluctuations in quit motivation over time, the current study used ecological momentary assessment (EMA) to examine contextual predictors of momentary quit motivation, most notably perceived self-regulation. The sample (n = 84) intentionally excluded smokers actively trying to quit and those who plan to never quit, focusing on the group of smokers who are interested in but not yet committed to quitting. Participants completed one week of EMA, where they responded to random prompts 7x/day and logged each cigarette. At each prompt, they completed measures assessing cigarette craving, affect (positive and negative), quit motivation, and perceived momentary self-regulation (i.e., willpower self-efficacy, distress intolerance, craving uncontrollability). Participants also completed a nightly prompt where they reported their quit intentions. Results revealed that the newly developed 4-item index of quit motivation had strong psychometric properties, and demonstrated fluctuations over time and across situations. Quit motivation was higher when craving was ideographically lower than usual, and when willpower was higher than usual. People with higher distress intolerance reported higher quit motivation. In addition, stronger quit motivation and less variability in quit motivation was associated with higher quit intentions. The current study suggests that quit motivation does shift dynamically and speaks to the potential importance of targeting craving and willpower self-efficacy to enhance people’s motivation to quit.

Most smokers know that smoking is a health hazard, yet despite half of smokers attempting to quit each year, less than 10% are successful ¹. Increasing both quit attempts and cessation success rates are important goals which both require an understanding of motivation and self-regulation. Without self-regulation—the processes needed to pursue and attain goals ²—smokers are unlikely to attempt cessation, nor are they likely to successfully regulate the craving, physical withdrawal and negative affect that accompany quit attempts ^3,4. Importantly, active quit attempts are typically preceded by quit intentions (i.e., a stated resolution or plan such as “I’m going to quit smoking next week”; Berli et al., 2015; Herzog et al., 2015), which are in turn preceded by motivational processes such as desire (e.g. “I’d like to quit smoking”; Schwarzer & Luszczynska, 2008). Considering evidence that quit motivation and quit intentions vary over time ^5–8, an understanding of motivation and self-regulation underlying quit intentions will involve examining these factors dynamically.

Variability in Quit Motivation

Over two-thirds of smokers report that their motivation to quit changes daily ⁶, perceptions which are confirmed by daily diary studies ⁸. Indeed, greater fluctuations in motivation predict cessation lapse in smokers undergoing a quit attempt ⁹. However, it is currently unclear what factors coincide with momentary quit motivation fluctuations. Affect and craving are also motivational concepts, where people are more likely to approach smoking when negative affect and craving are elevated ³. Yet, an experimental study found that participants randomized to view either negative affect or cigarette images reported greater motivation to quit than participants who viewed neutral images ¹⁰. These conflicting ideas suggest the importance of examining real-life fluctuations in affect and craving as related to quit motivation.

In addition, little is known about how quit motivation fluctuates for people who are considering quitting. Roughly 68% of smokers indicate that they want to quit ¹. Understanding motivation variability in people considering quitting is important because these smokers may be amenable to interventions to stabilize desire to quit and ultimately attempt cessation if the parameters of motivation variability were better understood ¹¹.

Perceived Self-Regulation Abilities

When people have higher cessation self-efficacy¹² they also report greater quit intentions ^5,9. Here we propose that when smokers feel generally more self-efficacious at self-regulation—exerting willpower, tolerating distress, and controlling cravings—they will feel more motivated to quit. These three aspects reference general (i.e., perceived willpower as a general sense of being able to exert control over behavior ¹³), emotion-specific (i.e., withstanding negative emotion) and smoking-specific (i.e., withstanding or controlling cigarette cravings) aspects of self-regulation. Self-efficacy for exerting willpower and perceived self-efficacy for tolerating distress both fluctuate dynamically over time ^13–15. Theoretically, shifts in self-regulation alter motivation ^16,17, which suggests that when people generally feel less capable of self-regulation, they may also likely to feel less motivated to quit smoking.

Motivation Predicting Intention

According to the Health Action Process Approach (HAPA) theory ¹⁸, motivation is a pre-intentional process that occurs before people form intentions to change a health behavior. For example, motivation to quit smoking reflects the desire to be smoke-free (e.g., “I wish I weren’t a smoker anymore”), and intention is a stated plan to enact the desire, such as “I will stop smoking on Monday.” A prior study found that higher motivation predicted stronger intentions to quit ¹⁰, and we propose these relationships are likely to be found in daily life (e.g., via ecological momentary assessment [EMA]). In addition, as prior work has shown that instability in motivation disrupts intentions ¹⁹, it may be that greater variability in momentary quit motivation is associated with lower quit intentions.

Current Study

The purpose of this study was to examine the dynamics of momentary quit motivation in smokers who are considering (but not actively attempting) quitting. EMA is an ideal method for this aim as it involves asking people to report on their thoughts, feelings, and situations throughout their day ²⁰. We predicted that motivation to quit smoking would be elevated alongside higher negative affect and higher craving ¹⁰. We also predicted that people would generally feel more motivated to quit when reporting greater perceived momentary self-regulation (greater willpower, lower distress intolerance, higher craving controllability). Finally, we expected that higher quit motivation and lower variability in quit motivation would predict greater nightly quit intentions. All data and syntax are available here: https://osf.io/9jszb/?view_only=4025104c8d314db5951845aa47a8774d

Method

Recruitment and Participants

All procedures were approved by the Institutional Review Board at [University Name]. Daily smokers who had smoked at least ten nicotine cigarettes (not e-cigarettes) per day for the last year were recruited via local advertisements. People who indicated they never planned to quit, those who reported a current active quit attempt, and those who reported smoking e-cigarettes were excluded. Only participants between the ages of 25 and 65¹ were recruited, because younger smokers are typically less interested in quitting ²¹. Finally, participants had to be able to read in English, willing to download an application to their Android or iPhone smartphone², and indicated no diagnoses of heart disease, lung cancer, or chronic obstructive pulmonary disease.

In total, 88 participants completed the study. Most (n = 58) completed baseline measures and EMA orientation in the lab, the remainder (n = 29) were onboarded online due to the COVID-19 pandemic. Four participants were excluded, one for not completing any EMA notifications and three for technology problems. The final sample size was 84, 45.8% women and 90.4% White, with an average age of 40.65 (SD = 10.01). The participants smoked an average of 16.33 (SD = 5.94) cigarettes per day, with a baseline contemplation ladder score of 5.87 (SD = 1.92) which suggests that most participants were thinking about quitting but not quite ready to quit. The average Fägerstrom Test of Nicotine Dependence (FTND) ²² score was 5.19 (SD = 1.92), suggesting moderate nicotine dependence. There were no differences in demographic or smoking variables between those who started the study in person versus those who began the study online.

Procedure

After providing consent, all participants completed individual difference measures via Qualtrics. Participants were then guided to download the LifeData application to their Android or iPhone smartphone, where they completed a brief orientation session to the EMA protocol, and they were taught how to log cigarettes on the app. The one week of EMA data collection began immediately thereafter.

For one week, participants received about 7 random prompts per day between the hours of 9:30am and 9:30pm, and were asked to respond to the prompt within five minutes³. Participants were also asked to log all of their cigarettes. On one-third of the logged cigarettes, participants completed the same set of questions as the random prompts. Notably, at the end of each random prompt, participants were asked “are you smoking right now?” If they said yes, the session was converted to a smoking session; these converted smoking sessions were combined with the data from logged cigarettes to represent all sessions in which smoking was active. In addition, at 9:35pm each night, participants were prompted with nightly assessments of quit intentions.

EMA Measures

Momentary Prompts

Emotion.

Participants indicated the degree to which (0 = not at all to 6 = extremely) they were currently feeling ten different emotions (lonely, sad, scared, nervous, angry, irritable, joyful, excited, calm, relaxed). These emotion adjectives were similar to those used in other studies ¹⁵, and grouped reliably into positive (within α = .74; between α = .85) and negative affect subscales (within α = .74; between α = .86).

Craving.

Participants were asked to rate their craving to smoke and their desire to resist smoking on 0 to 100 visual analogue scales.

Smoking Quit Motivation.

Participants were asked four items assessing smoking motivation: (1) Right now, I wish I weren’t a smoker. (2) Right now, I want to quit smoking. (3) Right now, I am willing to do whatever it takes to quit smoking, and (4) Right now, I’d like to avoid cigarettes. These items were given on 1 (strongly disagree) to 7 (strongly agree) Likert-type scales.

Although these items were used to assess momentary quit motivation in a prior experimental study ¹⁰, they had not been psychometrically evaluated as an index of quit motivation. Descriptive statistics and inter-item correlations of the items confirmed high between- and within-person correlations at the item level (see Supplementary Table 1). Interclass correlations (ICCs), which provide an estimate of dynamic variability, were high (between .65 and .71) compared to typical levels for intensive longitudinal data (.20 to .40; ²³) which suggests greater between-person than within-person variability. Reliability estimates of the 4-item quit motivation index were strong at both the between-level (α = .97) and within-level (α = .78). Finally, a multilevel confirmatory factor analysis (MCFA) using the semTools package in R ²⁴ demonstrated excellent fit at both levels (Χ²(4) = 125.190, p < .001; CFI = 0.977, SRMR_between = 0.014, SRMR_within = 0.032). These preliminary analyses verified use of the 4-item index as a viable scale, and we used an average score on the four items at each prompt in subsequent analyses.

Momentary Self-Regulation Self-Efficacy.

We assessed self-efficacy in willpower, lack of self-efficacy in tolerating distress and lack of self-efficacy in controlling craving as indices of momentary self-regulation. Willpower was assessed a two-item scale (e.g., “Right now, I have ____ willpower;” and “If something tempting came across my path right now, I would have the strength to resist it”¹³ ) rated from 0 (not at all) to 6 (extremely). Distress intolerance was assessed with a previously validated 3-item index (e.g., “I want to stop what I’m doing right now so I can feel better” ¹⁴) with items rated from 1 (strongly disagree) to 7 (strongly agree). Finally, craving uncontrollability was measured with the item “I don’t have any control over my current cigarette craving” rated from 1 (strongly disagree) to 7 (strongly agree). This item was used successfully in both smokers and dieters in a prior study ¹³.

Context.

Finally, participants were asked to indicate their location (home, work, another’s home, bar or restaurant, school, in transit, or other), and activity (in class, working, traveling, internet/texting, housework, leisure activity, exercising, interacting with others, nothing, other) and who they were with (no one/alone, spouse or romantic partner, friend, family member, acquaintance, coworker, other). Participants were also asked to indicate (yes/no) if (a) others were smoking nearby, (b) they had drank alcohol since the last prompt, (c) they were currently eating, (d) they were currently drinking caffeine, or (e) they had experienced a stressor in the prior 15 minutes.

Nightly Quit Intentions

Participants completed a contemplation ladder regarding quitting with 0 = no thought of quitting and 10 = taking action to quit ²⁵, and the single item Motivation to Stop Smoking (MTSS) scale ²⁶.

Data Analytic Plan

We first examined between-subject correlations among quit motivation, other averaged momentary variables (affect, craving, momentary self-regulation), smoking experience assessed at baseline (e.g., cigarettes per day, nicotine dependence, baseline contemplation ladder and motivation to stop smoking scales), and the mean-squared successive difference (MSSD) in momentary quit motivation. The MSSD is an index of variability that accounts for temporal dependence as well as the amplitude and frequency of assessment ²⁷. The overall MSSD was calculated by first finding the MSSD in quit motivation for each day, ignoring overnight lags, and then averaging the daily MSSDs. Within-subject correlations among momentary variables are presented in Supplementary materials.

We then moved to understanding predictors of momentary quit motivation using multilevel modeling. The first model examined if positive and negative affect, craving, and momentary self-efficacy in self-regulation (willpower, distress intolerance, craving uncontrollability) predicted momentary quit motivation. Predictors were person-mean centered to represent a deviation from a person’s typical level. Because person-mean centering removes the between person element, a separate term representing the average or typical level of each predictor was also included in each model.

We also ran an exploratory model to see which contexts were associated with fluctuations in quit motivation. These included session type (random = 0; smoking = 1), dummy-coded versions of social context (alone = 0, not alone = 1), location (home = 0, not at home = 1) activity (engaging in class work, paid work or housework = 0, non-work activities = 1). We also added yes (1) vs no (0) reports of nearby smokers, whether the person had consumed alcohol, was eating, was drinking caffeine, and if they had experienced a stressor since the prior prompt.

The last set of models examined whether momentary quit motivation assessed during the day predicted quit intentions (contemplation ladder, MTSS) assessed at night. The first two models examined momentary quit motivation and self-regulation variables as predictors, using both person-mean centered and average person-level predictors. The third and fourth models used daily MSSD motivation and self-regulation as predictors; these did not involve person-mean centering because the MSSD already reflected variability.

Results

The overall response rate to random prompts was 70.75% (SD = 20.29%). Participants also logged an average of 44.58 cigarettes per person. In total there were 1799 smoking sessions (M = 21.37 per person, including both logged cigarettes and after converting 19.93% or n = 594 of the random sessions to smoking sessions) and 2387 non-smoking random sessions (M = 28.37) included in analyses.

Correlations of average momentary quit motivation (across the entire week of EMA) and MSSD of momentary quit motivation (Table 1) revealed that average quit motivation was associated with greater intolerance of distress, and MSSD of quit motivation was associated with greater negative affect. In addition, greater baseline contemplation ladder and MTSS scores were associated with greater momentary quit motivation and lower MSSD of momentary quit motivation. Interestingly, average momentary quit motivation was not correlated with MSSD of momentary quit motivation, r = −.05, p = .63.

Table 1.

Between-person correlations of average momentary quit motivation and motivation MSSD with other variables

	M (SD)	Average Momentary Quit Motivation	MSSD Momentary Quit Motivation
Positive Affect	2.62 (.86)	−.09	−.05
Negative Affect	0.62 (0.59)	−.05	.30**
Craving	54.49 (16.68)	.17	−.09
Willpower	3.00 (.92)	−.20	−.14
Distress Intolerance	3.02 (.94)	.37**	.14
Craving Uncontrollability	3.82 (1.33)	.20	−.05
Quitting self-efficacy	3.07 (1.06)	.36**	−.002
FTND	5.19 (2.16)	.04	.07
Cigs/Day	16.34 (5.93)	.05	−.04
Contemplation Ladder	5.87 (1.92)	.59**	−.34**
MTSS	3.90 (1.50)	.60**	−.22*
		M (SD) = 3.65 (1.39)	M (SD) = 28.44 (114.51)

^*p < .05

^**p < .001.

Predictors of Momentary Quit Motivation

People had higher motivation to quit when their craving was lower than usual and when willpower was stronger than usual (see Table 2). Within-person negative affect, positive affect, distress intolerance and craving uncontrollability did not significantly predict motivation. At the between-subjects level, people with greater negative affect reported less quit motivation and people with greater distress intolerance were more motivated to quit.

Table 2.

Momentary quit motivation predicted by affect, craving, and regulation self-efficacy

		Quit Motivation
Level	Predictors	b	SE	t	p
1	Momentary Negative Affect	−0.05	0.04	−1.08	0.280
1	Momentary Positive Affect	0.01	0.02	0.28	0.776
1	Momentary Craving	−0.00	0.00	−5.54	< 0.001
1	Momentary Willpower	0.19	0.03	6.49	< 0.001
1	Momentary Distress Intolerance	0.04	0.02	1.71	0.087
1	Momentary Craving Uncontrollability	0.02	0.01	2.19	0.029
2	Avg. Negative Affect	−0.58	0.27	−2.17	0.030
2	Avg. Positive Affect	0.00	0.17	0.02	0.985
2	Avg. Craving	−0.01	0.01	−0.55	0.581
2	Avg. Willpower	−0.14	0.19	−0.71	0.478
2	Avg. Distress Intolerance	0.76	0.17	4.35	< 0.001
2	Avg. Craving Uncontrollability	0.07	0.14	0.47	0.636
	Marginal R2 / Conditional R2	0.184 / 0.842

Note. Level 1 = within-subjects/momentary, person-centered; Level 2 = between-subjects, person-mean.

The full table of contextual factors associated with quit motivation are presented in Supplemental Table 3. There were only two significant predictors. First, people reported higher average motivation at home compared to when not at home (b = −.06, SE = .03, p = .04). People also reported lower quit motivation after experiencing a stressor in the last 15 minutes (b = −.22, SE = .04, p = < .001). No other context variable significantly predicted quit motivation.

Predicting Nightly Quit Intentions

When daily quit motivation was higher, people reported higher nightly intentions on both contemplation ladder scores and nightly MTSS (see Table 3). Higher momentary craving uncontrollability predicted lower nightly intentions on the MTSS but not the contemplation ladder. At the between-subjects level, people who reported higher daily quit motivation reported higher nightly quit intentions on both the contemplation ladder and MTSS. No other predictor variables were significantly associated with nightly quit intentions. None of the daily MSSD scores predicted quit intentions (see Supplementary materials).

Table 3.

Daily Quit Motivation and Self-Efficacy Predicting Nightly Quit Motivation

		Contemplation Ladder				Nightly MTSS
Level	Predictors	b	SE	t	p	b	SE	t	p
1	Daily Quit Motivation	1.42	0.62	2.27	0.024	0.43	0.07	6.14	<0.001
1	Willpower	−0.44	0.61	−0.73	0.467	0.05	0.07	0.75	0.454
1	Distress Intolerance	0.52	0.65	0.79	0.427	0.02	0.07	0.24	0.810
1	Craving Uncontrollability	−0.71	0.46	−1.54	0.125	−0.12	0.05	−2.29	0.022
2	Avg. Daily Quit Motivation	2.04	0.74	2.77	0.006	0.75	0.09	8.26	<0.001
2	Avg. Willpower	−1.60	1.29	−1.24	0.217	−0.04	0.16	−0.27	0.785
2	Avg. Distress Intolerance	0.55	1.10	0.50	0.619	−0.01	0.14	−0.06	0.954
2	Avg. Craving Uncontrollability	−1.66	0.90	−1.84	0.067	0.06	0.11	0.55	0.580
	Marginal R2 / Conditional R2	0.091 / 0.686				0.408 / 0.809

Note. Level 1 = within-subjects/momentary, person-centered; Level 2 = between-subjects, person-mean; MTSS = Motivation to stop smoking.

Discussion

The goals of the current study were to examine the dynamics of momentary motivation to quit smoking, by confirming that quit motivation varies over time and across contexts, establishing that quit motivation is associated with self-efficacy for self-regulation (e.g., willpower, distress tolerance, craving controllability), and confirming that quit motivation is predictive of quit intention in daily life.

Quit Motivation Variability

Quit motivation does vary, consistent with people’s self-perceptions ^6,10, and recent EMA data with smokers trying to quit ⁹. However, the ICCs were high compared to what is typical in intensive longitudinal data, with less within-person variability than other momentary variables (e.g., willpower, emotion) though in the same range as other indicators of momentary beliefs ²⁸. However, greater average quit motivation and less variability in quit motivation were both associated with greater quit intentions reported at baseline, supporting that variation in quit motivation still matters and is associated with cessation-related outcomes ⁹.

We found few contextual predictors of fluctuations in motivation other than that people reported greater quit motivation at home, and lower quit motivation after experiencing a stressor. Future work could look at other types of contextual predictors (e.g., perceived social support, others smoking nearby; Waring et al., 2020) on motivation to quit. In addition, future research should assess and control for daily use of other substances including e-cigarettes as these might influence both craving and motivation to quit.

Quit Motivation & Self-Regulation

In contrast to previous experimental research, which found that negative affect and craving were associated with more motivation to quit ¹⁰, in this study we found that quit motivation was unrelated to affect and was lower at times when participants were drawn toward smoking (i.e., moments of heightened craving). Perhaps when people felt increased craving, their motivation to quit lowered in an attempt to reduce ambivalent motivational responses ³⁰, which are often interpreted as uncomfortable ³¹.

We predicted that greater quit motivation would be associated with higher perceived self-regulation, and found that people reported more motivation to quit at times that craving felt more uncontrollable, suggesting that smokers may wish they were not reliant on using smoking as a coping technique. We also found that smokers felt stronger desire to quit in moments where they reported greater willpower self-efficacy. The theory of planned behavior ³² posits a bidirectional relationship between perceived behavioral control (i.e., self-efficacy) and attitude toward the behavior, and would predict that a positive attitude toward quitting (i.e., greater quit motivation) should be associated with greater self-efficacy for quitting specifically. The current findings suggest that self-efficacy for a broader regulatory concept (i.e., willpower) might be linked with motivation to quit. Future research would benefit from examining causal order to follow up these momentary associations. For example, does shifting willpower self-efficacy actually boost subsequent motivation?

At the between subjects level, we also found that people who feel less tolerant of their distress reported greater desire to quit. As distress intolerance is associated with higher smoking rates ³³, perhaps people who struggle with distress tolerance wish they could quit smoking, but do not feel capable of doing so. Considering that perceived difficulties in tolerating distress elicited by smoking withdrawal show stronger associations with smoking than general distress tolerance ³³, future research could explore individual differences in distress tolerance with momentary perceptions of quitting self-efficacy.

Motivation and Intention

Our results confirmed the theoretical and empirical links between motivation and intention ^7,34 by demonstrating that people who generally report greater quit motivation indicated stronger intentions, and on days when people reported higher quit motivation than their own average, they also reported stronger intentions. However, greater variability in motivation was unrelated to intention at the daily level.

Limitations

Considering that the sample was primarily White and Southern, the dynamics of motivation may differ for people in non-White racial/ethnic groups or in regions where smoking is less socially accepted. We also recognize that the one-week time frame for EMA may not be sufficient to capture shifts in motivation. The motivation to quit smoking represents a larger life change than other more proximal health-related motivations (e.g., motivation for exercise) and thus it may be that a longer time scope is ultimately better for assessing motivational shifts. A longer time scope would also allow for assessing the impact of bigger life events (e.g., job stressors, holiday periods) on the dynamics of quit motivation.

Strengths & Conclusion

Strengths of the study included psychometric evaluation of a multi-item index of quit motivation. We also focused on smokers who are theoretically open to quitting, excluding smokers actively attempting to quit ⁹ and smokers who believe they will smoke forever ³⁵. This group could most benefit from motivational interventions, to help move them toward a cessation attempt, and is a sample worthy of future study.

The overall goal of the study was to explore the dynamics of quit motivation in daily life, to set aside the assumption that motivation to quit is stable. We confirmed that motivation does vary, and that intentions to quit are stronger following higher daily quit motivation. We likewise found enhanced quit motivation when cigarette craving was lower and willpower was higher than usual, suggesting that interventions to reduce craving and increase willpower self-efficacy may boost quit motivation and subsequent quit intentions. For example, smelling pleasant olfactory cues may be a simple but effective way to curb craving ³⁶ and potentially boost willpower, as positive affect has likewise been linked to increased willpower self-efficacy ¹³. In addition, the beliefs that people hold about willpower may be essential to helping people feel more agentic; when people feel that willpower is unlimited they tend to report higher levels of willpower, less mental fatigue and they persist more at challenges ³⁷. Results of this study demonstrate the need to continue understanding individual differences in in quit motivation and to explore more proximal shifts in the motivation to quit smoking.