To evaluate the temporal evolution of motivation and self-efficacy to quit smoking among health science students using a culturally adapted application in the Tunisian dialect.

This prospective observational study was conducted in 2023 across several health sciences institutions in Sousse, Tunisia, including public and private nursing schools and the Faculty of Medicine. The study adhered to standard guidelines for observational research and was approved by the Ethics Committee of Tunisia (registration number HS12–2022).

Students were recruited through institutional announcements in March 2023. Eligible participants met the following criteria: (1) age ≥ 18 years; (2) daily smokers for at least one year; (3) first-year health sciences students (medicine or nursing); (4) ownership of an Android smartphone compatible with the app in Tunisian dialect; (5) willingness to use the assigned mobile app; and (6) provision of informed consent online.

Exclusion criteria included: (1) severe psychiatric or neurological disorders; (2) current participation in medically supervised smoking cessation programs; (3) concurrent use of other smoking cessation apps; and (4) refusal to complete required assessments.

Sample size calculation was based on the accessible population and prior data from Lüscher et al., 2019 [13]. To detect an effect with 80% power at a 0.05 significance level, at least 86 participants were required. Anticipating a 44% attrition rate, 192 participants were recruited. After one month, 107 participants completed the study, exceeding the minimum required sample. Completion required using the app at least once per week and completing all questionnaires.

The mHealth intervention was developed specifically for this study rather than being adapted from existing applications, drawing on social cognitive theory (Bandura) and the transtheoretical model (stages of change). Development followed a three-phase iterative process, combining evidence-based strategies with features tailored to the culture and relevant to Tunisian health science students.

The application was culturally adapted to the Tunisian context by translating it into Tunisian dialect/French, following the various stages of cross-cultural translation. Participants were encouraged to use the application daily, with a recommended frequency of use.

And the fidelity of the intervention was monitored through app usage data (connection frequency, features consulted).

Participants were assessed at baseline (Day 0) and after four weeks of app use, based on [14]. Outcomes included smoking-related self-efficacy and motivation to quit.

Sociodemographic Factors: Collected at baseline, including age, sex, residence, origin, family situation, and socioeconomic status.

Nicotine Dependence: Assessed using the six-point Fagerström Test. Participants were categorized as low (0–4), moderate (5–7), or high dependence (8–10).

Previous Quit Attempts: Participants reported the number of serious quit attempts in the past 12 months.

Smoking Self-Efficacy: Measured using a 9-item scale assessing confidence to resist smoking in high-risk situations (Likert 1–5). Total scores range from 9 to 45, with higher scores indicating greater self-efficacy.

Two subscales were assessed: intrinsic self-efficacy, reflecting confidence in managing internal challenges such as nicotine cravings, withdrawal symptoms, and emotional triggers; and extrinsic self-efficacy, reflecting confidence in handling external challenges such as social pressure, exposure to smoking environments, and academic or work-related stressors. Both dimensions are critical predictors of successful long-term cessation.

Motivation to Quit (Q-MAT): A validated questionnaire assessing intrinsic and extrinsic motivation, self-determination, and confidence in quitting success. Items scored on a 7-point Likert scale; higher scores reflect stronger motivation.

Data analysis was performed using a per-protocol approach, including only participants who completed all follow-up assessments (n = 192).

Data were analyzed using SPSS. Descriptive statistics summarized participant characteristics, smoking habits, and previous quit attempts. Paired t-tests assessed changes in self-efficacy and motivation between baseline and follow-up. Linear regression models examined associations between changes in self-efficacy and motivation and demographic factors (age, sex, nicotine dependence). Statistical significance was set at p < 0.05, and 95% confidence intervals were reported.

The assumptions of linear regression were systematically verified before the analysis:

Note: Due to high multicollinearity between self-efficacy, resistance to external stimuli, and resistance to internal stimuli (r = 0.89-0.99), we retained only self-efficacy as a representative of this psychological dimension to avoid statistical issues.

The study was approved by the Ethics Committee of CHU Farhat Hached Sousse (HS12–2022). Institutional permissions were obtained prior to data collection. Participants were informed about the study objectives and conditions through an online consent form. Confidentiality and anonymity of data were maintained throughout.

Participant flow chart. Of 258 interested smokers, 192 met the eligibility criteria and used the app. At the one-month follow-up, 107 participants completed the final assessment, while 85 withdrew or stopped responding (Fig 1).

Among the 192 participants, the average age was 18.1 ± 1.2 years, with a sex ratio of 1.82 (64.6% male). The majority lived in urban areas (65.6%), had living parents (73.4%), and had a fairly satisfactory socioeconomic status (62%). Most lived in university housing (58.6%) or with their families (27.1%) (Table 1).

Mean daily cigarette consumption remained stable between D0 and M1 (14.3 ± 8.2 vs 13.2 ± 8.1 cigarettes/day, p = 0.541), with no significant difference. Similarly, Fagerström test scores showed no significant change (2.3 ± 1.2 vs 2.4 ± 1.1, p = 0.223). The distribution of dependence levels shifted slightly, with moderate dependence increasing from 39.1% to 51.4%, while low and high dependence decreased from 19.3% to 14.0% and from 41.6% to 34.6%, respectively (Table 2).

Of the 192 students recruited, 107 completed the study after using the app for one month.

A key finding of this study is the significant increase in motivation observed from the first month of the intervention, which is clinically relevant given that early improvements in motivation have been associated with longer-term smoking cessation outcomes. This increase likely reflects the effectiveness of the initial components of the intervention, including testimonials from healthcare professionals, visual representations of health risks, and personalized motivational messages. Consistent with previous research, this finding suggests that digital interventions may be particularly effective in supporting individuals during the transition from contemplation to action [17].

What stands out in this study is the high level of motivation maintained during the follow-up period. Many smoking cessation interventions report a decline in motivation over time as initial enthusiasm fades and the challenges of behavior change become more apparent [18,19]. Our results suggest that well-designed mobile apps can not only elicit rapid engagement but also sustain motivation over the medium term. We operationalized medium-term motivation as sustained motivation assessed one month after the intervention, recognizing that the initial 30-day period represents a critical window of vulnerability to relapse. This timeframe aligns with behavioral science literature, which shows that initial habit disruption and self-efficacy consolidation typically occur within the first 3–6 weeks of quit attempts. This finding is consistent with Rajani et al. (2021), who observed significant increases in motivation among adult smokers using gamified apps over a four-week period [14].

Unlike motivation, self-efficacy developed more gradually. Participants seemed to gain confidence in their ability to resist cravings and navigate risky situations through successful withdrawal experiences, thereby strengthening their behavioral skills. This gradual development of self-efficacy aligns with Bandura’s social cognitive theory, which posits that self-efficacy is built through mastery of concrete experiences [6]. Recent meta-analyses confirm that both motivation and self-efficacy are among the best predictors of successful withdrawal attempts [20,21].

Correlation analysis revealed several important patterns regarding the determinants of smoking behavior. The moderate to strong negative correlations observed between nicotine dependence (measured by the Fagerström scale) and psychosocial variables (overall motivation r = –0.59, self-efficacy r = –0.54) confirm that dependence is a significant barrier to behavior change, which is consistent with the findings of Kim et al. and Cupertino et al., who highlighted the predictive role of self-efficacy in successful cessation attempts [22,23].

The high correlations between motivation, self-efficacy, and other psychosocial factors (r = 0.77 to 0.89) suggest a strong interdependence among these variables. This synergy justifies the use of multi-component approaches, which have been shown to be more effective, particularly when personalized mobile interventions target multiple motivational levers in young smokers [24,25].

Interestingly, weak correlations between behavioral variables (e.g., cigarettes/day, age of onset) and psychosocial factors (|r| < 0.23) suggest that smoking intensity does not necessarily correlate with a smoker’s psychosocial profile. This finding highlights the importance of multidimensional assessments when designing smoking cessation interventions, especially for younger populations [26].

The temporal dynamics observed in this study where motivation increases rapidly followed by a gradual strengthening of self-efficacy are consistent with theoretical models in which motivation facilitates the initiation of change, while self-efficacy supports maintenance and relapse prevention [16]. Similar patterns have been reported in studies targeting young adult and university populations, suggesting that early engagement through motivational strategies is critical for this demographic. Traditional cessation programs often face barriers in this group, including limited use of pharmacotherapy and counseling, time constraints, discomfort with clinical settings, stigma, and, for health sciences students, concerns about confidentiality and professional identity. Our findings indicate that mHealth interventions can overcome these limitations by providing anonymous, accessible, and personalized support that enhances motivation quickly and builds self-efficacy over the medium term, offering a practical approach where conventional programs are underutilized or less effective [21].

The strong correlation (r = 0.76) between motivation and self-efficacy suggests a reciprocal relationship: increased motivation leads to greater engagement in skills-building modules, while progressive successes in self-efficacy further reinforce motivation [14].

Our observed attrition rate of 44% for the mobile app is similar to patterns reported in the recent systematic review by Fang and al, 2023, which noted that many eHealth smoking cessation interventions especially mobile health approaches exhibit decreasing user engagement over time, even when short-term effectiveness is demonstrated. Although Fang et al. did not quantify average dropout rates across all included studies, they highlighted that intervention effects, including sustained abstinence and engagement, tend to diminish as follow-up length increases, reflecting challenges in maintaining long-term participation in digital cessation programs. This consistency suggests that high attrition is a common issue in mHealth smoking cessation interventions and underscores the importance of strategies to improve sustained engagement [20].

This study supports the theoretical model that motivation and self-efficacy are key cognitive and motivational drivers for quitting. Clinically, it demonstrates that culturally adapted mobile applications can complement traditional support, particularly in settings with limited access to face-to-face counseling services. The results indicate that the gradual reinforcement of self-efficacy, combined with an early motivational boost, is a central mechanism for facilitating progression toward action and maintenance.

Self-reported measures may be influenced by social desirability or recall bias. Attrition could introduce bias if participants who dropped out differed from those who did not. Integrating mobile tools into health student training could strengthen motivational counseling skills and support smoking cessation, particularly in settings with limited access to in-person support.