Adherence to treatment to help quit smoking: effects of task performance and coping with withdrawal symptoms

Participants were 125 workers at the University of Granada (50 men and 75 women) with a mean age of 46.91 years (SD = 8.15). They were recruited between 2009 and 2013 at an occupational health clinic providing smoking cessation treatment. Participants smoked a mean number of 19.86 (SD = 8.95) cigarettes daily. The mean score on the Fagerström test for nicotine dependence was 4.62 (SD = 2.24). Participants entered the study if they were 18 years of age or older, had an employment contract with the University of Granada, wanted to voluntarily participate in the treatment, and correctly filled in the pre-treatment evaluation measures. Participants were excluded if they were diagnosed with a serious mental disorder (bipolar and/or psychotic disorder, etc.), had a concurrent dependence on other substances (cocaine, heroin, alcohol, etc.), or were regularly taking medications that were incompatible with the pharmacological treatment used in the therapy. Participants were assessed at baseline with the measures outlined below. As part of the treatment patients underwent cognitive-behavioral therapy, which included the performance of tasks at home, and pharmacological therapy with varenicline. Participants were informed about the aims of the study and provided signed informed consent. Ethical approval was obtained from the Ethics Committee, Research University of Granada, Spain.

An initial evaluation of the smokers was performed in a single session at the beginning of the program, in which the instruments described below were administered. The evaluation, treatment, and 3-, 6- and 12-month follow-up sessions of the program were implemented individually.

The program consisted of three phases. The first phase included a personalized number of sessions until abstinence was reached. During the first phase the cognitive-behavioral therapy was initiated by working towards developing stimulus control and self-control, and reducing tobacco consumption. Once the smoker had reduced tobacco consumption by 80%, pharmacological treatment with varenicline was started. We chose this scheme for two reasons. First, this was a way to ensure that the participant was sufficiently motivated to receive the free pharmacological treatment and quit smoking. Second, this way the participant starts to develop his or her self-control and management of symptoms. Varenicline is a nicotinic receptor partial agonist that effectively aids smoking cessation. The treatment starts with a daily dose of 0.5 mg for three days, and increases to a dose of 1 mg twice a day until the end of the treatment. The duration of the treatment with varenicline was 12 weeks. Adherence was defined as accurately following the prescribed regimen. Prescription of varenicline was administered in line with the Food and Drug Administration’s guidelines [28, 29]. The cost of the medication was covered by the Prevention Service (Area Labor Medicine) of the University of Granada. If the participants did not reach abstinence, they abandoned the program. Participants who reached abstinence proceeded to the second phase of the program which consisted of 6 sessions and aimed to maintain abstinence. It included training in problem solving, coping, and behaviors alternative to smoking. Participants also performed tasks at home. These tasks included reporting their smoking behavior, cravings, coping, and problem solving. The third phase consisted of three follow-up sessions (at 3, 6, and 12 months). Participants were telephonically contacted by an independent assessor (blind to the study purpose and methods) at each endpoint (3, 6 and 12 months after the start of the program). The purpose was to monitor their compliance with the treatment and their willingness to participate in the follow-up face-to-face assessments of smoking abstinence. Abstinence was assessed with a self-report of smoking behavior which was cross-validated with measurement of patients’ co-oximetry hemoglobin levels. Participants’ outcomes were coded as “relapse” or “abstinence”. Abstinence was defined as not having smoked even once since the day the participant quit after gradual reduction of nicotine intake and pharmacological treatment. Relapse was defined as having smoked for 7 consecutive days in the past 3, 6 and 12 months, respectively [30‐32].

Semi-structured interview for smokers[33]. This instrument provides information about socio-demographic data, preferred brand of cigarettes, number of years smoking, usage of other substances, and past attempts to quit.

Fargerström Test for Nicotine Dependence[34]. This instrument evaluates the intensity of physical addiction to nicotine and has a consistent factorial structure [35]. In this study we used the Spanish version of the test [36].

Smoking Processes of Change Scale (SPC) [37, 38]. This questionnaire measures 10 basic processes of change. Participants indicated the frequency with which they have engaged in or experienced 40 activities or events within the last month on a 5-point Likert scale from (1) never to (5) repeatedly. This instrument has good psychometric properties. In this research we used the Spanish version of the instrument [39].

Coping with Withdrawal Symptoms Interview (CWSI). This instrument was designed specifically for this research. Each item had 4 response alternatives, on a 4-point Likert scale from (1) never to (4) repeatedly. Smokers were asked about (1) craving in the past month, (2) coping with craving, (3) presence of anxiety, (4) depression, (5) sleeping problems, (6) eating problems (excess appetite), (7) physical problems like stomach pain as a result of drug therapy, or other physical problems, (8) whether they had done physical exercise in order to cope with withdrawal symptoms and (9) whether they perceived any benefits after quitting smoking. Exercise was measured following the criteria of the International Physical Activity Questionnaire (IPAQ) [40, 41]. In particular, participants answered 9 questions on a 5-point Likert scale about the intensity, frequency, and duration of physical activity they did in the past month. Based on their averaged responses to these questions, participants were classified in three groups: frequent exercise, moderate exercise, and inactive. Based on their responses on all interview questions, participants were classified into three coping groups: lack of coping and recognition of the problem, i.e., ineffective coping (score > 4), moderate coping (score 6 to 12), and effective coping (score 13 to 17).

Discrete-time survival methods were used to analyze how the variation in risk of smoking relapse over time was related to the socio-demographic, smoking-, and treatment-related variables. The survival time of participants who did not smoke during the observation period was set to the end of the data collection window [42]. Event status was coded as 0 = relapse (smoked prior to termination time) or 1 = still abstinent at termination time. A predictor was retained in the model if it improved the overall goodness of fit of the model. The effects of the continuous predictors were displayed by plotting survival functions using Kaplan–Meier graphs [43] and estimating the median life-time, the time at which half the sample had experienced the event and half had not [44]. Last, Cox proportional hazard regression models were used to calculate the hazard rate ratios [14, 45, 46]. The survival analysis was conducted with the SPSS software package.

At 1-month follow-up, 75 participants had maintained abstinence during treatment. Therefore, the abstinence rate was 60%. At 3-month follow-up, 70 participants had maintained abstinence during treatment (abstinence rate = 56%). At 6-month follow-up, 56 participants had maintained abstinence during treatment (abstinence rate = 44.8%). At the 12-month follow-up, 45 participants had maintained abstinence during treatment (abstinence rate = 36%) (see Table 1).

Survival analysis results (Table 2) revealed that the abstinence duration was not different across groups determined by socio-demographic characteristics (age, gender, education level, and occupation).

Survival analysis results (Table 3) showed that the abstinence duration for the different categories of the variable use of alcohol and/or tranquillizers differed significantly (Log Rank =17.679; p =0.001), i.e., using another substance had a significant influence on the probability to remain abstinent. Figure 1 shows the survival curve depending on the use of another substance. The group that used no other substance showed greater survival than the group with alcohol use; the group with alcohol use in turn showed greater survival than the group that used tranquilizers. The p-values in Table 3 indicate that there were no significant differences in abstinence duration with respect to the variables cigarettes brand, nicotine concentration mg/cigarette, number of cigarettes smoked daily, score on the Fagerström test, and years of smoking.

Survival analysis results (Table 4) revealed that the following variables had a significant influence on survival time: performance of treatment tasks, Log Rank = 68.330, p = 0.001, number of treatment sessions, Log Rank = 8.283, p = 0.041, coping with withdrawal symptoms, Log Rank = 49.750, p = 0.000, and physical exercise, Log Rank = 23.376, p = 0.001. Figures 2, 3 and 4 illustrate the survival curves for abstinence duration for the different levels of these variables. In particular, Figure 2 shows that participants in the effective task performance and moderate task performance groups have a better survival curve than the ineffective task performance group. Table 4 shows that participants who completed 16-20 treatment sessions had a better survival curve than those who completed 11-15, 6-10 or 1-5 treatment sessions. Figure 3 shows that participants who were classified in the groups effective coping and moderate coping had a better survival curve than those classified into ineffective coping. Figure 4 shows that participants in the frequent and moderate exercise groups have a better survival curve than participants who were classified as never engaging in physical exercise. Past attempts to quit and compliance with the pharmacological therapy had no significant influence on abstinence duration.

We performed a Cox regression analysis to identify risk factors associated with the probability of survival. We fitted a model including all factors related to tobacco consumption and treatment which were significant in the survival analysis. We included two additional quantitative variables: processes of change and number of days it took for the smoker to achieve abstinence. Table 5 shows that the variables performance of treatment tasks and coping with withdrawal symptoms were significant (p = 0.01 and 0.013, respectively). The negative sign of the beta-coefficients (B) indicates the direction of the relationship; i.e., these variables are protective factors for abstinence from smoking (Hazard ratio >1). Processes of change, days to achieve abstinence, other drug used, number of treatment sessions, and physical exercise were not significant (p >0.05) in the regression model. Table 6 shows a comparison between the levels of the variable performance of treatment tasks. Specifically, the group which did not perform the tasks had 3 times higher risk of smoking relapse than the group performing tasks moderately [Hazard ratio; (H.R = 3.032)]. No differences in survival are observed between the group performing tasks moderately and the group performing tasks effectively. Also Table 6 shows a comparison between the levels of the variable coping with withdrawal symptoms. Specifically, the group which coped ineffectively had 2.4 times higher risk of smoking relapse than the group which coped moderately [Hazard ratio (H.R = 2.410)]. No differences in survival were observed between participants who coped moderately and effectively.