Background
One preventable cause of early mortality is cigarette smoking [
1]. Although the tobacco smoking rate is reducing overall, more than two-thirds of recent deaths in developing countries are caused by diseases related to smoking tobacco. Based on a report of the World Health Organization (WHO), 22% of the world’s population aged 15 years and above are smokers, and almost 6 million people die from exposure to tobacco smoke or from tobacco use [
2]. According to a study from 2012, the prevalence rate of current daily smoking in Iran is respectively 11.3% (21.4% of men and 1.4% of women) and 12.5% (23.4% of men and 1.4% of women). Furthermore, it was reported that the mean number of cigarettes smoked every day by Iranian smokers was 13.7 sticks [
1].
The large number of population still use cigarette smoking in spite of its risky impacts. It is recognized that quitting cigarette smoking results in several health benefits, such as reduced mortality risk due to cardiovascular diseases [
3,
4]. Because of the strongly addictive nature of nicotine, relapse after stopping is common. Furthermore, there are some social benefits of smoking that prevent smokers from quitting, like enhanced feelings of relaxation and a sense of control [
5‐
7]. Persons who smoke for a long time are commonly not influenced by the long-term benefits of quitting cigarette smoking, especially when the diseases that are associated with smoking have not developed obviously [
8‐
10].
Interpersonal communication has an important effect on smoking. From the perspective of social cognition, the decision of whether or not to smoke is influenced by the response of peers in one’s social environment. For instance, individuals may be discouraged from smoking cigarettes if they believe that smoking is perceived as a negative behavior by the public and that they may face social disapproval by doing so. As such, the negative social consequences to smoking, such as peer refusal, are closely associated with a decreased probability of continued smoking [
11,
12]. Thus, individuals who received negative consequences may have less of a tendency to smoke cigarettes [
13].
Successful approaches to the cessation of cigarette smoking are often based on behavioral change models [
14]. The transtheoretical model (TTM) is known as one of the most important models in the field of preventive health behavior [
15,
16]. This model, developed by Prochaska, assesses an individual’s readiness to act on a new, healthier behavior and provides approaches for change to guide the individual [
17]. The TTM is composed of four constructs: stages of change, processes of change, self-efficacy, and decisional balance and temptations [
18‐
20]. The TTM is based on the assumption that people are at different stages of motivational readiness for engaging in healthier behaviors and that intervention approaches are most effective when they are matched to a person’s current stage of change [
15,
21]. Decisional balance, which is the focus of the present article, refers to the idea that pros and cons are important in the decision-making process for behavioral change.
Based on the TTM, several scales were developed to assess to what degree interventions can cause variations in individuals’ behavior. One of the TTM-based scales that measure lifestyle changes like smoking is the Decisional Balance Inventory (DBI), which measures positive thoughts (pros) and negative thoughts (cons) that might occur to an individual who is deciding whether or not to smoke [
22]. The initial scale comprised 24 items, but later on, Pallonen et al. (1998) developed a DBI with 12 items [
23]. In developed countries, the English version of the DBI has been validated in many studies [
24‐
27], but unfortunately, there is limited literature on this topic in developing countries like Iran .
There is need for a questionnaire on health behavior changes that can be applied in various cultural settings. Culturally and linguistically competent scales like the DBI consider cultural values, beliefs, and practices that vary among diverse people. So, it is essential to re-assess the validity and reliability of this scale in a specific culture, like the Iranian one. Therefore, the aim of this study is to examine the psychometric properties of the Persian version of the DBI for smoking cessation in Iranian adolescents and young adults. Consistent with other studies, albeit in a different context [
24‐
27], we expect the Persian version of the DBI to have good psychometric properties.
Design and data collection
A cross-sectional validation study was carried out in Nowshahr, Iran. A convenient sample of male smokers between 16 and 24 years of age who worked in three factories in Nowshahr, Mazandaran participated in the study. The inclusions criteria were (a) being a current smoker who has smoked at least 100 cigarettes total, (b) not having participated in any effort to quit, (c) planning to quit smoking in the next 30 days, and (d) having the ability to read and write in Farsi. We recruited individuals who were in this stage to quit smoking because we were planning to perform an intervention for such a group. According to the TTM, every stage of change needs its own approach to ensure a successful intervention [
18]. Therefore, we believed that the best suited sample for this study would be a sample of participants who were in a preparation stage to quit smoking. Otherwise, we would have had to create a number of different interventions for the individuals in various stages of change, that would have been difficult due to limited time and resources.
The sample size was estimated a priori. The sample size was determined based on the number of items in the scale multiplied by 10 (12 × 10 = 120) [
29]. A convenience sampling method was applied to recruit the respondents. An introductory letter was sent to four factories through personal contacts of the researchers. A positive reply to cooperate on this study was received from three factories. Formal consent from the factories’ manager and participants was required prior to the study. We asked the managers of the selected factories for permission to perform a study in their factory. The self-administered paper-and-pencil questionnaire was conducted during work time in the presence of a researcher, who explained the purpose and procedure of the study. Participants were assured their answers were anonymous and confidential, and that they could leave the study at any given time that they want. Afterwards, participants informed consent was achieved. The questionnaire took 25–30 min to complete.
Reliability
In order to assess the reliability of the DBI, the internal consistency was tested applying the Cronbach’s alpha coefficient. The alpha values equal to .70 or higher were considered acceptable [
30]. Furthermore, intraclass correlation coefficient (ICC) was estimated for assessing the stability of DBI. The scale was re-administered to 40 smokers below 25 years of age 1 week after the first completion. ICC values of .40 or above are considered acceptable (r’s between .81 and 1.0 are excellent, between 0.61 and .80 are very good, between .41 and .60 are good, between .21 and .40 are fair, and between .0 and .20 are poor) [
29]. The analyses were performed using the statistical program SPSS for Windows version 23.0 and Amos 24.0.
Discussion
The aim of the current study was to perform a psychometric evaluation for the translated Persian version of an additional component of the TTM, the Decisional Inventory Index (DBI) for smoking cessation. This scale measures movement through the stages of change and delivers insight into mechanisms through which individuals try to change their risky behaviors. In general, the results demonstrated that the translated DBI is a suitable and valid questionnaire that can be used for assessing smoking behavior among adolescent and young adult smokers who speak Persian. Developing theory-based questionnaires can be considered a main precondition for the evaluation of any intervention program. Therefore, we consider the results from this study to be useful for adolescent and young adult who are part of a cigarette smoking control plan.
The Cronbach’s alpha and the ICC were acceptable and showed good reliability and stability for the DBI. Furthermore, the CVI and the CVR showed a reasonable content validity. The EFA results were consistent with those found by the original developer of the DBI. This indicates that the DBI is effective for presenting multiple aspects of the health concerns affected by smoking. These findings are also similar to studies that have been conducted by other investigators in other contexts [
39,
40].
As expected, the present study showed a three-factor solution for the Persian version of the DBI, including social pros as well as coping pros and cons. The three factors were able to predict 55.4% of the observed variance. This result is somewhat higher than that found by Velicier (1985), in whose study a two-factor solution (i.e., pros and cons) accounted for 41% of the observed variance [
22], and it is also higher than that found by Pallonen et al. (1998), in whose study a three-factor solution (i.e., social pros, coping pros, and cons) accounted for 50% of the total variance [
23]. In another study among both smokers and nonsmokers by Hoeppner et al. (2012), a four-factor solution (i.e., two pro factors, two con factors) was obtained that explained 45% of the variance [
41].
The results also demonstrated that the questionnaire is able to discriminate between the perceived benefits and barriers involved in making the decision to quit smoking. Decisional balance is a key construct of the TTM, and the results of the DBI imply that pros and cons are comparatively significant parts of the model. With regard to changing health and risk behaviors among target groups, it is important to emphasize the pros and cons for that specific behavior.
We also performed the CFA to determine if there was coherence between the data and the theoretical structure. The CFA provided good fit indices for the present model, and the convergent validity of the subscales of the DBI was acceptable. These findings are consistent with studies conducted in different cultural backgrounds [
22,
23,
42], which have indicated that the DBI is reliable when it is applied in Persian-speaking smokers. Our findings also demonstrated that the model from the original scale is similar to our model [
23]. Furthermore, the internal consistency of the scale as measured by the Cronbach’s alpha revealed an acceptable reliability for all subscales, which was consistent with previous studies [
22,
23]. Furthermore, after examining 40 male young adult smokers over a one-week period, our findings clearly indicated that the DBI has good stability in the short term; however, it has yet to be seen whether the DBI is still stable in the long term.
Limitations
The present study has also some limitations. One limitation has to do with the accuracy of the participants’ answers, because all measures were self-reported. Another limitation of the current study is related to its generalizability and its sample size. The present sample was limited to a convenient sample of 120 male adolescent and young adult smokers, and it is unknown whether we would achieve the same outcomes if a large representative group of both male and female smokers were recruited. As such, the current study is unable to assess gender differences with regard to the psychometric properties of the DBI. Future studies should aim to include both male and female smokers in order to assess whether motivations for quitting smoking are similar between the genders and whether gender influences amenability to treatment. Furthermore, this study included only adolescents and young adults who were working in a factory, thus excluding students. Future studies should also assess the psychometric properties of the Persian version of the DBI in an Iranian adolescent and young adult student sample. Finally, the sample of the current study was ethnically homogenous (just Farsi); further studies need to consider the relationship between different Iranian ethnicities (e.g., Gilak, Turkish, Kurdish, Baluchi, and so on) and the DBI.
Acknowledgments
We would like to thank all participants for their contributions to this study, as well as the experts without whose support this study would not have been finished. The authors are also thankful for help received from the Department of Health Education & Promotion, School of Health, Tehran University of Medical Sciences (TUMS).