The use of planned behavior theory in predicting cigarette smoking among Waterpipe smokers

The prevalence of waterpipe smoking in the Middle East appears to be very high, especially among students and young adults [1]. One study conducted in some colleges of the Qassim University, Saudi Arabia revealed that total prevalence of waterpipe use was 40% [2]. Additionally, a study from four Jordanian universities showed that current waterpipe use prevalence was 30% among university students and 56% had or were continuing to use waterpipe [3]. While waterpipe continues to be popular in the Middle East, it is becoming increasingly popular in non-Middle Eastern countries [1]. For instance, in the United States, a survey of 152 universities indicated that the current rate of waterpipe use was 8.4% and ever using waterpipe was 30.5% [4].

In Saudi Arabia, waterpipe smoking is very common among adolescents and college students although studies have shown a large variation in the incidence of waterpipe use. The overall high school student smoking rate was 21.7%; and the proportion of all smokers using just waterpipe was 37.4%. In addition, 16.5% of smokers used both waterpipe and cigarettes [5]. A recent Saudi Arabian national survey indicated that overall prevalence of current tobacco smoking was 12.2% among those 15 years or older. Daily waterpipe smoking occurred in 4.3% of the total population and 1.4% current daily smokers of cigarettes, cigar, and waterpipe [6]. In the last two decades, waterpipe smoking has increased among adolescents and young adults [7] in a way that is alarming to public health officials [1].

To attract users, waterpipe tobacco is moist and flavored with many fruit flavors, but this does not make it safe to the smokers’ health. In fact, the toxicants in waterpipe tobacco/smoke are similar to those found in cigarettes [8]. The American Lung Association [9] has called waterpipe tobacco smoking “an emerging deadly trend”. In the last three decades, more evidence about health effects of waterpipe use has appeared in the literature. Studies have shown that waterpipe smoking is one of the risk factors for several tobacco-related health problems including lung cancer, cardiovascular diseases, periodontal disease, and low birth weight. Waterpipe use may lead to lung cancer [10] and smoking waterpipe has acute physiological effects on smokers which contradicted the claim that waterpipe use might be a safe alternative to cigarette smoking [11]. Waterpipe smoking is associated with markers of atherosclerosis which might increase the risk of coronary heart disease [12]. Waterpipe use may also negatively affect pregnancy outcomes. Mothers who smoked waterpipe more than one time a day were 2.4 times more likely to have a low birth weight baby compared to non-smoking mothers [13].

Waterpipe and cigarette smoking have been found to be associated with each other as cigarette smokers were four times more likely to be waterpipe users than non-cigarette smokers and waterpipe smokers were likely to be former daily cigarette users [14]. In this study, Theory of Planned Behavior (TPB) was used as a framework to predict being a cigarette smokers among current waterpipe users.

According to the TPB, the behavior is a result of the intention to do the behavior and the perceived behavioral control of the behavior. The intention can be predicted by the attitude towards behavior, the subjective norm regarding the behavior, and the perceived behavioral control over doing the behavior. The relationship between intention and actual behavior is not perfect but intention can act as a proxy for the actual behavior.

The purpose of this study was to predict being a cigarette smoker among current waterpipe users in Riyadh, Saudi Arabia using theory of planned behavior constructs. The study examined how attitude, subjective norm, and perceived behavioral control could predict intentions to smoking cigarettes and actual cigarette smoking behavior. In the last 30 years, the TPB has been one of the dominant theories used in understanding health-related behavior [15]. The TPB has been used successfully in predicting and understanding a number of health-related behaviors including tobacco smoking [16], alcohol consumption, and substance use [17].

Of the 406 participants, 70.9% were exclusive waterpipe smokers and 29.1% smoked both waterpipe and cigarettes. The sample mean age was 26.97 (S.D. 6.82); however, the exclusive waterpipe users mean age, 27.82 (S.D. 7.02), was significantly higher than the mean age of those who used both waterpipe and cigarettes 24.89 (S.D. 5.82), t (404) = 4.01, p < .001. Sample characteristics in Table 1 show also that the mean of waterpipe initiation age was under 21 years of age. Because our sample was relatively young, the mean years of waterpipe use was six years. Our data show that the mean duration of a waterpipe smoking session was 88.38 min and the largest group of participants were single, college or post graduate, or government employees. Those who smoked only waterpipe had more income compared with those who smoked both cigarettes and waterpipe. In addition, less educated participants appeared to use both cigarettes and waterpipe whereas educated participants smoked waterpipe only. Nevertheless, government employees seemed to smoke waterpipe only; whereas waterpipe and cigarette users were likely to be students.

Table 2 illustrates the association between TPB constructs and background variables. The table represents simple Pearson product moment correlations between the TPB variables and other study variables. The higher scores on the TPB constructs represent more favorable attitude, subjective norm, perceived behavioral control, and intention toward using cigarettes. Those with attitude, subjective norm, perceived behavioral control, and intentions favorable toward cigarette use tended to be younger and to have started waterpipe at younger age. In addition, tobacco withdrawal symptoms had significant positive association with all theory constructs. The more favorable of attitude, subjective norm, perceived behavioral control, and intentions toward cigarette smoking were, the higher tobacco withdrawal symptoms score was. Possibly, since those who are more favorable to cigarette smoking are also more likely to be cigarette smokers, it may well be that those more favorable have higher withdrawal symptoms because they are getting more nicotine and, hence, are more addicted.

Favorable attitude for cigarette use was significantly associated with less years of cigarette use. In other words, as years of cigarettes use increased, there would be less favorable attitude toward cigarettes. Furthermore, strong association was found between number of years waterpipe use and number of years cigarette use r(404) = .67, p = <.0005. Even when age was controlled for, there was a significant association of years of use of both waterpipe and cigarettes r(404) = .28, p = <.0005. In addition, favorable attitude and subjective norm toward cigarette use were associated with less years of waterpipe use. Strong association was found between intentions and the number of cigarettes smoked per week. Higher favorable intentions toward cigarettes use was associated with more cigarettes consumption per week.

Table 3 shows descriptive statistics for categorical study variables showing the means for each Theory of Planned Behavior variable within each subcategory of a categorical study variable. Participants who were single were significantly higher than other groups to have favorable attitude, subjective norm, perceived behavioral control, and intentions toward cigarettes. Education levels had no significant association to any TPB constructs. Furthermore, smokers with the least income had significantly higher attitude, subjective norms, perceived behavioral control, and intention scores favorable to smoke cigarettes compared to those with high income.

In this study, the TPB was applied to predict behavioral intention and cigarette smoking behavior among current waterpipe users in Riyadh, Saudi Arabia. To our knowledge, this is the first study that utilized a theory-based approach to understand cigarette smoking intention and cigarette smoking behavior among waterpipe users. The study provided support for the utility of the TPB to predict cigarette smoking behavior among current waterpipe users, although the subjective norm to smoke cigarettes was not an essential element for that prediction. This may be due to the fact that all participants were current waterpipe smokers, it might be that a person’s beliefs about whether friends and people of importance to him had less influence or pressure to engage or not to engage in cigarette smoking because a person was already involved in one type of smoking behavior— habitual waterpipe tobacco use. In addition, previous research suggested that subjective norm was more likely to be a predictor of smoking behavior among adolescents, whereas attitudes were more likely to be a predictor of smoking behavior among adults [23]. As our sample study were adults and they have already engaged in one smoking behavior— waterpipe— subjective norm was not an important direct predictor for intentions toward cigarettes or actual cigarettes use.

Behavioral intention to smoke cigarettes could be predicted by the TPB constructs. The results indicated that for current waterpipe users, attitude (regardless of negative attitude for cigarettes) toward cigarette smoking and perception of the ease and difficulty of smoking cigarettes directly influenced intention to use cigarettes, explaining 46% of the variance in cigarette smoking intention—F (3405), p < 0.001. Attitude and perceived behavioral control were the strong predictors of the intentions, whereas subjective norm had no direct influence in the intention to smoke cigarettes. This is consistent with some studies that utilized the TPB which found that subjective norms were either the weakest predictor of intentions [24‐26], or were non-significant in the prediction of intention [27, 28].

Our study also showed that those who used both waterpipe and cigarettes were significantly higher in all TPB constructs than those who were exclusive waterpipe users. This is logical since they did smoke cigarettes, they would have favorable attitude towards cigarettes, preconception about self and others of smoking cigarettes, and strong beliefs of their ability to smoke cigarettes; however, this may raise a question about why those who smoked both waterpipe and cigarettes had more perceived behavioral control of cigarette smoking than exclusive waterpipe smokers. One explanation is that since they had experienced smoking cigarettes and knew its negative influences, their feelings of control over not smoking cigarettes could be high due to currently smoking waterpipe. Waterpipe could be acting as a substitute for cigarettes which degraded the craving for cigarettes and provided them with more control. Exclusive waterpipe users seemed to have less predisposition to smoke cigarette compared to those who were not exclusive waterpipe users which was expected. Current waterpipe users with cigarette smoking or past cigarette smoking history had more favorable attitude, higher social approval, and higher self-beliefs of smoking cigarettes than those who never smoked cigarettes.

The findings of this study could be useful in prevention/intervention programs that are intended to prevent or reduce tobacco smoking behaviors. Prevention and intervention programs might be directed to the attitude and perceived behavioral control by targeting underlying behavioral and control beliefs. Though additional research to enumerate what these beliefs are would be useful.

One of the limitations of this study design is lack of the ability to determine causal relationships; our results are only correlational. Also this study was based on self-reported responses which might be susceptible to recall bias. In addition, the results of this study may not be generalized to female waterpipe smokers as the study sample included only males but it could benefit other geographical areas that share similar cultural backgrounds.

Regardless of these limitations, this study has some advantages. The study went through two levels of randomization, one at the level of the waterpipe lounges and another one within the sections of each waterpipe lounge. The randomization process may reduce the potential bias and the sample was likely to be highly representative of the population that was studied. Another advantage of this study was the electronic approach used to collect data. The iPad tablets with Internet access which were used to input and store data electronically not only saved paper, but sped up the process, produced cleaner data, and reduced chances of error in transcribing responses since no transcribing was necessary.

The study indicated that the TPB successfully predicted cigarette smoking behavior among waterpipe users. The TPB could provide solid explanations of intention to start cigarette use. Subjective norm was not a predictor of the intention but it had indirect effects on intention through attitude and perceived behavioral control. These findings might usefully be integrated in an intervention program aimed at reducing or preventing smoking initiation, especially among young adults since our sample was relatively young. Further investigation to understand the mutual relationship between waterpipe and cigarette use is needed to establish an effective approach to curb enthusiasm for smoking tobacco.