Prevalence of Tobacco Smoking and Associated Risk Factors Among Public Sector Employees in Kuwait: A Cross-Sectional Study

Introduction There needs to be more evidence about the tobacco products utilized by individuals who smoke and the primary determinants contributing to the development of smoking behavior. Our study aimed to assess the prevalence and factors associated with using one or more tobacco products among employees from various ministries in Kuwait. Methods A cross-sectional study was conducted among employees in different ministries in Kuwait from December 27, 2018, to January 3, 2019. A questionnaire about smoking status and socio-demographic variables was used. Results There was a total of 1057 participants in this study. Of the participants, 26% (n=275) reported using at least one tobacco product. The proportion of smoking of at least one tobacco product was higher among men (n=243, 46.5%) than women (n=32, 6%). Among smokers, 1.5%, 5.9%, and 18.6% reportedly use only three, two, and one tobacco products, respectively. Of the study participants, 26% were smokers, 20.3% were exclusively cigarette smokers, and 21.8% reportedly started cigarette smoking at the age of 15 years or less. Male compared to female workers had higher odds of being smokers of at least one tobacco product (adjusted OR= 15.3, 95% CI= 10.0-23.4). Participants were significantly (p= 0.009) more likely to use at least one tobacco product if their monthly income in Kuwaiti Dinars ranged from 501-1000 KD (adjusted OR= 1.9, 95% CI= 1.2-3.0) or 1501-2000 KD (adjusted OR= 2.3, 95% CI= 1.2-4.5) compared to those who had monthly income range 500 KD or less. Conclusion The male gender and high income of the participants were significant predictors of the use of at least one tobacco product. Anti-smoking campaigns, mass media interventions, and increasing tobacco product taxes may minimize this population's tobacco consumption.


Introduction
Tobacco is the most important avoidable risk factor for non-communicable diseases. Annually, almost 6 million people worldwide die because of diseases attributed to tobacco use, and this number is expected to increase to 7.5 million by 2020, accounting for 10% of all deaths. In 2015, smoking caused more than one in ten deaths worldwide, killing more than 6 million people with a global loss of nearly 150 million disabilityadjusted life-years [1]. Lung, oral, and nasopharyngeal cancers are some of the significant tobacco consumption-related cancers. Unless immediate steps are taken to reduce smoking rates, the number of deaths due to tobacco use is expected to rise to 10 million per year over the next 30-40 years, and 70% of these deaths are likely to occur in developing countries [2].
Nicotine is an active ingredient of tobacco smoke, affecting multiple body organ systems. In the central nervous system, nicotine reaches the brain within a few seconds of inhalation and changes one's mood.
Smoking increases the atrophy of the optic nerve, which impairs vision [3]. In the respiratory system, inhaling nicotine leads to the inability of the lungs to filter out toxic chemicals and renders smokers more susceptible to respiratory infections [4]. Additionally, tobacco smoking increases the blood's low-density lipoprotein (LDL) level and reduces the high-density lipoprotein (HDL). This eventually leads to the accumulation of fatty substances in arteries and causes atherosclerosis, which increases the risk of blood clots, recurrent coronary heart disease, and heart attacks [5]. Furthermore, inhaling different kinds of tobacco products reduces one's appetite, making the affected person unable to get all the needed nutrients for good health. Consequently, tobacco use by a large proportion of a country adds a burden to the national economy by increasing costs in health expenditure and other indirect costs due to tobacco smoking-related diseases.
In the Middle East and Africa, cigarette consumption increased by 57% between 1990 and 2009 [6]. Furthermore, it is estimated that the prevalence of smoking among men is nearly ten times higher than that of women worldwide [7]. The prevention and treatment of tobacco addiction have been targeted by the World Health Organization (WHO) as the priorities for intervention in developing countries. Recently, Global Framework Convention on Tobacco Control set forth the goal to reduce tobacco use prevalence by 30% by 2025 [8].
There is limited evidence about the range of tobacco products adults use and the leading factors to smoking habit in the Middle East, including Kuwait use. Cigarettes, cigars, pipe, and shisha are some of the most common modalities used in indulging in tobacco products used in Kuwait, with more females (69%) smoking shisha compared to males (57%) [9]. Another relatively recent study on tobacco use among male industrial workers in Kuwait reported a prevalence of 34.8% [10]. However, little is known about smoking behavior patterns, including various tobacco consumption modalities in Kuwait.
Understanding the current tobacco smoking burden, including the prevalence of smoking and tobacco consumption through different smoking modalities among adults in Kuwait, may provide scientific evidence for effectively controlling and preventing tobacco-related diseases. Therefore, this study sought to i) assess the prevalence of tobacco smoking and concurrent use of multiple tobacco products in public sector workers and ii) examine the factors associated with the smoking status in the study population.

Ethical approval
Health Sciences Center Ethics Committee approved the study protocol and instrument for Students' Research under the Institutional Review Board (IRB reference number: 4051/2018). As noted above, written informed consent was obtained from all the participants. The participants were assured about the complete confidentiality of their personal information and the anonymity of their responses. It was explained to them that their participation was voluntary and they could withdraw from the study at any point.

Questionnaire validity and reliability
A self-administered questionnaire comprising 39 items was used to collect data (

Measures
Socio-demographic variables: Questions on self-reported age, gender, nationality, and income were included.
Instruments for Tobacco use: Questions to assess the consumption of different tobacco products, including smoking onset and the amount consumed daily. Questions to assess the shift between using different tobacco products were also included, e.g., "What is the order of your smoking habits since the beginning?".
Comorbidities: Questions about self-reported or physician-diagnosed tobacco-related morbidities were also included.

Sampling, sample size, and data collection
The data were collected from December 27, 2018, to January 03, 2019. A proportionate random sampling technique was used to recruit participants in the study. There was a total of 15 ministries in Kuwait with over 160,000 employees. A systemic random sampling was used to choose eight ministries. The total number of employees in those chosen ministries was around 60,000 employees. In order to calculate the sample size, Slovin's formula was used n = N/(1 + N e2), where n = number of samples, N = Total population, and e = margin of error, (0.03), which resulted in an n = 1090. Four data collection teams, comprising two medical students (one with three members), were constituted to accomplish the fieldwork. Workers were approached during their break periods. The objectives of our study were explained to them, and they were invited to participate. Consenting workers were requested to fill out the questionnaire and sign the consent form attached at the front of the questionnaire. Completed questionnaires were checked immediately to spot any oversight by the workers.

Data Analysis
Descriptive statistics, including proportions (%), means, standard deviations, medians, and interquartile ranges, were computed as required to examine the distribution study variables, i.e., demographics, smoking practices, and comorbidities. Chi-square analysis was used to test for the statistical significance (p≤ 0.05) of the association between smoking status and socio-demographics and comorbidities. Multivariate logistic regression analysis was also used to identify the variables independently and significantly (p< 0.05) related to smoking status. The estimated adjusted odd ratio (OR) and their 95% confidence interval (CI) were used to interpret the final multivariate logistic regression model. Data were managed and analyzed using the SPSS statistical analysis software (version 25).  (Figure 1, Table 2). Regarding years of work experience, 359 (34.2%) were between 1-4 years, 415 (39.5%) were between 5-14 years, and 227 (26.4%) were between 15 or more years of employment. The distribution of variables, including weekly working hours, perception of home, work environment, and sleep quality, are given in Table 3.

Distribution of tobacco products used among workers in various ministries
Of 1057 participants, 275 (26%) were smokers of at least one product. Among smokers, 16 Figure 2 shows the order of indulging in using tobacco products among smokers. Most started smoking cigarettes, followed by shisha (47.6%). Among the smokers, 15.9% started with cigarettes and then moved to e-cigarettes. Nearly an equal proportion (15.1%) of smokers first began with shisha smoking before they indulged in cigarette smoking. Some other orders of a multiplicity of tobacco products used by smokers in this study are also shown in Figure 2.

Chi-square analysis of the association between socio-demographics, comorbidities, and smoking status
The socio-demographic variables that were significantly associated with smoking of at least one product included: gender (p< 0.001), nationality (p= 0.049), total monthly income in KD (p= 0.003), and ministry of employment (p< 0.001). Age, marital status, time of work, hours of work per week, description of work as stressful, disturbed by work, description of home as stressful, heart disease, high blood pressure, diabetes, obesity, upper respiratory tract infection, peptic ulcer disease, and sleep disturbances were not significantly associated with smoking of at least one product (

Prevalence of tobacco smoking
Our study assessed the prevalence and factors associated with the use of tobacco (cigarette, shisha, and ecigarette) in both genders, 21 years old or older, public-sector workers of eight of 14 ministries (Commerce and Industry, Defense, Finance, Foreign Affairs, Interior, Education, Health, and Communication) of Kuwait. In our study, the prevalence of current smoking was higher among males (46.5%) than females (6.0%)evidence of substantial increase over the previously reported corresponding figures of (34.4%) and (1.9%) among males and females, respectively, in Kuwait [9]. Most of the smokers in this study were older than 51 years (27.0%), which overlaps with the results of current smoking in the United States as it was highest in the young adult population [11]. In our study, the prevalence of concurrent use of three, two, or one tobacco products was 1.5%, 5.9%, and 18.6%, respectively. Of the participants, 20.3% were exclusive cigarette smokers. Although tobacco consumption including cigarette smoking has been declining since the mid-1960s, cigarettes remain by far the most commonly used tobacco product as reported in the United States [11]. This is also associated with a report in China, whose people smoke more than 40% of all cigarettes globally [12]. Despite anti-tobacco consumption, legislation was done and promulgated from time to time, yet tobacco consumption through different modalities of tobacco use continues to be high in Kuwait. Therefore, mass education through print and electronic media could reduce tobacco consumption in Kuwait.
On the other hand, the results showed that the percentages of using two and three tobacco products were relatively low (5.9% and 1.5%, respectively). In addition, 8.9% of the workers were shisha smokers, and 5.8% were e-cigarette smokers. In our study, the percentage of shisha smokers was almost the same as reported by Gaza strip refugee students and in Lebanon. However, the proportion of shisha smokers in our study (94 participants, 8.9%) was much smaller than that reported from Syria (30%) [13]. People in the Middle East are using shisha as a cultural tradition, and the onset of smoking is strongly influenced by similar practices by peers and families [14]. The prevalence of using e-cigarettes in our study was almost similar to that reported in New England, New York, and New Jersey. The main reason cited by e-cigarette users in the United States was that e-cigarettes help in smoking cessation [15]. However, this aspect needs further investigation.

Multivariate logistic regression
A significant and independent association existed between males and using one or more tobacco products. These results agree with the findings of a previous study, which reported that males were more likely to use tobacco products in Kuwait [9].
In addition, the high income of the participants was a strong and significant predictor of using one or more tobacco products. Thus, participants with higher incomes, presumably, are more willing to buy and indulge in using one or multiple tobacco products. This result is congruent with a previous study, which found that people with relaxed budgets are more flexible in spending on purchasing tobacco products compared to those with a low income [16]. According to reports from Gambia, individuals with low income and limited wealth were more likely to develop a smoking habit. The prevalence of smoking among this demographic group may be attributed to the stressful nature of their lives, which may lead to the use of smoking as a coping mechanism. This observation highlights the potential impact of socioeconomic factors on health behaviors and underscores the need for targeted interventions to address the root causes of health disparities [17]. However, another study in the United Kingdom showed an inverse relationship between high income and smoking tobacco products. They concluded that people with high incomes are more educated than those with lower incomes, and they recognized the adverse effects of smoking on their health [18].

Changing in smoking behaviors/modalities
The present study showed that smokers usually shift between different tobacco products. This change in smoking behavior may be due to trying to quit smoking (i.e., shifting from cigarettes to e-cigarettes) or only to change the taste (i.e., trying different flavors of e-cigarettes). Either way will probably lead to acquiring the habit of using more than one tobacco product. Interestingly, one participant reported that he smokes ecigarettes secretly during official working hours (indoors), cigarettes with tea and coffee, and shisha when gathering with friends. In the present study, most smokers (47.6%) began with cigarettes and then shifted to shisha. This percentage is similar in the Gulf countries, especially Saudi Arabia [19].
Similarly, in Iran, most university students still believe that shisha is the healthiest choice among tobacco products and is one of their traditional forms [20]. Others were going to shisha smoking, preferring different flavors like fruits, candy, and chocolates. For the same reason, shisha smoking is practiced by women in the United States [21]. According to the findings in our study, a portion of the individuals who were observed smoking cigarettes initially (15.9%) subsequently transitioned to using e-cigarettes. Conversely, it was observed that a subset of the participants (15.1%) initiated their tobacco use with shisha smoking, but later transitioned to cigarette smoking. The shift from using cigarettes to e-cigarettes was less frequent among the final-year pharmacy students in Serbia (9.9%) than in our present results.

Limitations and Recommendations
Some of the limitations of this study should be noted while interpreting the results. First, smoking status was self-reported; therefore, the true prevalence of smoking in Kuwait may be somewhat underestimated, particularly among females. This opinion may be attributed to cultural norms that dictate that smoking is morally unacceptable for women. As such, females may experience societal pressure to report their smoking status due to the perception of shame associated with this behavior. Second, the low prevalence of selfreported tobacco smoking among females precluded the stratified analysis for males and females to examine the gender-specific risk factors. Therefore, recall or information bias might have occurred as females who were smokers might be reluctant. This aspect warrants attention in future research to ensure the questionnaire's confidentiality, for instance, by incorporating anonymous data collection, private interviews, and non-judgmental language. Some examples of confidential questionnaires could be using coded identifiers instead of personal information, such as name or address, to maintain anonymity or providing the option for participants to complete the questionnaire online or through mail to avoid face-to-face interviews that might discourage candid responses.
Another limitation of our study is that the cause-effect relationship cannot be drawn from the observed associations as the nature of this study is cross-sectional. In addition, our study found that individuals with high incomes had a higher likelihood of smoking. However, the specific type of tobacco product preferred by this group was not identified, as well as their costs. Additionally, there needs to be more knowledgeable concerning the tobacco product preferences of low-income employees. Lower-income individuals may prefer cigarettes over other types due to their lower cost. Therefore, future studies should aim to determine the preferred type of tobacco product for each group depending on their income better to understand the relationship between income and tobacco product preference. We also recommend that future studies incorporate nicotine, tar, and carbon monoxide measurements for each tobacco product to identify which types pose more significant health risks. Elevated levels of these substances can result in severe illnesses. Finally, future research endeavors in this field should consider the limitations of our study while also considering the recommendations presented above. Doing so will promote a more profound comprehension of smoking behavior across diverse socioeconomic strata.

Conclusions
A higher prevalence of smoking among males (46.5%) and females (6.4%) was recorded than previously reported estimates in Kuwait. Among the participants, 18.6% used one tobacco product exclusively, while 7.4% used two or three. The high income of the participants was a significant predictor of the use of one or more multiple tobacco products. Anti-smoking campaigns through mass education using electronic and print media may help reduce the prevalence of tobacco smoking. Increasing taxes on tobacco products may discourage the purchase and use of tobacco products in this study population, particularly in low-income individuals.

Additional Information Disclosures
Human subjects: Consent was obtained or waived by all participants in this study. Health Sciences Center Ethics Committee issued approval 4051. Health Sciences Center Ethics Committee approved the study protocol and study instrument for Students Research. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.