Predictors and health outcomes of tobacco smoking among the population of Gaza: A large-scale study

Tobacco smoking, a significant public health concern globally, is associated with a rise in noncommunicable diseases and preventable deaths, with pronounced impacts in conflict zones like Gaza. A study in Gaza focused on individuals over 40 years of age, aiming to identify predictors of tobacco use and its links to diseases like coronary artery disease, chronic lung disease, and stroke. The research, based on the Gaza NCD study data with 4576 participants and a 96.6% response rate, found an overall tobacco smoking prevalence of 19.4%, with higher rates among men. Adjusting for various factors, the study revealed significant associations between cigarette smoking in men and adverse health outcomes, such as coronary artery disease and chronic lung disease. However, after adjusting for independent variables, shisha smoking in men showed no association with these health outcomes. In conflict-affected regions like Gaza, this large-scale study sheds light on predictors of cigarette and shisha smoking and their impact on health outcomes, offering valuable insights for researchers, public health officials, healthcare professionals, and policymakers. The findings aid in predicting smoking prevalence, addressing current health challenges, and mitigating potential health and financial burdens associated with tobacco use in conflict zones.


Background
Tobacco smoking is considered a global public health concern that is associated with signi cant morbidity and premature mortality. 1,2The number of smokers worldwide increased to 1.1 billion in 2019, with tobacco smoking leading to 7.7 million deaths. 3The prevalence of tobacco smoking, including cigarette and shisha smoking, is particularly increasing in the Middle East and has been alarming in some Arab countries. 4obacco smoking remains one of the leading causes of preventable deaths and is a primary risk factor for the increase in noncommunicable diseases (NCDs) and tobacco-speci c morbidities affecting mainly the cardiovascular and respiratory systems. 5In fact, the adverse health effects of smoking include but are not limited to coronary artery disease, chronic obstructive pulmonary disease and lung cancer. 6The burden of treating these smoking-related diseases contributes to a substantial economic strain on the healthcare system.This is evident in the increased costs for medical treatments, hospitalizations, medications, and the necessary care for addressing these chronic medical conditions. 7oking is also driven by various socioeconomic and psychological factors.In fact, tobacco smoking has been found to be more prevalent among those with lower socioeconomic status (SES). 8The high prevalence of smoking among disadvantaged groups is suggested to be in uenced by low social support, low levels of awareness and education, reduced motivation to quit, a stronger addiction to tobacco use, lack of self-e cacy, and the impact of tobacco industry marketing, particularly due to low tobacco prices. 9bacco smoking in the Arab region has also been a signi cant concern for public health o cials, health professionals, and policy makers. 4In fact, the prevalence of tobacco smoking has been particularly high in Arab men compared to women. 10However, there has been very little global research addressing the use of tobacco and nicotine dependence among civilians in con ict-affected areas, despite key vulnerabilities such as the presence of various socioeconomic stressors.Most studies have addressed this topic within adolescent and young adult populations.Indeed, a study performed in the Republic of Georgia revealed that nicotine dependence was signi cantly associated with multiple mental health disorders in older con ict-affected men. 11Moreover, the Palestinian Central Bureau of Statistics (PCBS), in its preliminary results of the "Smoking and Tobacco Consumption Survey, 2021", reported an obvious increase in the prevalence of smoking among individuals aged 18 and above in the West Bank. 12 such, many efforts have been made to address the predictors of smoking, given its association with the challenging burden of disease, disability, and death.Among these predictors, we mention socioeconomic status, cultural norms, age, and gender.
Past studies have examined the prevalence of smoking and awareness of smoking-associated health risks among young people in Gaza, speci cally focusing on university students.These studies were relatively small in size. 13,14e objectives of this study were to identify the predictors of tobacco smoking among a large representative sample of individuals who were at least 40 years of age and who were living in Gaza and to assess the associations between tobacco smoking and noncommunicable diseases, mainly coronary artery disease (CAD), chronic lung disease (CLD) and stroke.

Study Design
This was a secondary data analysis of the Gaza NCD study database.The 2020 cross-sectional study collected data on a representative sample of 4576 individuals aged 40 years and above selected from 2493 households across Gaza's ve governorates through an intervieweradministered household survey.Prior to data collection, interviewers visited each household for the study and obtained verbal informed consent.
In each household, one eligible male and one eligible female were interviewed and included in this study.The aim of collecting this dataset was to assess and identify potential solutions for noncommunicable diseases (NCDs) in a densely populated area affected by prolonged armed con ict, such as Gaza. 15The study received approval from the Imperial College Research Ethics Committee (reference 20IC5733), the American University of Beirut Institutional Review Board, and the Gaza Helsinki Committee (reference PHRC/HC/483/19).All methods conducted for the manuscript adhered to applicable guidelines and regulations.
Current cigarette smoking status was de ned according to the World Health Organization (WHO) de nition.This de nition encompasses individuals who have smoked 100 or more cigarettes over their lifetime and currently engage in daily smoking or smoking on some days.Current shisha smoking was de ned as smoking Shisha either daily or on some days.Our primary objective was to identify predictors of both cigarette and shisha smoke exposure.The selection of these predictors was based on known risk factors for smoking documented in the literature, as well as the availability of relevant data within the Gaza dataset.
Educational levels were classi ed into four groups: illiterate, basic education (able to read and write, elementary, preparatory), intermediate education (secondary, associate diploma, bachelor's degree), and higher education (higher diploma, master's degree, PhD).
We de ned physical activity as the initiation or increase in physical activity within the past year and categorized it as "yes" or "no." Body mass index (BMI) was initially classi ed into six categories: underweight, normal, overweight, obese class I, obese class II, and obese class III.Subsequently.We recategorized the data into three categories: normal/underweight, overweight, and obese.
Marital status was categorized into two groups: married and unmarried.
Other predictors, such as having worked during the past 30 days, having health insurance and receiving cash assistance, were all dichotomized as yes/no.The presence of coronary artery disease was assessed if participants reported a history of heart attack or angina.It was dichotomized as yes/no.Chronic lung disease was assessed by determining whether participants had a history of chronic obstructive pulmonary disease (COPD), asthma, or respiratory allergies and was dichotomized as yes/no.Stroke was assessed by determining whether participants had a history of stroke and was dichotomized as yes/no.Independent risk factors for the aforementioned health outcomes included smoking cigarettes and shisha, which are the same independent variables for the primary outcome, as well as hypercholesterolemia and hypertension.
Hypercholesterolemia and hypertension were assessed by determining if participants had previously received a diagnosis of high cholesterol or high blood pressure (or systolic blood pressure > 140 mmHg or diastolic blood pressure > 90 mmHg), and they were dichotomized as yes/no.

Statistical analysis
Demographics and other variables of the study were summarized using frequency distributions.The prevalence of cigarette and shisha smoking were computed along with their 95% con dence intervals.Univariable and multivariable logistic regressions were used to identify the predictors of cigarette smoking and shisha smoking.Unadjusted and adjusted odds ratios are presented along with their 95% con dence intervals.Hosmer and Lemeshow tests were used to assess the goodness-of-t of the models, and the predictive power of the models was assessed using receiver operating characteristic (ROC) curves.Similar analyses were performed to assess the potential association between tobacco smoking and noncommunicable diseases.IBM-SPSS (version 29, Armonk, NY, USA) was used for the data analysis.Statistical signi cance was set at the 5% level.

Results
A total of 4,576 participants were included in the cross-sectional study, resulting in a response rate of 96.6%.Of the participants, 46% were males, and 38% were above 60 years of age.For more details, Abu Hamad et al. provided a description of this sample. 15e prevalence of current cigarette smoking was 17.1% (95% CI: 16.0%-18.2%),and for shisha smoking, it was 3.5% (95% CI: 2.9%-4.0%).Among the participants, 783 individuals (19.4%) were identi ed as current cigarette or shisha smokers, with only 1.2% reported as users of both (refer to Fig. 1), but there was a marked sex difference.Given the higher prevalence rates among men (36.6% for cigarettes and 6.8% for shisha) than among women (0.8% for cigarettes and 0.6% for shisha), our analysis focused exclusively on male participants (see Fig. 2).The unadjusted and adjusted odds ratios of cigarette and shisha smoking are shown in Tables 1 and 2.  1).
Table 1 and Table 2 show the associations between the independent variables and the risk of smoking cigarettes and smoking Shisha rice in men.We calculated the adjusted odds ratios while considering adjustments for age, BMI, education, increase in physical activity over the past year, health insurance, marital status, cash assistance, and working status during the previous 30 days.Multivariate analysis of cigarette smoking (Table 1) revealed that older individuals were less likely to smoke [OR = 0.161, 95% CI (0.11, 0.24)].Individuals with intermediate education, in contrast to those who are illiterate, exhibit a lower likelihood of smoking [OR = 0.528, 95% CI (0.31, 0.90)].Increased physical activity over the past year was associated with a reduced likelihood of smoking [OR = 0.762, 95% CI (0.62, 0.93)].Moreover, individuals with a higher body mass index had a lower chance of smoking [OR = 0.762, 95% CI (0.62, 0.93)], while those with health insurance were less inclined to smoke [OR = 0.764, 95% CI (0.59, 0.99)].Having worked during the previous 30 days, having cash assistance and marital status were not associated with cigarette smoking.The goodness-of-t of the model using the Hosmer-Lemeshow test showed a good t (p value of 0.394) and the ability to predict smokers and nonsmokers correctly in 71.7% of the patients (Fig. 3).In the Shisha model, after adjusting for all the variables in the model (Table 2), only a few variables showed a signi cant association.Older men [OR = 0.084, 95% CI (0.03, 0.28)] and those receiving cash assistance [OR = 0.566, 95% CI (0.37, 0.86)] were less likely to smoke shisha.Additionally, our ndings revealed no association between smoking shisha rice and variables such as BMI, education, increase in physical activity during the past year, having health insurance, marital status, or working status during the previous 30 days.The goodness-of-t of the model using the Hosmer-Lemeshow test showed a good t (p value of 0.269) and the ability to predict smokers and nonsmokers correctly in 69.9% of the patients (Fig. 4).
Among men, the prevalence of NCDs was 13.3% for CAD, 5.9% for stroke, and 10.1% for CLD (Fig. 5).Three separate models for men were created, one for each of these conditions.We calculated the adjusted odds ratios for variables such as age, education, increase in physical activity during the past year, BMI, marital status, work during the past month, health insurance, cash assistance, hypercholesterolemia, hypertension, cigarette smoking, and shisha smoking.The goodness-of-t of the CAD, CLD, and stroke models was assessed using the Hosmer-Lemeshow test, which indicated good t, with p values of 0.134, 0.523, and 0.591, respectively.Our analysis revealed that cigarette smoking is signi cantly associated with a history of CAD and CLD.However, no signi cant association was found between smoking cigarettes and stroke incidence.Moreover, there was no association between smoking shisha rice and any of the three aforementioned health outcomes (Table 3).

Discussion
This study identi ed signi cant predictors of tobacco smoking (including cigarettes and shisha) in the Gaza Strip.This study with a representative sample aimed to assess the burden of smoking among adults aged above 40 years in Gaza.
In the context of cigarette smoking, a greater likelihood of smoking was associated with being male.Historically, smoking has been more socially accepted among men than among women.Furthermore, ndings from neuroimaging data indicate that smoking triggers reward pathways in men more than in women. 17Additionally, advertising agencies have traditionally targeted male consumers using role models such as actors and athletes.Given the near absence of smoking among women in the population of Gaza, we focused our analysis exclusively on men.Smoking is nearly nonexistent among women in Gaza, and if it does occur, it is not openly reported due to societal taboos.Women might also be more aware of the negative impact of smoking, especially its impact on pregnancy. 18 our study, smoking was also found to be less prevalent among older adults.This can be explained by the fact that older individuals might have quit smoking due to health conditions that make it inadvisable or due to its negative impact on their overall wellbeing.Another plausible explanation is that individuals aged 70 years and above are more likely to be healthy and are initially nonsmokers. 19oncommunicable diseases such as cardiovascular diseases and chronic lung disease.Several key strengths of the study can be highlighted.
First, the study boasts a reasonably large sample size, ensuring that the ndings are based on a substantial number of participants.This approach enhances the statistical power and reliability of the results.A notable strength is the representative nature of the sample, which ensures that the prevalence rates of smoking and associated risk factors accurately re ect the population of interest.Another strength of the study lies in its ability to assess multiple outcomes and predictors.By examining the relationship between smoking and various diseases, this study provides a comprehensive understanding of the risks associated with smoking.

Conclusion
In conclusion, our study revealed several important predictors of smoking habits.Notably, we found a positive correlation between cigarette smoking and coronary artery disease (CAD) and chronic lung disease, while no signi cant correlation with stroke was identi ed.This study represents a valuable asset for researchers, policymakers, and healthcare professionals engaged in efforts to prevent and control smoking among the population in Gaza.This information will be essential for informing the reconstruction of the healthcare system in Gaza following the war.

Table 1
Cigarette smoking model in men For men, at the bivariate level, younger age, nonobesity status, and lack of health insurance were signi cantly associated with increased odds of current cigarette smoking.Additionally, being younger, being working during the past 30 days, and receiving no cash assistance were associated with greater odds of current shisha smoking.Those between 40 and 50 years of age were the most likely to smoke cigarettes (48.7%), whereas individuals above 70 years of age were the least likely (17.3%) to smoke cigarettes.A similar pattern was observed for shisha smoking, with 12.2% between 40 and 50 years of age and 1.2% among those above 70 years of age, both showing statistical signi cance (p value 0.001).Additionally, individuals with a lower/normal BMI (59.5%), who had a basic education (42.1%), and who lacked health insurance (43.7%) were more likely to smoke cigarettes (Table

Table 3
Association between noncommunicable diseases (CAD, CLD and stroke) and smoking in men.