Tobacco smoking among young adults has become a global health concern1 as it affects different physiological systems and increases the risk of different forms of cancers2. Research shows that it reduces the life expectancy of male smokers by about 12 years and that of females by 11 years3. The ill health effects of smoking, particularly lung cancer, are most prevalent among people who start smoking at an early age and continue smoking into adulthood4. Smoking causes many deaths among smokers, and also among non-smokers who frequently are exposed to smoke1.

There are about 1.07 billion smokers worldwide, of whom 908 million are men and 162 million are women, with the majority from low- and middle-income countries1. In the USA, nearly 8 in100 adults aged 18–24 years smoke cigarettes5. In England, about 14% of adults smoke, which is the main risk to their health6. The smoking prevalence among Bangladeshi adults is about 19%1. A recent study among university students of Jahangirnagar University, Bangladesh, shows that 60% of the students smoke7. In another study conducted on university students of Sylhet division found that the prevalence of smoking is 37%, while about half of the male students were smokers8.

From the literature, the majority of university students start smoking due to the influence of friends. Students who have more smoker friends are more likely to become smokers7-9. They also start smoking by imitating their smoker family members7 and through curiosity7-9. The prevalence of cigarette smoking was found higher in students who live away from home and have higher living expenditures8. Very few students report that a feeling of maturity, symbol of manliness and unhappy family environment as reasons for smoking9. In the case of resident students, prevalence is higher than in non-resident students as they are away from their family and home and may receive different influences than students who live with their families. The objectives of this study were to explore the prevalence and risk factors associated with cigarette smoking among resident students of a university in Bangladesh.


Study setting and participants

A cross-sectional study was carried out among the resident university students of Patuakhali Science and Technology University, Bangladesh, from August to October 2019. There are six dormitories (halls) in this university and data were collected from each of the dormitories. The only inclusion criterion for the participants was being enrolled as a resident student of the respective hall. The sample size was calculated based on a single population proportion formula with the following considerations: prevalence of cigarette smoking (37%) from a previous study carried out among the university students in Sylhet division of Bangladesh8, 95% confidence level, 5% margin of error, and 10% non-response rate. A total of 360 students were asked to participate and completed data were collected from 355 students (response rate 98.6%).

All study procedures were carried out following the guidelines of the Helsinki Declaration (1975). Written ethical approval was obtained from the Institutional Ethical Committee of Patuakhali Science and Technology University (Ref. PSTU/IEC/2019/03). Written informed consent was obtained from the participants before data collection. Participants were informed that their participation was voluntary and they had the right to withdraw from the study at any time. They were also assured of the confidentiality and anonymity of their data.

Data collection

Data were collected using a pre-tested self-reported questionnaire. Two-stage cluster sampling was used to select the participants. Firstly, a list of each room number was prepared for each dormitory. Rooms were selected by a systematic random sampling technique. Secondly, two students were selected from each of the selected rooms by using the random sample method. After selecting the participants, a consent form was given to them to read, sign and return back to the data collector. After having the signed consent form, the participants were requested to fill in the questionnaire, which took approximately 8–10 minutes for the participants to complete.

Operational definitions


A student who smoked at least one cigarette stick per day over a period of 6 months was considered a smoker in this study. This definition was also used in a previous study8.


A student who never smoked a cigarette or was a smoker previously but stopped for at least one year before was considered a non-smoker in this study. This definition was also used in a previous study8.

Statistical analysis

Descriptive statistics were performed to calculate the frequencies and percentages of different sociodemographic variables. Chi-squared test was used for testing the association between sociodemographic variables and outcome variables (smoking status), and the association was considered significant at p<0.05. Both bivariate and multivariate logistic regression models were used to identify the potential factors of cigarette smoking among the students and the significance level of p-values was set at <0.05. Crude and adjusted analyses are presented. Data were analyzed using IBM SPSS 23 (windows version).


Table 1 shows the descriptive statistics of the study participants. The average age of the students was 21.5 years (SD=1.7) and the majority of the participants were male. The cigarette smoking prevalence for the 355 participants was 32.6%. As seen in Table 1, almost half of the male students were smokers (49.1%) while only 2.4% of females were smokers with the prevalence higher for those aged 21–23 years (36%). Smoking prevalence was also higher among the students from the engineering faculty (46.4%) and 4th year students (42.6%). There was no association found between the parental educational status and smoking behavior of the students. More than half of the students (54.7%) that spent more (>5000 BDT; exchange rate 1000 BDT about 12 US$) money on living expenses were found to be smokers compared with others. The analysis also shows that smoking behavior was significantly associated with parental smoking, peer smoking status, and living with smokers.

Table 1

Sociodemographic characteristics of the students by their smoking status, Bangladesh 2019 (N=355)

Smoking status
nSmoker n (%)Non-smoker n (%)
Age (years)
18–2010827 (25.0)81 (75.0)4.150.125
21–2318968 (36.0)121 (64.0)
24–265821 (36.2)37 (63.8)
Male230113 (49.1)117 (50.9)80.39<0.001
Female1253 (2.4)122 (97.6)
Study area
Agricultural science25474 (29.1)180 (70.9)5.500.064
Engineering2813 (46.4)15 (53.6)
Business studies7329 (39.7)44 (60.3)
Year of study
First8425 (29.8)59 (70.2)5.850.210
Second8127 (33.3)54 (66.7)
Third5519 (34.5)36 (65.5)
Fourth6829 (42.6)39 (57.4)
Master’s6716 (23.9)51 (76.1)
Father’s education level
0–43713 (35.1)24 (64.9)0.190.909
5–96221 (33.9)41 (66.1)
≥1025682 (32.0)174 (68.0)
Father’s occupation
Agriculture-based3712 (32.4)25 (67.6)0.560.967
Labor52 (40.0)3 (60.0)
Business10632 (30.2)74 (69.8)
Job14950 (33.6)99 (66.4)
Others5820 (34.5)38 (65.5)
Mother’s education
0–4177 (41.2)10 (58.8)1.310.518
5–915446 (29.9)108 (70.1)
≥1018463 (34.2)121 (65.8)
Mother’s occupation
Homemaker27176 (28.0)195 (72.0)11.160.001
Working outside home8440 (47.6)44 (52.4)
Permanent residence
Rural area20764 (30.9)143 (69.1)0.690.404
Urban area14852 (35.1)96 (64.9)
Household monthly income (BDT)
<150005919 (32.2)40 (67.8)5.120.077
15000–3000016545 (27.3)120 (72.7)
>3000013152 (39.7)79 (60.3)
Monthly expenditure (BDT)
<4000442 (4.5)42 (95.5)41.91<0.001
4000–500020556 (27.3)149 (72.7)
>500010658 (54.7)48 (45.3)
Parental smoking
Yes9441 (43.6)53 (56.4)6.950.008
No26175 (28.7)186 (71.3)
Smoker friend
Yes297114 (38.4)183 (61.6)26.92<0.001
No582 (3.4)56 (96.6)
Smoker roommate
Yes8765 (74.7)22 (25.3)92.57<0.001
No26851 (19.0)217 (81.0)

[i] BDT: Bangladeshi Taka (1000 BDT about 12 US$).

Results from the regression analysis showed that there was a significant association between mother’s occupation and cigarette smoking of the students. Students whose mothers were employed had higher odds of using cigarettes than students whose mothers were housewives (crude OR=2.3; 95% CI: 1.4–3.8). It remained higher after adjusting for covariables (AOR=3.4; 95% CI: 1.7–6.8). Respondents who spent >5000 BDT per month for living expenses had a higher risk of being a smoker (crude OR=25.3; 95% CI: 5.8–110.2) and it remained higher after adjusting for the variables (AOR=34.1; 95% CI: 7.1–163.1). The association was highly significant in both cases. The analysis also showed that students who are currently sharing their room with a smoker had a higher risk of being a smoker (crude OR=12.5; 95% CI: 7.0–22.2) and the odds of being a smoker remained higher after adjusting for the variables (AOR=9.7; 95% CI: 4.9–19.0), both associations were statistically significant (Table 2).

Table 2

Factors associated with cigarette smoking among students, Bangladesh 2019 (N=355)

OR95% CIpAOR95% CIp
Mother’s occupation
Working outside home2.31.4–3.80.0013.41.7–6.8<0.001
Monthly expenditure (BDT)
Parental smoking
Smoker friend
Smoker roommate

a Variables adjusted for all the variables in the Table.


In our present survey on cigarette smoking behavior among the resident university students in Bangladesh, the overall prevalence of cigarette smoking was 32.7%, lower than in previous studies conducted among the university students in Sylhet division10 and medical students in Bangladesh11. However, the prevalence was lower than given in the Global Adult Tobacco Survey (GATS) report, where the overall prevalence of tobacco smoking in Bangladesh was 23.0%12, suggesting a higher prevalence of smoking among university students in Bangladesh than among young adults in general. This study also reported that nearly half of the male students smoke, in line with the GATS12 and the study done in Sylhet10. The prevalence was higher than that reported in nearby countries such as India (20.4%)13, Pakistan (26.1%)14, Nepal (33.6%)15 and Malaysia (42.1%)16. The present study found that the prevalence of cigarette smoking among students who come from an urban area is higher than for those who come from a rural area. The most important independent factor associated with cigarette smoking was having smoking friends and/or smoking roommates; similar results were reported from different regions of the world17. We also found that smoking among university students was correlated with parental smoking and this finding is consistent with other studies17. The prevalence of smoking by mother’s profession as well as by monthly expenditure was significantly different, which may be related to higher levels of family income as students may have more money available10. This study also found that the risk of being a smoker is 9.7 times higher for those who have a smoker roommate compared with those with a non-smoker roommate – without excluding the likelihood that smokers may select each other as roommates. The risk of becoming a smoker was found to be 3.4 times higher among students whose mothers worked outside the home than those whose mothers were housewives. Respondents whose parents were smokers were more likely to be smokers.


Our present study is not without limitations. Firstly, it was limited to a sample of resident university students and did not include students from all of Bangladesh’s universities. In addition, negative responses from students who smoke may have underestimated the prevalence of smoking. Smoking risk and predictors among university students could be measured more precisely if samples were to be collected from different universities in Bangladesh. Lastly, as this study is cross-sectional, we can only assess associations and not causality.


The findings of our study show that cigarette smoking prevalence is higher among resident university students than in the general population and identified a number of parameters associated with higher odds of being a smoker. Our findings will contribute to knowledge of cigarette use among Bangladeshi university students. These findings may also be useful for policymaking and for the tobacco-free campus movement.