eveloping countries have reported a more rapid rise in the prevalence of tobacco consu-mption among the youth than developed countries. 1,2 Researchers have investigated such multi-factorial phenomenon revealing various individual and environmental correlates of youth tobacco use onset. These variables have included age, gender, ethnicity, race, family structure, attachment to family and friends, personal and parental socio-economic status, school factors, lifestyle, stress, self-esteem and other personality characteristics, knowledge and attitudes, and parental and peer smoking. [3][4][5][6][7] Smoking is considered a major preventable cause of morbidity and mortality, causing over four million deaths a year. 8 This figure may increase to 10 million deaths per annum by 2030; 70% of which will be in the developing countries. 9 Cough is a symptom that affects a large proportion of the general population 10 and can cause a deterioration of an affected subject's quality of life.
Cough is a common symptom in various respiratory disorders, such as asthma, chronic bronchitis and bronchiectasis. 11 Immunization remains one of the most important public health interventions and a cost effective strategy to reduce both the morbidity and mortality associated with infectious diseases. The uptake of vaccination services is dependent not only on provision of these services but also on other factors including knowledge and attitude of mothers, 12,13 density of health workers, 14 accessibility to vaccination clinics and availability of safe needles and syringes.
Garment Workers' Situation: Bangladesh Perspectives
The readymade garment (RMG) industry in Bangladesh has been expanding rapidly since the late 1970, accounting for about 76% of the country's total export earnings in 1999, making Bangladesh one of the 12 largest apparel exporters in the world 15 and by 2006 providing jobs for 4.5 million people, 80% of whom are women. 16 Despite such encouraging role of RMG sector in Bangladesh, the worker's life is still one the poorest in this country. 17 Many of these workers have no access to the existing healthcare system due to their long working hours and financial constraints. Not surprisingly, the productivity of the Bangladeshi garment work force is low even by South Asian standards. 18 Around nine-tenths of the workers go through an illness or disease such as headache, anemia, fever, chest, stomach, eye and ear pain, cough and fever, diarrhea, dysentery, urinary tract infections and reproductive health problems. Safe conditions in the garment industry are very crucial for the worker's health and productivity. 19 However, this area is largely unexplored as there are very few studies in this regard in Bangladesh. 20 Under these circumstances, this study focused on the prevalence as well as associated factors of smoking, cough and BCG vaccination coverage among garment workers in Bangladesh.
This study was carried out among 5829 garment workers in Dhaka city where 90% of the factories exist. The researchers used cross sectional study design to conduct the study. The participants were from 12 selected garment factories. As the reported prevalence of TB (pulmonary tuberculosis) in this high risk group was 9.6 per 1000 21 with 95% confidence interval the estimated sample size was calculated from: n =1.96 2 Xpq/d 2 (Here, p=9.6 per 1000; q=990.4 per 1000; d (relative precision) = 12% of current prevalence. So, estimated sample size was 2762). Further considering a 10% non-response with a design effect (taken at an approximate of 2), the final sample size was 5829.
Cluster sampling technique was employed for selection of the study sample. Considering the large sample size, factories having more than 500 workers were enlisted from the BGMEA (Bangladesh Garments Manufacturing Association). Each factory was considered as a cluster. From the selected clusters, workers willing to participate in the study were included. Data collection was carried out during the first half of 2009. Six trained female interviewers collected the data through face to face interviews with a pre-tested questionnaire. Pre-test of the questionnaire was carried out in five garment workers. Each of the five workers was separately interviewed with the draft questionnaire. The questionnaire was finalized incorporating the feedback from the pre-test. Written informed consent from the participants as well as permission from the garment authority was taken before carrying out the interview.
This study used three dependent variables, i.e., cigarette smoking, cough and BCG vaccination. Each of these three variables was measured with two categories (0=no and 1=yes). To calculate the prevalence of cough, disease history was taken by standard guide lines 22 . Chest X-ray and sputum for AFB were done in those having a history of cough for more than 21 days. Several independent variables were categorized under three headings viz. socio-economic, working environment and living conditions. Participants' socio-economic variables included age (initially collected as reported and later grouped as ?24, 25 -34 and ?35 years), gender (0=male and 1=female), education (later grouped as 0=no education, 1=1-5 years, 2=6-10 years and 3=>10 years of education), marital status (0=unmarried, 1=married and 2=divorced/ widowed/ others), job (initially open ended but later categorized as 0=sewing, 1=quality and 2=others) and salary (later categorized as 0= ?2000, 1=2 001-4000 and 2= ?4001 BDT).
The variables within working environment included room size (later categorized as 0= ?4000, 1=4001-6000, 2=6001-8000 and 3= ?8001 sq. ft.), number of persons working in a room (categorized as 0= ?100, 1=101 -200, 2=201-300 and 3= ?301), status of light-air in the working room (0=sufficient and 1=not sufficient) and sitting arrangement (0=sufficient and 1=insufficient). The variables within living environment consisted of living place (0=slum and 1=not slum), number of persons in living room (later categorized as 0= ?2, 1=3 -4, 2=5-6 and 3= ?8 persons), length of living room (later categorized as 0= ?6, 1=7-8 and 2= ?9 ft.) and width of living room (later categorized as 0= ?4, 1=5-7 and 2= ?8 ft.).
Data analyses were carried out in two stages: data was initially analyzed for distribution the frequency of background characteristics, working and living condition of the garment workers. In the second stage, data were analyzed to identify the differentiating patterns of smoking, cough and BCG vaccination coverage among garment workers. For this, multivariate logistic regression analyses were carried out for each of the dependent variables.
IV.
Out of 5829 garment workers, the under aged participants found was small 1.3% (not shown). The majority of the participants (54.6%) were ?24 years of age while 26.1% were 25-34 years and 6.9% participants were ?35 years of age (Table 1). Around two thirds of the participants (61.4%) were females, whereas 38.6% were males. Although more than 85% of the garment workers had some years of schooling, 82.2% had 1-10 years of schooling (34.1% had 1-5 years and 48.1% had 6-10 years), and only 4.8% had >10 years education. Around half (49.5%) were married. Most of the participants were working in the sewing section (71.3% 71.8%), while 14.6% and 14.1% 12.7% participants were working in quality and 'other' section respectively. The salary of garment workers varied from less than 2000 to more than 8000 BDT (1USD = 73 BDT) where around one third participants (31.6%) earned only <=2000 BDT, two-thirds (63.7%) earned 2001-4000 BDT and only 4.8% earned ?4001 BDT per month as salary.
The mean room size was 6823 (SD=1610.5) sq. ft. About half (47.1%) and 20.2% of participants were working within 6001-8000 and ?8001 sq. ft. room respectively (Table 2). Around one seventh participants were working in room accommodated for ?100 workers. Whereas 19.9% and 33.6% of participants were working in a room accommodated for respectively 201-300 and ?301 workers. About 35% workers worked with insufficient light and air. The sitting arrangement was insufficient for 48.8% 48.7% participants. c) Living Environment Only 13.6% of participants resided in the slum areas while more than four-fifths of the participants (86.4%) resided in other places (Table 3). Just over one fifth participants were living in a room with ?2 persons. Although the majority of the participants (51.2%) were living in a room with a number of 3-4 persons, 3.5% of participants were also found living with 7-8 persons in a single room. More than one-third of the participants had ?6 feet length of living room while more than two thirds of the participants lived in a 5-7 feet width room.
In this study 7.3% of the garment workers were cigarette smokers (not shown). Multivariate logistic regression analyses were done to find out associations of smoking. It appeared that garment workers aged 25-34 and ?35 years were less likely to smoke c ompared to garment workers aged ?24 years but significant relationship appeared with age 25-34 years (OR=0.576, p<0.01) (Table 4). Significantly, male participants were more likely to smoke (OR=24.465) compared to female counterparts. Garment workers with some years of education (1-5, 6-10 and >10 years education) were more likely to smoke but the result was not statistically significant. Although results were not significant, married and divorced/separated/others participants were more likely to smoke compared to unmarried workers. Workers working in quality section and 'other' sections were more likely to smoke compared to workers in the sewing section. Participants' salary was not significantly associated with smoking.
This study identified 12.3% of garment workers had cough (not shown). The odds ratio (Table 4) indicated that although age, gender, marital status and job status had no significant impact on cough but workers with >10 years education had significantly lower likelihood of cough (OR=0.594, p<0.05) compared to that of workers with no education. Workers with the salary of ?4001 BDT per month were significantly more likely to report cough (OR=1.209, p<0.05); however, it did not appear significant at p=0.02.
Multi-variate logistic regression was also carried out with the working environment to examine the impact on cough (Table 5). It appeared that room size of the factory had significant impact on garment workers' cough, i.e., participants working in a room with 6001-8000 and ?8001 square feet room we re respectively 1.978 and 1.643 times more likely to report cough compared to those working in a ?4000 square feet room size. Although more than 100 persons living in a room (101-200 and ?301 persons) appeared to have lower odds except 201-300 persons living in a room, the results were not statistically significant. Insufficient status of light-air in working room and insufficient status of sitting arrangement had significant impact on cough (OR=0.630, p<0.01 and OR=0.781, p<0.05 respectively) compared to the respective reference category.
This study further examined the impact of living environment on garment workers' cough (Table 6). It appeared that participants living in non-slum areas were more likely to have cough (OR=1.070, p>0.05) compared to participants living in slum areas but the result was not statistically significant. Significant relationship appeared in case of 5-6 persons living in a room (OR=0.593, p<0.05). Higher length of living room (7-8 and ?9 square feet) appeared to have higher odds compared to lower length of living room (?6 square feet). However, significant relationship appeared with the length of ?9 square feet (OR=1.926, p<0.001). Contrary to it, lower width living had lower odds in having cough. It appeared that participants living in a room with 5-7 and ?8 square feet were significantly less likely to have cough compared to that of ?4 square feet.
This study identified that 59.6% of the garment workers were vaccinated with BCG (not shown). Multivariate logistic regression analyses (Table 4) found that workers aged 25-34 and ?35 years were more likely to take BCG vaccine (OR=3.812, p<0.001 and OR=5.265, p<0.001 respectively) compared to workers who were ?24 y ears of age. Although workers' gender, marital status and job status had no significant impact on getting BCG vaccination, education and salary appeared to have a significant impact on it. Workers with 6-10 years of education were 1.409 (p<0.05) times more likely to take BCG compared to workers with no education; however, it did not appear significant at p=0.02. Workers While workers with monthly income of ?4001 BDT were less likely to take BCG vaccination (OR=0.741, p<0.001) compared to those with a monthly income of ?2000 BDT.
V.
This study identified the levels and patterns of smoking, cough and BCG vaccination coverage among garment workers in Dhaka city of Bangladesh. To identify the risk factors, socio-economic, working and living conditions were taken into consideration. This study revealed that >7% of the participants smoked cigarette. Although tobacco is perceived as being cheap, its actual cost, compared with food, education, and health care, is quite high, excessively so for those for whom basic survival is a daily struggle. Another study also indicates that tobacco is most commonly used by those who can least afford it. 23 The amount spent by the average male cigarette smoker in K 1997 would purchase 2,942 calories of rice per dayenough to make a difference between family members getting by or suffering from malnutrition. This is also true for the Bangladeshi garment workers as they earn a meager amount of money that incapacitates them to run the family well. This study further revealed a high prevalence of cough (12.3%) with low vaccination coverage (59.6%) among the garment workers.
This study found that participants' age had negative impact on smoking. This is similar to other studies. 7,[24][25][26][27] Numerous studies indicate that both in developed and developing countries the prevalence of smoking is increasing in youths and this is more rapid in the developing than the developed countries. 1,2 This study also revealed that garment workers with higher age were more likely to take BCG vaccination and male garment workers were more likely to smoke compared to female workers. Another study also revealed that in Addis-Ababa males were more likely to smoke compared to females. However, reports on gender differences in adolescents' smoking behavior in other countries are controversial, ranging from higher prevalence in boys 6,28 or in girls 29,30 to no significant difference. 31 The additional concern is gender, which is gradually shifting towards increased prevalence of smoking among females, notably among youths in Iran and many other countries. 3,7,32 Although no such inference can be drawn from this study in case of Bangladesh, concern rises because of the increasing pattern of smoking. 24,25 The findings of this study further revealed that male garment workers were more likely to get cough compared to female garment workers. However, in other studies a higher prevalence of nocturnal and non-productive cough was reported in women than in men. 33 Education reflects knowledge and skills which influences health-related behavior, whereas income is an indicator of the current material situation. 34 As expected the finding of this study revealed that garment workers with some years of formal education were more likely to take BCG vaccination compared to those with no education and were less likely to have the problem of cough. This study further revealed that cough among garment workers was associated with working as well as living environments of the workers. Unexpectedly, the finding of this study also revealed that insufficient light-air had significant lower odds of cough compared to that of sufficient light-air. If mechanical ventilation and air conditioning system is not well-maintained, it can become a reservoir or amplifier for micro-organisms. 35 Hence, such an outcome may not be attributed to sufficient light air rather it could be due to the maintenance of mechanical ventilation and air conditioning system.
The finding of this study also revealed that insufficient sitting arrangement had lower odds of cough than that of sufficient sitting arrangement. It is to note that crowding of any sort increases the possibility of respiratory infection because the number of microorganisms in the air a person breathes is much greater when larger numbers of people are crowded into small spaces. 36,37 However, lower likelihood for cough with higher number of persons living in a room may be because of other reasons including cleanliness and health awareness of the garment workers.
The finding of this study also indicated that garment workers with higher length of living room had higher odds for cough. However, this study found lower likelihoods for cough with longer width of the living room. It is to be noted that in Bangladesh, many of the garment factories are not purpose built, rather it is a rented accommodation with almost no/low facilities for the workers, and there being no health facility at all for them.
Volume XIII Issue IV Version I

| 013 |
| 2 |
| Year |
| Working environment | OR | SE |
| Room size at workplace (in sq ft)* | ||
| ?4000 | 1.000 | |
| 4001-6000 | 0.866 | 0.127 |
| 6001-8000 | 1.978*** | 0.140 |
| ?8001 | 1.643** | 0.188 |
| Number of person working in a room | ||
| ?100 | 1.000 | |
| 101-200 | 0.883 | 0.153 |
| 201-300 | 1.208 | 0.137 |
| ?301 | 0.794 | 0.155 |
| Status of light-air in working room | ||
| Sufficient | 1.000 | |
| Insufficient | 0.630** | 0.163 |
| Sitting arrangement | ||
| Sufficient | 1.000 | |
| Insufficient | 0.781* | 0.118 |
| Living environment | OR | SE |
| Living place | ||
| Slum in City | 1.000 | 1.000 |
| Not slum in City | 1.070 | 0.117 |
| Number of person in living room | ||
| ?2 | 1.000 | |
| 3-4 | 0.905 | 0.108 |
| 5-6 | 0.593* | 0.219 |
| 7-8 | 0.816 | 0.130 |
| Length of living room ( in ft) | ||
| ?6 | 1.000 | |
| 7-8 | 1.226 | 0.185 |
| ?9 | 1.926*** | 0.107 |
| Width of living room (in ft) | ||
| ?4 | 1.000 | |
| 5-7 | 0.407*** | 0.163 |
| ?8 | 0.382*** | 0.285 |
This study was supported from the TB-Leprosy Control Program, DGHS, Mohakhali, Dhaka, Bangladesh.
Bangladesh: a case control study. Pub Healt 1999; 113: 57-64.
Analysis of smoking behaviour among Iranian population: a cohort and period analysis. Asia Pac J Cancer Prev 2004. 5 p. .
An international comparison of tobacco smoking, beliefs and risk awareness in university students from 23 countries. Addiction 2002. 97 p. .
Cigarette smoking habits among schoolchildren. Chest 1996. 110 p. .
Smoking in adolescence: Do attitudes matter?. Addict Behav 2001. 26 p. .
Alternative projections of mortality and disease by cause, 1990-2020: global burden of disease study. Lancet 1997. 349 p. .
Case Series Analysis of Adverse Reactions to Vaccines: A Comparative Evaluation. Am J Epidemiol 1996. 143 (11) p. .
Bangladesh apparel sector: growth terns and the post-MFA challenges. Growth of Garment Industry in Bangladesh: Economic and Social Dimension, P Paul-Majumder, B Sen (ed.) (Dhaka, Bangladesh
Hungry for tobacco: an analysis of the economic impact of tobacco on the poor in Bangladesh. Tobac Cont 2001. 10 p. .
Habitual coughing and its associations with asthma, anxiety, and gastroesophageal reflux. Chest 1996. 109 p. .
Establishing social protection for women garment workers in Bangladesh. http://www.adb.org/Documents/Brochures/ Social-protection-Project-Briefs/BAN-Proj-Brief-Women-Garment-Workers.pdf), Accessed June 12. 2010. (Asian Development Bank)
Socioeconomic determinants of children's environmental tobacco smoke exposure and family's home smoking policy. Euro J Pub Healt 2009. 1997. 19 p. . (35. Environmental Protection Department)
Ethnic differences in adolescent smoking behaviors, sources of tobacco, knowledge and attitudes toward restriction policies. Addict Behav 2003. 28 p. .
Cardiovascular risk factors in Iranian adults according ducational levels: Isfahan Healthy Heart Program. Asia Pac J Pub Healt 2005. 17 p. .
Tobacco and the developing world. Br. Med. Bull 1996. 52 p. .
Smoking in young adolescents: an approach with multilevel discrete choice models. J Epidemiol Commun Healt 2002. 56 p. .
Tuberculosis among garments workers: magnitude of the problem in Bangladesh. Asian-Pacific Newsletter on Occupational Health and Safety 2007. 14 p. .
Psychosocial predictors of smoking during middle and high school. Addiction 2005. 100 p. .
Incidence of pulmonary tuberculosis in garments workers of Dhaka city. Ban Med Res Counc Bull 2005. 31 p. .
Late adolescent female smoking. J Adv Nurs 2000. 31 p. .
From a prospective study of chronic cough: diagnosis and therapeutic aspects. Arch Intern Med 1998. 158 p. .
Tobacco use among Iranian men, women, and adolescents. Eur J Pub Healt 2004. 14 p. .
Upward occupational mobility among female workers in the garment industry of Bangladesh. No. 153. Bangladesh Institute of Development Studies (BIDS), (Dhaka; Bangladesh
A longitudinal study of respiratory symptoms in a smoking, allergen skin-test reactivity, and serum IgE. Chest 1991. 99 p. .
Smoking, adolescents and health: Isfahan Healthy Heart Program-Heart Health Promotion from Childhood. Asia Pac J Pub Healt 2004. 16 p. .
Smoking behavior and its influencing factors in a national representative sample of Iranian adolescents: CASPIAN study. Prev Med 2006. 42 p. .
Prevalence of smoking and age of initiation in Alexandria. Egypt. East Medierr Healt J 2002. 8 p. .
Developing populations: the future health effects of current smoking patterns. Mortality from smoking in developed countries, (Oxford
Health workers and vaccination coverage in developing countries: an econometric analysis. Lancet 2007. 369 p. .
Acute respiratory infections. Disease control priorities in developing countries: a World Bank Book, D T Jamison, W H Mosley, A R Measham (ed.) (New York, NY
The Global Youth Tobacco Survey Collaborative Group. Tobacco use among youth: a cross country comparison. Tobac Cont 2002. 11 p. .
Measles vaccine coverage and factors related to uncompleted vaccination among 18-month-old and 36-month-old children in Kyoto. Japan. BMC. Pub Healt 2005. 5 p. 59.
Tobacco use-United States 1900-1999. Centers for Disease Control and Prevention 1999. 48 p. . (MMWR)