Transactions of the Royal Society of Tropical Medicine and Hygiene Advance Access published June 26, 2014
Tobacco smoking: a major risk factor for pulmonary tuberculosis – evidence from a cross-sectional study in central India V. G. Raoa,*, J. Bhata, R. Yadava, M. Muniyandia, M. K. Bhondeleya, M. A. Sharadab, V. K. Chadhab and D. F. Waresc
ORIGINAL ARTICLE
Trans R Soc Trop Med Hyg doi:10.1093/trstmh/tru082
a
Regional Medical Research Centre for Tribals (Indian Council of Medical Research), Nagpur Road, P.O. Garha, Jabalpur 482 003, Madhya Pradesh, India; bNational Tuberculosis Institute, Bangalore, Karnataka, India; cGlobal TB Programme, World Health Organization, Geneva, Switzerland
Received 23 January 2014; revised 23 April 2014; accepted 23 April 2014 Background: This paper provides information on the association of tobacco smoking and alcohol consumption with pulmonary TB (PTB) in central India. Methods: A community based cross-sectional TB prevalence survey was conducted in Jabalpur district of the central Indian state of Madhya Pradesh. The information on tobacco smoking and alcohol consumption was collected from individuals aged ≥15 years. Using logistic regression analysis, the risk factors for PTB were identified. Results: A total of 94 559 individuals provided information on tobacco smoking and alcohol consumption. Persons aged 35–54 years and 55 years and above had, respectively, a 2.19 (95% CI 1.57–3.07) and a 3.26 (95% CI 2.23–4.77) times higher risk of developing PTB compared to persons aged below 35 years. Males had a 2.35 (95% CI 1.66–3.32) times higher risk than females. Tribals (indigenous population) had a 2.32 (95% CI 1.68–3.21) times higher risk than non-tribal population. The adjusted prevalence odds ratio for mild, moderate and heavy tobacco smokers were 2.28, 2.51 and 2.74 respectively as compared to non-smokers. Alcohol consumption was not found to be a risk factor on multivariate analysis. Conclusion: Tobacco smoking is significantly associated with PTB in this central Indian district. Smoking cessation services need to be integrated into the activities of the TB control programme. Keywords: Alcohol consumption, India, Pulmonary tuberculosis, Risk factors, Smoking
Introduction Despite being a curable disease, TB is the world’s leading cause of death from a single infectious agent, accounting for a quarter of the avoidable adult deaths in the developing world.1 Though many biological, socio-economic and behavioural risk factors are known to be associated with the development of pulmonary TB (PTB), tobacco smoking is one of the most important risk factors for TB. It has also been evident for decades that there is a strong association between alcohol use and risk of TB. However, this is rarely mentioned among the challenges identified in the policy documents for TB control. Tobacco use is also a major cause of death in the world and is currently responsible for one in ten adult deaths worldwide.2 The annual death toll from direct tobacco use is over 5 million with more than 600 000 from exposure to second-hand smoke.3 If action is not taken to curb the spread of tobacco use, annual deaths are expected to reach 8.3 million by 2030, of which more than 80% will be in developing countries.4 A systematic review on the association between tobacco and TB indicates
that passive or active exposure to tobacco smoke is significantly associated with TB. Active smoking is also significantly associated with recurrent TB and TB mortality.5 Another systematic review on alcohol use and TB indicates that heavy alcohol use/alcohol use disorders constitute a risk factor for TB and re-infection with TB.6 India has the largest TB burden in the world accounting for almost one fifth of the global incidence. The prevalence of TB in the country has been estimated to be 249/100 000 population, and the mortality due to TB is 24/100 000 population.3 Tobacco smoking in the form of cigarettes and ‘beedis’ is highly prevalent in India, particularly amongst men (24.3%). A beedi consists of flaked tobacco rolled in a rectangular piece of dried Tendu leaf. It is predicted that annual tobacco deaths in India may exceed 1.5 million by 2020. Alcohol consumption is also prevalent in the country, again particularly amongst men (31.9%).7 Despite the current knowledge of the harm caused by tobacco use, its consumption continues to increase in many low- and middle-income countries with the epicentre of the tobacco epidemic shifting from high income countries to developing countries.
# The Author 2014. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved. For permissions, please e-mail: [email protected].
1 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
*Corresponding author: Tel: + 91 761 2370800/818; Fax: + 91 761 2672835; E-mail: [email protected]
V. G. Rao et al.
A recently conducted TB disease prevalence survey in central India showed the overall prevalence of pulmonary TB at 255.3/ 100 000 population.8 Currently there is no information on the effect of tobacco and alcohol consumption on the TB situation in the survey area. Though the associations between smoking and TB have been documented by various authors, this study provides an additional evidence of association between smoking and PTB from a large sample, including the tribal population, in central India.
Materials and methods A cross sectional TB prevalence study was conducted in Jabalpur district of the central Indian state of Madhya Pradesh from January 2009 to January 2010. The district, with more or less equal proportion of urban and rural population, was selected in discussion with the Central TB Division, Ministry of Health and Family Welfare, Government of India, as one of seven sentinel sites to be used by the Revised National TB Control Programme for TB prevalence surveys. The required sample size was estimated to be about 90 000 adults aged ≥15 years for an assumed prevalence of 240/100 000 bacteriologically positive PTB cases, with a precision of 20% at 95% confidence level, a design effect of 2, and coverage for examination of at least 90%. The district was considered the sampling universe for the study. A stratified cluster sampling design was adopted to select the sample. A house-to-house census was carried out and all permanent residents (stay ≥6 months) in the household were registered. Two sputum specimens (one spot, one early morning) were collected from all chest symptomatics and examined by smear microscopy for acid fast bacilli (AFB) and culture on Lowenstein Jensen (LJ) media. The details of the sputum collection and processing have been described in the published report of the survey.8 During the survey, in addition to demographic data, extra information on tobacco smoking and alcohol consumption was collected from the study population and recorded on individual survey cards by investigators who had been trained in survey methodology at the Regional Medical Research Centre for Tribals (RMRCT) in Jabalpur, Madhya Pradesh State, central India. The information collected included exposure to these factors: duration of the habit, current/past history, consumption per day (number of cigarettes/beedis or quantity of alcohol in ml) and was cross checked by the field supervisor. Smokers were categorised on the basis of the number of cigarettes/beedies smoked per day: mild (1–10 cigarettes or beedis/day), moderate (11–20/day), and heavy (.20/day) smokers. Alcohol consumers were categorised on the basis of the quantity of alcohol consumed per day: mild (up to 200 ml/day), moderate (200–500 ml/day) and heavy (.500 ml/day) consumers. The duration of smoking and alcohol consumption was classified into three categories, namely ≤10 years, 11–20 years and .20 years.
Composite index of smoking and alcohol consumption The composite index of smoking was calculated considering their quantity and duration of tobacco smoking. The scores were given for tobacco smoking: zero for non-smoker; one for those smoking ≤10 years, two for 11–20 years and three for .20 years; one for
2 of 8
Data analysis All the completed cards were scrutinized, checked and computerized by trained data entry operators. Prevalence of disease was compared between exposed and non-exposed groups. The prevalence odds ratio (POR) was used as the effect measure. The POR was calculated by measuring the ratio of the prevalence odds of the exposed and non-exposed groups for each risk factor. The crude POR was obtained by univariate analysis. To eliminate the effect of confounding, POR amongst smokers compared to nonsmokers was estimated, stratified by alcohol consumption and vice versa, by each age group and sex. For the variables, found significantly associated with PTB on univariate analysis at 5% level of significance, adjusted prevalence odds ratio (APOR) for pulmonary TB was obtained by multivariate analysis by using logistic regression with risk factors, i.e., age, sex, social class, smoking and alcohol consumption as composite variables, using SPSS 13.0 software (SPSS Inc., Chicago, IL, USA). Linear trend in dose and duration response of tobacco smoking and alcohol consumption was tested using x2 for trend.
Results Coverage Table 1 shows the prevalence of smoking and alcohol consumption in different age groups among men and women. Of 95 071 individuals (48 105 males and 46 966 females) screened for chest symptoms, information on tobacco smoking and alcohol consumption was obtained from 94 559 (99.5%) individuals. The information from remaining individuals could not be obtained as they were not present during successive three visits by the field team. There was equally high coverage among males (47 775/ 48 105; 99.3%) and females (46 784/46 966; 99.6%).
Prevalence of tobacco smoking Of the 94 559 individuals examined, 8448 were smokers with the overall prevalence of 8.9%. Of these, 8279 (98.0%) were current smokers and 169 (2.0%) were ex-smokers. The prevalence was significantly higher amongst males compared with females (p,0.01). The highest prevalence (6502/20 999; 31.0%) was seen amongst males aged 35 years and above. The commonest form of tobacco smoking was ‘beedi’ (7646/8448; 90.5%),
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Study area, sample size, sampling design and procedure
1–10 cigarettes or beedis/day, two for 11–20/day, and three for .20/day. Adding these scores, they were divided into four categories: zero for non-smoker; Score 1–3 as mild, 4 as moderate and .4 as heavy smokers. Similarly, the composite index for alcohol consumption was also calculated. Zero for non-alcohol consumers; one for those consuming alcohol ≤10 years, two for 11–20 years, and three for .20 years; one for ,200 ml/day, two for 200–500 ml/day and three for .500 ml/day alcohol consumption. Adding these scores, the total was divided into four categories namely zero for non-alcohol consumers; 1–3 as mild, 4 as moderate and .4 as heavy alcohol consumers. Informed written consent was obtained from all individuals included in the survey. The study was approved by the ethics committee of the RMRCT, Indian Council of Medical Research.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 1. Prevalence of smoking and alcohol consumption by sex and age groups Sex
Number eligible
Number examined
Prevalence of smoking n (%)
Prevalence of alcohol consumption n (%)
15–34
Male Female Total Male Female Total Male Female Total Male Female Total
26 903 25 626 52 529 14 941 14 248 29 189 6 261 7 092 13 353 48 105 46 966 95 071
26 776 25 558 52 334 14 807 14 183 28 990 6 192 7 043 13 235 47 775 46 784 94 559
1 902 (7.1) 16 (0.1) 1 918 (3.7) 4 525 (30.6) 14 (0.1) 4 539 (15.7) 1 977 (31.9) 14 (0.2) 1 991 (15.0) 8 404 (17.6) 44 (0.1) 8 448 (8.9)
1 092 (4.1) 9 (0.0) 1 101 (2.1) 2 027 (13.7) 22 (0.2) 2 049 (7.1) 626 (10.1) 21 (0.3) 647 (4.9) 3 745 (7.8) 52 (0.1) 3 797 (4.0)
35–54
^55
Total
followed by cigarette smoking (392; 4.6%), and 188 (2.5%) smoked both beedis and cigarettes. Of those who smoked, 5202 (61.6%) were mild smokers and remaining 3246 (38.4%) were moderate to heavy smokers. The majority of smokers (5973; 70.7%) had been smoking for .10 years (Table 2). The overall prevalence of tobacco smoking amongst the tribal population was 19.5% (1870/9610) and was significantly higher amongst males compared with females (1864/4710; 39.6% vs 6/4900; 0.1%; p,0.001).
Prevalence of alcohol consumption The overall prevalence of alcohol consumption was 4.0% (3797/ 94 559), and was significantly higher amongst males compared with females (p,0.01). The proportion of individuals consuming alcohol was highest amongst males aged 35 years and above (with 70.8% (2653/3745) of all males that consumed alcohol in this age group). Of those who consumed alcohol, 3045 (80.2%) were mild consumers and 752 (19.8%) were moderate to heavy consumers. The majority of individuals (2150; 56.6%) had consumed alcohol for a duration of .10 years. Of the total 94 559 individuals, 2841 (3.0%) both smoked tobacco and consumed alcohol (Table 3). The overall prevalence of alcohol consumption amongst the tribal population was 12.2% (1175/9610), and was significantly higher amongst males compared with females (1138/4710; 24.2% vs 37/4900; 0.8%; p,0.001).
Prevalence of TB and association with tobacco smoking and alcohol consumption The prevalence of TB by risk factors is presented in Table 2. The prevalence of TB was observed to be significantly higher amongst males than females and it increased with age. It was significantly higher amongst mild, moderate and heavy smokers compared to non-smokers, and amongst alcohol consumers compared to nonconsumers. The prevalence of TB was significantly higher amongst the tribal compared to non-tribal population. The stratified analysis of prevalence of TB among males in the different age groups by smoking and alcohol consumption also
showed significant association of smoking with TB (Table 3). However after controlling for smoking, no significant association was seen with alcohol consumption and TB (Table 4). Similar analyses were not undertaken for females due to the small number of TB cases in the individual sub-groups. On multivariate analysis, prevalence of TB was found to be significantly associated with sex, increase in age, social class and smoking, but not with alcohol consumption (Table 5).
Discussion Studies investigating the risk factors for TB have been conducted in a variety of settings,9,10 but considering the burden of TB in the country very few have been conducted in India.11,12 Various risk factors including tobacco smoking and alcohol use have been shown to play a role in the development of TB. Tobacco smoking and alcohol use is highly prevalent in India particularly among men.7 Though India has the largest TB burden of any country in the world, there are limited reports on the association of tobacco smoking and alcohol consumption with pulmonary TB from the country. This is the first report of its kind from a community based TB prevalence study conducted on a large sample in central India, including tribal groups in the sampled population. The prevalence of smoking among males (17.6%) was higher as compared to females (0.1%) with the overall prevalence of 8.9% in the present study. It is to be noted that the smoking rates reported here are comparatively lower than those reported in the National Family Health Survey-3 (40.2% in males) and Global Adult Tobacco Survey (16.9% Overall and 30.5% in males) for the state of Madhya Pradesh.7,13 One possible explanation for this observation is the well documented regional variations in smoking pattern in different parts of the state, with a substantially lower prevalence amongst women and men aged 15–34 years in Jabalpur district. This could also be due to recall bias resulting in underestimation of prevalence of smoking. Various studies have clearly shown that the prevalence of PTB is considerably higher among males than females and that prevalence increases with age.12,14 The findings of the present study are consistent with the findings of these earlier studies and
3 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Age group
V. G. Rao et al.
Table 2. Prevalence OR (POR) for selected risk factors for pulmonary TB (PTB) Variable
a
Non-TB
51 170
46 733 47 605
61 96 64
Prevalencea
POR (95% CI)
p value
109 356
1 3.27 (2.39–4.48)
0.001
52 273 28 894 13 171
117 331 484
1 2.85 (2.06–3.92) 4.16 (2.93–5.92)
0.001
165 56
84 784 9 554
194 583
1 3.01 (2.22–4.08)
0.001
139 49 30 3
85 972 5 153 2 942 271
161 942 1 009 1 095
1 5.88 (4.24–8.15) 6.31 (4.24–9.37) 6.85 (2.17–21.6)
0.001
139 11 29 42
85 972 2 464 2 853 3 049
161 444 1 006 1 359
1 2.76 (1.49–5.11) 6.29 (4.21–9.40) 8.52 (6.02–12.0)
0.001
183 30 7 1
90 579 3 015 640 104
202 985 1 082 952
1 4.92 (3.34–7.26) 5.41 (2.53–11.6) 4.75 (0.66–34.3)
0.001
183 8 13 17
90 579 1 134 1 485 1 140
202 701 868 1 469
1 3.49 (1.72–7.10) 4.33 (2.45–7.62) 7.38 (4.47–12.2)
0.001
Prevalence / 100 000.
show that ageing is a significant risk factor for TB as evident from the higher chance of developing TB in older age groups as compared to younger age groups. Men were observed to have an APOR of 2.23, again signifying a strong association of sex with TB, and is consistent with the findings of earlier studies.12,15 A possible explanation is that men are more exposed to the wider world than women, especially in rural areas of central India, with resultant greater social interactions with other people and greater risk of exposure to persons with TB and therefore have a higher chance of becoming infected. As observed in the present study, the higher prevalence of smoking and alcohol consumption among males, could also be additional factors for the higher prevalence rates observed amongst males. The study findings show that smokers had a higher risk of developing TB than non-smokers, even after controlling for the effects of age, sex, alcohol use and social class. This finding is in agreement with many studies conducted both in India and other areas.9,12,14,16 Previous studies have also reported that smoking increases the risk of relapse17 and of subsequent death from
4 of 8
the disease.18 The current study found that there was a strong dose response relationship between tobacco smoking and the occurrence of TB disease. This is in line with the strong dose-response relationship between smoking and TB demonstrated by many other authors.10,12,19,20 The findings of the present study also show a higher prevalence of TB amongst the tribal population. Similar findings have been reported amongst different tribal communities in the central Indian state of Madhya Pradesh.21–23 The present study also showed that tribals had a higher risk of developing TB than nontribals. This could be due to the higher prevalence of tobacco smoking observed in the present study. Higher prevalence of tobacco smoking and its association with TB has also been reported amongst the Saharia tribal community in Madhya Pradesh, central India.24 Potential mechanisms for increased susceptibility to TB among smokers could be both structural changes affecting lung function and altered immune response. These include a decreased level of circulating immunoglobulins, a depression of antibody responses
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Sex Females Males Age in years 15–34 35–54 ^55 Social class Non-tribal Tribal Smoking quantity Non smoker Mild Moderate Heavy Smoking duration Non-smoker ≤10 years 11–20 years .20 years Alcohol quantity Non consumers Mild Moderate Heavy Alcohol duration Non-consumers ≤10 years 11–20 years .20 years
PTB
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 3. Association of Pulmonary TB (PTB) with smoking and alcohol among men in different age groups Age
Alcohol consumption
Smoking
Non-TB
PTB n (per 100 000)
POR (95% CI)
p value
13–34
No
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
24 438 1 208 403 686 9 861 2 865 390 1 615 4 084 1 441 107 507 38 383 5 514 900 2 808
33 (134.9) 5 (412.2) 0 3 (435.4) 27 (273.1) 27 (933.6) 4 (1 015.2) 18 (1 102.3) 24 (584.2) 17 (1 166.0) 0 12 (2 312.1) 84 (218.4) 49 (880.8) 4 (442.5) 33 (1 161.6)
3.06 (1.19–7.86)
0.014
Yes 35–54
No
^55
No Yes
All age groups
No Yes
3.4 (2.01–5.88)
0.001
1.09 (0.36–3.23)
NS
2.01 (1.07–3.75)
0.026
NC 4.06 (2.85–5.78)
0.001
2.64 (0.93–7.48)
0.036
p value
NC: Not calculated.
Table 4. Association of PTB with alcohol consumption and smoking among men in different age groups Age
Smoking
Alcohol consumption
Non-TB
PTB No (per 100 000)
POR (95% CI)
13–34
No
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
24 438 403 1 208 686 9 861 390 2 865 1 615 4 084 107 1 441 507 38 383 900 5 514 2 808
33 (134.9) 0 5 (412.2) 3 (435.4) 27 (273.1) 4 (1 015.2) 27 (933.6) 18 (1 102.3) 24 (584.2) 0 17 (1 166.0) 12 (2 312.1) 84 (218.4) 4 (442.5) 49 (880.8) 33 (1 161.6)
NC
Yes 35–54
No Yes
^55
No Yes
All age groups
No Yes
1.06 (0.25–4.43)
NS
3.75 (1.30–10.8)
0.030
1.18 (0.65–2.15)
NS
NC 2.01 (0.954.25)
NS
2.03 (0.74–5.55)
NS
1.32 (0.85–2.06)
NS
NC: not calculated; NS: not significant; POR: prevalence OR.
to certain antigens, a decrease in CD4+ lymphocyte counts, an increase in CD8+ lymphocyte counts, depressed phagocyte activity, and decreased release of proinflammatory cytokines.25 Nicotine, an important component of cigarette smoke, stimulates
catecholamine and corticosteroid release resulting in the increase of CD8+ lymphocytes in the cellular-mediated system and thereby suppressing the host defense against infections.26 The other components of cigarette smoke such as acrolein,
5 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Yes
NC
V. G. Rao et al.
Table 5. Adjusted prevalence odds ratio (APOR) for selected risk factors for pulmonary TB (PTB) Variable
p value
1 2.35 (1.66–3.32)
0.001
1 2.19 (1.57–3.07) 3.26 (2.23–4.77)
0.001 0.001
1 2.28 (1.43–3.63) 2.51 (1.59–3.96) 2.74 (1.60–4.69)
0.001 0.001 0.001
1 1.13 (0.66–1.94) 1.45 (0.82–2.55) 1.14 (0.35–3.75)
NS NS NS
1 2.32 (1.68–3.21)
0.001
NS: not significant. Composite variables.
a
acetaldehyde, formaldehyde, free radicals produced from chemical reactions within the cigarette smoke, and nitric oxide may contribute to the structural alterations in the airway epithelial cells.27 Though the present study showed a significant univariate effect of alcohol consumption as a risk factor for PTB, it was not found to be a risk factor on multivariate analysis. A study in West Africa also did not find any association of alcohol use with TB. 10 Some workers have however, reported association of alcohol consumption with PTB. 12 Tobacco smoking is often associated with alcohol intake, presenting the potential for confounding. Studies that measured both tobacco and alcohol use found a significant adjusted effect for tobacco even after adjusting for alcohol use.5 Hence the present study suggests that tobacco smoking is significantly associated with PTB disease in central India. Tobacco smoking and TB are two large and mainly preventable epidemics. Though smoking prevalence has remained stable or declined in the developed world, it is on the rise in the developing world. In 2020, more than three-quarters of projected deaths due to tobacco smoking are expected to occur in developing countries.28 A nationally representative study of smoking and death in India has reported that smoking has been causing a large and growing number of premature deaths in India.29 It has been estimated that almost 40% of TB deaths in the country are associated with smoking.11 A prospective study by Pednekar and Gupta has also reported 32% of TB deaths attributable to beedi smoking in
6 of 8
Conclusion The present study indicates that tobacco smoking is significantly associated with PTB. Smoking cessation services need to be integrated into the activities of TB control programmes. To achieve wider public health impact, implementation of tobacco control measures aimed at the general population are also needed.
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Sex Female Male Age in years 15–34 35–54 ^55 Smokinga Non smoker Mild Moderate Heavy Alcohol consumptiona Non-consumers Mild Moderate Heavy Social class Non-tribal Tribal
APOR (95% CI)
India.30 New approaches to TB control including smoking cessation have been advocated to curb this growing epidemic. The potential benefits of smoking cessation in improving outcomes during the course of TB treatment have been demonstrated by Awaisu et al. in a study conducted at five chest clinics in Malaysia.31 Studies conducted in different settings have also recommended smoking cessation as part of the therapeutic plans for individuals with TB and have emphasized training of health professionals to promote smoking cessation.31,32 The International Union against Tuberculosis and Lung Diseases recently called for inclusion of brief smoking cessation advice in standard TB case management. 33 TB patients with a tobacco smoking habit need counselling and support to quit smoking. Considering these, and the findings of the present study, it is re-emphasized that health personnel working with TB patients, who are often not aware of the association between smoking and TB, should be trained in the provision of smoking cessation services to TB patients, their contacts and the community at large. The public and medical practitioners need to be educated about the association of tobacco smoking and TB. The utility of screening smokers for TB disease should be explored in the programmatic setting. There is also a need for further research on the impact of smoking cessation on the occurrence of TB, and other aspects like treatment outcomes, default rates and recurrence of TB. The results of such research will help to guide revision of TB control policy and treatment guidelines of the TB control programme. The limitations of the present study need to be considered while interpreting the results. The findings are based on a crosssectional survey and hence the study design itself has limitations (e.g., recall bias and the assumption that the amount of smoking remained the same during the entire period of exposure). The recall bias regarding smoking habits is more likely in individuals not suffering from TB compared to those with TB. Thus the OR for association of TB occurrence with smoking in this study is likely to be overestimated. However, the extent of overestimation is not expected to be so significant as to alter the finding that smoking is associated with a higher risk of developing TB. Also a standard protocol and questionnaire were used in the disease prevalence survey. As a result, the potential role of other risk factors such as nutritional status, poverty, overcrowding, poor housing, migration, HIV status and biomass fuel usage, were not investigated. While assessing the effects of the selected risk factors, the confounding effects of these other factors therefore could not be controlled for. Although this might have resulted in either an overestimation or underestimation of the true effects of the factors assessed in the study, the findings linking tobacco smoking to TB are consistent with the results found elsewhere. In view of this, future studies should include all of these potential risk factors for better understanding.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Authors’ disclaimer: FWD is a staff member of WHO. The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the decisions or policies of WHO. Authors’ contributions: RVG, BJ, YR and FWD conceived the study; RVG, BJ and YR designed the study protocol; RVG, BJ, YR and BMK actively participated in the field work for data collection and supervision; MM, BMK, and SMA did data entry and cleaning; FWD, MM, SMA and CVK carried out analysis and interpretation of the data; RVG, BJ, YR, MM, BMK, CVK and FWD drafted the manuscript. RVG, MM, SMA, CVK and FWD reviewed the manuscript. All authors read the final draft and provided input to finalize the manuscript. RVG is the guarantor of the paper.
Funding: This study was supported in part by the WHO, with financial assistance provided by the United States Agency for International Development under the WHO/TRC (NIRT) Collaborative Model DOTS Project, and the Indian Council of Medical Research, New Delhi. Competing interests: None declared. Ethical approval: This study was approved by the Ethics Committee of the Regional Medical Research Centre for Tribals, Jabalpur.
10 Lienhardt C, Fielding K, Sillah JS et al. Investigation of the risk factors for tuberculosis: a case–control study in three countries in West Africa. Int J Epidemiol 2005;34:914–23. 11 Gajalakshmi V, Peto R, Kanaka TS, Jha P. Smoking and mortality from tuberculosis and other diseases in India: retrospective study of 43000 adult male deaths and 35000 controls. Lancet 2003;362: 507–5. 12 Kolappan C, Gopi PG, Subramani R, Narayanan PR. Selected biological and behavioural risk factors associated with pulmonary tuberculosis. Int J Tuberc Lung Dis 2007;11:999–1003. 13 GATS India Report 2009–2010. Global Adult Tobacco Survey (GATS) India. International Institute for Population Sciences, Mumbai. Ministry of Health and Family Welfare, Government of India, New Delhi; 2009–2010. 14 Subramani R, Radhakrishna S, Frieden TR et al. Rapid decline in prevalence of pulmonary tuberculosis after DOTS implementation in a rural area of South India. Int J Tuberc Lung Dis 2008;12:916–20. 15 Gustafson P, Gomes VF, Vieira CS et al. Tuberculosis in Bissau: incidence and risk factors in an urban community in sub-Saharan Africa. Int J Epidemiol 2004;33:163–72. 16 Gajalakshmi V, Peto R. Smoking, drinking and incident tuberculosis in rural India: population-based case-control study. Int J Epidemiol 2009;38:1018–25. 17 Thomas A, Gopi PG, Santha T et al. Predictors of relapse among pulmonary tuberculosis patients treated in a DOTS programme in South India. Int J Tuberc Lung Dis 2005;9:556–61. 18 Kolappan C, Subramani R, Kumaraswami V et al. Excess mortality and risk factors for mortality among a cohort of TB patients from rural south India. Int J Tuberc Lung Dis 2008;12:81–6. 19 Gambhir HS, Kaushik RM, Kaushik R, Sindhwani G. Tobacco smoking-associated risk for tuberculosis: a case-control study. Int Health 2010;2:216–22. 20 Prasad R, Suryakant, Garg R et al. A case-control study of tobacco smoking and tuberculosis in India. Ann Thorac Med 2009;4:208–10.
References 1 Malin AS, McAdam KPWJ. Escalating threat from tuberculosis: the third epidemic. Thorax 1995;50:S37–S42. 2 WHO. World Health Report 2003: Shaping the Future. Geneva: World Health Organization; 2003. 3 WHO. WHO Global Report: Mortality Attributable to Tobacco. Geneva: World Health Organization; 2012. 4 Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006;3(11):e442. 5 WHO. A WHO / The Union Monograph on TB and Tobacco Control: Joining efforts to control two related global epidemics. Geneva: World Health Organization; 2007. WHO/HTM/TB/2007.390. 6 Rehm J, Samokhvalov AV, Neuman MG et al. The association between alcohol use, alcohol use disorders and tuberculosis (TB). A systematic review. BMC Public Health 2009;9:450. 7 International Institute for Population Sciences (IIPS) and Macro International. National Family Health Survey (NFHS-3), 2005–06: India Mumbai: IIPS; 2007. 8 Rao VG, Bhat J, Yadav R et al. Prevalence of pulmonary tuberculosis - a baseline survey in central India. PLoS ONE 2012;7(8):e43225.
21 Bhat J, Rao VG, Gopi PG et al. Prevalence of pulmonary tuberculosis amongst the tribal population of Madhya Pradesh, central India. Int J Epidemiol 2009;38:1026–32. 22 Rao VG, Gopi PG, Bhat J et al. Pulmonary tuberculosis: a public health problem amongst Saharia, a primitive tribe of Madhya Pradesh, central India. Int J Infect Dis 2010;14:e713–e6. 23 Rao VG, Bhat J, Yadav R et al. Pulmonary tuberculosis among Bharia - a primitive tribe of Madhya Pradesh. Int J Tuberc Lung Dis 2010;14:368–70. 24 Rao VG, Gopi PG, Bhat J et al. Selected risk factors associated with pulmonary tuberculosis amongst Saharia tribe of Madhya Pradesh, central India. Eur J Public Health 2012;22:271–3. 25 Arcavi L, Benowitz NL. Cigarette smoking and infection. Arch Intern Med 2004;164:2206–16. 26 Miller LG, Goldstein G, Murphy M, Ginns LC. Reversible alterations in immunoregulatory T cells in smoking: analysis by monoclonal antibodies and flow cytometry. Chest 1982;82:526–9. 27 Marcy TW, Merrill WW. Cigarette smoking and respiratory tract infection. Clin Chest Med 1987;8:381–91. 28 Mackay J, Eriksen M. The Tobacco Atlas. Geneva: World Health Organization; 2002.
7 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Acknowledgements: The authors are grateful to Dr Neeru Singh, Director, RMRCT, Jabalpur for her encouragement and support throughout the study. The contributions of the State Tuberculosis Officer, the WHO/ RNTCP consultant, district health authorities, District TB Officer, health officials of Jabalpur Municipal Corporation and the officials from the Jabalpur Cantonment Board are gratefully acknowledged. We thank our study subjects for their cooperation and providing information on risk factors. Thanks are also due to the laboratory and field staff involved in the study. The assistance provided by Mr Shailendra Jain, Mr Narayan Soni from RMRCT Jabalpur and Mrs Malti V Joshi from NTI, Bangalore for entry, verification and analysis of the data is acknowledged. Acknowledgements are given to Dr Edouard Tursan d’Espaignet of the Tobacco Free Initiative, WHO, Geneva, for his comments and assistance in improving the final manuscript.
9 den Boon S, van Lill SW, Borgdorff MW et al. Association between smoking and tuberculosis infection: a population survey in a high tuberculosis incidence area. Thorax 2005;60:555–7.
V. G. Rao et al.
29 Jha P, Jacob B, Gajalakshmi V et al. A nationally representative case-control study of smoking and death in India. N Engl J Med 2008;358:1137–47.
intervention in tuberculosis care on treatment outcomes. Subst Abuse Treat Prev Policy 2011;6:26.
30 Pednekar MS, Gupta PC. Prospective study of smoking and tuberculosis in India. Prev Med 2007;44:496–8.
32 Pradeepkumar AS, Thankappan KR, Nichter M. Smoking among tuberculosis patients in Kerala, India: proactive cessation efforts are urgently needed. Int J Tuberc Lung Dis 2008;12:1139–45.
31 Awaisu A, Nik Mohamed MH, Mohamad Noordin N et al. The SCIDOTS Project: Evidence of benefits of an integrated tobacco cessation
33 Slama K, Chiang CY, Enarson DA. Introducing brief advice in tuberculosis services. Int J Tuberc Lung Dis 2007;11:496–9.
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
8 of 8
Tobacco smoking: a major risk factor for pulmonary tuberculosis – evidence from a cross-sectional study in central India V. G. Raoa,*, J. Bhata, R. Yadava, M. Muniyandia, M. K. Bhondeleya, M. A. Sharadab, V. K. Chadhab and D. F. Waresc
ORIGINAL ARTICLE
Trans R Soc Trop Med Hyg doi:10.1093/trstmh/tru082
a
Regional Medical Research Centre for Tribals (Indian Council of Medical Research), Nagpur Road, P.O. Garha, Jabalpur 482 003, Madhya Pradesh, India; bNational Tuberculosis Institute, Bangalore, Karnataka, India; cGlobal TB Programme, World Health Organization, Geneva, Switzerland
Received 23 January 2014; revised 23 April 2014; accepted 23 April 2014 Background: This paper provides information on the association of tobacco smoking and alcohol consumption with pulmonary TB (PTB) in central India. Methods: A community based cross-sectional TB prevalence survey was conducted in Jabalpur district of the central Indian state of Madhya Pradesh. The information on tobacco smoking and alcohol consumption was collected from individuals aged ≥15 years. Using logistic regression analysis, the risk factors for PTB were identified. Results: A total of 94 559 individuals provided information on tobacco smoking and alcohol consumption. Persons aged 35–54 years and 55 years and above had, respectively, a 2.19 (95% CI 1.57–3.07) and a 3.26 (95% CI 2.23–4.77) times higher risk of developing PTB compared to persons aged below 35 years. Males had a 2.35 (95% CI 1.66–3.32) times higher risk than females. Tribals (indigenous population) had a 2.32 (95% CI 1.68–3.21) times higher risk than non-tribal population. The adjusted prevalence odds ratio for mild, moderate and heavy tobacco smokers were 2.28, 2.51 and 2.74 respectively as compared to non-smokers. Alcohol consumption was not found to be a risk factor on multivariate analysis. Conclusion: Tobacco smoking is significantly associated with PTB in this central Indian district. Smoking cessation services need to be integrated into the activities of the TB control programme. Keywords: Alcohol consumption, India, Pulmonary tuberculosis, Risk factors, Smoking
Introduction Despite being a curable disease, TB is the world’s leading cause of death from a single infectious agent, accounting for a quarter of the avoidable adult deaths in the developing world.1 Though many biological, socio-economic and behavioural risk factors are known to be associated with the development of pulmonary TB (PTB), tobacco smoking is one of the most important risk factors for TB. It has also been evident for decades that there is a strong association between alcohol use and risk of TB. However, this is rarely mentioned among the challenges identified in the policy documents for TB control. Tobacco use is also a major cause of death in the world and is currently responsible for one in ten adult deaths worldwide.2 The annual death toll from direct tobacco use is over 5 million with more than 600 000 from exposure to second-hand smoke.3 If action is not taken to curb the spread of tobacco use, annual deaths are expected to reach 8.3 million by 2030, of which more than 80% will be in developing countries.4 A systematic review on the association between tobacco and TB indicates
that passive or active exposure to tobacco smoke is significantly associated with TB. Active smoking is also significantly associated with recurrent TB and TB mortality.5 Another systematic review on alcohol use and TB indicates that heavy alcohol use/alcohol use disorders constitute a risk factor for TB and re-infection with TB.6 India has the largest TB burden in the world accounting for almost one fifth of the global incidence. The prevalence of TB in the country has been estimated to be 249/100 000 population, and the mortality due to TB is 24/100 000 population.3 Tobacco smoking in the form of cigarettes and ‘beedis’ is highly prevalent in India, particularly amongst men (24.3%). A beedi consists of flaked tobacco rolled in a rectangular piece of dried Tendu leaf. It is predicted that annual tobacco deaths in India may exceed 1.5 million by 2020. Alcohol consumption is also prevalent in the country, again particularly amongst men (31.9%).7 Despite the current knowledge of the harm caused by tobacco use, its consumption continues to increase in many low- and middle-income countries with the epicentre of the tobacco epidemic shifting from high income countries to developing countries.
# The Author 2014. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene. All rights reserved. For permissions, please e-mail: [email protected].
1 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
*Corresponding author: Tel: + 91 761 2370800/818; Fax: + 91 761 2672835; E-mail: [email protected]
V. G. Rao et al.
A recently conducted TB disease prevalence survey in central India showed the overall prevalence of pulmonary TB at 255.3/ 100 000 population.8 Currently there is no information on the effect of tobacco and alcohol consumption on the TB situation in the survey area. Though the associations between smoking and TB have been documented by various authors, this study provides an additional evidence of association between smoking and PTB from a large sample, including the tribal population, in central India.
Materials and methods A cross sectional TB prevalence study was conducted in Jabalpur district of the central Indian state of Madhya Pradesh from January 2009 to January 2010. The district, with more or less equal proportion of urban and rural population, was selected in discussion with the Central TB Division, Ministry of Health and Family Welfare, Government of India, as one of seven sentinel sites to be used by the Revised National TB Control Programme for TB prevalence surveys. The required sample size was estimated to be about 90 000 adults aged ≥15 years for an assumed prevalence of 240/100 000 bacteriologically positive PTB cases, with a precision of 20% at 95% confidence level, a design effect of 2, and coverage for examination of at least 90%. The district was considered the sampling universe for the study. A stratified cluster sampling design was adopted to select the sample. A house-to-house census was carried out and all permanent residents (stay ≥6 months) in the household were registered. Two sputum specimens (one spot, one early morning) were collected from all chest symptomatics and examined by smear microscopy for acid fast bacilli (AFB) and culture on Lowenstein Jensen (LJ) media. The details of the sputum collection and processing have been described in the published report of the survey.8 During the survey, in addition to demographic data, extra information on tobacco smoking and alcohol consumption was collected from the study population and recorded on individual survey cards by investigators who had been trained in survey methodology at the Regional Medical Research Centre for Tribals (RMRCT) in Jabalpur, Madhya Pradesh State, central India. The information collected included exposure to these factors: duration of the habit, current/past history, consumption per day (number of cigarettes/beedis or quantity of alcohol in ml) and was cross checked by the field supervisor. Smokers were categorised on the basis of the number of cigarettes/beedies smoked per day: mild (1–10 cigarettes or beedis/day), moderate (11–20/day), and heavy (.20/day) smokers. Alcohol consumers were categorised on the basis of the quantity of alcohol consumed per day: mild (up to 200 ml/day), moderate (200–500 ml/day) and heavy (.500 ml/day) consumers. The duration of smoking and alcohol consumption was classified into three categories, namely ≤10 years, 11–20 years and .20 years.
Composite index of smoking and alcohol consumption The composite index of smoking was calculated considering their quantity and duration of tobacco smoking. The scores were given for tobacco smoking: zero for non-smoker; one for those smoking ≤10 years, two for 11–20 years and three for .20 years; one for
2 of 8
Data analysis All the completed cards were scrutinized, checked and computerized by trained data entry operators. Prevalence of disease was compared between exposed and non-exposed groups. The prevalence odds ratio (POR) was used as the effect measure. The POR was calculated by measuring the ratio of the prevalence odds of the exposed and non-exposed groups for each risk factor. The crude POR was obtained by univariate analysis. To eliminate the effect of confounding, POR amongst smokers compared to nonsmokers was estimated, stratified by alcohol consumption and vice versa, by each age group and sex. For the variables, found significantly associated with PTB on univariate analysis at 5% level of significance, adjusted prevalence odds ratio (APOR) for pulmonary TB was obtained by multivariate analysis by using logistic regression with risk factors, i.e., age, sex, social class, smoking and alcohol consumption as composite variables, using SPSS 13.0 software (SPSS Inc., Chicago, IL, USA). Linear trend in dose and duration response of tobacco smoking and alcohol consumption was tested using x2 for trend.
Results Coverage Table 1 shows the prevalence of smoking and alcohol consumption in different age groups among men and women. Of 95 071 individuals (48 105 males and 46 966 females) screened for chest symptoms, information on tobacco smoking and alcohol consumption was obtained from 94 559 (99.5%) individuals. The information from remaining individuals could not be obtained as they were not present during successive three visits by the field team. There was equally high coverage among males (47 775/ 48 105; 99.3%) and females (46 784/46 966; 99.6%).
Prevalence of tobacco smoking Of the 94 559 individuals examined, 8448 were smokers with the overall prevalence of 8.9%. Of these, 8279 (98.0%) were current smokers and 169 (2.0%) were ex-smokers. The prevalence was significantly higher amongst males compared with females (p,0.01). The highest prevalence (6502/20 999; 31.0%) was seen amongst males aged 35 years and above. The commonest form of tobacco smoking was ‘beedi’ (7646/8448; 90.5%),
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Study area, sample size, sampling design and procedure
1–10 cigarettes or beedis/day, two for 11–20/day, and three for .20/day. Adding these scores, they were divided into four categories: zero for non-smoker; Score 1–3 as mild, 4 as moderate and .4 as heavy smokers. Similarly, the composite index for alcohol consumption was also calculated. Zero for non-alcohol consumers; one for those consuming alcohol ≤10 years, two for 11–20 years, and three for .20 years; one for ,200 ml/day, two for 200–500 ml/day and three for .500 ml/day alcohol consumption. Adding these scores, the total was divided into four categories namely zero for non-alcohol consumers; 1–3 as mild, 4 as moderate and .4 as heavy alcohol consumers. Informed written consent was obtained from all individuals included in the survey. The study was approved by the ethics committee of the RMRCT, Indian Council of Medical Research.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 1. Prevalence of smoking and alcohol consumption by sex and age groups Sex
Number eligible
Number examined
Prevalence of smoking n (%)
Prevalence of alcohol consumption n (%)
15–34
Male Female Total Male Female Total Male Female Total Male Female Total
26 903 25 626 52 529 14 941 14 248 29 189 6 261 7 092 13 353 48 105 46 966 95 071
26 776 25 558 52 334 14 807 14 183 28 990 6 192 7 043 13 235 47 775 46 784 94 559
1 902 (7.1) 16 (0.1) 1 918 (3.7) 4 525 (30.6) 14 (0.1) 4 539 (15.7) 1 977 (31.9) 14 (0.2) 1 991 (15.0) 8 404 (17.6) 44 (0.1) 8 448 (8.9)
1 092 (4.1) 9 (0.0) 1 101 (2.1) 2 027 (13.7) 22 (0.2) 2 049 (7.1) 626 (10.1) 21 (0.3) 647 (4.9) 3 745 (7.8) 52 (0.1) 3 797 (4.0)
35–54
^55
Total
followed by cigarette smoking (392; 4.6%), and 188 (2.5%) smoked both beedis and cigarettes. Of those who smoked, 5202 (61.6%) were mild smokers and remaining 3246 (38.4%) were moderate to heavy smokers. The majority of smokers (5973; 70.7%) had been smoking for .10 years (Table 2). The overall prevalence of tobacco smoking amongst the tribal population was 19.5% (1870/9610) and was significantly higher amongst males compared with females (1864/4710; 39.6% vs 6/4900; 0.1%; p,0.001).
Prevalence of alcohol consumption The overall prevalence of alcohol consumption was 4.0% (3797/ 94 559), and was significantly higher amongst males compared with females (p,0.01). The proportion of individuals consuming alcohol was highest amongst males aged 35 years and above (with 70.8% (2653/3745) of all males that consumed alcohol in this age group). Of those who consumed alcohol, 3045 (80.2%) were mild consumers and 752 (19.8%) were moderate to heavy consumers. The majority of individuals (2150; 56.6%) had consumed alcohol for a duration of .10 years. Of the total 94 559 individuals, 2841 (3.0%) both smoked tobacco and consumed alcohol (Table 3). The overall prevalence of alcohol consumption amongst the tribal population was 12.2% (1175/9610), and was significantly higher amongst males compared with females (1138/4710; 24.2% vs 37/4900; 0.8%; p,0.001).
Prevalence of TB and association with tobacco smoking and alcohol consumption The prevalence of TB by risk factors is presented in Table 2. The prevalence of TB was observed to be significantly higher amongst males than females and it increased with age. It was significantly higher amongst mild, moderate and heavy smokers compared to non-smokers, and amongst alcohol consumers compared to nonconsumers. The prevalence of TB was significantly higher amongst the tribal compared to non-tribal population. The stratified analysis of prevalence of TB among males in the different age groups by smoking and alcohol consumption also
showed significant association of smoking with TB (Table 3). However after controlling for smoking, no significant association was seen with alcohol consumption and TB (Table 4). Similar analyses were not undertaken for females due to the small number of TB cases in the individual sub-groups. On multivariate analysis, prevalence of TB was found to be significantly associated with sex, increase in age, social class and smoking, but not with alcohol consumption (Table 5).
Discussion Studies investigating the risk factors for TB have been conducted in a variety of settings,9,10 but considering the burden of TB in the country very few have been conducted in India.11,12 Various risk factors including tobacco smoking and alcohol use have been shown to play a role in the development of TB. Tobacco smoking and alcohol use is highly prevalent in India particularly among men.7 Though India has the largest TB burden of any country in the world, there are limited reports on the association of tobacco smoking and alcohol consumption with pulmonary TB from the country. This is the first report of its kind from a community based TB prevalence study conducted on a large sample in central India, including tribal groups in the sampled population. The prevalence of smoking among males (17.6%) was higher as compared to females (0.1%) with the overall prevalence of 8.9% in the present study. It is to be noted that the smoking rates reported here are comparatively lower than those reported in the National Family Health Survey-3 (40.2% in males) and Global Adult Tobacco Survey (16.9% Overall and 30.5% in males) for the state of Madhya Pradesh.7,13 One possible explanation for this observation is the well documented regional variations in smoking pattern in different parts of the state, with a substantially lower prevalence amongst women and men aged 15–34 years in Jabalpur district. This could also be due to recall bias resulting in underestimation of prevalence of smoking. Various studies have clearly shown that the prevalence of PTB is considerably higher among males than females and that prevalence increases with age.12,14 The findings of the present study are consistent with the findings of these earlier studies and
3 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Age group
V. G. Rao et al.
Table 2. Prevalence OR (POR) for selected risk factors for pulmonary TB (PTB) Variable
a
Non-TB
51 170
46 733 47 605
61 96 64
Prevalencea
POR (95% CI)
p value
109 356
1 3.27 (2.39–4.48)
0.001
52 273 28 894 13 171
117 331 484
1 2.85 (2.06–3.92) 4.16 (2.93–5.92)
0.001
165 56
84 784 9 554
194 583
1 3.01 (2.22–4.08)
0.001
139 49 30 3
85 972 5 153 2 942 271
161 942 1 009 1 095
1 5.88 (4.24–8.15) 6.31 (4.24–9.37) 6.85 (2.17–21.6)
0.001
139 11 29 42
85 972 2 464 2 853 3 049
161 444 1 006 1 359
1 2.76 (1.49–5.11) 6.29 (4.21–9.40) 8.52 (6.02–12.0)
0.001
183 30 7 1
90 579 3 015 640 104
202 985 1 082 952
1 4.92 (3.34–7.26) 5.41 (2.53–11.6) 4.75 (0.66–34.3)
0.001
183 8 13 17
90 579 1 134 1 485 1 140
202 701 868 1 469
1 3.49 (1.72–7.10) 4.33 (2.45–7.62) 7.38 (4.47–12.2)
0.001
Prevalence / 100 000.
show that ageing is a significant risk factor for TB as evident from the higher chance of developing TB in older age groups as compared to younger age groups. Men were observed to have an APOR of 2.23, again signifying a strong association of sex with TB, and is consistent with the findings of earlier studies.12,15 A possible explanation is that men are more exposed to the wider world than women, especially in rural areas of central India, with resultant greater social interactions with other people and greater risk of exposure to persons with TB and therefore have a higher chance of becoming infected. As observed in the present study, the higher prevalence of smoking and alcohol consumption among males, could also be additional factors for the higher prevalence rates observed amongst males. The study findings show that smokers had a higher risk of developing TB than non-smokers, even after controlling for the effects of age, sex, alcohol use and social class. This finding is in agreement with many studies conducted both in India and other areas.9,12,14,16 Previous studies have also reported that smoking increases the risk of relapse17 and of subsequent death from
4 of 8
the disease.18 The current study found that there was a strong dose response relationship between tobacco smoking and the occurrence of TB disease. This is in line with the strong dose-response relationship between smoking and TB demonstrated by many other authors.10,12,19,20 The findings of the present study also show a higher prevalence of TB amongst the tribal population. Similar findings have been reported amongst different tribal communities in the central Indian state of Madhya Pradesh.21–23 The present study also showed that tribals had a higher risk of developing TB than nontribals. This could be due to the higher prevalence of tobacco smoking observed in the present study. Higher prevalence of tobacco smoking and its association with TB has also been reported amongst the Saharia tribal community in Madhya Pradesh, central India.24 Potential mechanisms for increased susceptibility to TB among smokers could be both structural changes affecting lung function and altered immune response. These include a decreased level of circulating immunoglobulins, a depression of antibody responses
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Sex Females Males Age in years 15–34 35–54 ^55 Social class Non-tribal Tribal Smoking quantity Non smoker Mild Moderate Heavy Smoking duration Non-smoker ≤10 years 11–20 years .20 years Alcohol quantity Non consumers Mild Moderate Heavy Alcohol duration Non-consumers ≤10 years 11–20 years .20 years
PTB
Transactions of the Royal Society of Tropical Medicine and Hygiene
Table 3. Association of Pulmonary TB (PTB) with smoking and alcohol among men in different age groups Age
Alcohol consumption
Smoking
Non-TB
PTB n (per 100 000)
POR (95% CI)
p value
13–34
No
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
24 438 1 208 403 686 9 861 2 865 390 1 615 4 084 1 441 107 507 38 383 5 514 900 2 808
33 (134.9) 5 (412.2) 0 3 (435.4) 27 (273.1) 27 (933.6) 4 (1 015.2) 18 (1 102.3) 24 (584.2) 17 (1 166.0) 0 12 (2 312.1) 84 (218.4) 49 (880.8) 4 (442.5) 33 (1 161.6)
3.06 (1.19–7.86)
0.014
Yes 35–54
No
^55
No Yes
All age groups
No Yes
3.4 (2.01–5.88)
0.001
1.09 (0.36–3.23)
NS
2.01 (1.07–3.75)
0.026
NC 4.06 (2.85–5.78)
0.001
2.64 (0.93–7.48)
0.036
p value
NC: Not calculated.
Table 4. Association of PTB with alcohol consumption and smoking among men in different age groups Age
Smoking
Alcohol consumption
Non-TB
PTB No (per 100 000)
POR (95% CI)
13–34
No
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
24 438 403 1 208 686 9 861 390 2 865 1 615 4 084 107 1 441 507 38 383 900 5 514 2 808
33 (134.9) 0 5 (412.2) 3 (435.4) 27 (273.1) 4 (1 015.2) 27 (933.6) 18 (1 102.3) 24 (584.2) 0 17 (1 166.0) 12 (2 312.1) 84 (218.4) 4 (442.5) 49 (880.8) 33 (1 161.6)
NC
Yes 35–54
No Yes
^55
No Yes
All age groups
No Yes
1.06 (0.25–4.43)
NS
3.75 (1.30–10.8)
0.030
1.18 (0.65–2.15)
NS
NC 2.01 (0.954.25)
NS
2.03 (0.74–5.55)
NS
1.32 (0.85–2.06)
NS
NC: not calculated; NS: not significant; POR: prevalence OR.
to certain antigens, a decrease in CD4+ lymphocyte counts, an increase in CD8+ lymphocyte counts, depressed phagocyte activity, and decreased release of proinflammatory cytokines.25 Nicotine, an important component of cigarette smoke, stimulates
catecholamine and corticosteroid release resulting in the increase of CD8+ lymphocytes in the cellular-mediated system and thereby suppressing the host defense against infections.26 The other components of cigarette smoke such as acrolein,
5 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Yes
NC
V. G. Rao et al.
Table 5. Adjusted prevalence odds ratio (APOR) for selected risk factors for pulmonary TB (PTB) Variable
p value
1 2.35 (1.66–3.32)
0.001
1 2.19 (1.57–3.07) 3.26 (2.23–4.77)
0.001 0.001
1 2.28 (1.43–3.63) 2.51 (1.59–3.96) 2.74 (1.60–4.69)
0.001 0.001 0.001
1 1.13 (0.66–1.94) 1.45 (0.82–2.55) 1.14 (0.35–3.75)
NS NS NS
1 2.32 (1.68–3.21)
0.001
NS: not significant. Composite variables.
a
acetaldehyde, formaldehyde, free radicals produced from chemical reactions within the cigarette smoke, and nitric oxide may contribute to the structural alterations in the airway epithelial cells.27 Though the present study showed a significant univariate effect of alcohol consumption as a risk factor for PTB, it was not found to be a risk factor on multivariate analysis. A study in West Africa also did not find any association of alcohol use with TB. 10 Some workers have however, reported association of alcohol consumption with PTB. 12 Tobacco smoking is often associated with alcohol intake, presenting the potential for confounding. Studies that measured both tobacco and alcohol use found a significant adjusted effect for tobacco even after adjusting for alcohol use.5 Hence the present study suggests that tobacco smoking is significantly associated with PTB disease in central India. Tobacco smoking and TB are two large and mainly preventable epidemics. Though smoking prevalence has remained stable or declined in the developed world, it is on the rise in the developing world. In 2020, more than three-quarters of projected deaths due to tobacco smoking are expected to occur in developing countries.28 A nationally representative study of smoking and death in India has reported that smoking has been causing a large and growing number of premature deaths in India.29 It has been estimated that almost 40% of TB deaths in the country are associated with smoking.11 A prospective study by Pednekar and Gupta has also reported 32% of TB deaths attributable to beedi smoking in
6 of 8
Conclusion The present study indicates that tobacco smoking is significantly associated with PTB. Smoking cessation services need to be integrated into the activities of TB control programmes. To achieve wider public health impact, implementation of tobacco control measures aimed at the general population are also needed.
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Sex Female Male Age in years 15–34 35–54 ^55 Smokinga Non smoker Mild Moderate Heavy Alcohol consumptiona Non-consumers Mild Moderate Heavy Social class Non-tribal Tribal
APOR (95% CI)
India.30 New approaches to TB control including smoking cessation have been advocated to curb this growing epidemic. The potential benefits of smoking cessation in improving outcomes during the course of TB treatment have been demonstrated by Awaisu et al. in a study conducted at five chest clinics in Malaysia.31 Studies conducted in different settings have also recommended smoking cessation as part of the therapeutic plans for individuals with TB and have emphasized training of health professionals to promote smoking cessation.31,32 The International Union against Tuberculosis and Lung Diseases recently called for inclusion of brief smoking cessation advice in standard TB case management. 33 TB patients with a tobacco smoking habit need counselling and support to quit smoking. Considering these, and the findings of the present study, it is re-emphasized that health personnel working with TB patients, who are often not aware of the association between smoking and TB, should be trained in the provision of smoking cessation services to TB patients, their contacts and the community at large. The public and medical practitioners need to be educated about the association of tobacco smoking and TB. The utility of screening smokers for TB disease should be explored in the programmatic setting. There is also a need for further research on the impact of smoking cessation on the occurrence of TB, and other aspects like treatment outcomes, default rates and recurrence of TB. The results of such research will help to guide revision of TB control policy and treatment guidelines of the TB control programme. The limitations of the present study need to be considered while interpreting the results. The findings are based on a crosssectional survey and hence the study design itself has limitations (e.g., recall bias and the assumption that the amount of smoking remained the same during the entire period of exposure). The recall bias regarding smoking habits is more likely in individuals not suffering from TB compared to those with TB. Thus the OR for association of TB occurrence with smoking in this study is likely to be overestimated. However, the extent of overestimation is not expected to be so significant as to alter the finding that smoking is associated with a higher risk of developing TB. Also a standard protocol and questionnaire were used in the disease prevalence survey. As a result, the potential role of other risk factors such as nutritional status, poverty, overcrowding, poor housing, migration, HIV status and biomass fuel usage, were not investigated. While assessing the effects of the selected risk factors, the confounding effects of these other factors therefore could not be controlled for. Although this might have resulted in either an overestimation or underestimation of the true effects of the factors assessed in the study, the findings linking tobacco smoking to TB are consistent with the results found elsewhere. In view of this, future studies should include all of these potential risk factors for better understanding.
Transactions of the Royal Society of Tropical Medicine and Hygiene
Authors’ disclaimer: FWD is a staff member of WHO. The authors alone are responsible for the views expressed in this publication and they do not necessarily represent the decisions or policies of WHO. Authors’ contributions: RVG, BJ, YR and FWD conceived the study; RVG, BJ and YR designed the study protocol; RVG, BJ, YR and BMK actively participated in the field work for data collection and supervision; MM, BMK, and SMA did data entry and cleaning; FWD, MM, SMA and CVK carried out analysis and interpretation of the data; RVG, BJ, YR, MM, BMK, CVK and FWD drafted the manuscript. RVG, MM, SMA, CVK and FWD reviewed the manuscript. All authors read the final draft and provided input to finalize the manuscript. RVG is the guarantor of the paper.
Funding: This study was supported in part by the WHO, with financial assistance provided by the United States Agency for International Development under the WHO/TRC (NIRT) Collaborative Model DOTS Project, and the Indian Council of Medical Research, New Delhi. Competing interests: None declared. Ethical approval: This study was approved by the Ethics Committee of the Regional Medical Research Centre for Tribals, Jabalpur.
10 Lienhardt C, Fielding K, Sillah JS et al. Investigation of the risk factors for tuberculosis: a case–control study in three countries in West Africa. Int J Epidemiol 2005;34:914–23. 11 Gajalakshmi V, Peto R, Kanaka TS, Jha P. Smoking and mortality from tuberculosis and other diseases in India: retrospective study of 43000 adult male deaths and 35000 controls. Lancet 2003;362: 507–5. 12 Kolappan C, Gopi PG, Subramani R, Narayanan PR. Selected biological and behavioural risk factors associated with pulmonary tuberculosis. Int J Tuberc Lung Dis 2007;11:999–1003. 13 GATS India Report 2009–2010. Global Adult Tobacco Survey (GATS) India. International Institute for Population Sciences, Mumbai. Ministry of Health and Family Welfare, Government of India, New Delhi; 2009–2010. 14 Subramani R, Radhakrishna S, Frieden TR et al. Rapid decline in prevalence of pulmonary tuberculosis after DOTS implementation in a rural area of South India. Int J Tuberc Lung Dis 2008;12:916–20. 15 Gustafson P, Gomes VF, Vieira CS et al. Tuberculosis in Bissau: incidence and risk factors in an urban community in sub-Saharan Africa. Int J Epidemiol 2004;33:163–72. 16 Gajalakshmi V, Peto R. Smoking, drinking and incident tuberculosis in rural India: population-based case-control study. Int J Epidemiol 2009;38:1018–25. 17 Thomas A, Gopi PG, Santha T et al. Predictors of relapse among pulmonary tuberculosis patients treated in a DOTS programme in South India. Int J Tuberc Lung Dis 2005;9:556–61. 18 Kolappan C, Subramani R, Kumaraswami V et al. Excess mortality and risk factors for mortality among a cohort of TB patients from rural south India. Int J Tuberc Lung Dis 2008;12:81–6. 19 Gambhir HS, Kaushik RM, Kaushik R, Sindhwani G. Tobacco smoking-associated risk for tuberculosis: a case-control study. Int Health 2010;2:216–22. 20 Prasad R, Suryakant, Garg R et al. A case-control study of tobacco smoking and tuberculosis in India. Ann Thorac Med 2009;4:208–10.
References 1 Malin AS, McAdam KPWJ. Escalating threat from tuberculosis: the third epidemic. Thorax 1995;50:S37–S42. 2 WHO. World Health Report 2003: Shaping the Future. Geneva: World Health Organization; 2003. 3 WHO. WHO Global Report: Mortality Attributable to Tobacco. Geneva: World Health Organization; 2012. 4 Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006;3(11):e442. 5 WHO. A WHO / The Union Monograph on TB and Tobacco Control: Joining efforts to control two related global epidemics. Geneva: World Health Organization; 2007. WHO/HTM/TB/2007.390. 6 Rehm J, Samokhvalov AV, Neuman MG et al. The association between alcohol use, alcohol use disorders and tuberculosis (TB). A systematic review. BMC Public Health 2009;9:450. 7 International Institute for Population Sciences (IIPS) and Macro International. National Family Health Survey (NFHS-3), 2005–06: India Mumbai: IIPS; 2007. 8 Rao VG, Bhat J, Yadav R et al. Prevalence of pulmonary tuberculosis - a baseline survey in central India. PLoS ONE 2012;7(8):e43225.
21 Bhat J, Rao VG, Gopi PG et al. Prevalence of pulmonary tuberculosis amongst the tribal population of Madhya Pradesh, central India. Int J Epidemiol 2009;38:1026–32. 22 Rao VG, Gopi PG, Bhat J et al. Pulmonary tuberculosis: a public health problem amongst Saharia, a primitive tribe of Madhya Pradesh, central India. Int J Infect Dis 2010;14:e713–e6. 23 Rao VG, Bhat J, Yadav R et al. Pulmonary tuberculosis among Bharia - a primitive tribe of Madhya Pradesh. Int J Tuberc Lung Dis 2010;14:368–70. 24 Rao VG, Gopi PG, Bhat J et al. Selected risk factors associated with pulmonary tuberculosis amongst Saharia tribe of Madhya Pradesh, central India. Eur J Public Health 2012;22:271–3. 25 Arcavi L, Benowitz NL. Cigarette smoking and infection. Arch Intern Med 2004;164:2206–16. 26 Miller LG, Goldstein G, Murphy M, Ginns LC. Reversible alterations in immunoregulatory T cells in smoking: analysis by monoclonal antibodies and flow cytometry. Chest 1982;82:526–9. 27 Marcy TW, Merrill WW. Cigarette smoking and respiratory tract infection. Clin Chest Med 1987;8:381–91. 28 Mackay J, Eriksen M. The Tobacco Atlas. Geneva: World Health Organization; 2002.
7 of 8
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
Acknowledgements: The authors are grateful to Dr Neeru Singh, Director, RMRCT, Jabalpur for her encouragement and support throughout the study. The contributions of the State Tuberculosis Officer, the WHO/ RNTCP consultant, district health authorities, District TB Officer, health officials of Jabalpur Municipal Corporation and the officials from the Jabalpur Cantonment Board are gratefully acknowledged. We thank our study subjects for their cooperation and providing information on risk factors. Thanks are also due to the laboratory and field staff involved in the study. The assistance provided by Mr Shailendra Jain, Mr Narayan Soni from RMRCT Jabalpur and Mrs Malti V Joshi from NTI, Bangalore for entry, verification and analysis of the data is acknowledged. Acknowledgements are given to Dr Edouard Tursan d’Espaignet of the Tobacco Free Initiative, WHO, Geneva, for his comments and assistance in improving the final manuscript.
9 den Boon S, van Lill SW, Borgdorff MW et al. Association between smoking and tuberculosis infection: a population survey in a high tuberculosis incidence area. Thorax 2005;60:555–7.
V. G. Rao et al.
29 Jha P, Jacob B, Gajalakshmi V et al. A nationally representative case-control study of smoking and death in India. N Engl J Med 2008;358:1137–47.
intervention in tuberculosis care on treatment outcomes. Subst Abuse Treat Prev Policy 2011;6:26.
30 Pednekar MS, Gupta PC. Prospective study of smoking and tuberculosis in India. Prev Med 2007;44:496–8.
32 Pradeepkumar AS, Thankappan KR, Nichter M. Smoking among tuberculosis patients in Kerala, India: proactive cessation efforts are urgently needed. Int J Tuberc Lung Dis 2008;12:1139–45.
31 Awaisu A, Nik Mohamed MH, Mohamad Noordin N et al. The SCIDOTS Project: Evidence of benefits of an integrated tobacco cessation
33 Slama K, Chiang CY, Enarson DA. Introducing brief advice in tuberculosis services. Int J Tuberc Lung Dis 2007;11:496–9.
Downloaded from http://trstmh.oxfordjournals.org/ at Southern Methodist University on August 15, 2014
8 of 8
