Association of mass media exposure with combustible and smokeless tobacco use among Indian population: findings from a panel survey

Combustible and smokeless tobacco consumption is the most significant preventable cause of death worldwide [1]. Being the second-largest tobacco consumer, India is one of the global players in the tobacco epidemic race [2]. In 2016, tobacco consumption (including smoking, smokeless tobacco and second-hand smoke) had alone contributed 6% DALYs (Disability-adjusted life-years) in India, and the burden was higher among the men (8.3% of total DALYs) than in women (3% of total DALYs) [3]. Existing evidence from India has shown the trends in the prevalence of different forms of tobacco consumption [4‐6]. Several studies linked combustible and smokeless tobacco consumption with family, friends, and peer influence (social support for tobacco use) [7]. A couple of studies have linked tobacco consumption with illiteracy and working stress [8, 9]. Other studies identified affordability and social acceptability [10], socioeconomic inequality [11], pro-tobacco media campaigns by tobacco companies to attract the population, especially the youngsters [12], as factors leading to a hike in smokeless and combustible tobacco consumption. Additionally, the powerful influence of films and advertisements featuring the macho image of characters who smoke has an everlasting impact on children and adolescents’ minds, leading them to adopt similar tobacco consumption behaviour [13]. Some studies have also shown the differential in smoking and smokeless tobacco usage across the Indian states and communities [5, 6]. A cross-sectional study from India had tried to reveal geographic variation in tobacco consumption and showed the importance of local contextual factors and policies that shape tobacco use [14].

Despite such prominent explanatory factors of tobacco consumption, a recent reduction in combustible tobacco has been noticed among Indian individuals [15, 16]. However, a different concern of increased smokeless tobacco consumption compared to smoking has come up [17]. It is worth noting that India, Bangladesh and Myanmar jointly contribute 71% of world smokeless tobacco consumers [18]. One study showed that a ban on smoking in public places in India had resulted in an increment in smokeless tobacco consumption [19]. Another research from three countries (Bangladesh, India, Nepal) of Southeast Asian regions revealed that tobacco companies’ misleading tobacco advertisements continuously drove smokers to the alternative of smokeless tobacco consumption [17]. The study noted that the marketing of tobacco products was done by promoting them as an inalienable part of the consumer’s lifestyle. Direct and surrogate advertising of these products through the media influences the consumers and encourages them to use them [20]. However, the Indian parliament had introduced the Cigarettes and Other Tobacco Products Act (COTPA), 2003, which came into effect from May 2004 to ensure that Indian people do not indulge in or increase their tobacco consumption by being influenced by the media content [21]. Ample evidence from India, since 2005, revealed that media exposure still plays an essential role in increasing combustible and smokeless tobacco usage among people [22, 23].

The present study examines the association between tobacco use and mass media exposure among the Indian population, considering the contextual factors that may affect the clustering of tobacco consumption at both community and state levels. The rationale for such analysis is as follows. First, minimal attempts have been made to understand the effect of the tobacco advertisement ban through mass media on the likelihood of combustible and smokeless tobacco use in the Indian context. One study using National Family Health Survey 2005–06 data (i.e., after COTPA act implementation) showed the association between smoking and smokeless tobacco use and mass media exposure among Indian adults aged 15–54 years [24]. However, it could not provide similar evidence for adolescents and the elderly. Second, rather than proceeding with a before-after impact assessment of the COTPA act, this study wants to utilize the panel nature of the India Human Development Survey (IHDS) dataset conducted in 2005 and 2012, to explore the association of mass media exposure and different tobacco consumption in the short-term (in the early phase of a 1-year leap) and the long-term (in the later phase of 7 years leap) period among the same individuals after the COTPA act implementation. Third, a dearth of evidence on the association of tobacco consumption and mass media exposure after considering the contextual level factors convince us to explore the variation in combustible and smokeless tobacco consumption clustering using a three-level multilevel approach.

Using two rounds of IHDS, this panel study examined the association of combustible and smokeless tobacco consumption with mass media exposure among the Indian population considering the extent of clustering and heterogeneous risk of tobacco consumption at the state and community levels. The results revealed no significant association between mass media exposure and combustible and smokeless tobacco consumption across the two rounds. While comparing both rounds using marginal predicted probability, this study further shows a minimal change in smoking behaviour and an increment in smokeless tobacco consumption from the short-term to the long-term period after COTPA act implementation. It was worth noting that, in the short-term and long-term phase after the COTPA act implementation, exposure to television, radio and newspaper was no longer associated with tobacco consumption. These findings are similar to a 2015 Indian study that showed how strategies like banning advertisements had efficiently nullified the association between mass media exposure and tobacco consumption [40]. However, the results of this study were also contradictory with another Indian study, using 2005–06 data for 15–49 aged women and 15–54 aged men [24]. This study highlighted the association of television and radio with a higher prevalence of tobacco chewing among men and newspaper reading with a lower likelihood of smokeless tobacco consumption among women [24].

Further, the current study observed the presence of clustering among individuals and a significant level of unobserved contextual risk of combustible and smokeless tobacco at the community and state levels. Community-level clustering was more pronounced as compared to the state-level in both rounds. Although, along with the nationally-implemented acts, India had witnessed different community-level initiatives (e.g. tobacco-free village) for tobacco control [41] and the state administration partnership helping various states win the tag of “smoke-free state”. Some studies contradict such association, providing evidence of increment in tobacco use in movies to promote such behaviour among youngsters at both state and community levels [42]. India has various entertainment sources across different communities and states and diverse cultures. The content shown in such entertainment sources might be the reason for promoting combustible and smokeless tobacco in India.

This study further revealed that education among individuals and the community had helped decrease combustible tobacco consumption. Besides, the smokeless tobacco consumption had increased from the short-term to the long-term phase of COTPA act implementation, and this result was consistent with a couple of studies [24, 43]. Smoking was higher among adults, and the elderly, whereas women were inclined towards smokeless tobacco consumption, and such results are consistent with an extant Indian study [10]. The high prevalence of smokeless tobacco consumption among Indian women occurred because it was culturally acceptable among some communities [42, 43] and was readily available due to its inexpensiveness. Further, the growing campaigns [12] and efforts of the government to air anti-tobacco television ads [44] adversely affect smoking behaviour across the country, making the tobacco industry more inclined towards the marketing of smokeless tobacco and introducing it as a quick replacement for combustible tobacco.

Exposure to radio has been a common means of communication and entertainment among people for many years, unlike, television which was seen as a newcomer and yet influential to every individual’s life [45]. Radio is a means of communication available in different languages and is readily accepted by individuals irrespective of their literacy status or age. Also, radio usage are common among some communities whose individual’s sit together and usually share their experience and behaviour. In such a situation, any pro- and anti-tobacco advertisements can influence many individuals in a community. Like radio, a newspaper is also a media type commonly seen among some communities, but more than this, it is an individual choice media which is common among the literate and the higher section of society. Although radio and newspaper exposure was not associated with tobacco consumption, a higher amount of community-level clustering in tobacco consumption among the Indian population may be explained by the effect of mass media on the communities. Besides, the variation in geographic level factors was also consistent with an Indian study [14].

One of the key strengths of this study is that rather than providing any impact assessment of the COTPA act, we have tried to examine the changes from the short term to the long-term period of COTPA act implementation on the combustible and smokeless tobacco consumption behaviour among Indian population using panel data. The study provided the opportunity to understand how the growing ban of tobacco advertisements on mass media after COTPA act implementation had reduced combustible tobacco consumption but paved the way for increment in smokeless tobacco marketing due to their inexpensive and readily available nature. The study also provided significant evidence that the risk of smoking and consuming smokeless tobacco varies significantly at the community and state levels. However, the study has its shortcomings too. Although past literature had brought forward the association between tobacco use and mass media exposure before and after the COTPA act implementation, the present study could not analyze such association due to the unavailability of information in IHDS data. Moreover, the study assumed that exposure to mass media involves involuntary exposure to advertisements promoted by commercial organizations through these media. However, to verify this assumption, one needs data on the media content type that an individual is exposed to, which was not possible in this study due to a lack of data. Primary studies considering the quality of content in the mass media can be conducted to address this limitation. Lastly, this study examines the correlation between tobacco consumption and mass media exposure, and the findings do not imply causality.

The present study found a minimal change in the significant effect of mass media on combustible tobacco consumption among the Indian population after the COTPA act implementation. However, an increment of smokeless tobacco consumption during the two rounds, along with higher community-level clustering in tobacco consumption, had indicated the growing burden of smokeless tobacco behaviour. In terms of research implications, the findings show that mass media exposure cannot be considered as a strong predictor of combustible tobacco consumption in the Indian population. However, there is a need to view the content of media exposure as the type of content usually changes with the type of  medium. In terms of policy implications, there is a need for policy reforms to curb the significant effect of factors that promotes smokeless tobacco consumption in India, along with health warning labels on all types of tobacco to increase awareness in the individuals [46]. Moreover, clustering implies that such policies need to be implemented in specific high-risk communities from high-risk states, thereby reducing the time and cost required for implementation.