2.1. Sample

The TCP India Pilot Survey is a cross-sectional survey conducted in 2006 in the states of Maharashtra and Bihar, as a lead-into the TCP India Project, a larger prospective cohort study which began in 2010. Study recruitment for the pilot survey was focused on urban and surrounding rural areas of the major cities of Mumbai and Patna, respectively. The TCP India Pilot Survey, as with all other surveys conducted by the International Tobacco Control Policy Evaluation Project (ITC Project), conducted in a total of 22 countries to date, was designed to evaluate and understand the psychosocial and behavioral effects of national-level tobacco control policies [14]. In India, the “ITC India Project” was renamed as the “Tobacco Control Policy (TCP) India Project” to avoid confusion with the “India Tobacco Company.”

The target population for the pilot survey included approximately equal numbers of adult current smokers (i.e., someone who reported smoking cigarettes or bidis at least weekly, n = 249), current smokeless tobacco users (i.e., someone who reported only current use of any smokeless tobacco product, n = 248), and non-users of tobacco (i.e., never users; n = 267) aged 18 years and older. Dual users of tobacco (i.e., those who reported current use of both smoked and smokeless tobacco) were excluded from the study. The surveys were conducted by face-to-face interviews in households through a survey involving a household enumeration process. In this enumeration, data on gender, age, and current tobacco use status were collected from the head of household for all members. A maximum of four respondents were interviewed from a household and priority was given to youth. Surveys were conducted in Hindi in Bihar and Marathi in Maharashtra. Actual survey time was an average of two hours per respondent (range 1.5–2.5 h); each respondent was given a small token at the end in appreciation of their time. Additional information on the research design and survey methods is reported in Raute et al. [15] and on the ITC Project Web site (www.itcproject.org). The study protocol and survey materials were approved by the Office of Research Ethics at the University of Waterloo, Canada, and the Institutional Review Board at the Healis-Sekhsaria Institute for Public Health, India.

2.2. Measures

The TCP India Pilot Survey included questions on a number of individual, environmental, and policy factors, standardized to the ITC Surveys conducted in other countries, but with minor modifications based on language and colloquial differences. Questions included demographics, tobacco use, policy-relevant domains (e.g., health warning labels, advertising/promotion, secondhand smoke restrictions), psychosocial factors (including knowledge, attitudes, and beliefs about different tobacco products), and intention to quit.

Participants were first asked the general ad awareness question: “In the last 6 months, how often have you noticed things that are designed to encourage smoking or which make you think about tobacco?” (Response options: never, once, once in a while, often, DK/cannot say). Next, they were asked the following individual questions about noticing tobacco advertising in specific locations: “In the last 6 months, have you noticed tobacco products being advertised in any of the following places?” (Response options: yes, no, DK/cannot say; places listed included on television, on radio, on posters, on billboards, in newspapers or magazines, in cinema halls, on shop/store windows or inside shops/stores where you buy tobacco, on or around street vendors, in the workplace, on public transportation vehicles or stations, in restaurants, cafeterias, or tea bars. Then participants were asked questions specific to the level of awareness of tobacco promotion in select locations: (1) “In the last 6 months, have you seen or heard about any sport or sporting events that are sponsored by or connected with either tobacco brands or tobacco companies?” (Response options: yes, no, DK/cannot say); (2) “In the last 6 months, have you seen or heard about any music, theater, art, or fashion events that are sponsored by or connected with either tobacco brands or tobacco companies?” (Response options: yes, no, DK/cannot say); (3) “In the last 6 months, have you seen or heard about any religious events that are sponsored by or connected with either tobacco brands or tobacco companies?” (Response options: yes, no, DK/cannot say); (4) “In the last 6 months, have you noticed any of the following types of tobacco promotion?” (Response options: yes, no, DK/cannot say, offers listed included free sample of tobacco, special price offers for tobacco, free gifts or special discount offers on other products when buying tobacco, clothing or other items with a tobacco brand name or logo, competitions linked to tobacco); and (5) “Thinking about the entertainment media, like movies, TV programs, magazines, in the last 6 months, about how often, if at all, have you seen people smoking in the entertainment media?” (Response options: never, once in a while, often, DK/cannot say). Response options of ‘often’, ‘once in a while’, ‘once’, and ‘yes’ were coded as noticing any advertisement or promotion for tobacco (1); all other responses were coded as not noticing (0).

Participants were also asked questions regarding their beliefs and attitudes about tobacco use. Specifically, participants were asked: (1) “Do you think smokeless tobacco use is good for your health?” (Response options: good, neither good nor bad, not good, DK/cannot say); (2) “Do you think smoking is good for your health?” (Response options: good, neither good nor bad, not good, DK/cannot say); and (3) “How soon after waking do you usually smoke?” (Response options: <5 min, 6–30 min, 31–60 min, 61+ mins, not sure). Respondents were also asked, individually: “In the last month, how often, if at all, did you: Think about how much you enjoy smoking?; Think about the harm your smoking might be doing to you?; Seriously consider quitting smoking?; Think about the cost of smoking?” (Response options: never, sometimes, often, DK/cannot say). Finally, respondents were asked: “In the last month, have you butted out a cigarette or bidi before you finished it because you thought about the harm of smoking?” (Response options: yes, no, not sure).

2.3. Statistical analyses

Chi-square tests were used to examine differences in demographic characteristics (age, sex, income, education, current tobacco use status (smoker, smokeless tobacco user, non-user of tobacco) and state and locale of residence (Bihar urban or rural, Maharashtra urban or rural). A summary index was created indicating the total number of places respondents reported noticing tobacco product advertising and promotion (i.e., level of awareness of tobacco advertising and promotion) in the last six months. The summary index consisted of the general ad awareness question, the tobacco advertising questions, and the tobacco promotion questions described in the Measures section. Linear regression was then used to test associations between current tobacco use habit (smoker, smokeless user, non-user of tobacco) and this level of awareness. The analyses were adjusted for the following variables (categorical): age in years (18–24, 25–39, 40–54, 55 and older); sex (male, female); education (standardized to low [illiterate or primary], moderate [middle or secondary], high [college or above]), household income level (low [<5000 INR per month], middle [5000–15,000 INR per month, high [>15,000 INR per month]), and state and locale of residence (Maharashtra urban or rural, Bihar urban or rural). Two-way Analysis of Variance was employed to evaluate differences in mean responses to the respondent’s perception of overall health by tobacco use status and geographic region (state and urban vs. rural). All analyses were conducted using SPSS 21.

3.1. Sample characteristics

Sample characteristics are presented in Table 1. A higher percentage of respondents were recruited from the rural areas of each state compared with the urban areas, with equal distributions of smokers, smokeless tobacco users, and non-users of tobacco across the respondents. Two-thirds of the sample were male, and half of the sample was of low education. More than half of the sample was of low income as well. Statistically significant differences were observed when examining respondents from different regions surveyed (Bihar urban or rural, Maharashtra urban or rural). Notably, significant differences were detected with a greater percentage of smokeless tobacco users living in urban Maharashtra (38%), a slightly greater percentage of smokers living in rural Bihar (38%), and a greater percentage of non-users living in urban Bihar (52%) (Chi-square statistic p < 0.01). Additionally, differences were noted in age, education, and income of respondents between the different geographic regions.

3.2. Observations of pro-tobacco advertising and promotion

All respondents were asked questions about awareness of advertising and promotion of tobacco products for the past six months. The responses are presented in Table 2. Frequencies of noticing tobacco advertising and promotional materials differed between smokers, smokeless tobacco users, and non-users largely by the vendor or location; for example, smokeless tobacco users noticed tobacco ads on shop/store windows or inside shops where they buy tobacco (borderline statistical significance, Chi-square statistic p-value = 0.06), while non-users were significantly more likely to report noticing tobacco ads on posters (Chi-square statistic p-value = 0.003). Few people reported seeing or hearing about a sporting, music, theater, art, or religious event being sponsored by a tobacco company or connected to a tobacco brand. While few people also reported noticing specific types of tobacco promotions, including free samples, special offers, special discounts, or competitions linked to tobacco (<6%), while a slightly greater percentage reported noticing clothing or other items with tobacco brand names or logos (9–14%). Linear regression analysis found that although tobacco use status, age, income, and locale were not significantly associated with the level of awareness of tobacco advertising (results not shown), females were less likely to report greater levels of awareness (β = −0.126, p < 0.01), while respondents with a higher education were more likely to report greater levels of awareness of tobacco advertising (β = 0.234, p < 0.01).

Results from this study found that over two-thirds (74%; n = 562) of respondents reported seeing any form of pro-tobacco advertising in the last six months, with no differences observed between smokers (74%), smokeless tobacco users (74%), and nonsmokers (73%). This finding demonstrates a need for greater enforcement and compliance with Section 5 of COTPA 2003, which prohibits advertising of tobacco products in both direct and indirect forms [7]. The analyses also found that more than half of the respondents (54%) reported seeing pro-tobacco advertising on store windows or inside of shops where they buy tobacco and one-third (32%) reported noticing displays on or around street vendors. This highlights an area that needs additional specific regulation for both cigarettes and other tobacco products.

3.3. Perceptions of risk

The findings from this study, as presented in Table 3, show that in general smokers had a negative opinion of smoking and believed that smoking is bad for their health. However, most smokers and smokeless tobacco users in the sample believed they are in good health (96% smokers and 98% smokeless tobacco users reported ‘good’, ‘very good’, or ‘excellent’ when asked about their overall health) and over half did not perceive themselves as addicted to tobacco (p-value <0.01). In addition, most tobacco users believe that the government should do more to control tobacco use and have a negative opinion about tobacco companies. Tobacco users in the sample were also split on their responses to statements such as: “You enjoy tobacco too much to give it up” and “If you had to do it over again, you would not have started using tobacco.” Although the differences between smokers and smokeless tobacco users were not significant, the results suggest that although tobacco users know that smoking is not good for their health and is addictive, their objective beliefs about tobacco use may be rationalized by their visceral reactions to the products themselves.

When asked to rate their overall health with the question: “Now a question about your overall health. In general, how would you describe your health?” on a scale of 0 (poor) to 4 (excellent), all respondents in the survey rated their health highly (overall mean = 3.26). Stratified analyses (Table 4) showed that non-users, on average, had higher perceptions of their overall health (mean = 3.49), followed by smokeless tobacco users (mean = 3.24); smokers, on average, rated their overall health lower compared with these two groups (mean = 3.04).

When evaluating an association between perception of overall health and tobacco use status while also accounting for locale of respondent (i.e., urban or rural area of Maharashtra, urban or rural area of Bihar), the post hoc tests revealed a significant difference in perception of overall health between smokers, smokeless tobacco users, and non-users (Two-Factor ANOVA F-statistic 22.413, p < 0.001). However, the estimated marginal means and profile plots did not reveal an interaction between the two factors, suggesting that respondents in urban and rural areas of Maharashtra and Bihar do not have significantly different perceptions about their overall health (Two-Factor ANOVA F-statistic 0.970, p = 0.444). Additional data summarizing the responses to knowledge and belief items about specific tobacco products and their use among the TCP India Pilot Survey respondents have been reported in greater detail in Surani et al., Sansone et al. and Raute et al. [15–17].