You are here

Article Text

Article menu

Download PDFPDF

Original research
Neighbourhood tobacco retail access and tobacco use susceptibility in young adolescents in urban India
  1. Ritesh Mistry1,
  2. Michael J Kleinsasser2,
  3. Namrata Puntambekar3,
  4. Prakash C Gupta4,
  5. William J McCarthy5,
  6. Trivellore Raghunathan2,
  7. Keyuri Adhikari4,
  8. Sameer Narake4,
  9. Hsing-Fang Hsieh1,
  10. Maruti Desai4,
  11. Shervin Assari6,
  12. Joseph Alberts1,
  13. Mangesh S Pednekar4
  1. 1 Department of Health Behavior and Health Education, University of Michigan, Ann Arbor, Michigan, USA
  2. 2 Department of Biostatistics, University of Michigan, Ann Arbor, Michigan, USA
  3. 3 Department of Research, Healis Sekhsaria Institute for Public Health, Navi Mumbai, Maharashtra, India
  4. 4 Healis Sekhsaria Institute for Public Health, Navi Mumbai, Maharashtra, India
  5. 5 Department of Health Policy and Management, University of California Los Angeles, Los Angeles, California, USA
  6. 6 Department of Family Medicine, Charles R Drew University of Medicine and Science, Los Angeles, California, USA
  1. Correspondence to Dr Ritesh Mistry, Health Behavior and Health Education, University of Michigan, Ann Arbor, MI 48109, USA; riteshm{at}umich.edu

Abstract

Background Neighbourhood tobacco retail access may influence adolescent tobacco use. In India, we examined the association between neighbourhood tobacco retail access and cognitive risks for tobacco use during early adolescence.

Methods In 2019–2020, a population-based sample (n=1759) of adolescents aged 13–15 years was surveyed from 52 neighbourhoods in Mumbai and Kolkata. Neighbourhood tobacco retail access was measured as the frequency of visits to tobacco retailers, mapped tobacco retailer density and perceived tobacco retailer density. We estimated associations between neighbourhood tobacco retail access and cognitive risks for tobacco use (perceived ease of access to tobacco, perceived peer tobacco use and intention to use tobacco).

Results There was high neighbourhood tobacco retail access. Tobacco retailer density was higher in lower income neighbourhoods (p<0.001). Adolescent frequency of tobacco retailer visits was positively associated with cognitive tobacco use risks. Mapped tobacco retailer density was associated with perceived ease of access in Kolkata but not in Mumbai, and it was not associated with perceived peer tobacco use nor intention. Perceived tobacco retailer density was associated with perceived ease of access and perceived peer use, but not with intention. In Kolkata, higher perceived retailer density and frequency of tobacco retailer visits were negatively associated with perceived ease of access.

Conclusions Efforts to reduce neighbourhood tobacco retail access in India may reduce cognitive tobacco use risk factors in young adolescents. The frequency of tobacco retailer visits and perceived tobacco retailer density increased cognitive risks, though there were some exceptions in Kolkata that further research may explain.

  • global health
  • advertising and promotion
  • environment
  • priority/special populations
  • prevention

Data availability statement

Deidentified data may be available upon reasonable request.

https://doi.org/10.1136/tobaccocontrol-2021-056915

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Over 1.3 million people die annually in India from tobacco use.1 2 The prevalence of current tobacco use has been declining and recent estimates show that 32.8% of adults3 and 8.5% of adolescents4 report current use. The most common tobacco products used in India are smokeless products such as paan or paan masala with tobacco, khaini, mishri, snuff, snus and others, but smoking products like cigarettes, bidis, pipes and hookah are also widely available and used. A key policy strategy to reduce India’s health burden from tobacco is to reduce access to tobacco products, particularly among adolescents, who are vulnerable because greater access can lead to early tobacco experimentation and regular use.5–7 The Government of India through the Cigarettes and Other Tobacco Projects Act has restricted where tobacco can be sold and advertised by banning the sale of tobacco near educational institutions and banning tobacco advertisements including at retail locations.8

    Adolescents in India who use tobacco usually acquire it from retail sources abundant in urban and suburban areas.9 Retail tobacco access may increase future adolescent tobacco use risk by influencing cognitive factors associated with experimentation and initiation of regular tobacco use. For example, retail tobacco access may increase perceptions about tobacco availability, exposure to tobacco advertisements and exposure to tobacco-related behaviours among community members, which could influence cognitive factors such as perceived ease of access to tobacco products, perceived peer tobacco use norms and intentions to use tobacco. These cognitive factors have been used to assess tobacco use susceptibility10 and have been shown to increase tobacco use risk.11 12 These cognitive factors are of particular interest when assessing susceptibility to future tobacco use in tobacco-naïve young adolescents.

    Research suggests that both access to tobacco retailers and the retail environment itself influence tobacco use in youth. For example, a study of high schoolers in India reported increased risk of smokeless tobacco use in school neighbourhoods with higher tobacco retailer density and point-of-sale advertisement density,9 which were more prevalent in low-income school neighbourhoods. Another study in India found that adolescents who said it was possible to buy tobacco near school were much more likely to use tobacco.13 Compliance with retail laws about tobacco promotion and sales to minors in India is very low and concerning,14–20 given that greater compliance to tobacco retail laws in India appears to be associated with reduced tobacco use risk in adolescents.15 Research from other countries also suggests that neighbourhood tobacco retail access may increase risks of tobacco use.5 21 22 Cross-sectional studies, for example, have found that greater tobacco retailer density surrounding schools was associated with smoking susceptibility23 and experimental smoking.24 Smoking susceptibility was measured as intention to smoke, lower likelihood to refuse a cigarette and acquiring cigarettes from retail sources.11 Adolescent reports of visiting convenience stores also increased the risk of tobacco use.25

    To further understand the role of neighbourhood tobacco environmental factors on adolescent tobacco use, the current study assessed the association between neighbourhood tobacco retail access (measured as tobacco retailer visit frequency, mapped tobacco retailer density and perceived tobacco retailer density) and cognitive risk factors for tobacco use in young adolescents living in urban India.

    Methods

    The study was conducted in two geographically, economically and culturally distinct large metropolitan areas in India (Mumbai and Kolkata). Mumbai is a major cultural, financial and entertainment centre with an economically and culturally diverse population from across India. Kolkata is a cultural centre of East India, where economic growth has been slow compared with Mumbai. The Kolkata population is culturally diverse but less so than Mumbai.26 27

    In 2019–2020, we conducted a population-based household survey of adolescents aged 13–15 years (n=1759) and their parents/main caregivers (parents) from 52 neighbourhoods in Mumbai and Kolkata. We used the Urban Frame Survey28 sampling frame from the National Sample Survey Organisation (NSSO) of the Indian Ministry of Statistics and Programme Implementation to draw a multistage random sample of neighbourhoods and households. NSSO Urban Frame Survey sampling units (called ‘investigator unit’) were selected using a stratified random sample accounting for socioeconomic status. Household-level response rates were 93% in Kolkata and 86% in Mumbai. Parents/main caregivers provided informed consent and permission for their adolescent to participate, and adolescents provided informed assent. Each participant completed a computer-assisted personal interview (~20 min) and audio computer-assisted self-interview (~20 min).29 We used a process of cognitive and pilot testing to ensure that survey questions were culturally relevant and were understood by participants as intended by the research team.

    Cognitive risk factors of tobacco use

    Tobacco was defined as smoking (cigarette, bidi, cigar, chutta or dhumti, hookah or waterpipe, chillum, pipe), chewing (gutka, pan masala, betel-quid, khaini, mawa, zarda) or applying (mishri, gul, bajjar, snuff, tobacco toothpaste, tobacco tooth powder and other products) forms. Adolescent perceived ease of access to tobacco was measured by asking, ‘Do you think it would be easy or hard for you to get tobacco products if you wanted them?’ (four-level Likert response scale from ‘Very easy’ to ‘Very hard’). Perceived peer tobacco use norms was measured by asking, ‘How many kids your age do you think use tobacco?’ (four-level Likert response scale from ‘None of them’ to ‘All of them’). To measure intention to use tobacco we asked, ‘At any time during the next 12 months, do you think you will [smoke/chew or apply] tobacco in any form?’ (four-level Likert response scale from ‘Definitely yes’ to ‘Definitely not’). These questions were adapted from the Global Youth Tobacco Survey-India and our prior research in India.9 29 30 For logistic regression models described below, we defined perceived ease of access as: very easy (1) versus not very easy (somewhat easy, somewhat hard and very hard) (0); perceived norms as some or more peers use tobacco (1) versus none (0); and intention to use tobacco as definitely not (0) versus some or more (probably not, probably yes and definitely yes) (1).

    Tobacco retailer visit frequency index

    We created an index based on how often participants reported visiting general stores, pan kiosks, tea shops/stalls and street vendors, which are the most common types of tobacco retailers. Participants responded on a four-level Likert scale from ‘Never’ to ‘More than once a week’ for each retailer type. Using a single-factor generalised partial credit model (GPCM),20 an item response theory model, responses were used to create a latent retailer visit frequency index measuring the frequency of visits to tobacco retailers. The GPCM was used to estimate each participant’s probability of endorsing each response about how often they visit each tobacco retailer type. We averaged the probability across the four types of retailers. The GPCM comparative fit index values for the Mumbai (0.954) and Kolkata (0.946) samples were >0.90 indicating good fit.21

    Mapped neighbourhood tobacco retailer density

    To measure mapped neighbourhood tobacco retailer density, we used procedures adapted from ones used in our past research.9 In 2019, field investigators walked along all roads within each sampled neighbourhood (n=52) to collect Geographic Information System (GIS) data about the location of all tobacco retailers using the Environmental Systems Research Institute’s ArcGIS Collector software. Neighbourhoods were defined as the sampled NSSO designated investigator unit areas which were expanded to an upper limit of 1 km2 by creating boundaries of each neighbourhood drawn along natural (waterways, forest, parks, hills) and built (major roads, highway railroads, land use) environmental features. Tobacco retailers were defined as any shop or vendor selling any form of tobacco within the neighbourhood boundaries. Mapped neighbourhood tobacco retailer density was measured as a count of all tobacco retailers within each neighbourhood boundary. The neighbourhoods were then ranked in tertiles of tobacco retailer density. We used spatial joins in ArcGIS Pro V.2.8.0 to count the number of tobacco retailers within neighbourhoods.

    Perceived tobacco retailer density

    We measured perceived tobacco retailer density near home and school with answers to, ‘How many stores and shops that sell tobacco are there within walking distance from your home?’ and ‘…from your school?’ (three-level Likert response scale from ‘None’ to ‘A lot’).

    Covariates

    We included adolescent age (13–15), gender (male/female) and religion (Hindu, Muslim, Other) as covariates. We did not include household socioeconomic status as a covariate because the inclusion of parental education and household income did not improve model fit for any outcome. We did not include neighbourhood socioeconomic status and population size because they were accounted for in the sampling design and statistical analysis.

    Statistical analysis

    Statistical analysis was performed in R V.4.0.5 using Tidyverse V.1.3.1 for data processing and lme4 V.1.1 for multilevel logistic regression analysis. The source code is available at https://githubcom/ATP-India/access. All analyses accounted for the complex survey design and weights,31 which were adjusted for sampling, non-response and poststratification. Factors in constructing survey weights included block selection, block refusal, household selection, household refusal, screening response, eligibility, interview response and child selection. Probability weights were trimmed for outliers21 and normalised to the sample size. Missing values (for 160 participants in Mumbai and 60 in Kolkata) were imputed using multivariate imputations by chained equations using a fixed seed for reproducibility.32 All analyses were performed separately for Mumbai and Kolkata.

    First, unweighted and weighted frequencies and proportions were calculated for study measures. Second, Pearson’s χ2 test with the Rao and Scott adjustment for complex survey design was used to test for city differences (alpha=0.05). Additionally, summary statistics were produced from neighbourhood-level tobacco retailer counts and distributions by retailer type. Third, generalised linear models, accounting for survey design, were used to estimate the associations between tobacco retail access exposure variables and each cognitive risk factor outcome while controlling for covariates. ORs were considered statistically significant if corresponding 95% CIs excluded one.

    Results

    The weighted adolescent sample (table 1) had a modal age of 14 years in both cities. Hinduism was the most common religion, followed by Islam. About 13% of adolescents in Mumbai and 47% in Kolkata reported very easy access to tobacco. About 53% reported some or more peer tobacco use in Mumbai and 36% in Kolkata. Intention to use tobacco was relatively low at 16% in Mumbai and 13% in Kolkata.

    View this table:
    Table 1

    Adolescent characteristics (n=1759: Mumbai=843, Kolkata=916)

    Overall, there was substantial neighbourhood tobacco retail access. In Mumbai and Kolkata, respectively, about 27% and 19% of adolescents reported ‘A lot’ of tobacco retailers near home (table 1). There were on average 99 (range: 14–374) mapped tobacco retailers within each sampled neighbourhood in Mumbai and 95 (range 55–203) in Kolkata (table 2). In Mumbai, the most common tobacco retailer type was general stores, followed by tobacco street vendors and pan/bidi kiosks. In Kolkata, pan/bidi kiosks were the most common type, followed by general stores and tea shops/stalls. There were about three times as many tobacco retailers in the lowest socioeconomic status neighbourhoods as compared with the highest (p<0.001). The frequency of tobacco retailer visits was high (table 3). A substantial proportion of adolescents reported weekly or more visits to general stores (Mumbai 75%, Kolkata 60%), pan/bidi kiosks (Mumbai 12%, Kolkata 23%), tea stalls (Mumbai 12%, Kolkata 27%) and street vendor (Mumbai 32%, Kolkata 19%). The tobacco retailer visit frequency index was higher in Mumbai than Kolkata (p<0.001).

    View this table:
    Table 2

    Neighbourhood tobacco retailer density (n=52: Mumbai=26, Kolkata=26)

    View this table:
    Table 3

    Frequency of adolescent visits to retailers (n=1759: Mumbai=843, Kolkata=916)

    We used multilevel logistic regression adjusted for study covariates to estimate associations between measures of tobacco retail access and study outcomes (table 4). The frequency of tobacco retailer visits was positively associated with each cognitive risk factor of tobacco use in Mumbai. In Kolkata, frequency of tobacco retailer visits was only associated with perceived norms about tobacco use but was not associated with other study outcomes. Mapped tobacco retailer density was positively associated with perceived ease of access only in Kolkata when comparing the lowest density with the middle density category, but not the highest category. Perceived tobacco retailer density near home and near school were positively associated with perceived norms about peer tobacco use in both cities. Perceived retailer density near school was positively associated with intention to use tobacco in Mumbai when comparing ‘None’ to ‘A lot’ of retailers. There were some unexpected findings in Kolkata; for example, higher frequency of visits to tobacco retailers and perceived retailer density were negatively associated with perceived ease of access.

    View this table:
    Table 4

    Associations between neighbourhood tobacco retail access measures and cognitive tobacco use risk factors, controlling for sociodemographic measures (n=1759: Mumbai=843, Kolkata=916)

    Discussion

    Overall, we found that young adolescents had high neighbourhood tobacco retail access which was associated with cognitive risk factors for tobacco use. The results suggest that the frequency of visits to tobacco retailers and perceptions about neighbourhood tobacco retailer density of tobacco may be predictive of cognitive tobacco use risk. There was a positive association between mapped tobacco retailer density and ease of access to tobacco in the Kolkata sample, but in both cities, there were no associations with perceived peer tobacco use norms and intention to use tobacco.

    We found very high exposure to neighbourhood tobacco retailers and substantial variations by neighbourhood socioeconomic status. Lower socioeconomic status neighbourhoods had disproportionately higher tobacco retailer density. This socioeconomic disparity in tobacco retailer density is consistent with prior research in India9 33 and tracks with a higher prevalence of tobacco use in lower socioeconomic status groups.34 Because upper income individuals are more likely to use cigarettes while lower income individuals are more likely to use bidis and smokeless tobacco, the socioeconomic patterning of tobacco use is likely to influence the types of tobacco products available in high compared with low socioeconomic status neighbourhoods. Most adolescents reported weekly or more visits to tobacco retailers. Visits to retail environments may influence adolescent tobacco use risk because adolescents may be exposed to tobacco advertisements, and to people selling, buying and using tobacco.25 35 36 These social influences may increase risk of tobacco use, for example, through cognitive processes such as perceptions about tobacco access, tobacco use norms and intention to use.

    The regression results involving Mumbai data were in the expected direction. In Kolkata, however, perceived tobacco retailer density and frequency of visits to tobacco retailers were negatively associated with perceived ease of access to tobacco, the opposite of what we expected. These results occurred even though mapped retailer density in Kolkata was positively associated with ease of access to tobacco, as expected. The counterintuitive results in Kolkata may reflect unmeasured differences between the two cities such as in the implementation of and compliance with tobacco control policies, as well as sociocultural contexts about tobacco use. For example, we found that peer tobacco use norms and intention were lower in Kolkata adolescents while perceived ease of access to tobacco was much higher relative to Mumbai adolescents, even though Kolkata adolescents visited tobacco retailers less frequently. Because adolescents in Kolkata were more likely to have parents who use tobacco (27% Mumbai vs 46% Kolkata fathers used tobacco) and were more likely to report that tobacco use was allowed in their homes (13% Mumbai, 47% Kolkata), Kolkata adolescents may have greater access to tobacco at home than Mumbai adolescents, independent of influences from neighbourhood tobacco retailers.

    The study results have policy implications. Despite laws regulating retail sale of tobacco in India, there remains substantial exposure to tobacco retail environments in adolescents.37 The governments of India should institute tobacco retail licensing38 39 as has been initiated in states like Himachal Pradesh40 and Jharkhand.41 Through tobacco retail licensing, governments could further regulate the retail sale of tobacco and improve enforcement of the current bans on the sale of tobacco products near schools and tobacco advertisements at retail locations.39 Regulators could require tobacco retail environments to communicate the dangers of tobacco. Young people who visit retailers frequently would then be repeatedly exposed to anti-tobacco messages. Schools and families could address adolescent behaviours through prevention efforts to limit exposure to tobacco at neighbourhood retailers.

    Regulations could be enacted to decrease tobacco retailer density.42 43 For example, lower tobacco retailer density has been found to be associated with lower risk of experimental and current tobacco use5 21 22 and may reduce perceived tobacco retail access which in our study was an important factor influencing perceived peer tobacco use norms in both cities, and ease of access to tobacco in Mumbai. Implementation of tobacco retail licensing laws can support monitoring of tobacco retailers to increase compliance with sale and marketing regulations.44 In addition, provisions could also limit the number of tobacco retail licences made available within an area. Licensing and zoning laws can reduce density, but implementation may increase disparities in tobacco access.40 45 Although the Government of India banned the sale and marketing of tobacco products within 100 yards of educational institutions, in our sample, more than 70% of adolescents in Mumbai and about 90% in Kolkata reported ‘a few’ or more tobacco retailers within walking distance from school. Enforcement of the ban is the responsibility of local law enforcement and school administrators.46 Though guidelines are provided about how to implement and enforce the ban,46 there continues to be low compliance with this law.16 Regulators must identify strategies to better enforce the current ban.

    The results have several implications for research. First, the unexpected opposite effects of frequency of visits to tobacco retailers and tobacco retailer density on perceived ease of tobacco access in Kolkata illustrate the importance of analysing contextual factors, including city-level differences in tobacco use norms, neighbourhood-level differences in tobacco access and family-level differences in tobacco use and rules about tobacco. Second, the differences in results between mapped and perceived tobacco retailer density require further research, particularly regarding measurement and intervention. Perceptions about retailer density may be influenced by many factors including what individuals consider to be ‘some’ or ‘a lot’ of tobacco retailers. The seeming ubiquity of tobacco retailers may be perceived as the norm or the average and captured with response options like ‘some’ instead of ‘a lot’. Researchers must be careful about how questions about perceived retail density are asked and understood by participants. In addition, further research is required to understand how discrepancies between perceived and actual tobacco retailer density may influence tobacco use risk. If discrepancies are influential or if tobacco use risk is more influenced by subjective than objective measures of tobacco retailer density (as was found in the current study), interventions may be needed to address perceptions about tobacco retailer density. Third, our findings help understand pathways that may explain links between neighbourhood tobacco retail access and adolescent tobacco use.5 22 24 For example, in the current study, neighbourhood retail access appeared to be associated with cognitive tobacco use risk factors, which may function as mediators between access and future tobacco use experimentation and initiation. Longitudinal research is necessary to understand the pathways between tobacco retail access, tobacco use-related cognitive effects and adolescent tobacco use.

    There are several strengths to this study. First, we objectively and subjectively measured tobacco retailer density using field GIS data collection and adolescent reports. Though some studies used GIS-based data to measure tobacco retailer density,9 47 48 most relied on secondary data and did not include subjective measures. Second, our study helps understand tobacco use within the Indian contexts. More research is needed in India because the population has a disproportionate number of tobacco users. This study provides new data on important tobacco control policy issues relevant to India.

    Several factors reduce the validity of the study. Although the data were population-based, the adolescent study data were collected cross-sectionally, thus limiting conclusions regarding the causality of associations between perceived retailer density, frequency of visits to tobacco retailers and tobacco use-related cognitive risk factors. For example, adolescent cognitive risks for tobacco use may influence perceptions about tobacco retailer density and frequency of visits to tobacco retailers rather than vice versa. Second, we did not measure compliance with tobacco retail laws, which may moderate the adverse effects of tobacco retailer density and frequency of visits to tobacco retailers. Third, we did not examine tobacco use per se because too few of these young adolescents reported ever or current tobacco use; instead, we focused on cognitive risk factors for tobacco use.

    Conclusions

    Adolescents had extremely high neighbourhood tobacco retail access. Tobacco retailer density was higher in lower socioeconomic status neighbourhoods. The frequency of visits to tobacco retailers and perceived tobacco retailer density increased cognitive risk factors for tobacco use. However, the objective mapped measure of tobacco retailer density was only associated with ease of access to tobacco in Kolkata but was not associated with other cognitive risk factors for tobacco use. Interventions that reduce neighbourhood access to tobacco retailers among adolescents in India may reduce their cognitive risk for tobacco use.

    What this paper adds

    • There was high tobacco retail access among young adolescents in Mumbai and Kolkata.

    • The density of tobacco retailers was higher in lower income neighbourhoods.

    • Adolescent frequency of tobacco retailer visits was positively associated with tobacco use susceptibility measured as perceived ease of access to tobacco products, perceived norms of peer tobacco use and intention to use tobacco in the future.

    • Mapped tobacco retailer density was associated with perceived ease of access in Kolkata but not in Mumbai.

    • Perceived tobacco retailer density was associated with perceived ease of access and perceived peer use, but not with intention to use.

    Data availability statement

    Deidentified data may be available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    The study was approved by the University of Michigan, the University of California, Los Angeles, and Healis Sekhsaria Institute for Public Health institutional human research ethics committees.

    References

    1. ↵
      2. Sinha DN ,
      3. Palipudi KM ,
      4. Gupta PC , et al
      . Smokeless tobacco use: a meta-analysis of risk and attributable mortality estimates for India. Indian J Cancer 2014;51 Suppl 1:S73–7.doi:10.4103/0019-509X.147477 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25526253
    2. ↵
      2. Jha P
      . Avoidable global cancer deaths and total deaths from smoking. Nat Rev Cancer 2009;9:655–64.doi:10.1038/nrc2703 pmid:http://www.ncbi.nlm.nih.gov/pubmed/19693096
    3. ↵
      2. Mathur P ,
      3. Kulothungan V ,
      4. Leburu S , et al
      . National noncommunicable disease monitoring survey (NNMS) in India: estimating risk factor prevalence in adult population. PLoS One 2021;16:e0246712.doi:10.1371/journal.pone.0246712 pmid:http://www.ncbi.nlm.nih.gov/pubmed/33651825
    4. ↵
      2. India G
      . Global youth tobacco survey (GYTS) fact sheet India. 2009, 2019.
    5. ↵
      2. Altman DG ,
      3. Wheelis AY ,
      4. McFarlane M , et al
      . The relationship between tobacco access and use among adolescents: a four community study. Soc Sci Med 1999;48:759–75.doi:10.1016/S0277-9536(98)00332-3 pmid:http://www.ncbi.nlm.nih.gov/pubmed/10190639
    6. ↵
      2. Chadda R ,
      3. Sengupta S
      . Tobacco use by Indian adolescents. Tob Induc Dis 2002;1:1–9.doi:10.1186%2F1617-9625-1-8
    7. ↵
      2. Cummings KM ,
      3. Hyland A ,
      4. Perla J , et al
      . Is the prevalence of youth smoking affected by efforts to increase retailer compliance with a minors' access law? Nicotine Tob Res 2003;5:465–71.pmid:http://www.ncbi.nlm.nih.gov/pubmed/12959784
    8. ↵
      2. COTPA MNH
      . And rules made THEREUNDER. 2021. New Delhi: Ministry of Health and Family Welfare, 2003.
    9. ↵
      2. Mistry R ,
      3. Pednekar M ,
      4. Pimple S , et al
      . Banning tobacco sales and advertisements near educational institutions may reduce students' tobacco use risk: evidence from Mumbai, India. Tob Control 2015;24:e100–7.doi:10.1136/tobaccocontrol-2012-050819 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23958643
    10. ↵
      2. Flay BR ,
      3. Petraitis J ,
      4. Hu FB
      . Psychosocial risk and protective factors for adolescent tobacco use. Nicotine Tob Res 1999;1 Suppl 1:S59–65.doi:10.1080/14622299050011611 pmid:http://www.ncbi.nlm.nih.gov/pubmed/11072406
    11. ↵
      2. Leatherdale ST ,
      3. Strath JM
      . Tobacco retailer density surrounding schools and cigarette access behaviors among underage smoking students. Ann Behav Med 2007;33:105–11.doi:10.1207/s15324796abm3301_12 pmid:http://www.ncbi.nlm.nih.gov/pubmed/17291176
    12. ↵
      2. Bold KW ,
      3. Buta E ,
      4. Simon P , et al
      . Using latent class analysis to examine susceptibility to various tobacco products among adolescents. Nicotine Tob Res 2020;22:2059–65.doi:10.1093/ntr/ntz216 pmid:http://www.ncbi.nlm.nih.gov/pubmed/31758190
    13. ↵
      2. Chatterjee N ,
      3. Todankar P ,
      4. Mandal G , et al
      . Factors associated with tobacco use in students attending local government schools in Mumbai, India. Asian Pac J Cancer Prev 2016;17:5075.doi:10.22034/APJCP.2016.17.12.5075 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28122437
    14. ↵
      2. Goel S ,
      3. Kumar R ,
      4. Lal P , et al
      . How compliant are tobacco vendors to India’s tobacco control legislation on ban of advertisments at point of sale? A three jurisdictions review. APJCP 2015;15:10637–42.doi:10.7314/APJCP.2014.15.24.10637
    15. ↵
      2. Mistry R ,
      3. Pednekar MS ,
      4. McCarthy WJ , et al
      . Compliance with point-of-sale tobacco control policies and student tobacco use in Mumbai, India. Tob Control 2019;28:220–6.doi:10.1136/tobaccocontrol-2018-054290 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29743339
    16. ↵
      2. Khargekar NC ,
      3. Debnath A ,
      4. Khargekar NR , et al
      . Compliance of cigarettes and other tobacco products act among tobacco vendors, educational institutions, and public places in Bengaluru City. Indian Journal of Medical and Paediatric Oncology 2018;39:463–6.doi:10.4103/ijmpo.ijmpo_136_17
    17. ↵
      2. Turner MM ,
      3. Rimal RN ,
      4. Lumby E , et al
      . Compliance with tobacco control policies in India: an examination of facilitators and barriers. Int J Tuberc Lung Dis 2016;20:411–6.doi:10.5588/ijtld.15.0376 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27046725
    18. ↵
      2. Jain ML ,
      3. Chauhan M ,
      4. Singh R
      . Compliance assessment of cigarette and other tobacco products act in public places of Alwar district of Rajasthan. Indian J Public Health 2016;60:107.doi:10.4103/0019-557X.184540 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27350703
    19. ↵
      2. Pimple S ,
      3. Gunjal S ,
      4. Mishra GA , et al
      . Compliance to Gutka ban and other provisons of COTPA in Mumbai. Indian J Cancer 2014;51 Suppl 1:60.doi:10.4103/0019-509X.147475 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25526251
    20. ↵
      2. Priyanka R ,
      3. Rao A ,
      4. Rajesh G
      . Compliance with statutory tobacco guidelines in and around schools in Mangaluru, India: an observational study. World J Dent 2017;8:10–13.
    21. ↵
      2. Henriksen L ,
      3. Feighery EC ,
      4. Schleicher NC , et al
      . Is adolescent smoking related to the density and proximity of tobacco outlets and retail cigarette advertising near schools? Prev Med 2008;47:210–4.doi:10.1016/j.ypmed.2008.04.008 pmid:http://www.ncbi.nlm.nih.gov/pubmed/18544462
    22. ↵
      2. Henriksen L ,
      3. Schleicher NC ,
      4. Feighery EC , et al
      . A longitudinal study of exposure to retail cigarette advertising and smoking initiation. Pediatrics 2010;126:232–8.doi:10.1542/peds.2009-3021 pmid:http://www.ncbi.nlm.nih.gov/pubmed/20643725
    23. ↵
      2. Chan WC ,
      3. Leatherdale ST
      . Tobacco retailer density surrounding schools and youth smoking behaviour: a multi-level analysis. Tob Induc Dis 2011;9:9–7.doi:10.1186/1617-9625-9-9
    24. ↵
      2. McCarthy WJ ,
      3. Mistry R ,
      4. Lu Y , et al
      . Density of tobacco retailers near schools: effects on tobacco use among students. Am J Public Health 2009;99:2006–13.doi:10.2105/AJPH.2008.145128 pmid:http://www.ncbi.nlm.nih.gov/pubmed/19820214
    25. ↵
      2. Sanders-Jackson A ,
      3. Parikh NM ,
      4. Schleicher NC , et al
      . Convenience store visits by US adolescents: rationale for healthier retail environments. Health Place 2015;34:63–6.doi:10.1016/j.healthplace.2015.03.011 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25955537
    26. ↵
      2. Chattopadhyay S
      . Introduction: the historical legacy of suburbs in South Asia. Urban History 2012;39:51–5.doi:10.1017/S0963926811000770
    27. ↵
      1. United Nations DoEaSA,
      . Population Divi- sion. The World’s Cities in 2018—Data Booklet. United Nations, 2018.
    28. ↵
      1. NSSO
      . Urban frame survey. Ministry of Statistics and Programme Implementation, 2017.
    29. ↵
      2. Mistry R ,
      3. Pednekar MS ,
      4. Gupta PC , et al
      . Longitudinal study of adolescent tobacco use and tobacco control policies in India. BMC Public Health 2018;18:1–8.doi:10.1186/s12889-018-5727-8 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29970049
    30. ↵
      2. Patel M ,
      3. Wj M ,
      4. Mistry R
      . Tobacco retailer proximity to schools & teen smoking: A geospatial analysis. Tobacco Related Diseases Research Program Biennial Meeting. Sacramento, CA 2007.
    31. ↵
      2. Lumley T
      . Complex surveys: a guide to analysis using R. John Wiley & Sons, 2011.
    32. ↵
      2. Buuren Svan ,
      3. Groothuis-Oudshoorn K
      . mice : Multivariate Imputation by Chained Equations in R. J Stat Softw 2011;45:1–67.doi:10.18637/jss.v045.i03
    33. ↵
      2. Schensul JJ ,
      3. Nair S ,
      4. Bilgi S , et al
      . Availability, accessibility and promotion of smokeless tobacco in a low-income area of Mumbai. Tob Control 2013;22:324–30.doi:10.1136/tobaccocontrol-2011-050148 pmid:http://www.ncbi.nlm.nih.gov/pubmed/22387521
    34. ↵
      2. Palipudi K ,
      3. Gupta P ,
      4. Sinha D
      . Tim McAfee, on behalf of the GATS Collaborative Group. social determinants of health and tobacco use in thirteen low and middle income countries: evidence from global adult tobacco survey. PLoS one 2012;7:e33466.
    35. ↵
      2. Henriksen L ,
      3. Feighery EC ,
      4. Wang Y , et al
      . Association of retail tobacco marketing with adolescent smoking. Am J Public Health 2004;94:2081–3.doi:10.2105/AJPH.94.12.2081 pmid:http://www.ncbi.nlm.nih.gov/pubmed/15569957
    36. ↵
      2. Pasch KE ,
      3. Nicksic NE ,
      4. Opara SC , et al
      . Recall of point-of-sale marketing predicts cigar and e-cigarette use among Texas youth. Nicotine Tob Res 2018;20:962–9.doi:10.1093/ntr/ntx237 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29069425
    37. ↵
      2. Khariwala SS ,
      3. Garg A ,
      4. Stepanov I , et al
      . Point-of-Sale tobacco advertising remains prominent in Mumbai, India. Tob Regul Sci 2016;2:230–8.doi:10.18001/TRS.2.3.3 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28217721
    38. ↵
      2. Pouranik NS ,
      3. Saraf S ,
      4. Wright K , et al
      . Tobacco retailer density and tobacco retailers near schools in two cities of East India, Ranchi and Siliguri. Indian J Tuberc 2021;68S:S14–22.doi:10.1016/j.ijtb.2021.07.003 pmid:http://www.ncbi.nlm.nih.gov/pubmed/34538386
    39. ↵
      2. Patil AR
      . Tobacco control law in India – origins and proposed reforms. 48. Bangalore, India, 2020.
    40. ↵
      2. Chauhan G
      . Licensing tobacco vendors in the state of Himachal Pradesh in India - challenges, opportunities and the way forward to implement the new legislation. Tob Induc Dis 2018;16.doi:10.18332/tid/84278
    41. ↵
      2. Ray D
      . Jharkhand government makes separate trade licence for tobacco a must. The Times of India. Mumbai, India: Bennett, Coleman & Co. Ltd 2020.
    42. ↵
      2. Pearson AL ,
      3. van der Deen FS ,
      4. Wilson N , et al
      . Theoretical impacts of a range of major tobacco retail outlet reduction interventions: modelling results in a country with a smoke-free nation goal. Tob Control 2015;24:e32–8.doi:10.1136/tobaccocontrol-2013-051362 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25037156
    43. ↵
      2. Dai H ,
      3. Hao J
      . The effects of tobacco control policies on retailer sales to minors in the USA, 2015. Tob Control 2018;27:258–60.doi:10.1136/tobaccocontrol-2016-053408 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28219976
    44. ↵
      2. Astor RL ,
      3. Urman R ,
      4. Barrington-Trimis JL , et al
      . Tobacco retail licensing and youth product use. Pediatrics 2019;143. doi:doi:10.1542/peds.2017-3536. [Epub ahead of print: 07 01 2019].pmid:http://www.ncbi.nlm.nih.gov/pubmed/30617237
    45. ↵
      2. Craigmile PF ,
      3. Onnen N ,
      4. Schwartz E , et al
      . Evaluating how licensing-law strategies will impact disparities in tobacco retailer density: a simulation in Ohio. Tob Control 2021;30:e96–103.doi:10.1136/tobaccocontrol-2020-055622 pmid:http://www.ncbi.nlm.nih.gov/pubmed/32826386
    46. ↵
      2. Mission NH
      . National tobacco control program: Guidlines/Manuals. New Delhi: Ministry of Health and Family Welfare, 2021.
    47. ↵
      2. Marsh L ,
      3. Vaneckova P ,
      4. Robertson L , et al
      . Association between density and proximity of tobacco retail outlets with smoking: a systematic review of youth studies. Health Place 2021;67:102275.doi:10.1016/j.healthplace.2019.102275 pmid:http://www.ncbi.nlm.nih.gov/pubmed/33526204
    48. ↵
      2. Valiente R ,
      3. Sureda X ,
      4. Bilal U , et al
      . Regulating the local availability of tobacco retailing in Madrid, Spain: a GIS study to evaluate compliance. Tob Control 2019;28:325–33.doi:10.1136/tobaccocontrol-2018-054269 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29991639

    Footnotes

    • Contributors RM conceptualised the paper, guided the spatial and statistical analysis and wrote the paper. MJK conducted the statistical analysis and drafted the data analysis section. TR provided guidance on the sampling, weighting and statistical analysis plan. JA conducted the spatial analysis. MSP oversaw research activities in India. MJK, NP, PCG, WJM, TR, H-FH, SN, MD, SA, JA and MSP critically reviewed the drafts of the paper. RM is responsible for the overall content as guarantor.

    • Funding The National Cancer Institute of the National Institutes of Health under award number R01CA201415 (Multiple PIs: RM and MSP) supported the research reported in this paper.

    • Disclaimer The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

    • Competing interests None declared.

    • Provenance and peer review Not commissioned; externally peer reviewed.