Introduction Evidence suggests inconsistent findings on the relationship between density of tobacco outlets around schools and risk of smoking among students. This study examines the density of tobacco outlets around secondary schools in New Zealand (NZ) and current smoking, experimental smoking, susceptibility to smoking, and attempted and successful tobacco purchasing.
Methods Smoking data came from the 2012 ASH Year 10 survey, a national survey of youth smoking in NZ. Geographic Information Systems were used to map tobacco retail outlets; a layer of secondary school locations was obtained from Koordinates.com. Logistic regression examined the relationship between density of outlets around schools and smoking behaviours, adjusting for individual-level and school-level confounders.
Results Of the 27 238 students surveyed, 3.5% (952) were current smokers, 4.1% (n=1 128) were experimental smokers, and 39.8% (10 454) of nonsmokers were susceptible to smoking. An inverse relationship was found between the density of tobacco retail outlets and current smoking. Current smokers were significantly more likely to attempt to purchase tobacco if the density of tobacco retail outlets around their school was high. Non-smoking students were more likely to be susceptible to smoking if the density of tobacco outlets around their school was high. There was no statistically significant association between density of tobacco outlets and successful purchasing, nor experimental smoking.
Conclusions Restricting the permitted density of tobacco retail outlets around schools should be part of comprehensive tobacco control. In this regard, both smokers and non-smokers support the introduction of increased regulation of the tobacco retail environment to achieve our national smoke-free 2025 goal.
- Advertising and Promotion
- Priority/special populations
- Public policy
Statistics from Altmetric.com
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.
The density of tobacco retail outlets is higher in areas where a larger proportion of the population are younger than 18 years, at a time when the risks of initiation of tobacco use and transitions to daily use are greatest.1 However, research has produced inconsistent findings on the relationship between density of outlets around a school and risk of smoking among students. More tobacco retail outlets around a school has been found to be associated with higher levels of experimental smoking,2 ,3 current smoking,4 ,5 susceptibility to future smoking6 and ease of purchasing cigarettes among students.7 Other studies have reported no association between retail density and current smoking.2 ,3 ,8 Newly published research showed that high retail outlet density around the home was associated with increased odds of ever smoking and current smoking, while high density around the school was associated with lower odds of smoking among students.9
There are various pathways through which the density of tobacco retail outlets may be associated with smoking behaviours. Greater outlet density may lead to more opportunities to purchase tobacco products,7 and may encourage greater competition between retailers, possibly driving cigarette prices down10 and increasing sales to minors.7 Greater density of outlets may support normalisation of smoking in a community,2 and may increase environmental cues to smoke, whether through tobacco advertising,4 or the mere presence of an outlet.11
These mixed findings may reflect differences among countries in determinants of youth smoking and the marketing environment for tobacco products. New Zealand (NZ) has strong restrictions on retail tobacco marketing, unlike countries where previous research has been undertaken, for example, the USA. Nevertheless, tobacco retailers are widespread in NZ,12 retail outlets are more densely located in areas of higher deprivation, and one-half of NZ secondary schools have at least one retailer within a 500 m walk.12 Although social sources remain the main form of tobacco access, 10% of individuals aged 14–15 years in NZ purchase their own tobacco from a shop.13
Youth smoking rates have been declining for some time, but to meet the NZ Government's goal of a smoke-free nation by 2025, further effective strategies are needed. Consequently, we wished to gather data to determine if restricting tobacco sales around schools is warranted in NZ. This study investigates whether the density of tobacco outlets around secondary schools in NZ (500 and 1000 m) is associated with current smoking, experimental smoking, susceptibility to future smoking among non-smokers, and attempted and successful tobacco purchasing.
Smoking data came from the 2012 ASH Year 10 (aged 14 and 15 years) survey, a national cross-sectional annual survey of youth smoking undertaken by Action on Smoking and Health NZ and funded by the NZ Ministry of Health. All 621 secondary schools with a Year 10 class were invited and 298 schools took part in the survey (48% response rate); 59 627 Year 10 students were enrolled in 2012 and 28 443 participated (48% response rate). Clustering was controlled for at the school level.
Information about tobacco retail density came from previously published research.12 This involved the collation of known tobacco outlets in 2012 from each of the District Health Boards, who enforce the smoke-free legislation in NZ. The database included 5059 small convenience stores, supermarkets, service stations, licensed premises and a range of other small retailers. A Geographic Information System layer of secondary school locations was obtained from the geographic data website Koordinates.com (accessed 4 June 2012). The position of each school was confirmed by visual inspection and was relocated if necessary. The points from which the walking zones were derived were centrally located on campuses.
Current tobacco smoking: Tobacco smoking was measured using two questions: How often do you smoke now? and How many cigarettes have you smoked in your entire life? Current smokers were defined as those who reported smoking in the past 30 days and smoked more than 100 cigarettes in their lifetime. Experimental smokers were those who had smoked in the past 30 days, but had not smoked 100 cigarettes in their lifetime. All other students were defined as non-smokers.
Susceptibility to future smoking: Non-smokers were asked about their future intentions to smoke using three questions previously validated to examine susceptibility to tobacco smoking:14 If one of your best friends offered you a cigarette, would you smoke it?; At any time during the next year do you think you will smoke a cigarette?; and Do you think you will try a cigarette soon? Respondents were classified as not susceptible if they answered definitely not to each question; all others were classified as susceptible.
Purchasing tobacco: All participants were asked to report if they had attempted to buy cigarettes or tobacco in a shop in the past month (yes, no). Current and experimental smokers were also asked how they usually got their cigarettes in the past month, with one option being I bought them from a shop.
Density of tobacco retail outlets: Polygons of 500 and 1000 m around secondary schools and based on the road network were used to define two boundary areas of tobacco retailer density. Distances of 500 and 1000 m away from the schools were selected as the most likely distances pupils would be willing to walk to reach a retailer.12 The median number of retailers within each of these polygons was calculated and three levels of density identified. All polygons across the country were classified as either above or below the median: zero density (no tobacco retailers), medium density (≤median) and high density (>median).
Student-level variables: Respondents provided demographic information including sex, age and ethnicity.
Family and peer-level variables: The smoking status of family members and peers was determined through the question: Which of the following people smoke? Participants were considered to have a family member as a smoker if they responded that any of the following people smoked: mother, father, other caregiver, older brother or sister. Peers were considered smokers if the student reported that their best friend or other close friends smoked.
School and neighbourhood demographics: School decile (a measure of socioeconomic status of the neighbourhoods from which the school draws its students; Ministry of Education, 28 November 2012) and school location (urban or rural) was obtained from the Census data collated by Statistics New Zealand (15 May 2012).
We assessed individual, family, peer, school and neighbourhood variables as possible confounding factors as they have been shown to influence smoking behaviour among NZ youth.15
Only questionnaires with completed answers on age, gender, ethnicity and smoking status were included in the data set for analysis. Logistic regression was used to examine the influence of the density of tobacco retailers on five outcome variables (current smoking, experimental smoking, susceptibility to smoking, attempted and successful tobacco purchase), as well as the influence of confounders (student-level, family and peer-level, and school-level variables). These three models were fitted for each of the two polygon areas around the schools: 500 and 1000 m.
A total of 27 238 students completed the survey and were included in the analyses. Of these, 3.5% (n=952) were current smokers, 4.1% (n=1128) were experimental smokers, and 39.8% (n=10 454) of non-smokers were susceptible to smoking. Of the 298 schools, 67 did not have a retailer within 1000 m of their school, 140 schools had at least one retailer within 500 m; 231 schools had at least one retailer within 1000 m. The median density of tobacco outlets around schools was 2 within 1000 m of the school, and 1 within 500 m of the school.
Having any tobacco retail outlets within 500 or 1000 m of a school was inversely associated with odds of current smoking among students, compared with having zero density within 500 or 1000 m, after adjusting for individual-level and school-level variables (table 1). There was no statistically significant relationship between retailer density and experimental smoking.
Non-smoking students were significantly more likely to be susceptible to smoking if the density of tobacco outlets around their school was high, compared with non-smoking students at schools with low outlet density, after adjusting for individual-level and school-level variables. This relationship held for retailer density within 500 and 1000 m.
Current smokers were more likely to report attempts to purchase tobacco where their schools had a high density of tobacco retail outlets within 1000 m, compared with students at schools with low outlet density, even after adjusting for individual-level and school-level variables. There was no association between density and successful purchasing in the fully adjusted analyses.
Consistent with previous literature,2 ,6–8 our study suggests that students attending schools surrounded by a higher density of tobacco outlets have greater odds of smoking susceptibility, and are more likely to try and purchase tobacco, compared with students at schools with zero density of outlets. Greater retailer density surrounding schools may encourage curiosity and temptation in relation to tobacco, may normalise smoking in the community as an acceptable and normal behaviour, and provide opportunities and triggers for youth to attempt to purchase tobacco. Restrictions on tobacco marketing in NZ have reduced most forms of tobacco promotion, but the large number of retail outlets represents a major form of promotion16 and undermines attempts to denormalise tobacco. Internationally, jurisdictions have implemented zoning restrictions limiting the distance that tobacco retailers can be located to a school.17 In NZ, there is strong support for a reduction of tobacco outlets around schools,18 and recent modelling has demonstrated that eliminating tobacco outlets within 2 km of a school would lead to a reduced smoking prevalence.19 Policymakers should consider restricting the density of tobacco retail outlets around schools to reduce smoking initiation and opportunities to purchase tobacco.
Retailer density was not related to experimental smoking and was inversely related to current smoking. Two studies have shown a relationship between density of outlets and current smoking, which was not corroborated by this current study.4 ,5 Inconsistent findings between these studies may be due to differences in zones around schools used, for example, polygons versus circles, and definitions of current smoking.4 ,5 There were also differences in how the density of outlets was calculated6 and the distances examined.7 However, our finding of an inverse relationship between current smoking and retailer density is consistent with a recent UK study9 and there are plausible explanations for this. First, established youth smokers often use multiple sources of tobacco,15 and density of outlets around the school may not be important for established smokers to obtain their tobacco, which could explain why successful purchasing of tobacco is not related to density around the school. Second, the finding may also be explained by increased enforcement in areas with high density of retailers around schools, resulting in higher compliance among retailers.
This is the first study in NZ to examine the density of tobacco outlets around schools in relation to smoking behaviours among students. Study strengths include a large random sample, a response rate of 48%, and the only known database of NZ tobacco retailers. We were able to examine a range of behavioural outcome measures, which gives additional insight into the nature of the risk posed by higher density of retail outlets. This study is limited by its cross-sectional nature, and it is possible the observed associations may have been influenced by residual confounding, though we did control for a range of individual-level and school-level confounders. The smoking data were collected in 2012 before the implementation of the point-of-sale tobacco display ban. Therefore, we cannot determine whether the associations observed were due to retail access to tobacco retailers or exposure to tobacco displays, or both.
This study has found that retailer density is associated with susceptibility to smoking in the future and trying to purchase tobacco, but not successful tobacco purchase, nor experimental or current smoking. NZ enforces illegal sales of tobacco to minors through Controlled Purchase Operations which may have contributed to some of the results in this study. This indicates the importance of enforcing smoke-free legislation to limit tobacco sales to minors. Since this research was conducted, NZ has implemented a tobacco retail display ban, so one of the few remaining triggers for tobacco purchase and normalisation of smoking is the retailer itself. Further regulation of the retail environment for tobacco could potentially reduce smoking prevalence, denormalise tobacco and ‘hasten the endgame’.20
What this paper adds
Current evidence suggests inconsistent findings on the relationship between density of tobacco outlets around a school and increased risk of smoking among students.
This research adds to the evidence base that the high density of tobacco retail outlets around a school is associated with an increase in susceptibility of smoking among non-smoking students, and the likelihood of purchasing their own tobacco among current smokers.
For New Zealand to achieve a smoke-free nation by 2025, the introduction of increased regulation of the tobacco retail environment is required, and reducing the density of tobacco outlets around schools would be one way of achieving this.
Teachers and students from many New Zealand schools gave their valuable time and effort during data collection. ASH appreciates the time and commitment of those secondary schools that participated in the survey. The ASH Year 10 Snapshot Survey forms part of the New Zealand Youth Tobacco Monitor (NZYTM). In 2012, the NZYTM Research Coordinating Group were Dr Rhiannon Newcombe, RM, Anaru Waa, Professor Richard Edwards, Dr Simon Denny. This group provided expert advice and input into the ASH Year 10 Snapshot Survey. Additional input into the survey questions was provided by Professor Janet Hoek and Professor Phil Gendall. The 2012 NZYTM was managed by the Health Promotion Agency (HPA), formerly Health Sponsorship Council (HSC) and particular thanks to Rose Patterson for project management and additional advice and input into the project. Data entry was provided by Canterbury Business Solutions.
Contributors LM, LR and RM designed the research. CD developed the original retailer database. AA analysed the data and CC wrote the results. LM developed the manuscript. All authors have reviewed and approved the final version of the submitted manuscript.
Funding ASH New Zealand receives funding from a combination of Ministry of Health contracts, membership fees and donations from organisations. The Cancer Society Social and Behavioural research unit is supported by the Cancer Society of New Zealand and the University of Otago. LR is supported by a New Zealand Lottery Health Scholarship.
Competing interests None declared.
Ethics approval Ethical approval for the ASH Year 10 Snapshot Survey was granted by the New Zealand Ministry of Health Multiregional Ethics Committee in 2007.
Provenance and peer review Not commissioned; externally peer reviewed.