Article Text

Download PDFPDF

Proximity of IQOS and JUUL points of sale to schools in Israel: a geospatial analysis
  1. Daniel Elbaz1,
  2. Yael Bar Zeev1,
  3. Carla J Berg2,
  4. Lorien C Abroms2,
  5. Hagai Levine1
  1. 1 Braun School of Public Health and Community Medicine, The Hebrew University of Jerusalem and Hadassah Medical Organization, Jerusalem, Israel
  2. 2 Department of Prevention and Community Health, Milken Institute School of Public Health, George Washington University, Washington, District of Columbia, USA
  1. Correspondence to Prof. Hagai Levine, Braun School of Public Health and Community Medicine, The Hebrew University of Jerusalem and Hadassah Medical Organization, Jerusalem 91120, Israel; hlevine{at}hadassah.org.il

Abstract

Significance Exploring proximity and density of heated tobacco product (HTP) and electronic nicotine delivery system (ENDS) points of sale (POS) to schools is critical for understanding youth marketing exposure and informing policy and enforcement to protect youth. This study examined IQOS and JUUL POS (prominent HTPs and ENDS), specifically their proximity to and density around schools in Israel.

Methods Using geospatial analysis and IQOS/JUUL website data, distance matrices were used to calculate distance from each school in Israel (grades 1–12) to the nearest POS and number of POS within 1 km, accounting for schools’ neighbourhood socioeconomic status (SES) ranking.

Results An average of 8.7 IQOS POS and 5.2 JUUL POS were within walking distance (1 km) from schools. Average distances from schools to nearest IQOS and JUUL POS were 954 m (median=365 m) and 1535 m (median=579 m), respectively. The percentages of schools with at least one IQOS or JUUL POS within 1 km were 86% and 74%, respectively. The average numbers of POS within 1 km of schools in low-SES, middle-SES, and high-SES neighbourhoods were 7.5, 9.9, and 7.6 for IQOS and 4.1, 5.9, and 5.5 for JUUL, respectively. Median distances from schools in low-SES, middle-SES, and high-SES neighbourhoods to nearest POS were 428 m, 325 m, and 403 m for IQOS and 1044 m, 483 m, and 525 m for JUUL.

Conclusions Youth experience high environmental exposure to IQOS and JUUL POS, particularly IQOS. POS were more densely located near schools in middle-SES neighbourhoods. Thus, regulating HTP and ENDS POS near schools and in certain neighbourhoods is key to reducing youth population impact in Israel and elsewhere.

  • nicotine
  • surveillance and monitoring
  • socioeconomic status
  • public policy
  • electronic nicotine delivery devices

Data availability statement

Data are available upon reasonable request.

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

As cigarette use has declined over the past decade in many parts of the world, emerging tobacco/nicotine products, including both heated tobacco products (HTPs) and electronic nicotine delivery systems (ENDS), have become prominent globally,1 with their use increasing among both adults and youth.2 3 Youth who use HTPs and ENDS may be exposed to high levels of nicotine, which has implications for addiction and harms to health, particularly the cardiovascular system, respiratory system and brain.4–8

Israel is a particularly noteworthy context for HTPs and ENDS, as it was among the first countries in 2016, where IQOS, the leading HTP globally, entered the market. In addition, JUUL (an ENDS) entered the Israeli market in the beginning of 2018, accompanied by a vast amount of public advertising and numerous points of sale (POS).9 In Israel, IQOS has specialty stores which sell all IQOS products, while IQOS HEETS and JUUL are sold in numerous POS alongside tobacco products, such as convenience stores, stores within gas stations, kiosks and grocery stores.10

Use of HTPs and ENDS is expanding rapidly in Israel. In 2019, the Israeli Ministry of Health began surveillance of youth HTP and ENDS use, revealing that more than 25% of youth had used an HTP or ENDS at least once, greater than the prevalence of youth cigarette smoking. Furthermore, 17% of youth who had tried an HTP or ENDS device have not previously smoked cigarettes, and one out of six youth who smoke cigarettes had initially used HTPs or ENDS.9 11 12

Surveys on JUUL conducted in the USA have shown a usage rate of 9.5% between ages 15 and 17 years, those under the age of 21 years were shown to have higher odds of ever and current use. The usage patterns themselves suggest that youth are not only experimenting with JUUL but are using it on a regular basis.13 In early 2020, youth awareness of HTPs and IQOS specifically was 5.6%, 6.4%, 9.1% in England, Canada, and the USA, respectively, with 38.6% of youth who recognised the products being interested in trying them.14

A growing concern is that HTPs and ENDS may be marketed directly to youth using tactics previously used to market cigarettes.15 Tobacco industry documents have shown that youth have been a specific target for marketing campaigns, as underage smoking has been viewed as critical to the tobacco industry future.16

Proximity of POS that sell tobacco to schools has been recognised as an important factor influencing adolescent exposure to tobacco marketing, smoking initiation and smoking behaviour.17–19 For example, simulation research examining exposure to advertising within tobacco retailers in the USA (Texas) showed that stores located near schools (within 305 m) contained more tobacco advertisements, relative to those not proximal to schools.17 A study in California found a positive association of cigarette POS within walking distance of schools and school-specific cigarette smoking prevalence and incidence of youth initiation.18 In Scotland, a 2016 study found that greater density of tobacco retailers around homes and schools was associated with increased odds of lifetime and current smoking among adolescents.19

Moreover, a greater density of tobacco POS has been shown to be associated with other neighbourhood characteristics, not surprisingly the proportion of the neighbourhood comprised of youth,20 21 as well as other targeted subgroups such as those of low socioeconomic status (SES).22 This body of research has led many jurisdictions to attempt to reduce exposure to cigarette retail among vulnerable populations; in particular, to protect youth, many have implemented minimal distance requirements between tobacco POS and schools (eg, 100 m in China; 150–300 m common in the USA).23 24 This literature both suggests that HTP and ENDS POS locations may be used to target specific subgroups (eg, youth), as well as demonstrating the potential of minimal distance requirements to reduce exposure to HTP and ENDS marketing.25–28

Worldwide, research regarding tobacco and nicotine product retail availability and population impact faces challenges due to limited access to comprehensive regulatory lists containing the location of HTP and ENDS POS. A recent study in Australia indicated that, even when such lists are available, they may contain large amounts of inaccurate data (ie, including retailers not selling tobacco and excluding retailers selling tobacco).29 In Israel, there is no required license for POS that sell tobacco products, HTPs or ENDS, and thus, no surveillance regarding the distribution of IQOS and JUUL in general, nor near schools. Recently, IQOS and JUUL have made their POS locations in Israel publicly available online to facilitate consumer access.30 31 This publicly available information may be used to assess the spatial distribution of IQOS and JUUL in relation to schools.

In summary, the current study is informed by a socioecological framework32 and the literature regarding macrolevel influences, specifically tobacco retail exposure, on tobacco use among specific populations, particularly young people and those from low SES backgrounds.33 While this literature is vast, the majority of this research has been conducted in the USA or Canada,34 and there has been relatively little research examining alternative tobacco/nicotine product retail availability, particularly outside of the USA.34 35 Thus, this study addresses the potential of IQOS and JUUL retail availability to influence IQOS and JUUL use among young people35 36 and aims to contribute to the aforementioned limitations of the existing literature. This study may also help inform policymakers regarding the potential impact a sales ban, at different radii surrounding schools, would have. IQOS (HTP) and JUUL (ENDS) POS were specially chosen due to both companies being well-known brands in the Israeli market and because their websites declare that their products are intended to serve as an alternative to cigarette use by adults but should not be used by minors.30 31

Drawing from publicly available IQOS and JUUL retail location data provided online by IQOS and JUUL in Israel, we examined: (1) proximity to the nearest POS; and (2) number of POS (density) within meaningful proximal distances, specifically adjacent (within 400 m) and within walking distance (within 1000 m).34–36 In addition, given the literature indicating greater density of tobacco retail in low-SES neighbourhoods,37–40 we examined school neighbourhood SES in relation to these indicators of POS proximity and density.

Methods

Study design

This study entails a cross-sectional geospatial analysis. Data involved in these analyses include: (a) IQOS and JUUL POS locations (advertised to the public on companies’ websites, downloaded in June 2019); (b) locations of all schools, including elementary, middle and high schools (grades 1–12), provided by the Israeli Ministry of Education (obtained in August 2017); and (c) SES of neighbourhoods surrounding school provided by a private company Points Location Intelligence. We did not attempt to conduct a separate analysis per school type (elementary/middle/high) as classification of schools varies across different municipalities, that is, in some municipalities elementary school included only grades 1–6 versus other municipalities where it includes grades 1–8. The SES data set is based on the National Population Statistics Survey released in 2017 by the Israeli Central Bureau of Statistics (CBS)41 and is widely used by the private sector in Israel, the Israeli Ministry of Health for its ‘'National Program for Quality Indicators in Community Healthcare’ as of 2016,41 the Israeli Center for Disease Control, and the Israel Department of Economics and Administration. SES ranking of neighbourhoods is based on population attributes within the following categories: demographics, education, employment, pension and lifestyle (eg, number of cars per household).

Data analysis

First, we characterised each school’s neighbourhood SES level. Points Location Intelligence data use statistical areas of 0.99 km2 on average with a median size of 0.49 km2 and are based largely on parameters from the CBS which divides Israel into small homogeneous statistical areas with an average of 3000 people per statistical area. The SES rank is calculated by the CBS based on 14 different sociodemographic and economic factors such as the proportion of people aged 27–54 years old who have an academic degree and average monthly income per person. From the CBS ranking, population SES was recoded from a 1 to 10 non-linear ordinal scale: low (1–3), medium (4–7) and high (8–10). Statistical areas irrelevant for SES ranking due to being categorised as unpopulated areas received a value of 0 on the original scale and were removed from SES relevant analysis. Each school received the socioeconomic ranking of the statistical area within which they are located. Of the 4580 schools, 4462 schools were located within a statistical area with an SES ranking; the 118 schools missing these data were excluded from SES relevant analyses.

Next, we examined our primary research aims. School coordinates (defined a point approximately at the centre of each school) and the number of IQOS and JUUL POS both adjacent to (defined as maximum distance of 400 m) and within walking distance (defined as a maximum distance of 1000 m or 1 km) of schools were calculated.34–36 42 In addition, a parallel analysis of the minimal distance of POS surrounding schools was conducted. Figure 1 demonstrates a comparison of both measures on a map of POS surrounding one high school in Israel.

Figure 1

Map of IQOS and JUUL POS surrounding one high school in Tel Aviv: the number of POS adjacent (400 m) to schools, within walking distance (1000 m) of schools and the distance to the nearest POS. *Adjacent POS (400 m; inner circle): five IQOS, four JUUL. POS within walking distance (1000 m; outer circle): 35 IQOS, 24 JUUL. The distance to the nearest POS: 100 m to IQOS POS (I) and 225 m to JUUL POS (J). POS, points of sale.

Using linear distance matrices (k=1), we calculated the distance from each school to the nearest IQOS and JUUL POS. We then calculated the proportion of schools with at least one POS within different radii. We assessed the exposure of schools to IQOS and JUUL POS by stratifying the distance to the nearest POS, the number of POS within walking distance and the number of POS adjacent to each school, by the surrounding neighbourhood SES ranking.

In addition, with the results of the distance matrices, we calculated the proportion of schools with at least one POS within different radii (ranging from 100 to 2000 m) to explore the potential impact of different minimal distance requirements between POS and schools.

Non-parametric tests were chosen due to the non-normal distribution of the data. The association between SES and the calculated continuous variables was initially tested with the Kruskal-Wallis H test. A significant Kruskal-Wallis H test was followed by a post hoc analysis, Dunn’s multiple comparison test, and Vargha and Delaney’s A effect size statistic.

The programs used for statistical analysis, geospatial analysis and mapping included R V.3.6.2, RStudio V.1.1.456 and QGIS V.3.6.3 for Microsoft Windows 10.

Results

Overall, 3267 IQOS POS and 1996 JUUL POS were identified. Of the 4580 total schools, 4462 had information regarding school neighbourhood SES; 1467 (33%), 2502 (56%), and 493 (11%) were in low-SES, middle-SES, and high-SES neighbourhoods, respectively.

Within a radius of 1 km from schools, we found an average of 8.73 IQOS POS (SD=10.63, median=5, IQR=1–12) and 5.19 JUUL POS (SD=7.38, median=2, IQR=0–8; table 1). The average distances from schools to the nearest IQOS and JUUL POS were 954 m (SD=2218 m, median=365 m, IQR=206–703 m) and 1535 m (SD=2957 m, median=579 m, IQR=299–1544 m; table 1), respectively.

Table 1

Exposure to IQOS/JUUL POS among schools in Israel (n=4580)

The proportion of schools with at least one POS within different radii is presented in figure 2. For example, the proportions of schools with an IQOS or JUUL POS within a 1 km radius were 86% and 74%, respectively.

Figure 2

Percentage of schools with at least one POS of IQOS/JUUL within a radius of X. POS, points of sale.

The average numbers of POS within walking distance (1 km) of schools in low-SES, middle-SES and high-SES neighbourhoods were: 7.49, 9.91, and 7.64 IQOS POS and 4.10, 5.92, and 5.46 JUUL POS, respectively (table 2); that is, low-SES schools generally had the lowest average number of POS.

Table 2

Association between the number of POS within 400 m/1000 m of educational centres and the socioeconomic status

Aligning with table 2 findings, the median distances from schools in low-SES, middle-SES, and high-SES neighbourhoods to the nearest POS were 428 m, 325 m, and 403 m for IQOS, respectively, and 1044 m, 483 m, and 525 m for JUUL, respectively. The distances from schools to the nearest POS were significantly farther for those located in low-SES neighbourhoods in comparison with middle-SES neighbourhoods (table 3).

Table 3

Association between the distance (metres) from educational centres to the nearest IQOS/JUUL POS and the surrounding neighbourhood socioeconomic status

Discussion

Our findings indicate high spatial access to IQOS and JUUL POS among schools in Israel and the potentially targeting of middle-SES populations. Specifically, the majority of the schools in Israel had at least one POS of IQOS (82%) and JUUL (66%) within walking distance (1 km), and nearly half had at least one POS adjacent to them (400 m). Each school had on average 8.7 IQOS POS and 5.2 JUUL POS within walking distance. Schools located in middle-SES neighbourhoods were found to be closer to at least one IQOS and JUUL POS and to have a greater number of POS within walking distance and adjacent to them.

Comparisons of findings in tables 1–3 suggest differences between number of POS within ‘walking’ distance versus distance to the nearest POS. Despite differences in these metrics, both indicators consistently demonstrated considerable school exposure to POS and increased exposures in middle-SES versus low-SES neighbourhoods.

These findings highlight similar concerns about youth exposure to POS carrying traditional and novel tobacco products globally,17–19 as well as the potential targeting of youth among tobacco, HTP, and ENDS companies and retailers. Moreover, analysing exposure at different radii might provide insights about the potential impact of specific minimum distance requirements from schools that could be implemented. In Israel, new legislation was passed in 2019 that limits marketing of HTPs, ENDS and tobacco products within POS. However, recent research in Israel has demonstrated that many POS continue to market HTPs and ENDS despite the new legislation.43 44 Results of our study highlight the need for additional legislation in Israel to counter increased IQOS and JUUL exposure among youth by limiting HTP and ENDS POS near schools. A rise in indicators of dependence on HTPs and ENDS among youth in Canada, England and the USA emphasises that time is of the essence.45

Our findings show increased accessibility to IQOS and JUUL POS in middle-SES to high-SES (vs low-SES) neighbourhoods, thus aligning with recent research in the USA that has shown increased JUUL use by young adults from higher SES households.46 Vape shops in the USA were also found to be farther away from public schools in areas with a higher proportion of the population in poverty.47

Current results emphasise the importance of collecting and analysing the location of HTP and ENDS POS on a countrywide scale and the potential utility of industry-generated and distributed data regarding their POS. Such research in Israel (and in other countries) has been challenging due to the limited and potentially inaccurate data available on a broadscale basis. It is important to note that the data used in the current study—derived from industry websites—may also have limitations (eg, lack of precision or specificity). Nonetheless, this study underscores the necessity of high-quality data accessible to public health officials and researchers to support national efforts to develop and evaluate policies aimed at reducing youth use of HTPs, ENDS and tobacco products more broadly. Unfortunately, Israel is currently without mandatory retailer licensing requirements to sell HTPs, ENDS and tobacco products, which represents a key first step in establishing such data sources required for regulation. This concern is further compounded by increasing use of sophisticated big data analytics to reach consumers.48 As an increasing number of advanced data collection and analytical methods are implemented by companies (including the tobacco industry) to promote their products and nicotine-based products evolve, public health advocates must be equally equipped with the methods and technology necessary to protect the public.

Strengths and limitations

Strengths of this study are the comprehensive inclusion of all schools in Israel and contextualising them by neighbourhood SES, using Israel’s gold standard data sources for schools and neighbourhood demography. However, there may be some limitations to the data sources for IQOS and JUUL POS, derived from the industry websites; these lists might not have included all relevant POS or may have included POS no longer selling the respective products. While these concerns are mitigated by the fact that the companies themselves maintain these data to facilitate consumer access to products and retailers, there is nonetheless a need to interpret findings with caution. Additionally, we used the centre of each school (rather than the front entrance) to calculate distances and proximity to these POS as these were the data available to us. While this may create some level of error, it may also help account for the likelihood that more than one school entrance may be frequently used by students. We also were not able to determine whether a single POS sold both IQOS and JUUL, as POS were mapped based on coordinates, thus precluding us from ascertaining if any overlap was true because both were sold within a single POS, two adjacent POS or two POS within the same shopping centre.

Moreover, other factors may have played a role in the associations between school and POS locations that were documented. For example, additional attributes of each school (eg, number of students, predominately Arab students, predominately Jewish ultra-Orthodox students) and neighbourhood (eg, smoking, IQOS and JUUL use rates) were not analysed in this study. In addition, we did not compare proximity and density with other ‘control’ locations (eg, non-youth serving facilities), so the extent to which exposure to IQOS and JUUL POS among schools differs from other facilities within communities is uncertain. Finally, exposure to IQOS and JUUL POS may be different than exposure to tobacco POS in Israel—and these associations may differ across countries. For example, while both the USA and Israel share a wide variety of types of areas of living, including high density cities, farmland, suburbs, rural and forests/deserts, the relative size of each differs. In addition, in Israel, recent legislation banned public advertising of HTPs and ENDS, with 2020 marking when a display ban and plain packaging came into effect; therefore, we may assume that marketing exposure, especially among youth, may be different in Israel than in other countries.

Conclusion

A high percentage of schools in Israel have IQOS and JUUL POS within walking distance. Like tobacco, the high level of spatial access to IQOS and JUUL POS among schools in Israel may lead to increased HTP and ENDS use by youth. Legislation to prevent youth HTPs and ENDS use should be guided by a spatial analysis of the exposure of schools to HTP and ENDS POS. For example, legislation limiting the proximity of HTP and ENDS POS to a radius of at least 400 m would affect the exposure of 54% of schools to IQOS POS and 35% of schools to JUUL POS. In comparison, legislating a minimal radius of 100 m would impact only 8% of schools to IQOS POS and 4% of schools to JUUL POS.

What this paper adds

What is already known on this subject

  • Use of heated tobacco products (HTPs) and electronic nicotine delivery systems (ENDS) among youth is becoming increasingly common globally.

  • HTP and ENDS use among youth might lead to adverse health outcomes, including nicotine addiction.

  • Limiting distribution of HTPs and ENDS near schools is a potential method to reduce youth exposure and youth initiation.

What important gaps in knowledge exist on this topic

  • Many countries around the world, including Israel, do not require HTP and ENDS points-of-sale (POS) licensing and therefore lack a national database listing HTP and ENDS POS locations.

  • Thus, alternative ways of assessing youth environmental exposure and distribution of HTP and ENDS near schools warrant exploration.

What this paper adds

  • Using HTP and ENDS POS location data publicly advertised by the companies themselves feasibly allows examination of their spatial distribution relative to schools.

  • Analysis of IQOS and JUUL (prominent HTPs and ENDS) POS proximity to and POS density around schools indicated high levels of potential exposure among Israeli youth.

  • Schools in middle-income neighbourhoods were particularly exposed to IQOS and JUUL POS, compared with high-income and low-income neighbourhoods in Israel.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

References

Footnotes

  • Contributors DE was the lead author for this manuscript. DE lead all stages of the research working with lab director HL throughout. HL was the guarantor of the study and accepts full responsibility for all aspects of the study. YBZ, CJB and LA contributed equally both in planning the research and editing throughout the writing process of the manuscript.

  • Funding This research was funded with a grant from the Israeli Medical Association for Smoking Prevention and Cessation; and part of a grant by the National Cancer Institute (R01CA239178-01A1; MPIs CJB, HL).

  • Competing interests YBZ has received fees for lectures from Pfizer, Novartis and GSK Consumer Health (distributors of smoking cessation pharmacotherapy in Israel) in the past (2012–2007/2019). YBZ volunteers (without pay) as chair of the Israeli Medical Association for Smoking Prevention and Cessation. LA receives royalties for the sale of Text2Quit and is a shareholder in Welltok. HL had received fees for lectures from Pfizer Israel (distributor of a smoking cessation pharmacotherapy in Israel) in 2017. HL has been on the board (2017–2021) and chairman (2019–2021) of the Israeli Association of Public Health Physicians, Israel Medical Association.

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