INTRODUCTION
Globally, tobacco usage is the largest preventable cause of mortality and disability1. In 2010, the World Health Organization (WHO) estimated that tobacco use, primarily smoking, led to 4.9 million premature deaths per year1. This increased to 7.1 million in 2016. Assuming that present cigarette consumption rates remain stable, that figure will reach eight million by 20302. The regions with the most smokers are South-East Asia Region (SEARO) and the Western Pacific, contributing about 6.4 million and 4.7 million smokers, respectively.
Indonesia is a major contributor to SEARO’s tobacco pandemic3. Within Indonesia, the smoking prevalence among youth aged 13–18 years is 38.3%. This is substantially higher than the rates seen in other SEARO countries, such as Malaysia (20.6%), Thailand (17.2%), and Myanmar (17%)3. Smoking prevalence among Indonesian youth aged 10–18 years grew from 7.2% in 2013 to 9.1% in 20184.
Tobacco companies invest millions of dollars annually to market cigarettes through television, radio, cinema, newspapers, magazines, point of sale, outdoor displays, and the internet5. Indonesia is among the few ASEAN countries that have not completely prohibited tobacco advertising, promotion, and sponsorships (TAPS)6. Several studies have found a correlation between TAPS and youth smoking behavior7,8. Empirical evidence indicates that advertising and promotion by tobacco companies significantly affect adolescents to initiate tobacco use. Adolescents exposed to tobacco promotion frequently perceive the advertisements as attractive. Tobacco advertisements render smoking attractive, potentially heightening adolescents’ inclination to smoke9. WHO considers banning TAPS to be a necessary part of a global movement to overcome the tobacco epidemic1.
Studies conducted in Indonesia have documented outdoor cigarette advertisements (OCA) near public places, such as schools, health facilities, and religious sites10-12, the most recent was in 20208. Significant correlations existed between teenage smoking prevalence and the density and proximity of outdoor tobacco advertising in Indonesia, although no such relationships were seen among adults. This underscores the necessity of implementing a ban on outdoor tobacco advertising, particularly in proximity to schools, in Indonesia and elsewhere. Other assessments of TAPS and cigarette retailers (CR) around schools in Indonesia are few13. This study provides a recent estimate of the current state of tobacco advertisements and retailers around selected schools in Indonesia. Specifically, it describes the density and proximity of OCA and CR near Indonesian schools and summarizes co-located students’ self-reported OCA exposure.
METHODS
Study design
Our study was cross-sectional, conducted between September 2022 and April 2023 in three selected districts (Serang, Banyuwangi, and East Lombok) and a city district (Padang). These districts all have some form of smoke-free area regulations. Serang and East Lombok have loose tobacco control policies. Padang and Banyuwangi have tight tobacco control policies. Specific local regulations ban tobacco advertisements14.
Student population and sample
Our student survey sample was selected using multistage cluster sampling. First, within each of the four districts, we selected the three subdistricts with the highest number of schools and greatest population density. The selected subdistricts were Padang Barat, Padang Timur, and dan Koto Tangah (in Padang city); Cikande, Kramatwatu, and Kragilan (in Serang district); Muncar, Genteng, and Banyuwangi (in Banyuwangi district); and Aikmel, Pringgabaya, and Selong (in East Lombok district). Then, we identified all junior and high schools in these subdistricts. Next, we selected schools using proportional random sampling. The total number of schools in Padang, Serang, Banyuwangi, and East Lombok was 110, 108, 117, and 171, respectively. After selecting schools, we randomly selected two classes (each grade level at both junior and senior high school) in each school. All students in the selected classes were interviewed. Interviews were conducted in Bahasa. After missing data were excluded, 6715 students were interviewed.
Cigarette retailers, outdoor cigarette advertisements, and survey data collection
We identified all cigarette retailers and outdoor cigarette advertisements in the study areas. Global Positioning System (GPS) devices were used to map schools, OCA, and CR locations, and Google Maps was used to identify and validate area boundaries. We recruited 37 local enumerators for the various districts and cities. A project researcher trained enumerators. Enumerators walked and rode motorcycles to collect the needed GPS data.
The enumerators also conducted the student surveys. Surveys were conducted in the schools, and parental permission was obtained before gathering survey data.
Data were collected using KoboToolBox15. Field coordinators and supervisors’ quality assured the data. The field coordinator checked the enumerators’ daily data collection count and reported to the supervisor. The supervisor verified the number of data entries in the database system. To ensure accuracy and validity, the quality control team reviewed a 10% random sample of the OCA and CR data collected by the enumerators, and reinterviewed 10% of each enumerator’s students.
Variables
For the district OCA and CR survey, variables compiled included the total number of CR and OCA and their types. OCA types were posters, banners, billboards, powerwalls, stickers, store name boards, and others. CR types included stores, kiosks, street retailers, mini markets, supermarkets, and others.
We measured OCA and CR school proximity, as well as calculated OCA and CR density. We measured OCA proximity as the closest OCA to the school in meters. OCA density was the total number of OCA in km2 within three radii: 0–100 m, 0–300 m, and 0–500 m. CR proximity was measured as the closest CR to the school in meters. CR density was the total number of CR within the three radii in km2.
The student survey included five self-reported student exposures to tobacco products (Supplementary file). We assessed exposure to tobacco products based on four types of exposure, old media (television, newspaper, or magazine), new media (internet), OCA, and CR.
RESULTS
Cigarette retailers
We identified 21460 retailers. Of these, 30.4% were CR. Across all locations, the proportion of CR varied between 24.8% and 40.7%. East Lombok had the highest proportion of CR (40.7%). Banyuwangi had the highest density, 9.96 per km2. Based on the number of children and adolescents, Banyuwangi also had the highest density at 2.6 per 1000 adolescents. Among the four study areas, kiosks were the most common type of retailer (Table 1).
Table 1
Distribution of the number of schools and cigarette retailers and outdoor cigarette advertisement characteristics, by district, September 2022 to April 2023, Indonesia
Outdoor cigarette advertisements
OCA were observed at 13660 points (range per district: 1918–4472). Of the OCA types, banners were the most common around schools (62.1%). Posters were second (32.8%). Padang had the highest percentage of cigarette banners (62.8%.) Banyuwangi had the highest percentage of posters (38.0%) (Table 1).
In East Lombok, 60.2% of all schools had at least 1 cigarette retailer point within a 100 m radius. More than 90% of schools had at least one within a 0–300 m radius. This is only slightly better than in Banyuwangi, where 76.9% of schools had at least one within a 0–100 m radius, 98.4% of schools had at least one within a 0–300 m radius, and all schools had at least one within a 0–500 m radius. In Serang, the results were similar to Banyuwangi’s where 71.3% of schools had at least one within a 0–100 m radius, and more than 90% of schools had at least one within 0–300 m. Padang City was slightly better than the other three districts; half of the schools (51.8%) had at least 1 cigarette retailer point within a 0–100 m radius, 88.2% of schools had at least one within a 0–300 m radius, and 93.6% of schools had at least one within a 0–500 m radius (Table 2).
Table 2
Density of cigarette retailers and outdoor cigarette advertisements near schools by district, September 2022 to April 2023, Indonesia
In all study locations, the density of CR/km2 increased as the distance from schools decreased. The density from the study schools ranged from 20.5–29.8 CR/km2 within 0–500 m, 29.4–36.0 within 0–300 m, and 33.4–67.8 within 0–100 m. The increase in CR/km2 was higher in Banyuwangi and Serang, than in Padang and East Lombok (Table 2).
The OCA density pattern was the same as the CR density pattern, i.e. as the distance from the schools decreased, the OCA density also increased. At a radius of 0–100 m, 71.3% of the schools in Padang, 70.9% in Banyuwangi, and 63.2% in East Lombok had at least 1 outdoor cigarette advertisement. Serang was the one exception, where only 40.9% of schools had one. Within a radius of 0–300 m, more than 88% of all schools in the study locations had at least 1 outdoor cigarette advertisement. Within 0–500 m, more than 90% of the schools had at least one, and every school in Banyuwangi had at least one. The density pattern differed somewhat in Padang, where at a radius of 0–100 m, the OCA/km2 slightly decreased compared to the density at a radius of 0–300 m. Within the 0–100 m radius, the highest OCA/km2 was in Banyuwangi (137.6), followed by East Lombok (97.5), Serang (88.7), and Padang (48.1) (Table 2).
Students were more exposed to tobacco products through the internet (range per district: 18.6–66.0%) than through television (range per district: 20.6–38.3%), magazines, or newspapers (range per district: 0.8–7.6%). However, exposure to tobacco products from the internet was lower than that from OCA (range per district: 75.2–85.7%) and CR (range per district: 80.2–95.5%). The highest tobacco exposure the students reported was OCA and CR (Table 3).
Table 3
Distribution of students’ self-reported tobacco product exposure, September 2022 to April 2023, Indonesia (N=6715)
DISCUSSION
General findings
Our study found that 30.4% of retailers were CR. The number of OCA observed was 13660 points. The density of CR and OCA per km2 increased as the distance from schools decreased. Students’ self-reported exposure to tobacco products showed that exposure from OCA and CR was higher than that from the internet, television, magazines, or newspapers. This means that the students were surrounded by various TAPS, which increases the possibility of initiating or maintaining smoking behavior. Tobacco advertising is a form of TAPS that is used by the cigarette industry to promote their product, enhance sales, and expand their youth market. They use both conventional and social media to do so. Therefore, if anti-TAPS policies do not include a comprehensive ban, the tobacco industry will find alternative means of promoting their products16.
Comparison with other studies and possible mechanisms
Our study found that 30.4% of retailers were CR. In all four districts, kiosks were the most common type of retailer. This resembles results in Bali, Indonesia, where 77.8% of CR were kiosks13. CR play a crucial role in increased access to tobacco products17. The presence of CR around schools may prompt children and adolescents to smoke and increase their perception of the availability of tobacco products13,18. In Western Australia, a ban on the display of tobacco products reduced smoking prevalence19. Many people, especially children and youth, are exposed to the presence of cigarette products, which makes them interested in buying the products20. The presence of CR around educational institutions was found to lead to a higher risk of tobacco-related problems in Thai youth21.
In our study, the density of CR per km2 increased with decreasing distance from schools. We also found that 52–77% of the schools in all districts examined had at least 1 cigarette retailer point within a 100 m radius. Similar results were found in Bali13, Banyuwangi22,23, as well as Depok24. The proportion in our study was higher than that found in Thailand and elsewhere in Indonesia25. In Depok, West Java, Indonesia, a 40% higher density of CR was observed in areas near educational facilities24. In Ontario, Canada, smoking behavior was found to be significantly influenced by access to tobacco products. A major contributing factor to youth smoking is the high density of CR around schools20. The high density we found probably increases youth smoking prevalence in Indonesia26. The higher density of CR around schools indicates that CR were allowed to display tobacco products near schools and undoubtedly these displays are part of the tobacco industry’s programs to enhance the sales of their tobacco product specifically targeted at youth27,28.
The total number of OCA observed was 13660 and in any one district ranged between 1918 and 6472. This means that the students were surrounded by various TAPS. A previous study in Semarang found 3453 OCA points29. In our study, banners were the most common form of OCA around schools. In Banda Aceh, Indonesia, banners were also the most commonly found type of OCA30. However, in Surabaya, Indonesia, billboards were the most frequent type encountered OCA11. Surabaya is the capital of East Java Province and is the second-largest metropolitan city in Indonesia. This metropolitan environment may explain the dominance of billboards.
Similar to our results for CR, we found that as the distance from the schools decreased, OCA density increased. Likewise, in Semarang, Indonesia, OCA density was higher near schools29. In addition, a study conducted by Sebayang et al.31 in Surabaya and Banyuwangi, Indonesia, found that the density of outdoor tobacco advertising increased as the distance to an educational facility used by children and adolescents decreased. Research among Thai youth identified a high risk for tobacco-related problems due to their exposure to tobacco product advertising near educational institutions25.
According to students’ self-reported exposure to tobacco products, exposure from OCA and CR was higher than exposure through the internet, television, magazines, or newspapers. These results differ from a 2017 Indonesian study which identified high TAPS exposure from the Instagram (29.6%). Among offline types of exposure to TAPS, OCA (54.4%) was the second-highest exposure following television (74%)8. This difference might be because our research was conducted after the COVID-19 pandemic, between September 2022 and March 2023. During the pandemic, the government imposed restrictions on activity and travel32,33. Thus, children and adolescents, bored from being at home, tended to increase their outdoor activities, which more frequently exposed them to tobacco products through OCA and CR.
Limitations
Our research does not represent all of Indonesia, due to differences in the geographical, social, and economic characteristics of the regions and because only four districts were studied. The sampling strategy used here involved selecting districts with high population densities and high school densities, which may contribute to a higher likelihood of OCA and CR in those areas. Additionally, the cross-sectional study design does not allow for the establishment of causal relationships between the variables studied. The use of self-reported information may introduce potential biases in this study. Further research is needed to examine broader and more representative areas in Indonesia.
Policy implications
This study demonstrated that TAPS-related policies in Indonesia remain weak. The implementation of Presidential Regulation No. 109 of 2012, Article 39, and Law No. 32 of 2002, Article 46, concerning broadcasting, needs to be strengthened to eliminate cigarette advertising and promotions in communities, especially near children and adolescents. The government should regulate tobacco access, particularly for vulnerable groups, including children and adolescents. Restricting the proximity of CR and OCA to school premises can reduce the exposure to tobacco products among children and adolescents25, and thus reduce the prevalence of smoking among adolescents and children. To strengthen these efforts, comprehensive tobacco control measures, including a prohibition of outdoor tobacco advertising, are necessary34,35.
