Cross-Country Comparison of Bans on Internet Tobacco Advertising, and Search Interest in Vaping Products

Abstract

Background: The rise in vaping use across high-income countries is driven partly by marketing, advertising, and flavors promoted by e-cigarette producer firms. The current study aim is to examine the potential variation in the relative research volume on vaping products between countries with highest level of tobacco banning internet advertising and those with no ban on tobacco internet advertising. Method: We used weekly relative search volume (RSV) data produced by Google Trends (GT). A total of eight countries included in the study, Saudi Arabia, United Arab Emirates, United States, Ireland, New Zealand, Canada, United Kingdom, and Australia. The countries were regrouped into ban in internet advertising countries, and the no bans in internet advertising countries. Results: The trend test indicates a statistically significant upward trend in GT vaping search volume across all the included countries. However, the percentage increase of search volume for the ads no ban group was higher than ban group. There is a statistically significant difference in median between the two group, (P < 0.0007), (median, 6.25 [IQR, 0.103] for the ban in internet advertising countries vs (median 7.5 Interquartile Ranges (IQR), [0.542] for the no bans on internet advertising countries. Conclusion: Countries with stringent bans on online tobacco advertising have shown lower levels of vaping-related search interest, indicating the potential effectiveness of such regulations. Our research underscores the importance of emphasizing comprehensive bans on tobacco advertising and sales online could help mitigate the upward trend in vaping interest.

Introduction

Vaping: A Global Public Health Challenge

E-cigarette use, commonly known as vaping, has emerged as a significant global public health concern.^1,2 Over the past decade, vaping devices have become the most prevalent tobacco products among young people worldwide, partly driven by aggressive marketing, advertising, and the appeal of various flavors promoted by e-cigarette manufacturers.^2,3 Consequently, there has been a substantial increase in youth vaping, especially in high-income countries, largely due to the extensive reach of advertising campaigns. ⁴ In 2020, the number of vapers worldwide reached 64 million, with more than half residing in high-income nations.^4,5 Recognizing the global tobacco epidemic, the World Health Organization (WHO) has prioritized initiatives aimed at promoting smoking cessation and reducing tobacco consumption worldwide. ⁶ To address the escalating prevalence of vaping, over 68 countries have implemented regulations, such as age restrictions, advertising limitations, bans on public vaping, and taxation measures. ⁷ Middle Eastern high-income countries, including the United Arab Emirates and Saudi Arabia, have also taken proactive measures by implementing regulations and imposing taxes on tobacco products to combat the growing vaping population.^7,8 However, further efforts are needed regarding bans on tobacco advertising and promotion and sponsorship.

Tobacco Advertising, Promotion and Sponsorship (TAPS)

The bans of tobacco advertising and promotion include any form of commercial communication, recommendation or action with the aim, effect or likely effect of promoting a tobacco product or tobacco use either directly or indirectly. ⁹ Whereas tobacco sponsorship includes any form of contribution to any event, activity or individual with the aim, effect or likely effect of promoting a tobacco product or tobacco use either directly or indirectly. ⁹ Tobacco companies employ various forms of tobacco advertising, promotion, and sponsorship (TAPS) to boost the consumption of their products. These activities aim to elevate tobacco use among existing users, attracting new consumers, and persuade former users to resume tobacco consumption. ¹⁰ A TAPS ban, even if initially considered comprehensive and forward-thinking, may become outdated or ineffective if it fails to adapt to evolving promotional methods. With the rapid transition to a digital media landscape, particularly the surge in online social media platforms, the tobacco industry has found new avenues to promote its products. Furthermore, although comprehensive TAPS bans in certain countries prohibit the tobacco industry from sponsoring tobacco depictions in movies and films, these regulations typically do not cover unsponsored tobacco depictions in entertainment media such as in movies, online videos and computer games. ¹⁰

Internet Ads

Laws governing the internet advertising of e-cigarettes vary across countries, reflecting diverse regulatory approaches to the burgeoning electronic cigarette industry. In some nations, such as Australia, Canada, New Zealand, and the United Kingdom, stringent measures have been implemented to prohibit online sponsored advertising of nicotine-containing e-cigarettes.^11-13 On the contrary, countries like Saudi Arabia, the United Arab Emirates, the United States, and Ireland do not impose a ban on online vaping advertising, fostering a more permissive environment for marketing and promotional activities related to e-cigarettes.^14-16 These variations and gaps of internet tobacco related policies and the availability of new channels of ads is expected to increase the number of people searching vaping related products through different social media platforms. ¹⁷ Hence, the present study aims to: (1) assess the temporal trend in the search interest in searching vaping products across multiple high-income countries with varying implementation levels of prohibition on the online vaping advertising, and then (2) examine the potential variation in the relative research volume on vaping products between countries with highest level of tobacco banning internet advertising and those with partial or no ban tobacco internet advertising. Our hypothesis assumes that countries implementing a comprehensive tobacco ban in internet advertising will experience a lesser volume of increase in the interest of the vaping population compared to countries with minimal or no restrictions on tobacco products internet advertising.

The Prevalence of Vaping Use

The Global Adult Tobacco Survey 2011 for Indonesia found that, while 10.9% of respondents were aware of electronic cigarettes, just 0.3% used them. ¹⁸ According to a recent assessment, 10% of UK smokers utilized electronic nicotine delivery devices, with the number of users increasing to roughly 1.3 million in 2013, up from 700,000 the year before. ¹⁹ According to a September 2013 data from the US Centers for Disease Control and Prevention, the number of middle and high school kids who tried e-cigarettes increased by double between 2011 and 2012, with an estimated 1.78 million students having done so by the end of that year. Furthermore, about 160,000 students who experimented with electronic cigarettes had never used traditional cigarettes. ²⁰ Serious questions concerning the overall effects of e-cigarettes on public health have been raised by their rising use. ²¹

Method

Google Trends Tool

Google Trends (GT) is an online open tool by google that provides data on the temporal trends of search queries and is a publicly available tool. It is used by the general population that has submitted the keyword phrases via the Google search engine, using weekly relative search volume (RSV). Data normalization was unutilized to ensure that values are standardized within a range of 0 to 100 scale which represent the relative popularity of a search term A value of 0 does not necessarily mean there were no searches; rather, it indicates extremely low search volumes that were not included in the results and 100 indicating the peak interest during the specified time period. For example, during a big vaping-related news event or regulatory announcement, a very popular search term like “vaping” or “e-cigarettes” frequently hits close to 100, indicating a significant increase in search volume. The normalization process also eliminates queries made within a short timeframe from the same internet protocol (IP) address and queries that contain special characters. While Google does not filter controversial topics, it does exclude sexually explicit related search terms. However, it does provide access to the normalized hits for any keyword entered, irrespective of filters.^22-24

GT enables users to examine the online interest in a single term or compare the online interest across up to five terms. It offers a wide range of combinations to compare different terms and regions. ²⁵ GT shows the change in interest in a topic over time and regional interest. RSV has been calculated weekly by GT based on the search terms that were entered in Google relative to the total search volume. ²⁶ GT doesn’t display total numbers of searches over time, instead it provides population-adjusted data reflecting the popularity of the search at a given time. ²⁷ The use of GT in public health research studies is increasing, which provides useful information about epidemiological surveillance, screening, and treatment options. ²⁸

Topic, Key Terms, Region Selection, and Tobacco Control Policy

All search queries entered into Google’s search engine become grouped based on all categories and web search option. To examine the population’s interest in vaping, we completed a series of search queries in GT between January 1, 2013, and July 07, 2023. We then downloaded the dataset of search interest in the terms, “vape”, “vaping”, “vape pen”, “vaping device”, and “vaporizer” from GT (https://www.google.com/trends),^29,30 comparing the search of vape products in high-income countries: United States of America, United Kingdom, Canada, Australia, Ireland, Saudi Arabi, United Arab Emirates and New Zealand. The most frequently mentioned key term across all countries was “vaping”, therefore it was used in the analysis for the current study. To ensure validity of the keyword selection, to avoid any potential variation of spellings and accents, we used the “+” feature. This feature includes the most commonly encountered misspellings of similar words. ³⁰

The rationale for the selection of these eight countries was the fact that half of the vape users living in a high-income country. ³⁰ Also, to gain a global perspective based on high-income countries in which Google and YouTube are commonly used search engines. Additionally, the high percentage of individuals using the internet in these high-income countries (90 % to 97% of the total population), enabled the study to present a trend line vividly due to the large volume of internet and google users in these countries.^31,32 We used individual search items and combined the GT results in a large table query, as listed in Table 1.

Table 1.

Google Trends for Vaping Relative Search Volume 2013 to 2023 ^a .

Year	Saudi Arabia	United Arab Emirates	United States	Ireland	New Zealand	Canada	United Kingdom	Australia
2013	0.67	1.14	10.25	1.08	0.67	2.75	2.92	1.83
2014	1.46	1.11	18.25	2.08	1.25	6.96	4.58	3.46
2015	2.89	3.48	26.44	3.83	2.19	13.08	8.92	5.56
2016	9.15	8.07	32.06	7.54	4.23	18.19	15.83	9.04
2017	20.15	12.25	39.08	12.72	8.55	24.22	22.83	13.25
2018	27.75	17.06	47.74	17.47	13.42	31.56	28.03	18.28
2019	34.87	22.77	53.70	22.04	18.40	37.32	32.17	22.70
2020	40.05	25.09	55.57	25.38	22.56	40.01	34.68	25.68
2021	44.12	29.16	58.11	29.82	28.26	43.94	37.97	32.06
2022	48.83	34.72	60.42	35.59	34.03	48.16	42.74	37.18
2023 b	50.88	37.35	61.76	38.89	37.66	50.77	45.84	40.32

^arelative search volume range from 0 to 100.

^bmissing months (August to December).

The selected sample countries span across different continents including North America, Europe, Oceania, and the Middle East. This geographic diversity might help in understanding the influence of cultural, economic, and regional factors on vaping research volume.^33-35 Cultural attitudes towards smoking and vaping vary significantly among these countries, providing a rich context for analyzing how these attitudes may impact research priorities and public health policies. Furthermore, countries such as Australia, Ireland, Saudi Arabia, and the United Arab Emirates have stringent regulations or complete bans on tobacco internet advertising. These regulations provide a distinct context to evaluate the impact of advertising bans on vaping research volume. Conversely, the United States, the United Kingdom, Canada, and New Zealand have more lenient regulations or no specific bans on tobacco internet advertising.^36-44 This contrast allows us to explore how differing regulatory environments influence the research focus and volume related to vaping products.

Statistical Analysis

To compare the median of combined vaping search volume across the two country groups, we had to standardize the search volumes for each group by country’s population size to have a common comparable score. ⁴² This involved dividing the total population by the search volume and subsequently converting the outcomes into z-scores for standardization. The z-scores were determined using the formula below, with “query V″ representing search term utilized in the analysis, and “n” is the population size for each country:

$$\mathrm{Z}{\text{score}}_{\mathrm{i}}=\frac{Search\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}Vi-\hspace{0.17em}\overline{Search\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}V}}{standard\hspace{0.17em}deviation\hspace{0.17em}ofsearch\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}V}$$

$$\overline{Search\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}V}=\frac{{{\displaystyle \sum }}_{i=1}^{n}Search\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}V}{n}$$

$$standard\hspace{0.17em}deviation\hspace{0.17em}ofsearch\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}V=\sqrt{\frac{{{\displaystyle \sum }}_{i=1}^n(Search\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}query\hspace{0.17em}Vi-\overline{\hspace{0.17em}{Search\hspace{0.17em}volume\hspace{0.17em}of\hspace{0.17em}queryV)}^2}}{n-1}}$$

The value for the populations used to calculate the search volumes per capita was the value in 2022, obtained from the world bank open data (https://data.worldbank.org/). The next step was to examine the mean difference of standardized search volume scores between the two generated groups. Due to violation of the normality test, we used the non-parametric Mann-Whitney U-test. The threshold for statistical significance was P <.05 (2-sided). We further calculated the Average Annual Percent Change (AAPC) with bootstrap 95% Confidence Intervals (CIs) to examine the characteristics of the trend over the study specific period from 2013 to 2023. AAPC was estimated by taking the current year value and subtracting that from the previous year’s value. Then, this difference was divided by the previous value and multiplied by 100 to obtain a percentage representation of the increase rate. The trend changes in search volume rates over time were assessed using the Mann-Kendall trend test. All analyses were performed using SAS 9.4.

Results

There has been consistent increase of vaping annual search volume for all included countries from 2013 to 2023. However, the percentage increase of search volume of no bans group was higher than bans group (Figure 1A and B). In 2013 the vaping search volume for no bans group Saudi Arabia, United States, United Arab Emirates, and Ireland were relatively low. However, by 2023, these volumes had increased for all four counties. Similar patterns were observed in the bans group. In New Zealand, Canada, United Kingdom, and Australia, vaping annual search volume were slightly low in 2013 but had increased by 2023. (Table.1, Figure 1A and 1B).

Figure 1.

Yearly, monthly and combined vaping relative search volume “Google Trends”.

Table 2 provides details on the average annual percent change rate (AAPC) of vaping search volume in GT. A consistent increase in AAPC was observed from 2014 to 2022 across all countries with slightly higher search volume among no bans group. The highest increase rates were recorded between 2014 and 2016, with Saudi Arabia showing the highest increase in 2015. Starting from 2018, there has been a noticeable deceleration in the increasing trend across all countries. The rate of increase varied, with the highest in the United Arab Emirates in 2018 and the lowest in the United States in 2019.

Table 2.

Average Annual Percent Change in Vaping Search Volume (Google Trends) 2014-2022.

	Saudi	United Arab Emirates	United States	Ireland	New Zealand	Canada	United Kingdom	Australia
	% a (95%CI)	% (95%CI)	% (95%CI)	% (95%CI)	% (95%CI)	% (95%CI)	% (95%CI)	% (95%CI)
2014	+97.95 (98.49, 97.41)	+213.51 (215.42, 211.6)	+44.88 (46.79,42.97)	+84.13 (84.47,83.79)	+75.2 (75.67,74.73)	+87.93 (89.85,86.01)	+94.76 (95.44,94.08)	+60.69 (61.37,60.01)
2015	+216.61 (218.02, 215.2)	+131.9 (134.74,129.06)	+21.26 (24.1, 18.42)	+96.87 (98.49,95.25)	+93.15 (93.96,92.34)	+39.07 (42.35,35.79)	77.47 (80.9,74.04)	+62.59 (64, 61.18)
2016	+120.22 (126.12, 114.32)	+51.8 (52.89, 50.71)	+21.9 (22.99,20.81)	+68.7 (71.18,66.22)	+102.13 (103.68,100.58)	+33.15 (34.7, 31.6)	44.22 (46.7, 41.74)	+46.57 (48.08,45.06)
2017	+37.72 (44.45, 30.99)	+39.27 (46.01, 32.53)	+22.16 (28.9, 15.42)	+37.34 (38.87,35.81)	+56.96 (58.95, 54.97)	30.31 (34.46,26.16)	22.78 (24.6, 20.96)	+37.96 (40.2, 35.72)
2018	+25.66 (30.89, 20.43)	+33.47 (35.83, 31.11)	+12.48 (14.84,10.12)	+26.16 (28.45,23.87)	+37.11 (38.71, 35.51)	+18.25 (23.51,12.99)	14.77 (16.17,13.37)	+24.18 (26.63,21.73)
2019	+14.86 (19.09, 10.63)	+10.19 (16.75, 3.63)	+3.48 (10.04, 3.08)	+15.15 (19.13,11.17)	+22.61 (25.52, 19.7)	+7.21 (10.63, 3.79)	7.8 (12.77, 2.83)	+13.13 (16.51,9.3)
2020	+10.16 (15.98, 4.34)	+16.22 (17.91, 14.53)	+4.57 (6.26, 2.88)	+17.49 (21.35,13.63)	+25.27 (28.88, 21.66)	+9.82 (13.24, 6.4)	9.49 (11.84, 7.14)	+24.84 (30.46,19.22)
2021	+10.68, (14.42, 6.94)	+19.07 (20.85, 17.29)	+3.98 (5.76, 2.2)	+19.35 (27.33,11.37)	+20.42 (25.32, 15.52)	+9.6 (12.82, 6.38)	12.56 (18.46, 6.66)	+15.97 (22.44, 9.5)
2022 b	+4.2 (7.38, 1.02)	+7.57 (9.51, 5.63)	+2.22 (4.16, 0.28)	+9.27 (13.21, 5.33)	+10.67 (15.2, 6.14)	+5.42 (10.86, 0.02)	7.25 (10.29, 4.21)	+8.45 (12.33, 4.57)

^apercentage and confidence intervals.

^bdue to missing months, the year 2023 was excluded.

The Mann-Whitney U test indicates that there is a statistically significant difference in median between the two group, the highest bans score in internet advertising countries, and the bans or no bans score internet advertising countries (P < 0.0007). The Mann-Whitney U test revealed that there are a significant difference in medianbetween the highest bans score in internet advertising countries and the no bans score on internet advertising countries (Tables 3 and 4; Figure 2).

Table 3.

Mean, Median, Standard Deviation, and Confidence Interval for Vaping Search Volume for the Two Country’s Groups 2013-2023.

Level	Number	Mean	Median	Std Dev	IQR a	Std Err Mean	Lower 95%	Upper 95%	Std Dev Lower 95%	Std Dev Upper 95%
Prohibition on the online advertising	128	−0.173	6.25	0.902	−0.103	0.079	0.0159	0.331	0.803	1.029
No prohibition on the online advertising	128	0.175	7.5	1.0644	0.542	0.094	0.011	0.3621	0.947	1.2143

^aInterquartile range.

Table 4.

Score Mean Differences Between High and Low Vaping Search Volume Score Group (Mann-Whitney U-Test) 2013- 2023.

Level	Count	Score Sum	Expected Score	Score Mean	(Mean−Mean0)/Std0	S	Z	P-Value
Prohibition on the online advertising	128	14794.0	16384.0	115.57	−2.699	17846	2.699	0.007
No prohibition on the online advertising	128	17846.0	16256.0	140.52	2.699

Figure 2.

Vaping relative search volume from 2013 to 2023.

Discussion

The current study found in a sample of high-income countries, a significant and consistent increase in the vaping GT search queries. Saudi Arabia, United Arab Emirates, United States, Ireland, New Zealand, Canada, United Kingdom, and Australia revealed a significant increase in vaping search interest over the last 10 years. However, after regrouping and standardizing search volume for the countries, the study’s finding indicated that there was a significant difference in vaping related search interest between the two groups. The group with no ban tobacco internet advertising encountered larger volume of vaping search interest compared to the group with the highest level of tobacco banning internet advertising. This surge in search interest in vaping could be attributed to the lenient online tobacco control policies in these countries. In fact, as of 2020, only one country has banned internet sales of tobacco products. This permissive environment potentially facilitates vaping users to direct their search for a vaping product with minimum restrictions. Notably, it’s worth acknowledging that 81% of Google searches in 2022 were geared towards facilitating online purchases, and a substantial portion of these searches were initiated by individuals aged between 18 and 34 years.^43,44 Consequently, it is reasonable to deduce that a significant portion of the GT search volume associated with vaping likely originates from young adults who are actively seeking vaping products for purchase. This insight demonstrates the vital role of online accessibility in fueling the interest in vaping among the younger demographic, thereby warranting a closer investigation of the potential attributed search health implications.

This finding suggests that imposing restrictions on tobacco internet advertising in these countries could potentially reduce individuals’ propensity to search for vaping products and limit their exposure to tobacco-related advertisements. Previous research has illuminated the influence of tobacco policies on information-seeking behavior related to smoking cessation and related topics.^44-46 However, these studies primarily focused on non-internet-related tobacco control policies like taxation and smoking cessation programs. ⁴⁶ In contrast, our study places specific emphasis on internet-related tobacco policies and their potential impact on the public’s interest in vaping. This is of paramount importance since minimal internet-related tobacco policies were in effect in these countries, potentially resulting in a significant portion of internet users being exposed to tobacco advertising when searching for vaping products. Thus, understanding the implications of internet-related tobacco policies on vaping behavior can provide valuable insights into the broader public health landscape.

Another reason might explain the difference in vaping search interest between the two country’s group, is the adoption of tobacco control policy in the online media platforms, and search engines such as Google. ⁴⁷ In line with World Health Organization tobacco control policies, Google have enforced an advertising policy regarding hazardous products or services, which includes the prohibition of advertising tobacco or any tobacco-containing products, items that are part of a tobacco product, and products or services that directly support or promote tobacco use. ⁴⁷ The benefits of these policies may become apparent primarily in nations where there are strict regulations surrounding tobacco advertising in the online world. However, there is an absence of concrete evidence indicating that Google’s voluntary tobacco control measures show any significant variation based on the regulatory landscape governing internet-based tobacco advertising within specific countries. Nevertheless, it is conceivable that the imposition of comprehensive internet tobacco bans across various online platforms and search engines within countries that enforce rigorous regulations on tobacco-related internet advertising could play a crucial role in the observed decline in search interest pertaining to vaping within these regions.^48-50

Limitations

The study possesses several limitations that warrant consideration. Firstly, the absence of demographic and other potential confounding factors in our analysis leaves room for the possibility of confounding bias. These unaccounted variables could influence search interest and their corresponding Google search behavior, potentially affecting the study’s outcomes. Secondly, our study’s focus on a specific sample of high-income countries restricts the generalizability of our findings to the broader spectrum of high-income nations worldwide. Variations in tobacco policies, cultural factors, and socioeconomic disparities among high-income countries may result in different trends and outcomes not captured by our study’s limited selection of countries. Furthermore, variations between the groups could be attributed to differences in vaping prevalence across the countries studied. This is important because without accounting for these differences, the study might misinterpret the reasons behind the variation in search behaviors. Consequently, caution should be considered when generalizing these findings to high-income countries not included in our analysis.

Conclusion

In conclusion, our study has identified significant and consistent growth in search interest in vaping over the past decade in several high-income countries. Notably, countries with stringent bans on online tobacco advertising have shown lower levels of vaping-related search interest, indicating the potential effectiveness of such regulations. Our research underscores the importance of aligning online tobacco control policies with international guidelines and emphasizes that comprehensive bans on tobacco advertising and sales online could help mitigate the upward trend in vaping interest, particularly among young individuals, thus safeguarding public health.

Ethical Statement

Ethics Approval and Consent to Participate

This study was performed in line with the principles of the Declaration of King Abdullah International Medical Research Center (KAIMRC). The Biomedical ethics committee in KAIMRC approved this study with the reference number (NRJ23J/109/04). All methods were performed in accordance with the relevant guidelines and regulations of KAIMRC.

Informed Consent

Informed consent were waived by the Biomedical ethics committee of King Abdullah International Medical Research Center (KAIMRC) in view of the retrospective nature of the study.

ORCID iD

Majed Ramadan https://orcid.org/0000-0001-9838-2155

Data Availability Statement

The data that support the findings of this study are publicly available. Data are available [Google Trends https://trends.google.com/trends].