**3. Results**

#### *3.1. Social Network Analysis of Study Linking Cancer and Vaping Published Research (October 2019)*

Figure 1 below shows a social network analysis of tweets from early October 2019, which relates to a time period when the first academic study was published linking vaping to cancer in mice. Figure 2 is zoomed into Group 2 and labels the influential users within the group. From group 3 onwards we have abbreviated the word 'Group' to 'G'.

**Figure 1.** Tweets related to e-cigarettes in October 2019.

**Figure 2.** Zooming into Group 2.

The network graph highlights that there were different clusters of discussion taking place on Twitter during this time with two large groups and several smaller groups indicating a number of communities that had emerged related to this topic. The ten largest groups were labelled. There was also a sizeable isolates group (Group 1), which indicated that a number of Twitter users were tweeting about e-cigarettes without mentioning or replying to other Twitter users. In Figure 2, we zoom into Group 2 in which discussions were formed around a number of influential user accounts such as President Donald Trump, the World Health Organisation, and Fox News. Alongside this network graph, we were also able to identify the most frequently used words, hashtags, websites, and most influential users in the network, which are described below.

#### *3.2. Most Frequently Occurring Words*

Table 1 below displays the most frequently occurring words during this time period. It highlights that mentions of e-cigarettes were made in conjunction with the scientific work that linked them to cancer.


**Table 1.** Most frequently occurring words.

The most frequently occurring words in Group 1 are displayed in Table 2 below. Here, it can also be seen that many of these words centred on the news story linking vaping to cancer.


**Table 2.** Most frequently occurring words in Group 1.

The most frequently occurring words in Group 2 (the second largest cluster of Twitter users) are highlighted in Table 3 below. It is important to note that the word-count numbers listed above would also contain hashtags that used those words.


**Table 3.** Most frequently occurring words in Group 2.

These words in Group 2 highlighted the polarisation on Twitter as certain hashtags were against the idea of any restrictions on vaping by governments, and Twitter users expressed this through the use of the hashtag "#wevapewevote", whereas those against e-cigarettes would use hashtags such as "#vapeban".

#### *3.3. Most Frequently Occurring Hashtags*

The most frequently occurring hashtags overall are summarized in Table 4 below. These hashtags highlighted division and polarisation on Twitter as some hashtags related to campaigns against vaping such as "vapeban", whereas other hashtags such as "wevapewevote" were pro-vaping. There were more Twitter users using pro-vaping based hashtags than against them.


**Table 4.** Most frequently occurring hashtags.

#### *3.4. Most Frequently Occurring Websites*

The top URL shared on Twitter during this time was entitled "Lung Damage From Vaping Resembles Chemical Burns, Report Says" published in the New York Times (*n* = 341) [18]. The other URLs consisted of an article by the Consumer News and Business Channel (CNBC) (*n* = 229) titled Researchers find e-cigarettes cause lung cancer in mice in first study tying vaping to cancer. Further websites also appeared such as an article titled Expert reaction to study on ecig vapour and cancer in mice, which was published by the Science Media Centre (*n* = 124). Interestingly, the article contained reactions from two professors with expertise in tobacco research, and both noted that the study linking e-cigarettes and cancer to mice was potentially seriously flawed in its relevance for human vapers. Another article titled Juul Is Sued by School Districts That Say Vaping Is a Dangerous Drain on Their Resources published by the New York Times was also shared (*n* = 88). These results are summarized in Table 5 below.


**Table 5.** Most frequently occurring hashtags.

#### *3.5. Influential Users Ranked by Betweenness Centrality*

Influential users were ranked by the betweenness centrality algorithm using NodeXL. Influential users consisted of CNBC, which had 3.3 million followers, and the news anchor of Columbia Broadcasting System's (CBS) Nightly Business Report, who had 31.3 thousand followers. The reason for the prominence of CNBC was because they had published an article on October 7th titled '*Researchers find e-cigarettes cause lung cancer in mice in first study tying vaping to cancer'.* A citizen who tweeted the study and noted the findings of the study for the reason for "not following trends" received over 400 retweets and 1.8 thousand likes also became influential and had 6.1 thousand followers. The user was characterizing e-cigarettes as a societal trend with dissent towards those who were using e-cigarettes. A user replied to this tweet indicating that the fear around e-cigarettes seemed irrational when traditional e-cigarettes were still being sold. Other influential accounts consisted of The Centres for Disease Control (CDC) with 1.24 million followers who tweeted about cases of lung damage of users of mostly illicit e-cigarettes containing the Tetrahydrocannabinol (THC), and Gregory Conley, with 18.1 thousand followers, who is a tobacco harm-reduction advocate and supporter of e-cigarettes who was tweeting at this time. Table 6 below provides a summary of the influential users in the network.


**Table 6.** Influential users ranked by betweenness centrality.

#### *3.6. Pro-Vaping and Anti-Vaping Themes*

Much of the content around this time related to Twitter users sharing the news story linking vaping and cancer in mice. Furthermore, our tweet clustering, word, and hashtag analysis revealed two further main themes taking place within the overall network related to Twitter users who were, broadly speaking, either pro-vaping or anti-vaping. In order to confirm this finding, we utilised content analysis to categorize tweets until thematic saturation occurred, and we found two additional main themes related to pro-vaping or anti-vaping views.

Twitter users who were supporting vaping tweeted how using vaping products had helped them stop smoking and also highlighted how the current issue around vaping was very specific to certain illicit types of vaping products that contained Tetrahydrocannabinol (THC). Arguments were also made towards vaping serving a function in harm-reduction as it was comparably safer than smoking. Twitter users against vaping requested for people to stop using e-cigarettes and for tougher legalisation on vaping products. Below, we provide anonymized tweets related to these two main themes.

*3.7. Pro-Vaping Tweet Extracts*

There were a number of Twitter users who would share their own positive personal experiences of using e-cigarette devices:

*"I was a heavy smoker for many decades and knew I'd be dead in my 50 s and never get a chance to see my grandchildren. But then 5 years ago I started to use e-cigarettes and I may be able to see my grandchildren now"*

Other Twitter users were concerned about the increased focus on younger people using e-cigarettes:

*"The dangers of vaping and youth is completely overblown! Car crashes are actually the leading cause of death in younger people"*

There were also Twitter users who noted the dangers of cigarettes compared to e-cigarettes:

*"Right then, they will ban e-cigs because it causes deaths, but regular cigarettes don't seem to be a problem, lol!"*

There were also Twitter users who criticised the study linking cancer and vaping:

*"If they re-ran the study and compared vaping and real cigarettes then vaping would be the safer alternative! #vapingsavedmylife"*

In the above tweet extract, the user employed the hashtag "vapingsavedmylife", which was used by Twitter users to indicate the life-saving potential of vaping.

#### *3.8. Anti-Vaping Tweet Extracts*

There were also Twitter users who called for people to quit vaping:

*"Please stop vaping! – read this news story Lung Damage From Vaping Resembles Chemical Burns, Report Says"*

Other Twitter users called for parents to inform their children that e-cigarettes were unsafe:

*"Let your children know that vaping is NOT SAFE, doesn't matter what their friends say, they've been brainwashed by big Tobacco!"*

There were also a number of general comments noting that e-cigarette use was likely to be dangerous:

*"I am not a medical professional, but most likely that vaping is not good for you"*

Another Twitter user noted:

*"I used to smoke 50 cigarettes a day and managed to just quit, that's the best way to quit. Vaping is not safe."*

From the tweet extracts above, we can see that there was a diverse range of views and comments noting that e-cigarettes were unsafe.
