What are people saying on social networking sites about the Australian alcohol consumption guidelines?

Posts provide valuable feedback during public consultation for health guidelines

Guidelines provide important information on key health behaviours that can influence the population, with public consultation forming an important part of guideline development.1 Public consultation provides transparency, while improving the quality, legitimacy and acceptability of guidelines to the public.1 Although the public are encouraged to provide formal feedback, they may also discuss and provide valuable feedback on popular social networking sites such as Twitter and Reddit. Social networking sites are universal, with 5.8 million Australians using Twitter and 110 000 using Reddit each month.2 Given that a number of these posts are publicly available, they can be used to answer research questions and track health behaviours,3,4,5 and may be an informative source of feedback on health guidelines during public consultation.6

More than 70% of Australians aged 15 years and over report drinking alcohol, and alcohol consumption is a major public health concern.7 The Australian Guidelines to Reduce Health Risks from Drinking Alcohol8 use evidence to provide recommendations to reduce drinking risks to a low level. The National Health and Medical Research Council (NHMRC) released draft guidelines for public consultation between 16 December 2019 and 24 February 2020 and published the final guidelines in December 2020 (Box 1).

To determine whether social networking site posts can be used as a source of feedback during public consultation for guidelines, we aimed to collect posts on social networking sites (Twitter and Reddit) related to the guidelines. Specifically, we aimed to identify: (i) which guideline recommendations the public are interested in (and discussed most); (ii) whether social networking site posts provided feedback that can be used for public consultation for guidelines; and (iii) any misinterpretations.

Although guideline‐related posts are made on many social networking sites, we limited our search to Twitter and Reddit as most posts on these sites are written publicly, with data accessible through the platforms’ application programming interfaces.9,10

We identified 545 tweets and 283 Reddit posts that explicitly referenced the guidelines. Two researchers (BR and DW) coded which guideline recommendations were referenced, whether the posts provided any feedback, and whether a poster had misinterpreted the guidelines (see the online Supporting Information for additional information).

How are the guidelines discussed on social networking sites?

Most tweets referenced the healthy adult drinking recommendation (guideline 1; 71%). Few tweets referenced the younger person (guideline 2; 3%) or pregnancy and breastfeeding (guideline 3; 4%) recommendations, and the remaining tweets referenced the guidelines in general. Furthermore, 218 (40%) tweets provided some form of feedback on the guidelines and we identified four main categories of themes (Box 2).

• The guidelines are too restrictive (30%; n = 66): tweets mentioned that no one should tell them how to drink and that the number of weekly drinks was far too low and unreasonable. Several tweets directly targeted the process of guideline development, stating they were based on flawed evidence or developed by researchers with conflicts of interest. Some tweets also offered alternative advice to the NHMRC based on the NHMRC’s own evidence.
• No alcohol is safe/less is best (17%; n = 38): tweets mentioned the link between alcohol and cancer or that the language used in the guidelines is not strong enough.
• The timing of public consultation (7%; n = 16): tweets highlighted that releasing the guidelines immediately before a holiday period is unfavourable.
• The guidelines are fine (8%; n = 17): tweets praised the guidelines as great, necessary and rigorous.

Nearly all Reddit posts referenced guideline recommendation 1 (99%), with a single post that also referenced recommendation 2 (1%), and no posts referred to recommendation 3. Nearly half the Reddit posts either included a reflection on how the guidelines impact one’s own life or provided feedback on the guidelines (48%; 134). Three main themes were identified from this feedback (Box 3):

• Zero alcohol is safe/less is best (29%; n = 37): posts highlighted that alcohol is carcinogenic, referenced the negative behavioural effects of alcohol, or stated that recommending consumption of any alcohol was political.
• The guidelines are too restrictive (21%; n = 28): posts either highlighted that the guidelines are pointless as everything is bad for you or that not drinking is boring.
• The guidelines are fine as they are (10%; n = 14): posts discussed and defended the methods used to generate the recommendations and were often written in response to others’ posts.

Are the guidelines being misinterpreted?

We identified misinterpretations within both the Twitter and Reddit posts, comprising 16% and 12% of the posts, respectively, and solely relating to guideline recommendation 1. A majority of misinterpretations on both social networking site platforms stated that the limit for recommendation 1 was “1.4 drinks per day” (rather than a maximum of 10 standard drinks per week). Further, several tweets reported that recommendation 1 had increased, with some providing alternative amounts. Overall, tweets that contained misinterpretations were written to over 850 000 followers.

Reddit posts also suggested that alcohol could be “saved up” and consumed on a single day or that the NHMRC was endorsing alcohol consumption (despite the recommendation explicitly stating otherwise).

Key messages and future directions

We identified feedback on both Twitter and Reddit for the guidelines. Despite the differences in the social networking site platforms, feedback predominantly pertained to recommendation 1 (healthy adults) and largely fell into two polarised categories: (i) the guidelines are too strict; and (ii) zero alcohol is safe. Given that the NHMRC recommends that guidelines deemed important to the public should be prioritised, the current research suggests that recommendation 1 was of interest to the public.

We identified that posts on both Twitter and Reddit misinterpreted guideline 1, with posters reporting that adults should have no more than 1.4 drinks per day. This appears to have originated from tweets and news stories that report adults should have “no more than 10 standard drinks per week, or roughly 1.4 drinks a day”. On Twitter alone this alternative recommendation was tweeted by accounts with over 300 000 followers. Addressing these misinterpretations now that the final guidelines are available may be beneficial to reduce dissemination of misinformation. Similarly, guideline developers could monitor references to their guidelines online and make plans to address public posts that include misinformation.

Our results were similar to those of research in the United Kingdom following the public consultation of the UK drinking guidelines, which found that guidelines are discussed online and were more negative than positive.6 Given how simple this method is and how quickly data can be collected, analysing social networking site posts should be considered as part of public consultation. Future guidelines could also consider actively posting or advertising on social networking sites during public consultation to promote discussion and feedback. Moreover, given the ubiquity of social networking sites, this approach could be applied to monitor the sentiment of how the public view other controversial health laws, such as pill testing, cannabis reform and lockout laws.

A strength of our method was that multiple social networking sites were used to collect diverse opinions from users. Indeed, while there were similarities between the two social networking sites, the “no alcohol is safe” theme tended to be more endorsed on Reddit than Twitter (29% v 17%). A limitation of using Twitter and Reddit is the lack of demographic information available. It is important to note that generally users are more likely to be educated, young and, in the case of Reddit, male.11,12 The posts collected in our study may therefore be more likely to be from young or educated users. One way to circumvent this in future research is to use Facebook posts (the most popular social networking site in Australia with a diverse range of users2). We did not utilise Facebook as, at the time, Facebook had restricted researcher access to data through their application programming interface following the Cambridge Analytica scandal.13,14 Future research should consider utilising Facebook posts if access is permitted again.

Popular social networking sites can serve as a valuable data source to provide an important and novel avenue to collect feedback on guidelines. We recommend that future health guidelines consider monitoring public responses online and as part of public consultation. Twitter and Reddit posts provide value to determine the acceptability and interpretation of guidelines by the public, which can supplement public consultation processes. Indeed, using the drinking guidelines as an example, we identified that social networking site users provide feedback and that misinterpretations can appear and be easily disseminated.