Introduction
As stated in Chapter 1, the promotion of drinking guidelines is a complex intervention. The CMOs announced the UK Government’s revised guidelines in January 2016 and the Government’s Alcohol Strategy5 states that their purpose is to:
Support people to in make informed choices about healthier and responsible drinking.
However, this is an incomplete account of the intervention and its aims. Subsequent promotional activity may be delivered by multiple stakeholders at different points in time, using different techniques and targeting outcomes that may be more or less proximal to alcohol consumption behaviours. These different promotional activities are likely to vary in their effectiveness, and changes in targeted outcomes arising from one activity may also be contingent on the timing and impact of other activities. As a result, identifying a single intervention time point or primary outcome measure for evaluation analyses is not straightforward.
Most previous studies evaluating the effectiveness of drinking guidelines as a public health intervention do not address this point. Many studies use simple before-and-after designs or analyses of annual time series and generic outcome measures, such as drinkers’ awareness of guidelines in general and their knowledge of specific guidelines. They also tend to evaluate the announcement of new or revised guidelines rather than the effects of specific promotional activities promoting those guidelines. As a result, there is little engagement with the actual aims, design or outcomes targeted by such activities. It is often unclear, therefore, whether promotion is ineffective, absent, poorly designed or targeted at outcomes that are not examined by evaluation studies. Exceptions to this include analyses of annual Danish promotional campaigns organised by the National Board of Health, which targeted different groups and used different messages in different years, and this is partly reflected in the interpretation of evaluation results.37,38 Similarly, Dixon et al.’s41 evaluation of a campaign in Australia that focused on drinking guidelines as well as alcohol and cancer in women surveyed samples of women on their knowledge of the alcohol–cancer link as well as more generic guideline-related knowledge and drinking behaviours before the campaign and at two time points afterwards.
For the purposes of this study, it is important that evaluation analyses start from a more detailed assessment of promotional activity. The use of a monthly time series within an interrupted time series design means that our results are likely to be sensitive to misspecification of the intervention point. For example, if the guidelines were announced in January 2016 but a large promotional campaign took place in September 2016, we would need to consider the appropriate analytical design to best estimate the effects of these two intervention points. It also clear that health promotion campaigns informed by theory are more effective than those with no clear theoretical basis and that behaviour change arises from multiple factors rather than a single outcome.65 Therefore, it is important to assess the content and characteristics of promotional campaigns to identify their likely effectiveness, the outcomes in which we would expect to see changes and whether or not we would expect these changes, either alone or in combination with those arising from other promotional activities, to lead to changes in alcohol consumption behaviour.
We therefore aimed to construct a timeline of promotional activity relating to the drinking guidelines, detailing when activities occurred and what they involved. We also sought to identify the cost of activities to inform a planned cost-effectiveness analysis of promotional activity as a whole. For the purpose of this study, we defined promotional activity as any deliberate and planned effect to communicate the drinking guidelines with a view to having a national-level impact of a scale detectable within our analyses, whether directly (e.g. via a mass media campaign) or indirectly (e.g. via education of health professionals). We were interested in capturing promotion of any aspect of the guidelines described in the final documentation.24 In addition to the guideline consumption threshold, this includes messages around drink-free days and the number of drinking days per week, the disputed cardioprotective benefits of moderate drinking, the cancer risks associated with low levels of alcohol consumption, risks for different adult age groups and drinking in pregnancy. The final example was not a focus of our study but was included in the data collection instruments for completeness.
Methods
Sample
We aimed to survey key stakeholders within organisations in five sectors:
- government and quasi-non-governmental bodies (e.g. Department of Health and Social Care and PHE)
- public health advocacy organisations (e.g. Institute of Alcohol Studies and Cancer Research UK)
- professional bodies (e.g. the Royal College of General Practitioners)
- non-health-related bodies (e.g. the police and the Local Government Association)
- industry and industry-related actors (e.g. Drinkaware and the Portman Group).
We generated a list of 23 potential organisations across these sectors and planned to request contacts within each to complete a survey on promotional activity by their organisation. However, during development of the survey, our monitoring of media content relating to guidelines and our informal conversations with stakeholders provided little indication that any organisation was promoting the guidelines in a way that might have a population-level impact. We therefore narrowed our initial survey sample to five key organisations that we judged to be most likely to be engaging in promotional activity. These organisations were also selected for the likelihood that they would also be aware, through their ongoing activities and networks, of promotional activity being undertaken by other organisations.
Known contacts in each of the five organisations were approached initially via e-mail in June 2017 and sent information about the study. The instructions in the e-mail asked individuals, or their nominated alternative, to follow a link to an online questionnaire containing questions about current or planned promotional activity. The questionnaire also asked participants to consent to recontact in 6 months to be asked about any further promotional activity.
Measures
The online questionnaire is provided in Appendix 2 and was informed by a similar tool created for documenting mass media campaigns related to tobacco.70 It includes questions about each major promotion or communication activity undertaken since January 2016. The questions covered:
- type of activity (e.g. advertising, distribution of health promotion materials)
- communication platform [e.g. television (TV), social media]
- which aspects of the guidelines were communicated (e.g. 14 units per week, number of drinking days)
- when the activity occurred and how frequently it was repeated
- any supplementary promotional activity (e.g. social media posts)
- intended target audience (e.g. national or local, men or women, particular age groups)
- whether or not the promotional content was interactive
- actual or estimated cost of the activity (if available)
- evidence of audience reached (if available)
- a request for the creative brief or other associated documents that could be shared with the research team.
The survey also asked about press statements released or interviews given by staff in the organisation in relation to drinking guidelines and whether or not respondents were aware of promotional activities by other organisations.
Analyses
Survey results were analysed descriptively.
The following planned analytical work did not take place because of the lack of promotional activity (see Results) and we include a description of it here for completeness. We planned for two members of the research team to review the content of each promotional activity using a structured tool adapted from Langley et al.71 The tool classifies the thematic, informational and emotional content of campaigns as well as the style of delivery. It was intended to aid identification of the targeted outcomes and likely effectiveness of the campaign with reference to prior research literature. As the tool was designed to look at the content of mass media campaigns, we planned to adapt and expand it iteratively to cover any other activities identified by the survey. We then planned to determine, in discussion with the Project Advisory Group and the PPI group, which promotional activities were important and appropriate to consider when designing the main evaluation analyses described in Chapter 4.
Results
None of the five organisations reported undertaking any substantial promotional activity relating to the revised drinking guidelines between January 2016 and June 2017. Furthermore, none of the organisations reported any knowledge of substantial promotional activity undertaken by other organisations or any awareness of planned activities by their own or other organisations.
Some organisations did report smaller-scale activities. These included news media activities, such as issuing press releases or giving interviews. These activities usually related to the guidelines being announced in January 2016, their formal adoption in August 2016 or other occasional media requests.
Given the lack of substantial activity identified by our respondents or the research team after 19 months of our 22-month study and the lack of indications of any future activity, we decided that surveying the remaining organisations was unwarranted and we discontinued this work after discussions with our Project Advisory Group and the funder.
Discussion
The main finding of this work was that there was no substantial active promotion of the revised drinking guidelines after they were announced in January 2016. This does not mean that no organisation promoted the guidelines at all. It is clear that the guidelines are promoted intermittently on social media by a range of organisations, that organisations such as PHE and Drinkaware provide the guidelines and related information on their websites and in other materials, and that news media routinely mention the guidelines in day-to-day reporting of alcohol-related and health-related topics (see Chapter 3). However, it is unclear that these activities would substantially affect the design or findings of our evaluation analyses as they are unlikely to either reach large audiences or lead on their own to changes in behaviour or in key influences on behaviour change.65 It is also unclear that they are of sufficient scale or sufficiently different in timing or nature to justify analysis of an additional intervention point beyond January 2016. Our monitoring of stakeholder websites and social media accounts suggests that most organisations changed their messaging on guidelines immediately after the January 2016 announcement and rolled out updated materials in the subsequent months. In the absence of large-scale promotional campaigns, it seems appropriate to regard this activity as part of a general accumulation of drinkers’ exposure to the revised drinking guidelines starting from January 2016 rather than seeking to identify additional intervention points.
Given the above, the remainder of this report proceeds on the basis that there was no significant activity promoting the revised guidelines after the announcement in January 2016. The implications of this for future research, policy and practice are discussed in the main discussion of the project findings in Chapter 5.