Reductions in alcohol consumption following participation in ‘Dry January’: an investigation of the psychological and social factors underpinning change
A thesis submitted in partial fulfilment of the requirements for the degree of Doctor of Philosophy
Temporary Alcohol-Free Challenges: What do we need to know?
Temporary Alcohol-Free Challenges have become increasingly popular. They offer potential for a relatively low-cost way of tackling alcohol-related harms, but to fulfil this potential it is crucial that we develop a deeper understanding of them. The popularity and longevity of Temporary Alcohol-Free Challenges ensures a living laboratory through which we could further our understanding of these complex interventions using sophisticated prospective observational and randomised studies to confirm the apparent benefits of taking part and to clarify the mechanisms through which participation leads to ongoing changes to alcohol consumption. We must also attempt to determine the likelihood of unintended negative consequences in order to mitigate them.
Establishing the extent to which Temporary Alcohol-Free Challenges help people to reduce or quit alcohol consumption, for whom, and how, would ensure that messaging and the external supports provided can be tailored to maximise the potential benefits of participation whilst minimising any harms.
Investigating reductions in consumption following participation in Dry January by heavy alcohol users
The study findings provide tentative support for a reduction in alcohol consumption among heavy alcohol users following participation in Dry January. This reduction does not appear to be fully attributable to seasonal variation in consumption.
Registering for Dry January and having access to online supports did not appear to enhance the benefits of participation. However, the reliability of this finding is limited by the very small sample of ‘unofficial’ participants. Use of more SETs following January predicted greater reductions in consumption, perhaps reflecting greater commitment to behaviour change.
During January, making a public commitment to participation and restructuring the social environment appeared to help people limit any alcohol consumption.
All results are limited by difficulties in recruitment, substantial attrition, and consequently underpowered analyses.
Nevertheless, the issues and limitations of the current study are also instructive and will be used to inform the design of the second prospective observational study.
Investigating the psychological and social determinants of reductions in alcohol consumption following participation in Dry January
Reductions in alcohol consumption at six months were observed among participants who registered for the ‘official’ Dry January campaign. This replicates findings from previous Dry January studies and research into other Temporary Alcohol-Free Challenges. It also demonstrates the applicability of previous findings to heavy alcohol users and addresses concerns that the campaign may attract and benefit those at the lowest risk of alcohol harm.
Whilst it is important to be clear that, though reduced, average alcohol consumption remained at levels indicating increasing/higher risk of alcohol harm (Public Health England, 2017). A substantial reduction in the proportion of participants at the highest levels of risk harm suggests that participation in Dry January is an effective way of reducing consumption for heavy alcohol consumers.
A modest reduction in consumption was also observed among people who did not take part in Dry January. ‘Official’ Dry January participation predicted lower consumption at six months compared to not taking part in any type of Dry January. This is consistent with previous research and suggests that whilst there may be some seasonal variation in alcohol consumption, for example a reduction following increased consumption over the Christmas and New Year period the changes linked with Dry January participation cannot entirely be attributed to this.
Similarly, whilst some of the reduction may be due to regression to the mean, as previously observed, this does not account for the full extent of the reduction reported by Dry January participants.
A reduction in consumption was also seen among ‘unofficial’ Dry January participants, although again this was to a lesser degree than that experienced by ‘official’ participants. This suggests that whilst an independent Dry January is associated with mid-term reductions in alcohol consumption these are enhanced by registering for the official campaign and having access to the structured online supports provided. This aligns with findings that a greater proportion of those who registered for Dry January experienced at least a 10% reduction in AUDIT-C at baseline compared to those who attempted an unsupported Dry January.
The study found that the proportion of people intending to reduce their alcohol use post Dry January increased over the month suggesting that some people who began Dry January not intending to make ongoing changes did so by the end of the month.
The study identified pathways between Dry January participation, more abstinent days during January, a greater increase in participants’ belief in their ability to refuse alcohol, and lower six month alcohol consumption. Increased belief in their ability to refuse alcohol has been observed following participation in Dry January and other Temporary Alcohol-Free Challenges. De Visser and colleagues (2016) found that complete abstinence during Dry January was linked with increased Alcohol Use Refusal Self-Efficacy – which itself predicted lower alcohol consumption at six months. Greater abstinence being linked to a greater increase in their belief in their ability to refuse alcohol supports arguments that changes to heavy alcohol use precedes change in Alcohol Use Refusal Self-Efficacy. These findings also lend support to suggestions that the relationship between Alcohol Use Refusal Self-Efficacy and alcohol consumption is a reciprocal one with abstinence predicting increased belief in their ability to refuse alcohol which in turn predicts reduced consumption.
A significant pathway was also observed between Dry January participation, belief in their ability to refuse alcohol, and lower six month consumption notwithstanding level of abstinence during January. Maintaining abstinence when participating in Dry January may therefore not be the only factor which contributes to an increase in someone’s belief in their ability to refuse alcohol.
Increases in Alcohol Use Refusal Self-Efficacy and frequency of use of the Try Dry app and reductions in alcohol consumer identity following January were also linked with reduced consumption at six months for ‘official’ participants. Use of the Try Dry app after January is likely to be more common among Dry January participants who are continuing to make changes to their alcohol use. Descriptive results indicate the components which facilitated self-monitoring, particularly monitoring of alcohol-free and alcohol consumption days, were those used most frequently following January.
At the end of Dry January, participants reported increased belief in their ability to refuse alcohol and identifying to a lesser degree with alcohol consumption. Whilst on average, this reverted slightly over the next few months such that Alcohol Use Refusal Self-Efficacy reduced and alcohol consumer identity increased slightly there was still an overall increase in the belief in their ability to refuse alcohol and reduction in alcohol consumer identity between baseline and six month follow-up.
These results suggest that for at least some participants, possibly those who continued to make/attempt to make changes to their alcohol consumption, the positive changes in Alcohol Use Refusal Self-Efficacy and alcohol consumer identity during January may have also continued.
Some evidence suggests that changes in alcohol consumer identity may occur after reductions in alcohol consumption and thus be a marker, rather than cause, of changes in alcohol consumption. The changes in identity observed post-January may therefore be a consequence of going alcohol-free or reduced consumption during January rather than its cause.
Summary of results
Participation in Dry January was consistently and robustly associated with a reduction in alcohol consumption six months later. Among Try Dry users, reductions were also evident nine months post-January. Reductions in consumption were also observed among those not participating in Dry January and those undertaking an ‘unofficial’ Dry January. There was convincing evidence of greater reductions in consumption for ‘official’ participants than those not participating in Dry January.
Additionally, there was reasonable evidence of greater reductions in alcohol consumption for ‘official’ compared to ‘unofficial’ Dry January participants. A substantial minority of ‘official’ participants using the Try Dry app experienced an increase in consumption following Dry January participation. People engaging in lighter alcohol use who infrequently engaged in heavy alcohol use were more likely to report increased consumption one or two months after Dry January.
Different psychological and social factors were linked with reductions in consumption for ‘official’ and ‘unofficial’ participants. Among ‘official’ Dry January participants, greater reductions were linked with higher baseline alcohol consumption, increases in Alcohol Use Refusal Self-Efficacy and, slightly less robustly, motivation to change during January.
Significant indirect pathways linked ‘official’ participation to reduced consumption via increased Alcohol Use Refusal Self-Efficacy and via level of abstinence and increased Alcohol Use Refusal Self-Efficacy. Post-January, increases in Alcohol Use Refusal Self-Efficacy, weakening of alcohol consumer identity, use of more SETs and greater frequency of use of the Try Dry app predicted greater reductions at six months.
Findings regarding ‘unofficial’ Dry January participants show that higher baseline alcohol consumption was consistently associated with greater reductions in alcohol consumption.
Relative to those not participating, people attempting Dry January had higher alcohol consumption, greater motivation to change, higher household income, and more strongly identified as an alcohol consumer at baseline. Some evidence also suggested that they were younger and had lower Alcohol Use Refusal Self-Efficacy at baseline. The same pattern was observed when comparing predictors of ‘official’ vs ‘unofficial’ participation, such that ‘official’ participants tended to have even greater consumption, were even more motivated to change, had an even higher household income, and an even stronger alcohol consumer identity at baseline. Again, some evidence indicated they were even younger with even lower baseline Alcohol Use Refusal Self-Efficacy than their ‘unofficial’ counterparts.
Practical implications
These findings have a number of practical implications.
- Evidence of enhanced benefits for ‘official’ participants can be shared to encourage people intending to have an alcohol-free January to register for the official campaign and make use of the online supports provided.
- Similarly, evidence of the benefits of Try Dry use post January should be communicated to potential participants to encourage ongoing app use. Approaches which recognise that people may not have found the app particularly useful during Dry January itself but may find it helpful when trying to lower their alcohol consumption post-January could also be considered.
- Having identified constructs which are significantly linked with reductions in alcohol consumption it is possible to develop elements of online supports which target them. For example, Voogt and colleagues found that providing advice on how to resist alcohol in situations considered high risk for alcohol refusal increased social pressure Alcohol Use Refusal Self-Efficacy in an intervention group relative to a control group (2014). Dry January organisers could consider incorporating similar, relatively simple, interventions in the Try Dry app.
- Understanding who participates in Dry January and who registers for the ‘official’ campaign provides some insight as to who the campaign is reaching and also who it is not reaching. Whilst likely to be influenced by mode of participant recruitment, the study findings suggest that the campaign may not be reaching or engaging with people from lower income households. People from lower socio-economic status backgrounds are disproportionately affected by alcohol harm even when consuming less alcohol than their more affluent peers (Bellis et al., 2016). As such, it is important for the campaign to effectively attract and connect with this population.
- Establishing why Dry January is failing to engage people from lower income households or barriers to participation is a necessary first step to making Dry January relevant to even more people. Conducting focus groups or interviews with individuals from lower socio-economic status households who have attempted to make changes to their alcohol consumption but who have never participated in Dry January would be the first step towards understanding, and ultimately addressing, these barriers.
Abstract
Dry January may help to reduce alcohol consumption and associated harms.
Participation in Dry January is linked with reductions in alcohol consumption six months later. Increased alcohol use refusal self-efficacy, belief in one’s ability to refuse alcohol, appears to mediate these reductions. Other mechanisms of change are yet to be identified.
This thesis aimed to establish the psychological and social changes underlying reductions in alcohol consumption among heavy alcohol consumers following participation in Dry January.
A mixed methods approach was used. First, a prospective observational survey study compared outcomes at six-months for ‘official’ Dry January participants (registered with access to online supports) with ‘unofficial’(unregistered) participants and people not participating (Chapter III). Alcohol consumption and relevant psychosocial constructs were measured. In Chapter IV, semi-structured interviews explored strategy use during and following Dry January. A second prospective study (Chapter V) addressed limitations of the first and incorporated findings from the interviews. Additional constructs were assessed including alcohol user identity and use of online supports. Finally, analysis of smartphone app data (Chapter VI) examined changes in consumption nine months post-January and investigated the impact of app engagement on these changes.
‘Official’ Dry January participation was linked with reductions in alcohol consumption six and nine months later. Greater reductions relative to people not participating suggest changes were not entirely attributable to seasonal variation in consumption. Greater reductions relative to ‘unofficial’ participants suggest access to online supports may enhance outcomes.
Increased consumption following January was evident in a substantial minority of ‘official’ app users.
Strategy use was dynamic and evolved post-January: minimal differences in approaches were observed between ‘official’ and ‘unofficial’ participants. Among ‘official’ participants, baseline consumption, motivation, alcohol use refusal self-efficacy, alcohol user identity, abstinence and use of strategies and the app post-January were associated with six-month consumption.
Harnessing relevant psychological and social factors may help maximise the benefits of Dry January.