Background Given the decline in alcohol consumption amongst young people, which has coincided with the rise in technological use, there is a need for research to investigate whether technological use might be impacting on how young people drink alcohol. This study explores how social media use, and changes in social media use over time might affect alcohol consumption among youths.

Methods Using data from Understanding Standing, a UK household longitudinal study, from participants aged 12–19 (N=3586) from the youth and main survey in Wave 3 (2011–2013), and followed up in Wave 6 (2014–2016), we explored whether belonging to social media sites and frequency of interaction on social media sites in the past week, were related to being a drinker and drinking frequency in the past four weeks respectively, at baseline. Calculating change scores over time, we assessed whether changes in social media belonging, were associated with changes in drinking status, and whether changes in frequency of use of social media sites were related to changes in drinking frequency, using multinomial logistic and linear regression model. Analyses were repeated with number of times binge drinking (5+ drinks) in the past four weeks, on a sub-sample of participants aged 16–19. Models adjusted for sex, age, household tenure, urban/rural area, educational/work status, baseline social media use/frequency and number of friends.

Results At baseline, those who belonged to a social media site were over four times more likely (OR=4.62 95% CI (3.61–5.91)) to be a drinker versus non-drinker. Those who chatted for longer on social media sites were more likely to have drank at least once a week compared with not drinking in the past month (none vs. 1–3 hours (2.31(1.45–3.68), 4–6 hours 2.60(1.44–4.72), 7 hours or more 5.92(2.86–12.28). Among participants aged 16–19, a similar gradient was found with higher odds of binge drinking three or more times in the past month, with greater chat frequency (e.g. no hours chatting vs. 7 hours or more 6.29(2.50–15.82)). Over time, becoming new social media users increased the odds of becoming a new drinker (1.77 (1.04–3.03)) versus no change in drinking status. Participants who became more frequent users of social media sites, were more likely to drink more frequently (β=0.13, 95% CI (0.09–0.17)), and binge drink (β=0.21, 95% CI (0.12–0.31)) more frequently across waves.

Conclusion Increasing social media use was associated with greater alcohol consumption; mechanisms behind this relationship should be investigated. Alcohol-related content on the internet should be monitored.