Newswise — The amount of alcohol consumed during a given drinking occasion is strongly associated with the duration of the occasion combined with the beverage type and serving size, according to a study reported in Alcoholism: Clinical and Experimental Research. Although previous research had indicated that alcohol consumption is influenced by the drinking context — for example, by the location, timing, or who was in the drinking group — it was not clear which characteristics are most strongly associated with alcohol consumption and how different factors combine to affect it. The new study aimed to identify which features, and combinations of features, are most predictive of the units of alcohol consumed during drinking occasions in Great Britain.

Researchers from the University of Sheffield, UK, analyzed data from over 18,000 adult drinkers in Great Britain who took part in an online market research survey in 2018. Participants completed a 7-day retrospective drinking diary and between them recorded the characteristics of over 46,000 drinking occasions (defined as a period of drinking with less than 2-hour gaps between drinks). The diaries captured multiple features of each drinking occasion, including drinking location/venue, time of day, the drinking group, occasion type (e.g. a quiet night in) and purpose, drink type and packaging, food eaten, and entertainment or other activities. For each occasion, the research team calculated the number of UK units (1 unit = 8 g of alcohol) consumed based on the reported number of servings, serving size, and drink type. They then used statistical modeling to identify the features most strongly associated with units consumed, using an analytic approach that accounts for the combined, as well as individual, effects of the occasion characteristics. Analyses were conducted for six groups based on participant age and sex, which are already known to influence consumption.

The strongest predictors of heavy alcohol consumption were long occasion duration, drinking spirits as doubles, and drinking wine. During light-drinking occasions, spirits were typically consumed as singles, suggesting that serving size is an important differentiator of overall consumption. Together with demographic characteristics, these factors accounted for up to 70% of the variance (i.e. the difference) in alcohol consumed per occasion. Longer drinking occasions — the most important predictor of units consumed — were in turn associated with starting drinking earlier in the day, drinking in both on-trade (e.g. pubs) and off-trade (e.g. at home) premises, and drinking with friends. Day of the week was a less important predictor of consumption than might be expected, with weekend drinking no heavier than weekday drinking after accounting for occasion duration.

These findings, that combinations of occasion duration and drink type are strongly predictive of alcohol consumption, could inform future drink prevention programs. Initiatives promoting shorter drinking occasions (or ‘when to call it a night’), more stringent exclusion of drunk people from on-trade venues, and restrictions on serving sizes could help to reduce consumption and drinking duration. Future research could contribute to developing and evaluating interventions in these areas.

Combinations of drinking occasion characteristics associated with units of alcohol consumed among British adults: An event-level decision tree modelling study

A.K. Stevely, J. Holmes, P.S. Meier (pages xxx).