Back

Understanding Purchasing Patterns of Alcoholic, Alcohol-Free, and Low-Alcohol Drinks: A Latent Profile Analysis

Rousham, O.; Stevely, A. K.; Holmes, J.

2025-12-04 public and global health
10.64898/2025.12.03.25341542 medRxiv
Show abstract

BackgroundAlcohol-free and low-alcohol (no/lo) drinks ([&le;]1.2% ABV) are increasingly popular in high-income countries. Their potential to reduce alcohol-related harm depends on who buys them, in what quantity, and their incorporation into overall drinking patterns. We aimed to (i) compare characteristics of purchases containing only no/lo drinks, only alcoholic drinks, or both, over time between 2018 and 2023; (ii) identify subgroups with distinct purchasing patterns in 2023; and (iii) describe purchasing and soiodemographic differences between these subgroups. DesignLatent profile analysis of cross-sectional household purchasing data. SettingGreat Britain, 2018 and 2023. ParticipantsNationally representative samples of 30,401 (2018) and 28,254 (2023) households. 4,975 households who purchased no/lo drinks in 2023 were included in the latent profile analysis. MeasurementsData included off-trade (i.e. shop) purchases categorised into: no/lo-only; alcohol-only; or no/lo alongside alcohol. Household characteristics were frequency of each purchase type; standard servings of no/lo drinks per adult; alcohol risk levels based on weekly units of alcohol purchased per adult in household (non-drinker: 0 units; low-risk:[&le;] 14 units; increasing risk: >14[&le;] 35 units; high-risk: >35 units [1 unit = 8g alcohol]); household age; social class; region; and ethnicity. FindingsFrom 2018 to 2023, the proportion of purchases that were alcohol-only fell from 97% (95% CI: 97%-97%) to 95% (CI: 95%-95%) while no/lo-only purchases rose from 1.4% (CI: 1.3%-1.4%) to 2.7% (CI: 2.7%-2.8%) and no/lo alongside alcohol purchases rose from 1.2% (CI: 1.2%-1.2%) to 1.9% (1.9%-2.0%). In 2023, no/lo-only purchases contained fewer servings (median = 6.9) than no/lo alongside alcohol purchases (median = 6.5 plus 24.5 alcohol units) and alcohol-only purchases (median = 24.6 units). No/lo-only purchases occurred earlier in the week, no/lo alongside alcohol purchases peaked on Fridays and Saturdays. Latent profile analysis identified three classes: (i) no/lo triers (53%) averaged 2.1 no/lo servings per adult annually with 95% purchasing no or low-risk levels of alcohol. (ii) Occasional purchasers (34%) averaged 7.5 servings with 20% purchasing alcohol at increasing or high-risk levels. (iii) Dual purchasers (13%) averaged 37.8 servings with 39% purchasing alcohol at increasing or high-risk levels. Dual purchasers and occasional purchasers were older than no/lo triers (60% [p<0.001], 54% [p=0.010], and 49% 55 years respectively). Dual and occasional purchasers were more often white than no/lo triers (97% [p = 0.014], 97% [p=0.0074], 94% respectively). ConclusionsIn Great Britain, most households purchasing no/lo drinks do so infrequently and purchase alcohol at low-risk levels; however, a smaller group of older, higher-risk households purchase no/lo drinks more frequently.

Published in Addiction (predicted rank #1) · training set

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.