Changes in food and drink purchasing behaviour in England during the first 3 months of the COVID-19 pandemic: an interrupted time-series analysis

Abstract

BackgroundRestrictions related to the COVID-19 pandemic affected many health behaviours, including diet. We aimed to examine changes in food and drink purchasing during the first 3 months of the COVID-19 pandemic in England. MethodsIn this interrupted time-series analysis, we used transaction-level purchasing data for food and drink items bought for at-home (n=1245 households) and out-of-home consumption (n=226 individuals) for London and the North of England (Kantar GB). Outcomes included household-level weekly purchased total energy, energy from specific products, alcohol volume, and frequency of out-of-home purchasing occasions. We compared purchases between March 16 and June 11, 2020 (pandemic restrictions, the intervention) and Jan 1, 2019, to March 15, 2020 (counterfactual). The effect of the intervention was modelled using 2-part negative binomial regression models adjusted for time, season, festivals, region, and sociodemographic characteristics. Subgroup analyses explored interactions between the intervention and sociodemographic characteristics and usual purchasing levels. FindingsThe marginal mean estimate of total take-home energy purchased was 17·4% (95% CI 14·9–19·9; 6130 kcal) higher during the intervention period than during the counterfactual period. Increases of 35·2% (23·4–47·0; 505 mL) in take-home volume of alcoholic beverages and 1·2% (0·1–2·4; 165 kcal) in foods and drinks high in fat, salt, and sugar were observed. Reductions in purchased energy from ultraprocessed foods (–4·0%, –5·2 to –2·8; –541 kcal), and out-of-home purchasing frequency (–44·0%, –58·3 to –29·6; –0·6 days) were observed. Highest socioeconomic status was associated with largest increases in total purchased energy (increase of 7217 kcal, 95% CI 5450–8985, vs 2479 kcal, 935–4023, among low socioeconomic status), while older age groups reported fewer changes in purchasing. Higher usual purchasing levels were associated with greater reductions during pandemic restrictions and vice versa, except for the absolute difference in alcohol purchasing, which increased the most for those with higher pre-pandemic purchasing (lowest 123·2 mL, 95% CI 71·3–175·0; highest 708·3 mL, 381·3–1035·3). InterpretationPandemic restrictions were associated with marked changes in purchasing, notably increases in energy and alcohol and decreases in ultra-processed-food and out-of-home purchasing, which differed by individual characteristics. Future research should ascertain if changes persisted, if these changes translate into changes in health, and whether other regions experienced similar effects. FundingNational Institute for Health and Care Research (NIHR) School for Public Health Research (SPHR).