key: cord-0815160-y231ag2c
authors: Anderson, Peter; Llopis, Eva Jané; O’Donnell, Amy; Kaner, Eileen
title: Impact of COVID-19 Confinement on Alcohol Purchases in Great Britain: Controlled Interrupted Time-Series Analysis During the First Half of 2020 Compared With 2015–2018
date: 2020-11-19
journal: Alcohol Alcohol
DOI: 10.1093/alcalc/agaa128
sha: 722b95695df6a6bdf3b08835f5849ab68199a101
doc_id: 815160
cord_uid: y231ag2c

AIMS: To investigate if COVID-19 confinement led to excess alcohol purchases by British households. METHODS: We undertake controlled interrupted time series analysis of the impact of COVID-19 confinement introduced on 26 March 2020, using purchase data from Kantar Worldpanel’s of 23,833 British households during January to early July 2020, compared with 53,428 British households for the same time period during 2015–2018. RESULTS: Excess purchases due to confinement during 2020 were 178 g of alcohol per 100 households per day (adjusted for numbers of adults in each household) above an expected base of 438 g based on averaged 2015–2018 data, representing a 40.6% increase. However, when adjusting for expected normal purchases from on-licenced premises (i.e. bars, restaurants, etc.), there was evidence for no excess purchases of grams of alcohol (a 0.7% increase). With these adjustments, beer purchases dropped by 40%, wine purchases increased by 15% and spirits purchases by 22%. Excess purchases increased the richer the household and the lower the age of the main shopper. Confinement was associated with a shift in purchases from lower to higher strength beers. CONCLUSION: During the COVID-19 confinement, the evidence suggests that households did not buy more alcohol for the expected time of the year, when adjusting for what they normally would have purchased from on-licenced premises.

Since early 2020, the population of Great Britain has been subject to a series of social distancing measures of varying intensity as part of the UK Government attempt to limit the spread of COVID-19 (Dunn et al., 2020) . Following unsuccessful efforts to contain and delay the virus during February, lockdown regulations accelerated from March onwards, with schools shut for all but the children of identified key workers, the closure of bars, restaurants and nonessential shops, and the introduction of 'stay-at-home' orders for the majority of the population, with restrictions on international travel, and social gatherings and events banned or severely constrained (HM Government, 2020) . Some of these measures were relaxed during May and June, permitting further indoor and outdoor gatherings between households. However, bars, pubs and restaurants remained closed until 4th July in England, 15th July in Scotland and 3rd August in Wales, although varied 'lockdown' measures continue to be implemented in response to specific spikes in the prevalence of new cases.

The full impact of long-term social distancing on the physical and mental health of people in Great Britain remains unknown. However, available evidence suggests that the combination of severe restrictions in social contact, reduced access to public services and adverse effects on financial and employment security (ONS, 2020) have led to a significant deterioration in UK mental health compared with pre-COVID-19 trends (Pierce et al., 2020) . There is a wellestablished link between alcohol use, stress and depression (Howell et al., 2010; Clay and Parker, 2020; The Lancet Gastroenterology and Hepatology, 2020) , causing some concern that lockdown may have also contributed to the increased levels of drinking during this period (The Lancet Gastroenterology and Hepatology, 2020); indeed, one cross-sectional study of alcohol use in COVID-19 social lockdown found that using alcohol to cope was associated with higher rates of consumption in lockdown (Wardell et al., 2020) . Furthermore, while in theory, there were reduced opportunities to consume alcohol outside the home, due to closure of consumption sites prior to July 2020, off-licence premises (i.e. shops selling alcohol) as well as supermarkets and pharmacies were permitted to remain open during even the early phase of lockdown as designated 'essential' businesses (Dunn et al., 2020) . Initial reports suggest that many people responded to the closure of pubs and restaurants by stockpiling alcohol for home consumption (Eley, 2020) .

Findings from previous surveys of alcohol use in the UK also suggest that consumption increased in many population groups during the COVID-19 pandemic. One study compared 2017-2019 data from the UK Household Longitudinal Study to those collected between 24th and 30th April, one month after the introduction of the stay-at-home order. It found that the prevalence of drinking on four or more occasions per week increased significantly during lockdown (14.2% in 2017-2019 compared with 23% in April 2020) and that there was also a rise in past-month occurrence of heavy episodic drinking, defined as women/men drinking 48/64 or more grams of alcohol on a single occasion (9.7% in 2017-2019 reported heavy episodic drinking compared with 16.6% in April 2020). Changes were largest in adults aged between 35 and 49 years, women and those with middle or high incomes (Daly and Robinson, 2020) . A cross-sectional survey of over 2000 UK residents by Opinium on behalf of the charity Alcohol Change UK found that more than a quarter (28%) of people agreed that they had drunk more alcohol than usual during lockdown, particularly younger drinkers, people in employment, those in higher socio-economic groups and heavier drinkers (Alcohol Change UK, 2020). People also reported drinking more frequently during lockdown, and one in five (19%) of those surveyed had drunk alcohol as a way to handle stress or anxiety, particularly parents of younger children and heavy drinkers. The Smoking and Alcohol Toolkit Survey found an increase in higher risk drinking, as measured by AUDIT-C, associated with lock-down (38 versus 25%), with an odds ratio of 1.85 (95% confidence interval = 1.67-2.06) (Jackson et al. 2020) . However, all surveys used self-report data to assess alcohol consumption and are thus potentially subject to recall bias and error (Greenfield and Kerr, 2008) . Purchase data provide a more objective measure of potential alcohol intake, based on product bar codes, and verified via digital receipts. Moreover, as such data capture precise information on which types of alcohol products have been bought by which types of households, and when, they permit greater specificity in the analysis of purchasing trends over time. For example, we report elsewhere on a small but significant rise in the purchase of lower strength beers in recent years, resulting in reductions in overall household purchases of grams of alcohol (Anderson, Jané Llopis, Rehm, 2020; . Finally, and importantly, given restrictions in on-licenced purchases prior to 4th July 2020, such as in bars, pubs, restaurants and nightclubs, household purchase data represent a particularly valuable source of information about alcohol consumption during lockdown.

Using a large, representative dataset, this study aimed to assess the impact of confinement measures in Great Britain on household purchases of alcohol using interrupted time series analyses. More specifically, we sought to address the following three questions.

(a) Did COVID-19 lockdown lead to a change in purchases of total grams of alcohol? (b) Did any change in purchases differ by household sociodemographic characteristics? (c) Did COVID-19 confinement lead to shifts in purchases of different strength beers?

We undertook controlled, interrupted time series regression analyses of the impact of the introduction of confinement in Great Britain (26 March 2020) on British household purchases of alcohol (including no-and low-alcohol products), using purchases made during the same calendar time period for 2015-2018 (averaged) as our control. We inferred that confinement would lead to excess purchases during confinement above the average level purchased during the same time period for 2015-2018, both to compensate for closure of bars, restaurants, etc.,-'on-licenced premises', and to use alcohol as an anxiolytic. Thus, we examine abrupt and persistent level changes in alcohol purchases, due to an event, the introduction of COVID-19 confinement.

Our data source is Kantar Worldpanel's (KWP) household shopping panel. KWP comprises ∼30,000 British households at any one time, recruited via stratified sampling, with targets set for region, household size, age of main shopper and occupational group. Households record all purchases (from all store types, including Internet shopping) brought into the home using barcode scanners. To be included in KWP's final datasets, households must meet quality control criteria (meeting thresholds for data recording and purchasing volume/spend (based on household size) every 4 weeks). Panellists also upload digital images of checkout receipts, which KWP use to verify the accuracy of scanner data. Households report the age of the main shopper, the number of adults in the household, income, social class and life stage of the household (e.g. with young family, retired, etc.). We grouped households into the following: The distributions of household socio-demographic characteristics for all households that made at least one alcohol purchase for the first 6 months of 2015 to 2018 (averaged) and 2020 are shown in Table 1 . Compared with the average of the first 6 months of each of 2015-2018, households for the first 6 months of 2020 tended to have a higher age of main shopper and a higher household income, but similar distribution of social class group and geographic area. We thus included the age of main shopper and household income as covariates in our models. We also undertook analyses separately for each household grouping, age, class group, household income and geographic area.

We obtained KWP data on all off-trade purchases of alcohol products in Great Britain, including non-alcoholic and low-alcohol products, for the 4 years covering 2015-2018, on which we have previously reported (O'Donnell et al., 2019; Anderson, Jané Llopis, Rehm, 2020; , and for the six and a quarter months between January and early July 2020. Because only items purchased and brought into the home are recorded, only off-trade alcohol purchases are included, that is, alcohol bought in supermarkets, other shops or via Internet shopping for consumption off the premises. Alcohol purchases are recorded daily. A detailed product description identifies the type and volume of the purchase using 19 drink categories, the brand and the alcohol by volume (ABV). Volume was combined with ABV to calculate grams of alcohol purchased.

We prepared data for the interrupted time series analyses by summing up daily records of purchases of alcohol in grams across all households, adjusted for the number of adults in each household, for each day during the first six and a quarter months of each year (2015-2018 and 2020) . To take into account differences in the numbers of households between 2015-2018 and 2020, we express purchases per 100 households for each of the time periods. To match days of the week across the 6 years (i.e. Monday for Monday, etc.), we re-set day 1 for each year as the first Sunday after 1st January (2015: 4th January; 2016: 3rd January; 2017: 8th January; 2018: 7th January; 2020: 5th January). Confinement was introduced on 26th March 2020, day 82 for our re-set day numbers. We restricted our analyses for the re-set days 8-182. Re-set day 8 was chosen to give a greater number of days after New Year, when alcohol purchases are minimal. Day 182 was chosen as this coincides with 4th July 2020, when pubs were re-opened in England, where over 86% of the included households were located. We defined confinement as the period from re-set day 82 to re-set day 182. For our potentially confounding covariates, we used the mean age of the main shopper per day and the mean household income (adjusted for numbers of adults in the household) per day.

We adhered to published guidance for undertaking interrupted time series analyses in the health (Bhaskaran et al., 2013; Bernal et al., 2018) and addictions (Beard et al., 2019) fields. To conduct the main interrupted time series analyses, we adopted similar methodology to that previously used in our assessment of the impact of the introduction of minimum unit price in Scotland (O'Donnell et al., 2019) , based on guidance for the use of controls in interrupted time series studies of public health interventions (Bernal et al., 2018) . As such, we generated a new series of dependent variables representing the differences between 2020 and 2015-2018 (averaged) for each corresponding re-set day. The dependent variables were as follows:

(a) differences between 2020 and 2015-2018 (averaged) in means of sum of grams of alcohol purchased per 100 households (adjusted per adult in the household) per day for all alcohol and by beverage category; (b) differences between 2020 and 2015-2018 (averaged) in the mean ABV of purchased beer per day; and (c) differences between 2020 and 2015-2018 (averaged) in the proportion of the volume of all purchased beer that had an ABV of 3.5% or less.

For the two potentially confounding covariates (mean age of main shopper and mean household income), we also generated a new set of independent variables, the differences between 2020 and the average of 2015-2018.

By creating the new series of differences for the dependent variable data, we largely removed seasonal variation in the data. For each dependent variable, we examined the distribution visually and with Q-Q plots and found all variables to be normally distributed. As recommended by a number of authors (Jebb et al., 2015; Beard et al., 2019) , we used a time series modeler function (IBM Corp, 2017) to estimate best fitting ARIMA models that: (a) specify degrees of differencing and transformations to ensure a stationary series; and (b) specify autoregressive and moving average orders. This eliminated the need to identify an appropriate model through trial and error (Makridakis et al., 1983; McLaughlin, 1984) . Figure 1 plots the residuals, residual autocorrelations and residual partial autocorrelations for both the original and modelled series (from the time series modeller function) for our main dependent variable, the differenced sum of grams of purchased alcohol per 100 households. The plots of the modelled series, ARIMA (0,0,2, 1,1,1) 7 , demonstrated a stationary series, with no evidence of autoregression. Our series was daily, and we set the seasonal function as daily, with a periodicity of 7 (i.e., a week). We repeated the time series modeller function separately for all dependent variables, including within subgroup analyses.

We examined abrupt level changes due to the single event, the introduction of confinement in Great Britain. The event variable was entered as a dummy variable coded with 0 for each day before the event and with 1 for each day from the event forwards. Thus, in our regression models, the dependent variables were ARIMA modelled purchased grams of alcohol, mean ABV of beer and proportion of beer with an ABV of 3.5% or less. The independent variable was the dummy variable confinement, with two covariates, the differences between 2020 and 2015-2018 (averaged) of mean age of main shopper and of mean household income.

The model we are testing is illustrated in Fig. 2 , where, with ARIMA terms (0,0,2) (1,1,1) 7 , the equation is as follows:

where we have:

(Seasonal AR (1) term) (seasonal difference (1) 

Y t is the dependent variable at day t; 1 is seasonal autoregressive operator at lag 1; B is the backshift operator; β intercept is the preevent intercept, in this case, the average of the differences between 2020 and 2015-2018 in purchases for each day across all days prior to the event; θ 1 is the non-seasonal moving average operator at lag 1, and θ 2 the non-seasonal moving average operator at lag 2; is the seasonal moving average operator at lag 1; α t is the error term; β confine is the impact of confinement, in this case, the change of mean daily purchase in grams per household for the time period since the confinement took effect versus before confinement; X t is confinement; β a is the coefficient of age and A t is age; and, β i is the coefficient of income and A t is income.

As results would be sensitive to the exact dates chosen for the event (introduction of confinement) and the end of the period of confinement chosen, we undertook sensitivity analyses for a range of alternative dates, detailed in the footnote to Table 3 .

We repeated the ARIMA models for each of the four separate groupings of households (by age of main shopper, household income, class group and area of Great Britain), with the results reported per 100 households.

All analyses were performed with SPSSv25 (IBM Corp, 2017). Table 2 Column B. Increases were 49% for beer, 31% for wine and 45% for spirits. However, when adjusting for the proportion of grams of alcohol normally purchased from on-licences (Giles and Robinson, 2019) ( Table 2 , Column E), the real change in excess purchases for all alcohol was 0.68% (95% CI: 0.65-0.71) ( Table 2 , column G), based on the off-licence purchases that would have had to have been made for not purchasing any on-licence alcohol, Table 2 , column F. The real changes were −40% for beer, +15% for wine and +22% for spirits. In other words, although the overall grams of alcohol purchased did not change, there was a relative shift in purchases from lower to higher strength products, when accounting for no on-licence purchases. Table 3 summarizes the results of the sensitivity analyses, based on a range of different start and end dates for an analysed confinement period. The excess off-licence purchases due to confinement range from 25.6% with a start date of the Prime Ministers' advice on 16th March to 13th May, when some partial relaxation of confinement occurred to 40.6% for a start date of 26th March, when confinement was actually put in place, to 4th July. The excess purchases for the different time periods match the plots of Fig. 3 .

Proportional excess off-licence purchases increased the younger the age of the main shopper, were considerably lower among the lowest income households, were lower among households in class groupings D and E (in class group E, the age of the main shopper was 65 years or more in over two-fifths of households) and tended to be higher in the north as opposed to the south of Great Britain, Table 4 . Figure 4 plots the mean ABV% of purchased beer and the proportion (%) of all beer purchased with an ABV ≤ 3.5% by study day. Prior to confinement, the mean ABV(%) of beer was lower during 2020 Table 2 . Estimate of excess grams of alcohol purchased (95% CIs), Column C, with percent increase due to confinement, Column D, by beverage category, Column A, with estimate of base for 2020, were purchases of 2015-2018 to apply, Column B; purchases, grams of alcohol per 100 households per day, adjusted for numbers of adults in each household. Accounting for proportion of grams of alcohol normally purchased from off-licences, Column E, and the excess purchases needed to compensate for this, Column F, the adjusted percent increase is shown in Column G than during 2015-2018 (−0.178), although not significant (95% CI: −0.40 to 0.045), but increased following confinement by 0.127 (95% CI: 0.012-0.241, P = 0.030). Prior to confinement, the proportion of beer with an ABV of 3.5% or less was higher during 2020 than during 2015-2018 (1.06% higher, 95% CI: 0.71-1.40), but decreased following confinement (1.48% lower, 95% CI: 1.30-1.66).

During the COVID-19 pandemic confinement, household purchases of alcohol increased by 41%, compared with expected purchases for the same time period averaged across 2015-2018, with no evidence of stockpiling. However, when adjusting for the proportion of alcohol normally purchased from only off-licences (71%), excess purchases did not change relative to other years (+0.7%), with decreases in purchases of beers and ciders and increases in purchases of wines and spirits. We used the 4-year period 2015-2018, averaged, as the control period, since we already had these data available (see O'Donnell et al., 2019) . Excess purchases, without adjusting for normal on-licence purchases, were higher the lower the age of the main shopper and the higher the household income, and much lower for households in class group E, which tended to be of older age.

Within the off-licence purchases, there was a shift to buying higher, rather than lower, strength beer. We are not aware of any published peer-reviewed studies that report objective data on alcohol purchases, such as household purchase data, during COVID-19 confinement in Great Britain. The trends reported here reflect the most recent HM Revenue and Customs data, which found increases in provisional year-to-date duty receipts for wine and spirits, of 8.3 and 8.6%, and decreases in beer and cider receipts of 22.9 and 19.8%, respectively (HMRC, 2020a (HMRC, , 2020b , similar to the direction of the changes of our results. Our results, however, show mixed consistency with the published results of the three surveys mentioned in the introduction (Daly and Robinson, 2020; Alcohol Change UK, 2020; Jackson et al. 2020 ). On the one hand, our results do not confirm the self-reported increases in alcohol consumption due to lockdown found in the surveys. However, we did find greater increases in off-licence purchases among younger and wealthier shoppers, consistent with the two groups that reported increased consumption during lockdown in the two surveys that reported such data (Alcohol Change UK, 2020; Daly and Robinson, 2020) .

Our analyses have several important strengths. Data were obtained from a large number of households, with a large number of daily data points before and after the introduction of confinement. Furthermore, data on purchases were objective, based on product bar codes, and verified via digital receipts. Given that on-licenced premises were closed for the time period 21st March to 4th July, the analyses will have captured almost all alcohol purchases. Any barcoded alcohol product brought into the home (including internetshopping) should have been measured.

By comparison with purchase data from the same calendar days for 2015-2018, we were readily able to estimate excess purchases due to COVID-19 confinement. Controlled interrupted time series analysis provides a strong methodology to evaluate the natural experiment of the introduction of confinement, and we adhered to best practice guidance for undertaking interrupted time series analyses (Bhaskaran et al., 2013; Bernal et al., 2018; Beard et al., 2019) .

One key limitation of our study is that we only measure alcohol purchases and not actual consumption data. Households could have been stockpiling and not consuming all the purchased alcohol. However, given that excess purchases maintained at a high level over the full confinement study period, stockpiling seems unlikely. Given that the proportional increases in excess purchases were much higher for younger rather than older shoppers, it is possible that younger people were buying for older people who remained at home, shielding due to greater vulnerability of COVID-19 illness. Nevertheless, there was an increase in excess off-licence purchases for shoppers aged 65 years or more (16%), and such patterns only represent a re-distribution of the excess off-licence purchases overall.

Our study has clearly demonstrated that British households did not buy more alcohol during the recent COVID-19 confinement, when accounting for what they would normally have purchased from onlicenced premises. Our findings are in line with HM Revenue and Customs duty receipts over the same time period. While reassuring from a public health perspective, our results contradict self-reported survey data. This discrepancy needs further investigation, including the extent to which self-reported data might reflect anxiety or disorientation during what is an unprecedented change to the normal activities of daily living.

P.A. conceptualized the paper and analyses and undertook the analyses. All the authors refined the various versions of the full paper and approved the final manuscript. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted. P.A. and A.O.D. are the guarantors.

Drinking in lockdown press release: new research reveals that without action lockdown drinking habits may be here to stay

Evaluation of alcohol industry action to reduce the harmful use of alcohol: case study from Great Britain

Impact of low and no alcohol beers on purchases of alcohol: interrupted time series analysis of British household shopping data

Understanding and using time series analyses in addiction research

The use of controls in interrupted time series studies of public health interventions

Time series regression studies in environmental epidemiology

Alcohol use and misuse during the COVID-19 pandemic: a potential public health crisis?

Problem drinking before and during the COVID-19 crisis in US and UK adults: evidence from two population-based longitudinal studies

COVID-19 policy tracker: a timeline of national policy and health system responses to COVID-19 in England London

Record month for UK supermarkets as consumers spend extra £1.9bn. Financial Times

Monitoring and Evaluating Scotland's Alcohol Strategy: Monitoring Report

Alcohol measurement methodology in epidemiology: recent advances and opportunities

Coronavirus Act 2020. LondonHer Majesty's Government

UK alcohol duty statistics

UK alcohol duty statistics

Anxiety sensitivity, distress tolerance, and discomfort intolerance in relation to coping and conformity motives for alcohol use and alcohol use problems among young adult drinkers

Released (2017) IBM SPSS Statistics for Windows, Version 25.0. Armonk, NYIBM Corp

Association of the Covid-19 lockdown with smoking, drinking, and attempts to quit in England: an analysis of 2019-2020 data

Time series analysis for psychological research: examining and forecasting change

Forecasting: methods and applications

Forecasting Techniques for Decision Making. Rockville, MDControl Data Management Institute

Immediate impact of minimum unit pricing on alcohol purchases in Scotland: controlled interrupted time series analysis for 2015-18

2020) Coronavirus and the social impacts on great Britain. London: Office for National Statistics

Mental health before and during the COVID-19 pandemic: a longitudinal probability sample survey of the UK population

The Lancet Gastroenterology & Hepatology (2020) Drinking alone. COVID-19, lockdown, and alcohol-related harm

Drinking to cope during the COVID-19 pandemic: the role of external and internal stressrelated factors in coping motive pathways to alcohol use, solitary drinking, and alcohol problems

We thank Kantar Worldpanel for providing the raw data. Kantar Worldpanel had no role in the study design, data analysis, data interpretation or writing of the manuscript, apart from reviewing the methods description as it pertains to the Kantar Worldpanel data. Professor Kaner is a National institute for Health Research (NIHR) Senior Investigator, and Director of the NIHR Applied Research Collaboration, North East and North Cumbria. Dr. O'Donnell is a National Institute for Health Research (NIHR) Advanced Fellow. The views expressed in this article are those of the authors and not necessarily those of NIHR, or the Department for Health and Social Care.

No funding was received in support of this study.

E.J.L., A.O.'D. and E.K. declare no competing interests and declare no financial relationships with any organizations that might have an interest in the submitted work in the previous three years. Within the previous three years, P.A. has received financial support from AB InBev Foundation outside the submitted work. All authors have completed the ICMJE uniform disclosure form at www. icmje.org/coi_disclosure.pdf and declare no support from any organization for the submitted work; all authors declare no other relationships or activities that could appear to have influenced the submitted work.

Not required.

Kantar Worldpanel data cannot be shared due to licensing restrictions.