key: cord-0717542-jilvhbqi
authors: Guignard, Romain; Andler, Raphaël; Quatremère, Guillemette; Pasquereau, Anne; du Roscoät, Enguerrand; Arwidson, Pierre; Berlin, Ivan; Nguyen-Thanh, Viêt
title: Changes in smoking and alcohol consumption during COVID-19-related lockdown: A cross-sectional study in France
date: 2021-04-07
journal: Eur J Public Health
DOI: 10.1093/eurpub/ckab054
sha: b3026b08fa9007738496b934a407d32f776f7b9a
doc_id: 717542
cord_uid: jilvhbqi

BACKGROUND: In many countries, lockdown measures were implemented to curb the COVID-19 pandemic. This situation may have an impact on mental health, tobacco smoking, and alcohol consumption. The aim of this research report is therefore to describe changes in tobacco and alcohol consumption in the general French population during the first two weeks of lockdown and identify any associated factors. METHODS: Self-reported changes in smoking and alcohol consumption following the lockdown implemented in France on March 17, 2020 were collected from 2,003 respondents aged 18 years and older in an online cross-sectional survey carried out from March 30 to April 1, 2020. Anxiety and depression levels were assessed using the Hospital Anxiety and Depression Scale. RESULTS: Among current smokers, 26.7% reported an increase in their tobacco consumption since lockdown and 18.6% reported a decrease, while it remained stable for 54.7%. The increase in tobacco consumption was associated with an age of 18-34 years, a high level of education, and anxiety. Among alcohol drinkers, 10.7% reported an increase in their alcohol consumption since lockdown and 24.4% reported a decrease, while it remained stable for 64.8%. The increase in alcohol consumption was associated with an age of 18-49 years, living in cities of more than 100,000 inhabitants, a high socio-professional category, and a depressive mood. CONCLUSION: The national lockdown implemented in France during the COVID-19 pandemic influenced tobacco and alcohol consumption in different ways according to sociodemographic group and mental health.

Tobacco smoking and alcohol use are among the leading preventable lifestyle risks for both premature death (i.e., 7.1 million for tobacco smoking and 2.8 million for alcohol use in 2017) and disease burden worldwide (1) . While the impact of tobacco smoking on SARS-CoV-2 infection remains largely unknown (2, 3) , recent research suggests that it could increase disease severity and mortality in infected patients (4) (5) (6) . A recent meta-analysis of 19 studies, including a total of 11,590 confirmed COVID- 19 patients, showed that current smoking was associated with a twofold increase in the risk of disease progression (7).

In January 2020, the World Health Organization (WHO) declared that the COVID-19 outbreak caused by the novel SARS-CoV-2 coronavirus was an international public health emergency with a strong risk of spreading to many countries around the world. The outbreak was subsequently reclassified as a pandemic on March 11, 2020 . In many countries, including France, lockdown measures were implemented to curb the COVID-19 pandemic, with drastic limitations imposed on travel and social interactions, including for work. The lockdown was introduced in France from March 17 to May 11, coverage of the crisis, and so on. Previous research showed that anxiety and depressive mood are associated with tobacco smoking (11) (12) (13) (14) and alcohol use disorders (15, 16) , as both substances are used to cope with stress or break boredom, for example. Furthermore, it has been shown that economic crises like the 2008 financial recession and unemployment are associated with a deterioration in mental health (17) , an increase in deaths from alcohol abuse (18) , and an increase in binge drinking (19) . Finally, the epidemiological context itself and the perceived risk of infection with a respiratory disease could have motivated some smokers to quit or reduce their consumption, while others may have been encouraged to increase their consumption or relapse after the media reported scientific debates about the potential protective effect of nicotine against COVID-19 infection.

Previous research conducted in the first weeks of the crisis showed that substance use in the general population has been impacted by the epidemic and the subsequent political measures, including the lockdown setting (20) (21) (22) . In this epidemic context and specifically during lockdown, the role played by lifestyle and socioeconomic status in behavioral changes linked to tobacco and alcohol use should be studied. Such studies are useful to adapt and target smoking cessation and alcohol use reduction programs and devices in the event that a second wave of the epidemic occurs and/or new population control measures are implemented. The purpose of this report is therefore to describe changes in tobacco and alcohol consumption during lockdown in the general French population and explore the sociodemographic and mental health factors associated with these changes.

Study design and sample Self-reported evolution of tobacco and alcohol use following the complete lockdown in France on March 17, 2020 was collected through an online survey carried out from March 30 to April 1, 2020, i.e., 2 weeks after the start of lockdown. A national sample of 2,003 respondents aged 18 years and older living in Metropolitan France (i.e. excluding overseas departments) was selected by the private company BVA Institute from a panel of internet users using the quota method for the following variables: sex, age, occupation, region, and size of urban area, reflecting the structure of the population from the 2016 French National Institute for Statistics and Economic Studies (INSEE) census (23). While the sample was similar to the general population structure according to quota variables, it differed from it in terms of level of education, as 44% of respondents reported being college graduate compared to 30% in the general population. Besides, the proportion of smokers (21.2%) and alcohol drinkers (66.7%) in the study sample was lower than that previously recorded in representative surveys in France (respectively, 28.7% among 18-85 years old in 2019 and 86.5% among 18-75 years old in 2017) (24, 25) .

The study was funded by Santé publique France, the French National Public Health Agency.

Smoking status at the time of the survey was collected with the question: "Do you currently smoke?" with three response options: "Yes, I smoke cigarettes, including roll-your-own cigarettes / Yes, but only other kinds of tobacco (cigars, pipes, waterpipes…) / No, I do not smoke". Respondents who reported that they currently smoke manufactured cigarettes, roll-your-own tobacco, cigars, pipes, or waterpipes were classified as smokers. Smokers were then asked about the number of cigarettes smoked per day (for cigarette users) as well as the evolution of their consumption with the question: "Compared to pre-lockdown, has your smoking consumption changed?" Three possible responses were possible: "It has increased / it has remained stable / it has decreased." Cigarette smokers who reported an increase were also asked about the number of cigarettes smoked per day before lockdown, which was dichotomized into less than 10 cigarettes per day versus 10 or more cigarettes per day.

The evolution of alcohol consumption was directly reported for the entire sample with the question:

"Compared to pre-lockdown, has your alcohol consumption, including beer, wine, cider, spirits, champagne, or other types of alcohol, changed?" Four possible responses were possible: "It has increased / it has remained stable / it has decreased / I never drink alcohol." Respondents who did not answer they never drank alcohol were considered as current drinkers and were asked their current frequency of use as well as the number of drinks per day using the two first questions of the AUDIT-C (26) : "How often do you currently drink alcoholic beverages, be it beer, wine, cider, hard liquor, champagne, or any other type of alcohol?" (daily / 4 to 6 times a week / 2 to 3 times a week / once a week / 1 or several times a month / less often / never) and "Currently, on the days you drink, how many drinks containing alcohol do you have, be it beer, wine, cider, hard liquor, champagne or any other type of alcohol?" Respondents who reported an increase in alcohol consumption were asked the same two questions for the pre-lockdown period, introducing the questions by "And before the lockdown…" Regular alcohol use was defined as at least 2 to 3 times a week; less frequent use was defined as non-regular use.

Sociodemographic variables used in the analyses were: sex, age group, size of urban area, education level, living alone, parent of a child aged 16 years or younger, socio-professional category, perceived financial situation, and overcrowded housing.

Size of urban area was recoded from the postal code indicated by the respondents for their place of residence, using data from INSEE.

Education level was obtained with the question: "What was the last degree you obtained?", and recoded in "less than high school graduate / high school graduate / college graduate".

The interviewee's occupation was categorized according to INSEE socio-professional classification as follows: 1) farmers (0.9% of the sample); 2) craftspeople, retailers, business owners (4.9%); 3) managers, senior-level professional occupations (20.8%); 4) professional workers (22.3%); 5) clerical workers (25.1%); 6) manual workers (14.0%); 7) inactive persons (12.0%). For the present study, categories 1 to 4 were classified together as "high socio-professional category," while categories 5 and 6 were grouped under "low socio-professional category." Retired respondents were coded as their last occupation category.

Perceived financial situation was assessed with the question: "How would you describe the current financial situation of your household/family?" Five possible answers were provided: "1) I'm comfortable; 2) I manage easily; 3) I need to be careful; 4) I manage with difficulty; 5) I cannot survive without debt (27). The first two answers were rated as "good financial situation" and the last two as "difficult financial situation," along with "need to be careful."

Overcrowded housing was defined as living in a house or apartment measuring less than 25 m2 for one person living alone or less than 18 m2 per person in other cases following the definition of INSEE (28).

Anxiety and depression levels were assessed using the 14-item Hospital Anxiety and Depression Scale (HADS), with seven 4-point (0-3) items for each of anxiety and depression and scores ranging from 0 to 21. A score between 8 and 10 indicates possible anxiety or depression, while a score above 10 suggests probable anxiety or depression (29) .

Sociodemographic characteristics, anxiety and depression levels, smoking intensity and alcohol use frequency were compared between males and females using Pearson chi-square tests. Bivariate analyses between reported changes in tobacco and alcohol consumption, sociodemographic variables, and anxiety and depression levels were conducted. Post-stratification weights were computed for the adjustment of percentages across the structure of the whole population from the 2016 INSEE census (23). Pearson chi-square tests of independence were used to compare the reported changes in tobacco and alcohol use according to sociodemographic and mental health variables. Every difference reported in the text corresponds to a significant difference at the 5% level.

Two multinomial logistic regressions were performed to disentangle the effects of the explanatory variables significantly associated with changes in tobacco or alcohol use at the 5% level in bivariate analyses (0: stable consumption; 1: increased consumption; 2: decreased consumption). Adjusted relative risk ratios (aRRR) as well as the 95% confidence intervals (95% CI) were computed.

The hypothesis according to which "light" consumers are mechanically more likely to have increased their consumption can be made. Thus, in sensitivity analysis, the reported increase in tobacco and alcohol consumption (compared with stable consumption) was cross-tabulated with the pre-lockdown number of cigarettes smoked and the pre-lockdown frequency of alcohol use.

All statistical analyses were performed using Stata SE, version 14.

The sociodemographic characteristics, anxiety and depression levels, smoking intensity and alcohol use frequency of the whole sample (n=2,003), males (n=954), and females (n=1,049) are shown in Table 1 . Females included in the study were younger than males (p<.001), were more often inactive or in a low socio-professional category (p<.001), had a higher level of anxiety (<.001) or depression (p=.026), and were less frequently regular alcohol users (p<.001). There was no significant difference between males and females on smoking status, however male smokers smoked slightly more cigarettes per day (12.6, standard deviation (SD) = 10.1, n=178) than female smokers (10.1, SD=7.0, n=196) (p=.008).

-Insert Table 1 - The sociodemographic characteristics, anxiety and depression levels, evolution of tobacco and alcohol consumption in smokers (n=422) and alcohol drinkers (n=1,344) are shown in Table 2 . Anxiety was probable for 29.0% of smokers and 20.3% of alcohol drinkers, while 24.6% of smokers and 17.4% of alcohol drinkers had probable depression according to the HADS.

Among current smokers, 26.7% reported an increase in their tobacco consumption since the beginning of lockdown (i.e., during the 2 weeks before the survey) and 18.6% reported a decrease, while it remained stable for 54.7%. Those reporting an increase in tobacco consumption since lockdown smoked 5.4 additional cigarettes per day on average (SD=4.7, n=93).

Among alcohol drinkers, 10.7% reported an increase in their alcohol consumption since the beginning of lockdown and 24.4% reported a decrease, while it remained stable for 64.8%. Among alcohol drinkers who reported an increase in consumption (n=143), 74.2% reported more frequent use and 33.5% an increase in the number of drinks per day of consumption (23.1% reported both changes).

Alcohol drinkers reporting an increase in the number of drinks per day of consumption drank 1.9 additional drinks per day on average (SD=1.2, n=49).

-Insert Table 2 Table 3 . Increased tobacco consumption was most frequent among individuals aged 18-34 years, college graduates, respondents living in overcrowded housing, those with high anxiety levels or depressed mood.

Decreased tobacco consumption was most frequent among individuals aged 65 years and older or high school graduates.

Bivariate analyses for alcohol use are shown in Table 4 . Increased alcohol consumption was most frequent among people aged under 50 years, people living in an urban area with more than 100,000 inhabitants, respondents living in overcrowded housing, parents of children aged 16 years or younger, respondents with high anxiety levels or depressive mood.

Decreased alcohol consumption was most frequently reported by [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] year-olds, people living in an urban area with more than 100,000 inhabitants, and respondents living in overcrowded housing.

-Insert Tables 3 and 4 

Among cigarette smokers who reported higher or stable tobacco consumption (n=303), there was no significant difference in the proportion of those who increased their consumption in terms of the number of cigarettes smoked before lockdown: 34.1% of smokers of less than 10 cigarettes per day reported an increase in consumption vs 31.7% of those smoking at least 10 cigarettes (p=.669).

Among alcohol drinkers who reported higher or stable alcohol consumption (n=1,021), there was no significant difference in the proportion of those who increased their consumption in terms of their frequency of use before lockdown: 14.6% of non-regular alcohol drinkers reported an increase in consumption vs 13.8% of regular alcohol drinkers (p=.720).

In France, lockdown during the COVID-19 pandemic has influenced tobacco and alcohol consumption.

In the present study, even if the consumption of both remained stable for the majority of respondents, a quarter of smokers reported higher tobacco consumption, with this proportion being similar to that observed in a previous study of a population of workers in the US (20) . Changes in alcohol consumption were more favorable, with around one in ten drinkers reporting higher consumption and one in five lower consumption, which could be due to the restrictions placed on social interactions. The results of the present study in France differ from previous research conducted in other countries, especially concerning alcohol use, for which more negative changes have been observed elsewhere (21, 22) . Such differences may be due to cultural factors and drinking patterns in different countries, as well as the political measures implemented to manage the epidemic, not to mention the methodological differences in the sampling and recruitment strategy that could lead to various profiles of respondents.

Increased tobacco consumption was associated with higher anxiety levels, age of 18-34 years, and a higher level of education (i.e., college graduates), while decreased tobacco consumption was associated with an intermediate level of education (i.e., high school graduates). Compared to stable consumption, increased alcohol consumption was associated with depressive mood, age of less than 50 years, living in urban areas, and high socio-professional category. Being aged under 50 years and living in urban areas were also associated with a decrease in alcohol consumption, which shows the heterogeneous behaviors within sociodemographic groups. Associations between mental health and smoking and alcohol use are in line with the available knowledge (11, 15, 16) as well as previous research on the impact of the COVID-19 epidemic (22) . However, the present study conducted during the first weeks of the French lockdown suggests a closer relationship between tobacco and anxiety on the one hand, and alcohol use and depression on the other hand. Moreover, the observed increase in tobacco and alcohol use appears to be more frequent among young people, city dwellers, and individuals from a high socio-professional category or with a high level of education. This cannot be fully explained by lower levels of substance use before lockdown, since the increase in tobacco and alcohol use was not related to pre-lockdown consumption levels in the sensitivity analysis. This could partly be due to the inequalities that emerged regarding working conditions: for the "essential workers" who mostly belonged to the lower socioeconomic groups and continued to travel to work, the lockdown may have had a more limited impact on their everyday lives (30) . On the contrary, some workers in the more advantaged groups may have benefited from remote working, leading to more dramatic changes in their lifestyle habits, and perhaps more opportunities to smoke and/or drink alcohol at home compared with the workplace. While remaining cautious about this assumption, this could have led to more unfavorable changes in smoking and alcohol behaviors among the most socially advantaged groups, compared with less advantaged ones, excluding those from the most vulnerable populations (e.g., homeless, incarcerated, or illicit drug users) that are usually not included in internet surveys (31, 32) .

Further research is needed to confirm these findings, including the role of specific mental health disorders as well as working conditions: a previous study suggested that telecommuting could be beneficial for health, including smoking and alcohol use, in a non-pandemic context (33) . Further studies should therefore assess the effect of lockdown on post-lockdown tobacco and alcohol use in order to detect whether the observed changes persist in the long term.

The limitations of the present analysis include the use of a non-probabilistic sample and the crosssectional design of the study. The sample included respondents with a high level of education compared with the general population. The limited statistical power of this research report did not allow us to study the relations with working conditions, in particular telecommuting. As no prelockdown assessment could be conducted, the validity of the data may have been influenced by recall bias. Finally, it is likely that the smokers and especially alcohol drinkers most at risk were not included in the survey, as such individuals are usually underrepresented in general population surveys (34, 35) .

Further, the coverage error could be larger when the sample is drawn from an access panel (36) . Thus, the proportion of smokers and alcohol drinkers in the study sample is lower than that previously recorded in representative surveys in France (24, 25) .

The strengths of the study include the joint analysis of sociodemographic and mental health factors associated with behavioral changes. Moreover, the analysis of factors associated with both higher and lower smoking and alcohol consumption can be used as a lever to better target the populations most at risk of adopting harmful practices but also encourage users who adopted positive behaviors to pursue their efforts. The initial trends observed in smoking habits were used to justify maintaining a smoking prevention campaign scheduled on World No Tobacco Day on May 31, 2020. Public health messages intended for smokers and at-risk alcohol drinkers were adapted, encouraging them to remotely use help services to stop smoking or reduce drinking in the context of the general lockdown.

Even more so than usual, the use of a benevolent and supportive tone was necessary. In this specific context in which anxiety and stress related to the health and social situation reinforced consumption, it was crucial not to make consumers feel guilty. These study results could be used again in the event of subsequent waves of the epidemic and/or a new lockdown in order to adapt the targeting and contents of smoking cessation and alcohol use reduction programs and devices.

As shown in the present study, the COVID-19-related lockdown in France gave rise to important changes in tobacco and alcohol consumption among the general population, with the trend being more unfavorable for tobacco than for alcohol. Changes are associated with factors such as anxiety and depression as well as age, socio-professional category, and place of residence. Researchers are encouraged to continue monitoring smoking and alcohol use beyond the lockdown period through longitudinal studies in order to assess whether these changes are lasting or not. This could lead to exploring the relationships between lifestyle factors, mental health, and substance use in the context of the COVID-19 pandemic. The impact of working conditions could also be better assessed using a cohort of workers. The characteristics of individuals who increased their alcohol consumption should be studied in greater detail. These results should finally encourage health professionals and decisionmakers to continue promoting smoking cessation and alcohol reduction during lockdown periods and providing telephone or online help services at a time when face-to-face assistance services may be less accessible.

This study was funded by Santé publique France, the National Public Health Agency.

survey field and exploratory analyses.

We declare that we have no competing interests. IB received occasional honoraria for presentations at meetings from Pfizer Ltd in the last 3 years that were not related to this article. All authors declare no financial links with tobacco or alcohol companies, gaming industry or e-cigarette manufacturers or their representatives.

 In many countries, lockdown measures were implemented to curb the COVID-19 pandemic.

This situation may have impacted tobacco smoking and alcohol consumption.

 In France, among current smokers, 26.7% reported an increase in tobacco consumption two weeks after lockdown and 18.6% reported a decrease.

 Among alcohol drinkers, 10.7% reported an increase in alcohol consumption since lockdown and 24.4% reported a decrease.

 Changes are associated with factors such as anxiety and depression as well as age, socioprofessional category, and place of residence.

 Researchers are encouraged to continue monitoring smoking and alcohol use beyond the lockdown period through longitudinal studies in order to assess whether these changes are lasting or not. 

Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks for 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study

Smoking habit and hospitalization for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-related pneumonia: The unsolved paradox behind the evidence

Risk factors for SARS-CoV-2 among patients in the Oxford Royal College of General Practitioners Research and Surveillance Centre primary care network: a cross-sectional study

COVID-19 and smoking: A systematic review of the evidence

Risk factors of critical & mortal COVID-19 cases: A systematic literature review and meta-analysis. The Journal of infection

Severity and Mortality associated with COPD and Smoking in patients with COVID-19: A Rapid Systematic Review and Meta-Analysis

Epidemic of COVID-19 in China and associated Psychological Problems. Asian journal of psychiatry

The psychological impact of quarantine and how to reduce it: rapid review of the evidence

A nationwide survey of psychological distress among Chinese people in the COVID-19 epidemic: implications and policy recommendations. General psychiatry

The Association of Cigarette Smoking With Depression and Anxiety: A Systematic Review

Cigarette smoking and depression comorbidity: systematic review and proposed theoretical model

Cigarette smoking among those with mental disorders in the US population: 2012-2013 update

Alcohol and depression

Comorbidity between substance use disorders and psychiatric conditions

The correlation between stress and economic crisis: a systematic review

The public health effect of economic crises and alternative policy responses in Europe: an empirical analysis

Alcohol Use During the Great Recession of

Change in tobacco and electronic cigarette use and motivation to quit in response to COVID-19. Nicotine & tobacco research : official journal of the Society for

Dietary Choices and Habits during COVID-19 Lockdown: Experience from Poland

Depression, Anxiety and Stress during COVID-19: Associations with Changes in Physical Activity, Sleep, Tobacco and Alcohol Use in Australian Adults

Tobacco use among adults : Five-year review of the National Tobacco Control Programme

Epub 2020/04/03. La consommation d'alcool en France métropolitaine en

Gender differences in the associations of gambling activities and suicidal behaviors with problem gambling in a nationally representative French sample

The Hospital Anxiety And Depression Scale. Health and quality of life outcomes

COVID-19 exacerbating inequalities in the US

Vulnerable Populations: Weathering the Pandemic Storm. American journal of preventive medicine

Characterizing the Impact of COVID-19 on Men Who Have Sex with Men Across the United States in April, 2020. AIDS and behavior

The Effects of Telecommuting Intensity on Employee Health

Adjustment for survey non-representativeness using record-linkage: refined estimates of alcohol consumption by deprivation in Scotland

A novel use of record linkage: resolving non-representativeness in health surveys and improving alcohol consumption estimates to inform strategy evaluation

Van Den Eijnden R. The utility of online panel surveys versus computer-assisted interviews in obtaining substance-use prevalence estimates in the Netherlands

Probable (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) 430 21