COVID-19, Coronavirus-Related Anxiety, and Changes in Women’s Alcohol Use

The coronavirus pandemic (COVID-19) has altered the way we live in fundamental ways: the way we work, interact with others, perform our daily routines, and engage in behaviors related to our health and well-being. Several studies suggest a surge in alcohol use since the outbreak. In a survey of 2,094 U.S. adults conducted in mid-April of 2020, 17% of respondents reported drinking more in the last few weeks compared to 11% who said they drank less [1]. Alcohol sales were 25% higher during the week ending April 4, 2020 compared to that same week in 2019, and liquor stores in large cities have reported substantial increases in sales [1,2]. Nationally, estimates indicate a 55% increase in retail alcohol sales between the second and third week of March 2020 [3]. At the same time, economic constraints as a result of the pandemic mean families have less discretionary income, and bar and restaurant closures and cancellation of sporting events provide fewer opportunities to drink outside the home.


Introduction
The coronavirus pandemic  has altered the way we live in fundamental ways: the way we work, interact with others, perform our daily routines, and engage in behaviors related to our health and well-being. Several studies suggest a surge in alcohol use since the outbreak. In a survey of 2,094 U.S. adults conducted in mid-April of 2020, 17% of respondents reported drinking more in the last few weeks compared to 11% who said they drank less [1]. Alcohol sales were 25% higher during the week ending April 4, 2020 compared to that same week in 2019, and liquor stores in large cities have reported substantial increases in sales [1,2]. Nationally, estimates indicate a 55% increase in retail alcohol sales between the second and third week of March 2020 [3]. At the same time, economic constraints as a result of the pandemic mean families have less discretionary income, and bar and restaurant closures and cancellation of sporting events provide fewer opportunities to drink outside the home.
The Centers for Disease Control (CDC) has suggested that people are using alcohol to cope with stress brought on by the pandemic [4]. In mid-April 2020, life satisfaction among Americans reached a 12-year low [5]. Potential stressors include unemployment, lost wages, inability to pay bills, and overall economic uncertainty [6]. People who have retained their jobs have had to adjust to working from home, and/or new social distancing and cleanliness standards at the workplace. Other contributing factors thought to be related to increased alcohol use are loneliness and social isolation, inability to exercise or take part in hobbies, and changes in work responsibilities. Parents, and women, are often juggling work and childcare in addition to caring for family members at high risk of COVID-19, with no clear idea of how long the situation will go on [7]. Such chronic ambiguity with respect to roles, relationships, and responsibilities is especially damaging to people's emotional and physical health [8]. A study of Australian women found drinking intentions and/or behavior were related to "having a stressful week" and because it "makes me feel relaxed" [9].
Women also make up half the American workforce and are disproportionately employed in occupations deemed essential, and it is widely recognized that essential workers are under an extreme amount of stress [10,11]. Stress, and work-family spillover, are associated with greater alcohol consumption among women [12,13]. Female workers in occupations not deemed essential are more likely than men to have experienced massive layoffs as a result of the pandemic [14]. Unemployment is associated with greater alcohol use and abuse [15][16][17]. Moreover, a study of 775 U.S. adults in mid-March of 2020 found anxiety specifically related to COVID-19 to be associated with greater alcohol use [18], and a study of women and men in Spain found that women experienced higher levels of depression, anxiety, PTSD, loneliness, and less spiritual well-being than did men during the COVID-19 lockdown [19].  [23,24]. Alcohol increases the risk of cancer, liver disease, dementia, depression, and suicide and the Global Burden of Disease suggests that no amount of alcohol is healthy [25][26][27][28][29]. Researchers are anticipating an increase in deaths of despair as a result of the coronavirus pandemic [30].
Although women drink less alcohol than do men [31][32][33], numerous studies show a convergence of women's and men's alcohol use in terms of prevalence, amount, and frequency, as well as alcohol-related problems and harms [34][35][36][37]. This convergence is being driven mainly by increases in alcohol use among women as opposed to declines among men [36]. Analysis of the National Epidemiological Survey on Alcohol and Related Conditions indicated that women experienced a 16% increase in past year alcohol use, a 58% increase in high-risk drinking, and an 84% increase in alcohol use disorder between 2002 and 2013 [37]. A meta-analysis of six U.S. surveys indicated a 10% increase in prevalence of alcohol use and 23% increase in binge drinking among women between 2000 and 2016, with no corresponding increase for men [38].
Women are more susceptible to alcohol-related health conditions than are men, as well as infertility, reproductive problems, and breast cancer [39][40][41]. Women have experienced the same plateau in life expectancy as men partly a result of increases in alcohol-related deaths and liver disease [42][43][44]. An analysis of U.S. death certificates filed between 1999 and 2017 showed an 85% increase in deaths due to alcohol among women compared to an increase of only 35% for men [24].
There have been recent calls for research on the effect of COVID-19 on alcohol consumption, harms, and policies [45,46]. The use and overuse of alcohol has increased substantially among women with negative effects on their health and emotional wellbeing. As the main caretakers of children, aging parents, and extended family members, women's use of alcohol can reverberate across the family system and between generations. One in ten children live with a parent who has an Alcohol Use Disorder (AUD); 1.4 million of those live with a single parent, most often with their mothers [47]. Alcohol is associated with child abuse, neglect, and interpersonal violence [48,49]. There is growing alarm as to how COVID-19 is affecting women with children, who are widely viewed as being under particularly stressful conditions since the pandemic. Jokes and memes about mothers' increased reliance on alcohol to cope with work, childcare, and managing their children's on-line education, have exploded on social media since the pandemic.
A developmental perspective on alcohol use stresses that "variability in drinking patterns is not constant across the lifespan, and that pressures to drink-or not to drink-are concentrated at certain stages in the course of a person's life" [50]. Indeed, a woman's use of alcohol over her lifetime or her drinking trajectory, is particularly sensitive to life course transitions. Some transitions are protective against alcohol use, including steady employment [51,52], marriage [53][54][55][56][57], and parenthood [54,[58][59][60]. Others, such as attending college and divorce, are associated with the greater use of alcohol (Chilcoat & Breslau 1996; [60]. Drinking tends to decline with age [55,61,62], although more recent studies suggest increased drinking among people in midlife and older, especially among women [37,38,63]. White, college-educated, and higher income women experienced the greatest increase in alcohol consumption in recent years and drink significantly more than do other women [34,[64][65][66]. Religiosity and organized religion are inversely related to alcohol use and alcohol-related problems [67,68]. The aim of this study is to assess changes in women's alcohol consumption since COVID-19, and to understand the role of coronavirus-related anxiety, and women's social and demographic characteristics. There is only one other similar study available currently. In a national study, Australian researchers found an increase in self-reported alcohol consumption among women and men during the pandemic (May 2020) than 2 to 3 years prior, but that the increase was substantially higher among women, and that having a child caring role was a strong predictor of an increase in alcohol consumption [69]. They did not examine the role of anxiety related to COVID-19.
Given large variation in rates of positive cases of COVID-19 and COVID-19-related hospitalizations and deaths across U.S. states, this study explores regional differences in these patterns. The survey contains a series alcohol use questions Delete from the National Institutes on Alcoholism and Alcohol Abuse adapted to capture drinking pre-and post-COVID-19, including frequency, number of drinks, and frequency of binge drinking [70], as well as the time of day alcohol is consumed and type of drink, and contains a broad range of sociodemographic variables. A coronavirus anxiety scale was designed by the researcher.

Materials and Methods
The study is based on an anonymous on-line Qualtrics survey of 546 women age 25 and older who reported drinking alcohol at least occasionally. Women under the age of 25 were excluded because they demonstrate a temporary upswing in drinking and binge drinking. Participants were recruited primarily through social media (Facebook, Twitter), but also by email, listserves, and newsletters. The survey was fielded between June 3 and June 30, 2020. Participants were directed to a website containing a consent form and a Qualtrics survey which they could access through their smartphones, computer, or other electronic device. Participants were encouraged to share the link to the study with others in their social network.
The main objective of the survey was to measure women's alcohol consumption pre-and post-COVID-19. Respondents were first asked about their current level of alcohol consumption. area, about how often do you have any kind of drink containing alcohol? By a drink, we mean half an ounce of absolute alcohol (e.g. a 12-ounce can or glass of beer or cooler, a 5-ounce glass of wine, or a drink containing 1 shot of liquor)? Responses ranged from every day to less than a few times a year. About how many alcoholic drinks would you say you have on a typical day when you drink alcohol? Responses ranged from one drink one drink to 12 or more drinks. Respondents were then asked about their frequency of binge drinking. About how often would you say you have 4 or more drinks containing any kind of alcohol in within a two-hour period? That would be the equivalent of at least 4 -12-ounce cans or bottles of beer, 4 -5-ounce glasses of wine, 4 drinks each containing one shot of liquor or spirits. Responses ranged from never to every day. Respondents were also asked about their typical pattern of alcohol consumption: On days you drink, when do you usually have your first drink? Responses included before lunchtime, around lunchtime, mid-afternoon, before dinner, with dinner, and after dinner or evening. Then, On days you drink, what type of drink do you have most often? Respondents could select from beer, wine, cocktails or mixed drinks, liquor or spirits (straight), and other. Finally, the women were asked, on days you drink, do you tend to switch from one kind of drink to another? Response categories were yes, I tend to switch from drinks with less alcohol to drinks with more, such as from beer to wine; yes, I tend to switch from drinks with more alcohol to drinks with less, such as from wine to beer; and, no, I tend to drink beverages with about the same amount of alcohol all day.
Respondents were then asked about their pre-COVID-19 level of drinking. Before COVID-19 and social distancing measures began in your area, would you say you drank quite a bit less than I do now, somewhat less than I do now, the same amount as I do now, somewhat more than I do now, or quite a bit more than I do now? Respondents who reported drinking "the same amount" were asked, if you drank the same amount as now, were there ever periods of time during COVID-19 and social distancing measures began in your area when you drank more or less alcohol than you did? Potential responses were yes, there were times when I drank more; yes, there were times when I drank less; and, no, I've been drinking about the same amount throughout the pandemic. Respondents who reported a change in their alcohol consumption answered the same series of alcohol consumption questions in relation to before COVID-19.
The Coronavirus Anxiety Scale, created for this study, contains 9 items designed to capture anxiety about COVID-19. Respondents were asked, since COVID-19 and social distancing measures began in your area, on a scale from 1 (never) to 5 (often), how often have you thought about the following: (a) me or a family member contracting the coronavirus, (b) me or a family member getting seriously ill or dying from the coronavirus, (c) others getting sick or dying from the coronavirus, (d) my personal finances and providing for my family, (e) the economy in general, (f) not being able to see friends and family, (g) the mental health and emotional well-being of friends and family, (h) my mental health and emotional well-being, and (i) the future of the country. The scale has a high level of reliability with a Cronbach's alpha of .83. The women in the sample exhibited a moderate to high level of anxiety with an average level of 3.6.
Respondents provided information on their social and economic characteristics. Age was recorded in years and is grouped in terms of the following: 25 to 29, 30 to 39, 40 to 49, 50 to 59, and 60 and older. Racial and ethnic identity is coded as White, Black/African American, Hispanic, Asian/Asian Indian, and other or more than one race. Due to small cell sizes, the latter categories were coded as non-White. Relationship status was coded as single, cohabiting, and married. Educational attainment is a threecategory measure: less than a bachelor's degree, bachelor's degree, to graduate or professional degree. Respondents were asked whether and how their employment situation has been affected by the pandemic: yes, I lost my job or was furloughed; yes, I work fewer hours; yes, I work more hours; yes, I work from home some or all of the time; or, no, my work life has stayed the same. Respondents reporting having lost their job, having been furloughed, or worked fewer or more hours were combined. Whether or not respondents had children was coded as a dichotomous variable (yes, no). Categories of religious affiliation included Catholic, Protestant, Jewish, Muslim, other religion, and no affiliation. Respondents recorded the region where they live: Northeast, Midwest, South, or West. Yearly gross household income in 2019 ranged from less than $25,000, $25,000 to $49,999, $50,000 to $74,999, $75,000 to $99,999, $100,000 to $149,000, $150,000 to $199,999, and $200,000 or more. Respondents with household incomes less than $75,000 were combined. The distribution of these variables can be found in Table 1. Table 2 describes women's perceptions of changes in their alcohol consumption since COVID-19. The results in Panel 1 indicate that women's alcohol consumption has increased since the beginning of the pandemic. Nearly two-thirds (64.5%) of women reported drinking quite a bit less (21.5%) or somewhat less (43.0%) before the pandemic than after. Only 13.0% of women reported drinking somewhat more or quite a bit more before COVID-19, and 23.4% reported drinking the same amount. Among those who reported drinking the same amount, 28.9% reported there were times since COVID-19 when they drank more, and 9.4% said there were times when they drank less. Panel 3 combines information from these two questions. Compared to pre-COVID-19, 15.2% of women reported ever drinking quite a bit less or somewhat less, 70.3% of women reported ever drinking quite a bit more or somewhat more, and 14.5% of women say they consistently drank the same amount pre-and post-COVID-19.  Note: Cells may not total to 100% due to rounding.  Table 3 examines changes in women's drinking in greater detail. First, there was a three-fold increase in daily drinking, from 4.6% pre-COVID-19 to 15.6% post-COVID-19. The volume of drinks consumed remained relatively consistent across time, although there was a slight decline in the percent of women who drank just one drink, from 32.4% to 26.9%. Women's frequency of binge drinking (drinking 4 or more drinks within a two-hour period) showed an interesting pattern with an increase in never (from 37.0% to 42.5%) and an increase in weekly (15.4% to 23.1%). There was also a shift in drinking from later in the day to earlier in the day. Pre-COVID-19, 4.4% of women reported having their first drink by mid-afternoon compared to 16.1% post-COVID-19. The type of alcohol consumed remained relatively consistent, with a slight increase in drinking cocktails, liquor, and beverages other than beer and wine. At both time points, most women drank beverages with the same amount of alcohol throughout the day (90.0% and 88.1%) as opposed to switching from one kind of drink to another. First, women who reported drinking more after the pandemic reported a significantly higher level of COVID-19-related anxiety (3.8) than did women who reported drinking the same amount (3.5) or less (3.6), although the difference in magnitude between groups is small. A higher percentage of younger women than older women reported drinking more since the pandemic. Women age 60 and older demonstrated the most stability in drinking; 53.6% reported drinking the same amount before and after COVID-19. A slightly higher percentage of non-White women reported drinking more. A significantly lower percentage of women with graduate or professional degrees reported drinking more (63.9%) compared to women with a bachelor's degree (74.9%) or less than a bachelor's degree (76.5%). Changes in employment were associated with changes in alcohol consumption. Women who lost their job, were furloughed, or who worked hours different than before the pandemic exhibited significantly less stable alcohol consumption before and after COVID-19 than did women whose work life remained the same. For example, only 7.0% in those There were no significant regional or income differences in changes in women's alcohol consumption.  Note: Cells may not total to 100% due to rounding. *Significant differences between groups at p < 05. Table 5 provides the results of a logistic regression, in the form of odds ratios, that assesses the independent effects of women's coronavirus-related anxiety and their sociodemographic characteristics on changes in women's alcohol consumption preand post-COVID-19. A test of the proportional odds assumption suggests that binary logistic regression (more versus same/ less) is preferred over a three-category ordered or multinomial logit model [71]. Model 1 indicates that women with higher coronavirus-related anxiety have higher odds of drinking more post-COVID-19 than pre-COVID-19. Every one-unit increase in anxiety is associated with 32.6% higher odds of drinking more after the pandemic as opposed to the same amount or less. Older women, namely women age 50 to 59 and sixty and older, have significantly lower odds (55.0% and 78.0%, respectively) of increasing their alcohol consumption compared to women in their twenties. Compared to women with less than a bachelor's degree, women with advanced degrees have 57.5% lower odds of increasing their alcohol consumption than women with less than a bachelor's degree and 48.5% lower odds compared to women with a bachelor's degree (results not shown). Catholic women have 62.4% lower odds of increased alcohol use compared to Protestant women. On the other hand, married women are over twice as likely to have increased their alcohol consumption compared to single women and cohabiting women (results not shown). Women's characteristics not associated with changes in alcohol use include race and ethnicity, changes in employment situation, region, and income. Additional analysis did not find a statistically significant relationship between employment status (full-time, part-time, and not employed), religious service attendance, and changes in women's alcohol use (results not shown). At baseline, having children was not associated with changes in drinking since the pandemic. However, the effect of children may be dependent upon women's relationships status. For example, the presumed stress of having a child could be compounded by traditional role responsibilities associated with marriage. Model 2 includes an interaction between relationship status and motherhood. Results indicate having children reduces the odds of increased alcohol consumption among married women by 69.3%. Children are not associated with changes in alcohol consumption among cohabiting or single women.

Discussion
Although women drink less alcohol than do men [31,32], numerous studies show that women's use of alcohol is catching up to that of men's in terms of prevalence, amount, and frequency, as well as alcohol-related illnesses and deaths [35]. The U.S. Department of Health and Human Services estimates there are 5.3 million women in the U.S. who are heavy drinkers or who "drink in a way that threatens their health, safety, and general wellbeing" (2018, p. 6). Alcohol-related cirrhosis increased 50% in women between 2009 and 2015 [43]. In an analysis of data from the Centers for Disease Control (CDC), nearly a third of women surveyed (23%) reported previous levels of alcohol use consistent with an alcohol use disorder, with 10% showing symptoms of an AUD in the past 12 months [32].
The results of this survey are consistent with studies suggesting a rise in alcohol use since COVID-19 and that anxiety about the pandemic plays a role, supporting anecdotal evidence and widespread media reports. Results regarding women's sociodemographic characteristics are for the most part in line with previous research on overall uasge. A lower proportion of older women than younger women increased their alcohol consumption. Religiosity and organized religion are generally associated with less alcohol use and alcohol-related problems [67,68], results indicate lower odds of increased drinking among only Catholic women. Although women with advanced degrees had lower odds of drinking more since the pandemic, they also had a higher level of alcohol consumption pre-COVID-19 (43% drank at least three to four times a week versus 28% of women with a bachelor's degree or less), consistent with prior work showing greater alcohol use among more educated women [55,65,72]. As noted above, race and ethnicity, changes in employment situation, employment status, region, and income, religious service attendance, were not associated with changes in women's alcohol use, in contrast to previous research indicating these are important variables to consider. However, previous research assessed these relationships with respect to current levels of drinking as opposed to this study which examines change over time.
The positive effect of marriage on increased alcohol consumption is interesting given that marriage has long been How to cite this article: Stewart considered a protective factor against substance use [37,53,54]. However, married women have been found to have higher rates of alcohol consumption than unpartnered women, at least among Whites [65]. Reczek et al. [57] examined the marital biographies of men and women from the Health and Retirement Study and found that marriage and remarriage reduced men's heavy alcohol use (3+ drinks at least one day each week) but increased women's and that divorce increased men's drinking but reduced women's. This work suggests that married women may be under greater stress since the pandemic, perhaps a result of having a husband at home more of the time. Gender roles and gender role attitudes are more traditional among married women than among cohabiting and single women, and married women may feel particularly responsible for "keeping things together" on the home front, even if working full time. The results may also be related to increased drinking among husbands since the pandemic, as women's drinking is highly sensitive to that of their spouse [57]. The effect of the pandemic on men's alcohol use is currently unknown with Research on the effect of the pandemic on men's alcohol use is just emerging. but financial strain and increased stress post-COVID-19 are certainly not exclusive to women. Results are also consistent with research showing lower levels of alcohol consumption among mothers, despite anecdotal evidence to the contrary [53,54,58].
It is important to emphasize that these results are not representative of the national population and are based on the experiences of a largely White, college-educated sample recruited primarily through social media. Nevertheless, this sample reflects women who have the highest levels of alcohol consumption relative to other women and who have experienced the greatest increase in alcohol use in recent years [34,64,66]. Women's reports are retrospective, and respondents may not accurately recall their behaviors and feelings before the pandemic, although the survey was constructed to minimize over-inflation of alcohol use since the pandemic by asking about their current usage prior to asking about previous levels. It is hoped that, in the absence of national studies, this study contributes to our understanding of changes in women's alcohol use as it relates to COVID-19. Studies conducted at the national level are needed, as are longitudinal assessments of women's alcohol use over time, and especially studies focusing on women of color, and less educated and lower income women.
Women's increased alcohol use since the pandemic is especially troubling given their level of alcohol use had already been on the increase. There are already many barriers for women getting treatment for alcohol overuse that prolong women's alcohol dependency and health risks. Guilt, shame, being perceived as a "bad mom," lack of childcare, the cost of treatment, and familial opposition, the lack of gender-specific treatment, physicians being slow to recognize AUDs in women, and for single mothers, the potential loss of custody [73][74][75][76]. Since the pandemic, many women have lost access to mental health services, substance use resources, and have been cut off from family and friends [30].
These findings demonstrate the importance of continued research on how COVID-19 is impacting women's alcohol use, factors underlying increased consumption, and the effect of drinking on women's social and emotional health and on the well-being of their families [77,78].