Abstract
Objective:
To estimate the extent to which drinking to cope with the COVID-19 pandemic and experiencing pandemic-related life stressors are associated with alcohol use escalation among young adults.
Methods:
Respondents in Los Angeles, CA, USA (N=2,130) completed prospective cohort study surveys before (baseline; October 2018-November 2019; mean age: 19.7[SD=0.4) and during (follow-up; May-August 2020) the COVID-19 outbreak. Past 30-day drinking days and number of drinks per drinking day were assessed from baseline to follow-up. At follow-up, participants reported drinking to cope with social isolation and pandemic-related stressors.
Results:
Pandemic-related stressor prevalence ranged from 5.5% (evicted/lost home) to 72.6% (worried about education) and 27.1% drank to cope with social isolation during the pandemic. Respondents who did (vs. did not) report pandemic-related coping drinking were more likely to increase past 30-day drinking days and drinks per drinking day from baseline to follow-up after adjustment for possible confounders. Employment loss/reduction, financial problems, and perceived likelihood of contracting COVID-19 or handling the pandemic poorly were each associated with increases in drinking days or drinks per drinking day.
Conclusions:
Experiencing certain life stressors and drinking to cope with social isolation may be associated with drinking escalation among young adults during the COVID-19 pandemic.
Keywords: COVID-19 pandemic, alcohol use, pandemic-related stressor, young adulthood
INTRODUCTION
Early adulthood (age 18-22 years) is a developmental period often marked by alcohol use escalation (Schulenberg et al., 2020). High alcohol use levels during young adulthood increase risks of alcohol use disorder and alcohol-related health consequences throughout the lifespan (Haber et al., 2016). Understanding the determinants of young adults’ alcohol use can inform alcohol use prevention, treatment, and policy that could substantially mitigate the alcohol-related public health burden.
Stress related to the economic and social consequences of the COVID-19 pandemic might intersect with typical developmental challenges of young adulthood, which could increase risk of alcohol use escalation. During young adulthood, people commonly traverse post-secondary educational environments, enter the workforce, expand their social network, move residences, and develop a vision for their long-term future (Arnett, 2019). During the pandemic, universities and other educational venues have closed, which may be a significant source of stress for many young adults (Sahu, 2020). Young adults in the workforce typically do not have stable careers and might be vulnerable to reduction or termination of employment and financial hardship during the pandemic (Lund et al., 2020). Consequently, views of future employment prospects might currently be bleak in many young adults. U.S. young adults have particularly high rates of SARS-CoV-2 infection (Centers for Disease Control and Prevention [CDC], 2020), some of whom may be concerned about contracting the virus. In addition to the detrimental effects of the abovementioned stressors, young adults’ alcohol use may increase during the pandemic because they may be drinking for the purpose of coping with the isolation and loneliness of social distancing. The COVID-19 pandemic combines stressors young adults experienced in other times in recent history, such as limited employment during economic recessions, with the unique experience of being instructed to socially isolate.
Cross-sectional evidence indicates that some young adults retrospectively report using substances more than they previously did in order to cope with their emotions during the COVID-19 pandemic (Czeisler et al., 2020). Another cross-sectional study found that college students’ alcohol consumption was higher after their university closed than drinking prior to school closure (Lechner, 2020). There is limited published prospective evidence on changes in young adult alcohol consumption before versus after the start of the COVID-19 global outbreak. One prospective study reported, on average, alcohol was consumed 1 day more per month by 3 of 4 adults (aged 30-80), and 1 day more of heavy drinking for 1 in 5 women (Pollard et al., 2020). Additional longitudinal research is warranted to examine whether increased drinking may be exacerbated by pandemic-related stressors and concerns among young adults.
The current study leveraged an ongoing prospective cohort study of young adults from Los Angeles, CA, USA that measured alcohol use before and after the COVID-19 outbreak, permitting prospective determination of increases in drinking levels. Whether young adults reported drinking to cope with social isolation during the pandemic or experienced several types of employment, education, residential, and financial life stressors caused by the pandemic were also assessed. The current study’s objective was to estimate the extent to which experiencing pandemic-related life stressors and drinking to cope with social isolation were associated with changes in alcohol consumption levels pre- vs. post-COVID-19 outbreak. Secondarily, we examined whether baseline drinking status moderated these associations in order to determine whether those that were already heavily drinking before the COVID-19 outbreak were more vulnerable to stressor-related alcohol use increases. Given that drinking to cope with the pandemic may indicate problematic drinking, we also conducted an additional cross-sectional analysis of associations of experiencing pandemic-related stressors with drinking to cope with the pandemic.
METHODS
Participants and Procedures
Participants were enrolled in an ongoing prospective cohort study on mental health and health behaviors originally recruited as students in 10 Southern California high schools in 2013 (Leventhal et al., 2015). The study sample includes data from participants who completed the two most recent survey waves conducted remotely via the Internet, labeled baseline (Median survey month[Range]=December 2018[October 2018-October 2019]) and follow-up (June 2020[May to August 2020]) assessments. Since the World Health Organization on March 11, 2020 has declared the COVID-19 outbreak a global pandemic (Cucinotta & Vanelli, 2020), the pre- and post-COVID-19 survey collection dates were outside of the beginning and official pandemic onset. Information on accrual and inclusion in this study’s analytic sample (N=2,130) depicted in sFigure 1 (see the Online Supplemental Material). All participants provided informed consent. The University of Southern California Internal Review Board approved the study.
Measures
Alcohol Use
Frequency and quantity.
At baseline and follow-up assessments, participants reported past 30-day number of days drinking at least 1 full drink (range=0-30) and number of standard drinks per drinking day (non-drinkers coded as ‘0 drinks’; Schulenberg et al., 2020). Pre-to-post COVID-19 outbreak change scores for both variables were computed (i.e., follow-up level subtracted by baseline level) and used as a continuous variable outcome in the primary analyses. Each observed change score was categorized into a binary variable, analyzed as outcomes in the supplemental analyses (i.e., increase[change score≥1] vs. stable[change score=0]/decrease[change score≤−1]).
Heavy, non-heavy, and non-drinker alcohol use status.
Using the Substance Abuse and Mental Health Services Administration (2017) definition of heavy drinking (≥5 days in past 30 days having drank ≥5 [males] or ≥4 [females] standard drinks in a row), past 30-day drinking status was classified as a trichotomous variable (i.e., non-drinkers [0 drinking days], non-heavy drinkers [≥1 drinking days without meeting heavy drinking criteria], heavy drinkers) at both baseline and follow-up assessments.
Experiences During COVID-19 Pandemic
Pandemic-related coping drinking.
At follow-up, the following item was administered, “To cope with social distancing and isolation are you doing any of the following? (check all that apply).” Participants who did (vs. did not) check the “drinking alcohol” response option were classified positive (vs. negative) for pandemic-related coping drinking.
Pandemic-related stress.
At follow-up, participants completed a survey module with this lead in text: “The following will assess how COVID-19 coronavirus has affected your daily life…” Subsequent survey items included; evicted from or lost home, lost job or work hours reduced, serious financial problems, and worried about future job prspects and impact on education (yes/no). Participants were queried on the item, “Have you had COVID-19 coronavirus?” (yes/maybe vs. no). Two continuous ratings were collected on 0-100 visual analogue scales: (a) “How likely do you think it is that you will contract COVID-19 coronavirus?” (‘No chance’ to ‘I will definitely get it’); and (b) “How well are you handling the COVID-19 coronavirus pandemic?” (‘I am handling it really well’ to ‘Not well at all’).
Covariates
A priori covariates were selected based on their associations with pandemic-related stressors and alcohol use (Czeisler et al., 2020; Lechner et al., 2021). Sociodemographic characteristics included baseline gender identity, sexual orientation, and race/ethnicity as well as education, living situation, financial situation, and age at baseline and follow-up (see Table 1 for response categories). At baseline and follow-up, behavioral health measures included the Generalized Anxiety Disorder scale (Chorpita et al., 2000) assessing past 2-week 7 anxiety symptoms (never[=0] to always[=3]; Cronbach’s αs=.93[baseline] and .91[follow-up]), the 10-item abbreviated Center for Epidemiologic Studies Depression Scale (Radloff, 1977) that assesses past-week frequency of 10 depressive symptoms each on 0-3 scales (αs=.79[baseline] and .77[follow-up]), and the 4-item UPPS-P impulsive behavior scale sensation seeking subscale (Cyders et al., 2014) measuring tendency to seek out novel or exciting experiences (strongly disagree[=0] to strongly agree[=3]; α=.82). Finally, across baseline and follow-up, 17 items measuring past 30-day use of non-alcohol substances were coded as three variables, indicating total numbers of cannabis products [e.g., combustible, vaporized cannabis], tobacco products [e.g., combustible cigarettes, vaping devices], and other substances [e.g., prescription painkillers, prescription stimulant pills].
Table 1.
Descriptive statistics for covariates in study sample
| Variables | N (%) or M (SD) |
|---|---|
| Baseline (Pre-Pandemic) | |
| Age, M (SD), y | 19.7 (0.4) |
| Gender identity, N (%) | |
| Female/Feminine | 1223 (59.2) |
| Male/Masculine | 792 (38.4) |
| Transgender/Non-binary/Other | 50 (2.4) |
| Sexual orientation, N (%) | |
| Straight | 1647 (80.3) |
| Bisexual | 173 (8.4) |
| Gay/Lesbian | 53 (2.6) |
| Othera | 177 (8.6) |
| Race/ethnicity, N (%) | |
| Hispanic | 978 (46.7) |
| Asian | 391 (18.7) |
| Black/African American | 100 (4.8) |
| Non-Hispanic White | 345 (16.5) |
| Otherb | 281 (13.4) |
| In 4-year college program (vs. other)c educational status, N (%) | 930 (48.2) |
| Live with parents/guardians (vs. other)d living situation, N (%) | 1425 (69.5) |
| Financial situation, N (%) | |
| Live comfortable | 932 (45.7) |
| Meet needs with a little left | 589 (28.9) |
| Meet basic expenses | 427 (20.9) |
| Do not meet basic expenses | 93 (4.6) |
| Sensation seeking score,e M (SD) | 6.06 (3.13) |
| Depressive symptoms score,f M (SD) | 10.16 (5.68) |
| Generalized anxiety symptoms score,g M (SD) | 5.84 (5.45) |
| Total no. of non-alcohol substances used in past 30 days | |
| Total no. of cannabis products use, N (%)h | 0.74 (1.22) |
| Total no. of tobacco products use, N (%)i | 0.50 (1.07) |
| Total no. of other substance use, N (%)j | 0.04 (0.23) |
| Follow-Up ( Mid-Pandemic) | |
| Age, M (SD), y | 21.2 (0.4) |
| Period surveyed during follow-up assessment period, N (%) | |
| May 2020 | 130 (6.3) |
| June 2020 | 1470 (70.8) |
| July or August 2020 | 476 (22.9) |
| In 4-year college program (vs. other)c educational status, N (%) | 971 (49.0) |
| Live with parents/guardians (vs. other)d living situation, N (%) | 1484 (74.9) |
| Financial situation, N (%) | |
| Live comfortable | 876 (44.4) |
| Meet needs with a little left | 610 (30.9) |
| Meet basic expenses | 423 (21.5) |
| Do not meet basic expenses | 63 (3.2) |
| Depressive symptoms score,f M (SD) | 00.18 (5.74) |
| Generalized anxiety symptoms score,g M (SD) | 5.54 (5.35) |
| Total no. of non-alcohol substances used in past 30 days | |
| Total no. of cannabis products use, N (%)h | 0.78 (1.24) |
| Total no. of tobacco products use, N (%)i | 0.33 (0.76) |
| Total no. of other substance use, N (%)j | 0.08 (0.34) |
N=2130; Participants with 2018-2019 (baseline) and 2020 (follow-up) survey data on study variables.
Other: 'Asexual', 'Pansexual', Queer', 'Questioning/unsure', or 'Other'
Other: ‘American Indian/Alaskan Native’, ‘Native Hawaiian/Pacific Islander’, or ‘Multiethnic/Multiracial’
Other: 'technical/vocational program', '2-year degree program', 'graduate/professional program', 'other type of degree program', or 'not enrolled in any education al program'
Other: 'in a dorm or other campus housing', 'In a fraternity/sorority house', 'Live alone', 'In an apartment/condo/house with a spouse/romantic partner, friends/roommates, or another relative/family friend', or 'Someplace else'
The 4-item sensation seeking subscale of the UPPS-P impulsive behavior scale (Sum score range=0-12). Not assessed at follow-up.
The 10-item abbreviated Center for Epidemiologic Studies Depression Scale (CESD; Sum score range=0-30).
The Generalized Anxiety Disorder 7-item scale (Sum score range=0-21).
Includes combustible cannabis, vaporized cannabis, edible cannabis, synthetic marijuana, and cannabis concentrate (Sum score range=0-5).
Includes combustible cigarettes, e-cigarettes with or without nicotine, other electronic vaping devices, cigars, cigarillos, and hookah (Sum score range=0-7).
Includes prescription painkillers, prescription stimulant pills, prescription sedatives, heroin, or ‘other’ drugs (Sum score range=0-5).
Statistical Analyses
Preliminary analyses calculated descriptive statistics of study covariates and alcohol use variables among the overall sample and by baseline heavy, non-heavy, or non-drinking alcohol use status. Primary analyses tested univariable linear regression models adjusting for all covariates that estimated the association of each pandemic-related experience variable with changes in past 30-day drinking outcomes from pre-to-post COVID-19 outbreak. Each model included a single pandemic-related regressor; separate models were tested for drinking day change and drinks per drinking day change outcomes. Each statistically significant association in the primary analyses was followed by moderation analyses that added the multiplicative pandemic-related experience (follow-up; mid-pandemic) × baseline (pre-pandemic) drinking status (i.e., non-drinker [reference], non-heavy drinker, heavy drinker) interaction term. For significant interaction terms, post-hoc analyses tested associations stratified by baseline drinking status. Additionally, cross-sectional binary logistic regression analyses were conducted to test whether pandemic-related stressors were associated with drinking to cope with social isolation during the pandemic. Regression model results are reported as unstandardized regression coefficients (Bs) for continuous outcomes and odds ratios (ORs) for binary outcomes with 95% confidence intervals (95%CIs). Analyses were conducted in Mplus 8 (Muthén & Muthén, 1998). Statistical significance was set at .05 (two-tailed). Full information maximum likelihood estimation was used to account for missing covariate data.
RESULTS
Descriptive Results
The analytic sample included 2,130 participants who took the follow-up survey and reported alcohol use data at both baseline and follow-up assessments (sFigure 1 details participant accrual and analytic sample exclusions). Depicted in Table 1, the sample was sociodemographically heterogenous, and most respondents (70.8%) completed the follow-up survey in June 2020; the median duration between baseline and follow-up was 552 (Range=508-589) days.
Presented in Table 2, the average number of past 30-day drinking days increased from baseline (M[SD]=2.21[4.31]) to follow-up (M[SD]=3.47[5.08]), with 39.0% of participants demonstrating an increase in past-month number of drinking days from baseline to follow-up. Number of drinks per drinking day increased from baseline to follow-up (M[SD]=0.62[3.16]). Overall, 27.1% of overall participants reported drinking alcohol to cope with the pandemic. Pandemic-related coping drinking was more common in baseline heavy drinkers (56.2%) compared to baseline non-heavy drinkers (39.2%) and non-drinkers (15.8%).
Table 2.
Descriptive results on alcohol use and drinking to cope with the pandemic, by baseline drinking status
| Alcohol-Related Variables | Baseline Past 30-Day Drinking Status | |||
|---|---|---|---|---|
| Overall Sample (N=2130) |
No Drinkinga (N=1232) |
Non-Heavy Drinkingb (N=719) |
Heavy Drinkingc (N=162) |
|
| M (SD) / N (%) | M (SD) / N (%) | M (SD) / N (%) | M (SD) / N (%) | |
| Alcohol Use Frequency in Past 30 days | ||||
| No. days, M(SD) | ||||
| Baseline | 2.21 (4.31) | 0.00 (0.00) | 3.65 (3.52) | 12.24 (6.15) |
| Follow-up | 3.47 (5.08) | 2.03 (3.89) | 4.82 (5.10) | 8.29 (7.70) |
| Change from baseline to follow-up | 1.29 (5.50) | 2.03 (3.89) | 1.18 (6.07) | −4.03 (9.42) |
| Directional of change baseline to follow-up, N(%)d | ||||
| Decrease | 309 (15.0%) | 0 (0.0%) | 219 (31.4%) | 85 (55.9%) |
| Stable | 951 (46.1%) | 720 (60.0%) | 187 (26.8%) | 39 (25.7%) |
| Increase | 805 (39.0%) | 481 (40.0%) | 291 (41.8%) | 28 (18.4%) |
| Drinks per Drinking Day in Past 30 Days e | ||||
| No. drinks, M(SD) | ||||
| Baseline | 1.66 (2.67) | 0.00 (0.00) | 3.29 (2.46) | 6.99 (2.26) |
| Follow-up | 2.26 (3.01) | 1.38 (2.45) | 3.20 (3.10) | 4.64 (3.72) |
| Change from baseline to follow-up | 0.62 (3.16) | 1.38 (2.45) | −0.06 (3.53) | −2.34 (3.97) |
| Directional of change baseline to follow-up, N(%)d | ||||
| Decrease | 401 (20.0%) | 0 (0.0%) | 308 (45.4%) | 93 (62.4%) |
| Stable | 878 (43.7%) | 720 (61.0%) | 129 (19.0%) | 28 (18.8%) |
| Increase | 730 (36.3%) | 461 (39.0%) | 241 (35.5%) | 28 (18.8%) |
| Drinking to Cope with Pandemic, N(%) f | 577 (27.1%) | 195 (15.8%) | 282 (39.2%) | 91 (56.2%) |
Alcohol use on 0 days in past 30 days.
Alcohol use on ≥1 days in past 30 days without meeting definition of heavy drinking (see superscript 'd').
Binge drinking (5 or more alcoholic drinks for males or 4 or more alcoholic drinks for females) on the same occasion on ≥5 days in the past 30 days.
Column percent is shown.
Participants that reported alcohol use on 0 days in past 30 days were coded as "0".
To cope with social distancing and isolation during the pandemic are you drinking alcohol? (yes/no).
Primary Results
Descriptive data and adjusted regression model results for pandemic-related stressors and alcohol-related outcomes are depicted in Table 3. Prevalence of pandemic-related stressful experiences ranged from 5.5% (evicted/lost home) to 72.6% (worried about education) among the overall sample. Respondents who did (vs. did not) report using alcohol to cope with pandemic were more likely to have increased number of drinking days in the past 30 days (B[95%CI]=1.97[0.91-3.28]) and drinks per drinking day (B[95%CI]=1.09[0.68-1.49]) from baseline to follow-up. Additionally, four pandemic-related stressor and concern variables were significantly associated with one or more alcohol use change outcomes after adjustment for time-invariant and time-varying covariates (Table 3). Employment loss/reduction during the pandemic was significantly associated with increases in past 30-day number of drinking days (B[95%CI]=0.72[0.23, 1.20]) and drinks per drinking day (B[95%CI]=0.47[0.17, 0.76]) from baseline to follow-up. Having serious financial problems was positively associated with increases in drinking days from baseline to follow-up (B[95%CI]=0.60[0.09, 1.11]). Each one standard deviation unit higher on perceived likelihood of contracting COVID-19 (B[95%CI]=0.39[0.13, 0.64]) and handling pandemic poorly (B[95%CI]=0.29[0.07, 0.51]) was significanlty associated with increasing past-month drinking days from baseline to follow-up.
Table 3.
Associations of pandemic-related experiences with quantitative outcomes of changes in alcohol use levels
| Regressors (Follow-Up; Mid-Pandemic) |
Descriptives in Overall Samplea,b |
Changes in Past 30-Day No. Drinking Daysa,c,d |
Changes in Past 30-Day No. Drinks per Drinking Daya,c,e |
||
|---|---|---|---|---|---|
|
|
|
||||
| N (%) / M (SD) | B (95% CI) | P | B (95% CI) | P | |
| Binary Exposures (Yes vs. No) | |||||
| Drinking to cope with pandemicf | 577 (27.1) | 1.97 (0.91, 3.28) | <.001 | 1.09 (0.68, 1.49) | <.001 |
| Evicted from home or lost home | 114 (5.5) | −0.36 (−1.55, 0.82) | .37 | −0.33 (−1.05, 0.40) | .37 |
| Lost job or work hours reduced | 923 (44.8) | 0.72 (0.23, 1.20) | .004 | 0.47 (0.17, 0.76) | .002 |
| Serious financial problems | 438 (21.3) | 0.60 (0.09, 1.11) | .03 | 0.06 (−0.35, 0.48) | .77 |
| Worried about future job opportunitiesg | 1379 (66.6) | −0.27 (−0.82, 0.25) | .29 | −0.24 (−0.57, 0.08) | .14 |
| Worried about impact on educationg | 1500 (72.6) | 0.02 (−0.63, 0.67) | .95 | 0.09 (−0.31, 0.48) | .67 |
| Possibly had COVID-19 | 126 (5.9) | 0.04 (−0.80, 0.92) | .93 | 0.45 (−0.17, 1.08) | .15 |
| Continuos Exposures (SD Units) h | |||||
| Perceived likelihood of getting the virusi | 36.2 (24.2) | 0.39 (0.13, 0.64) | .003 | 0.08 (−0.07, 0.24) | .29 |
| Not handling pandemic wellj | 37.1 (24.1) | 0.29 (0.07, 0.51) | .03 | 0.11 (−0.07, 0.29) | .23 |
N=2130; Participants with 2018-2019 (baseline) and 2020 (follow-up) survey data on study variables
Column % of "Yes" response for binary exposures and M (SD) for continuous exposures.
Adjusted univariable linear regression models with separate models for each pandemic-related regressor, controlling for all variables presented in Table 1 and baseline alcohol use frequency/quantity (B=unstandardized regression coefficient, 95%CI=95% confidence interval).
Change from baseline to follow-up (Range = −30 - 30).
Change from baseline to follow-up (Range = −10 - 10).
Drinking alcohol to cope with social distancing and isolation during the pandemic.
Response was coded by Yes ('strongly agree' or 'agree') or No ('strongly disagree' or 'disgree')
SD units mean regressors standardized, such that each increase in one SD is associated with the respective change in outcome.
How likely do you think it is that you will contract COVID-19 coronavirus? (0-100 visual analogue scale, with ‘no chance’ to ‘I will definitely get it’ as anchors).
How well are you handling the COVID-19 coronavirus pandemic? (0-100 visual analogue scale, with 'I am handling it really well' to ‘Not well at all’).
Moderation by Baseline Drinking Status
Some of the significant associations identified in the primary results were moderated by baseline drinking status (sTable 1). Follow-up stratified tests were conducted for significant interactions, which illustrate that the magnitude of associations between pandemic-related coping drinking and alcohol use escalation outcomes were amplified in baseline non-drinkers. For example, baseline non-drinkers, compared to baseline non-heavy drinkers and heavy drinkers, showed significantly stronger association between drinking to cope with pandemic and increases in drinking days and drinks per drinking day. Also, only baseline heavy drinkers had significant association of financial problems with increased number of drinking days, which was significantly stronger compared to baseline non-drinkers (Table 4).
Table 4.
Associations of pandemic-related experiences with increases in alcohol use outcomes, stratified by baseline drinking status
| Regressors (Follow-Up; Mid-Pandemic) |
Changes in Past 30-Day Drinking from Baseline to Follow-Upa,b | |||
|---|---|---|---|---|
| No. Drinking Daysc |
No. Drinks per Drinking Dayd |
|||
| B (95% CI) | P | B (95% CI) | P | |
| Binary Regressors (Yes vs. No) | ||||
| Drinking to cope with pandemic (follow-up) | ||||
| Baseline non-drinker | 2.25 (1.81, 2.98) | <.001 | 1.69 (1.32, 2.65) | <.001 |
| Baseline Non-heavy drinker | 1.05 (0.72, 1.76)* | <.001 | 0.77 (0.35, 1.37)* | <.001 |
| Baseline Heavy drinker | 1.48 (1.23, 1.97)* | <.001 | 1.03 (0.74, 1.60)* | <.001 |
| Serious financial problems (follow-up) | ||||
| Baseline non-drinker | 0.31 (−0.20, 0.83) | .18 | N/A | N/A |
| Baseline Non-heavy drinker | 0.60 (−0.16, 1.37) | .09 | N/A | N/A |
| Baseline Heavy drinker | 0.98 (0.48, 2.03)* | .01 | N/A | N/A |
N=2130
Adjusted univariable linear regression models with separate models for each pandemic-related regressor, adjusted for all variables in Table 1 and baseline drinking days and drinks per drinking day (B=unstandardized regression coefficient, 95%CI=95% confidence interval, N/A=not applicable).
Change from baseline to follow-up (Range = −30 - 30).
Change from baseline to follow-up (Range = −10 - 10).
B in respective group is significantly different (P<.05) than corresponding B in non-drinkers based on formal interaction tests in sTable 1.
Associations of Pandemic-Related Experiences with Drinking to Cope with Pandemic
Additional cross-sectional analyses of follow-up data identified associations between pandemic-related experiences and drinking to cope with pandemic (Table 5). Consistent with the primary results, we detected significant associations between four pandemic-related stressors (i.e., employment loss/reduction, financial problems, perceived likelihood of contracting COVID-19, handling the pandemic poorly) and increased odds of drinking to cope with social isolation during the pandemic. Additionally, concern about the pandemic’s impact on education was positively associated with drinking to cope with the pandemic (OR[95%CI]=1.30[1.01, 1.69]). Additional supplemental analyses (e.g., binary drinking outcome, potential effects of age transition) are presented in the Online Supplemental Material.
Table 5.
Associations of pandemic-related experiences with drinking to cope with the pandemic at follow-up
| Regressors (Follow-Up; Mid-Pandemic) | Drinking to Cope with Pandemic (Follow-Up; Yes vs. No)a,b,c |
||
|---|---|---|---|
| N (Row %)d | OR (95% CI) | P | |
| Categorical Variables, N (%) | |||
| Evicted from or lost home | |||
| No | 543 (27.7) | Reference | – |
| Yes | 27 (23.7) | 0.82 (0.38, 1.76) | .61 |
| Lost job or work hours reduced | |||
| No | 266 (23.4) | Reference | |
| Yes | 301 (32.6) | 1.52 (1.12, 2.06) | .008 |
| Serious financial problems | |||
| No | 420 (26.0) | Reference | – |
| Yes | 147 (33.6) | 1.32 (1.02, 1.75) | .03 |
| Worried about future job prospectse | |||
| No | 179 (25.9) | Reference | – |
| Yes | 392 (28.4) | 0.88 (0.58, 1.15) | .25 |
| Worried about impact on educatione | |||
| No | 138 (24.3) | Reference | – |
| Yes | 432 (28.8 | 1.30 (1.01, 1.69) | .04 |
| Possibly had COVID-19 | |||
| No | 540 (27.0) | Reference | – |
| Yes/Maybe | 37 (29.4) | 1.05 (0.54, 1.98) | .91 |
| Continuous Variables, M (SD) | |||
| Perceived likelihood of getting the virusf,h | N/A | 1.11 (1.01, 1.25) | .04 |
| Not handling the pandemic wellg,h | N/A | 1.30 (1.07, 1.58) | .008 |
N=2130
Drinking alcohol to cope with social distancing and isolation during the pandemic.
Adjusted univariable binary logistic regression models with separate models for each pandemic-related regressor, controlling for all variables presented in Table 1 and baseline alcohol use frequency/quantity (OR=odds ratio, 95%CI=95% confidence interval, N/A=not applicable).
Percentage reporting drinking to cope with pandemic by respective pandemic-related stressor.
Response was coded by Yes ('strongly agree' or 'agree') or No ('strongly disagree' or 'disgree').
How likely do you think it is that you will contract COVID-19 coronavirus? (0-100 visual analogue scale, with ‘no chance’ to ‘I will definitely get it’ as anchors).
How well are you handling the COVID-19 coronavirus pandemic? (0-100 visual analogue scale, with 'I am handling it really well' to ‘Not well at all’).
SD units mean regressors standardized, such that each increase in one SD is associated with the respective change in odds of the outcome.
DISCUSSION
To our knowledge, this is the first longitudinal study of associations between pandemic-related experiences and alcohol use escalation among young adults using prospective data collected before and after the COVID-19 outbreak. This study advances the field beyond previous studies on this topic that analyzed cross-sectional data after the start of the pandemic (Wardell et al., 2020) and prospective associations of changes in alcohol use with alcohol-related problems among an older sample of adults (Pollard et al., 2020). Here, we show that young adults experiencing certain life stressors during the pandemic were at modestly greater levels of drinking escalation relative to their pre-pandemic levels. Even though the associations were of modest magnitude, the high base rates of experiencing these life stressors during the pandemic raised the possibility that the population health impact on the young adult alcohol use burden might be appreciable.
The associations with alcohol use escalation were not consistent across each type of pandemic-related life stressor. Some pandemic-related stressors pertained to aspects of lifestyle security (e.g., employment status, education) or social functioning that may be especially relevant to young adults (Arnett, 2019), and were associated with alcohol use escalation. Young adults who lost their job or had to take a reduction in pay during the pandemic were at greater levels of increasing their drinking frequency and their drinks per drinking day. Moreover, young adults who reported drinking to cope with pandemic-related social isolation were more likely to have increased their drinking frequency and quantity since pandemic onset. Cross-sectional studies with retrospective reports of pre-pandemic drinking behavior have identified increased drinking frequency and quantity (Prestigiacomo et al., 2021). Additionally, a study of college students identified associations between loneliness, drinking to cope, and past-month alcohol use frequency during the pandemic (Mohr et al., 2021). This study provides prospective evidence of an association between pandemic isolation-related coping motives and increased alcohol use among young adults.
It is possible that experiencing pandemic-related stressors and drinking to cope with the pandemic are non-causal epiphenomena of developmental trajectories of escalating alcohol consumption among 19-to-21-year-old adults. It is plausible that people who are liable to report experiencing stressors during the COVID-19 crisis would have escalated their drinking levels (i.e., frequency and quantity) regardless of the pandemic. However, we accounted for this possibility by controlling for a host of potential sociodemographic and time-varying behavioral confounding factors. Furthermore, the null results of inter-survey-interval moderation sensitivity analyses also do not support the explanation that the associations between pandemic-related experiences and alcohol use escalation in our primary results are accounted for by confounding influences. If these pandemic-related experiences were a proxy for long-standing liability to greater escalating drinking trajectories, the associations with drinking level escalations would presumably be more robust among individuals with longer inter-survey intervals with more time for their alcohol use trajectories to mature.
There was considerable heterogeneity in drinking patterns from baseline to follow-up. On average, baseline heavy drinkers, compared to non-drinkers and non-heavy drinkers, demonstrated decreases in drinking frequency and quantity after the onset of the pandemic. Whether this reflects a conscious attempt to control drinking behavior during the pandemic (Cadigan et al., 2015), a regression toward the mean effect, or some other process related to decreased socialization is unknown. Regardless, baseline heavy drinkers were also relatively more likely to report drinking to cope with the social isolation of the pandemic, which was associated with how this group felt they were handling the pandemic in supplemental analyses (sTable 2). Heavy, possibly problematic, young adult drinkers may have been using alcohol as a coping mechanism prior to the pandemic, consistent with previous literature (Neighbors et al., 2007), and then also utilized drinking as a coping strategy to deal with social isolation during the pandemic. These results concur with the cross-sectional structural modeling results obtained by Wardell and colleagues among middle adulthood Canadian drinkers (Wardell et al., 2020), which showed that stress and negative affect coping motives within the context of the pandemic were each associated with alcohol consumption and drinking problems (Wardell et al., 2020). It should be noted that the drinking to cope with the pandemic was more strongly associated with alcohol use escalation among baseline non-drinkers (vs. non-heavy/heavy drinkers) who transitioned to drinkers at follow-up (Table 5). Further follow-up of the long-term alcohol use trajectories of young adults and the role of coping drinking is warranted.
This study has limitations. First, this study is limited by the self-reported nature of survey data, which may be subject to reporting error. Also, the analytic sample was from a single urban geographic region. Although the sample was sociodemographically heterogenous, it was in an urban location with high rates of COVID-19 (CDC, 2020b) with unclear generalizability to rural regions or locations with lower infection rates. Third, the psychometric properties of the pandemic-related stressors and drinking coping measures are unclear. The use of single-item indicators, while reducing participant survey burden, tend to have poorer reliability than multi-item measures. Finally, there were only two waves of data, the first wave tapping behaviors up to over a year prior to the COVID-19 pandemic. Long-term follow-up research is necessary.
CONCLUSIONS
In conclusion, this prospective study of individuals from Los Angeles, CA aging through young adulthood during the COVID-19 crisis found that pandemic-related life stressors and drinking to cope with the isolation and loneliness of social distancing were common and, in some cases, associated with increases in alcohol use escalation. Interventions addressing relevant life stressors experienced currently by young adults warrant consideration in efforts to mitigate the long-term effects of the COVID-19 pandemic on the alcohol-related public health burden.
Supplementary Material
ACKNOWLEDGEMENT
Funding: Research reported in this publication was supported by the National Cancer Institute (NCI) of the National Institutes of Health (NIH) under Award Number R01CA229617 (Barrington-Trimis/Leventhal) and by the National Institute on Drug Abuse (NIDA) under Award Number K24DA048160. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Role of Funder:
The funding agency had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Disclosures of Interest: None of the authors report a conflict of interest related to submission of this manuscript.
Access to Data and Data Analysis: JC had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
REFERENCE
- Arnett JJ (2019). Conceptual foundations of emerging adulthood. In: Murrary JLA (Ed.). Emerging Adulthood and Higher Education: A New Student Development Paradigm. New York, NY: Routledge. [Google Scholar]
- Cadigan JM, Martens MP, & Herman KC (2015). A latent profile analysis of drinking motives among heavy drinking college students. Addictive Behaviors, 51, 100–105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Disease Control and Prevention. (2020a). Coronavirus Disease 2019 (COVID-19): CDC COVID Data Tracker. Demographic Trends of COVID-19 cases and deaths in the US reported to CDC. Accessed on August 31, 2020. [Google Scholar]
- Centers for Disease Control and Prevention. (2020b). LA County Daily COVID-19 Data 2020. Available from: http://publichealth.lacounty.gov/media/coronavirus/data/index.htm.
- Chorpita BF, Yim L, Moffitt C, Umemoto LA, & Francis SE (2000). Assessment of symptoms of DSM-IV anxiety and depression in children: A revised child anxiety and depression scale. Behaviour Research and Therapy, 38(8), 835–855. [DOI] [PubMed] [Google Scholar]
- Cucinotta D, & Vanelli M (2020). WHO Declares COVID-19 a Pandemic. Acta Bio Medica, 19, 157–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cyders MA, Littlefield AK, Coffey S, & Karyadi KA (2014). Examination of a short English version of the UPPS-P Impulsive Behavior Scale. Addictive Behaviors, 39(9), 1372–1376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Czeisler MÉ, Lane RI, Petrosky E, Wiley JF, Christensen A, Njai R, & Rajaratnam SM (2020). Mental health, substance use, and suicidal ideation during the COVID-19 pandemic—United States, June 24–30, 2020. Morbidity and Mortality Weekly Report, 69(32), 1049–1057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haber JR, Harris-Olenak B, Burroughs T, & Jacob T (2016). Residual effects: young adult diagnostic drinking predicts late-life health outcomes. Journal of Studies on Alcohol and Drugs, 77(6), 859–867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lechner WV, Laurene KR, Patel S, Anderson M, Grega C, & Kenne DR (2020). Changes in alcohol use as a function of psychological distress and social support following COVID-19 related University closings. Addictive Behaviors, 110, 106527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lechner WV, Sidhu NK, Jin JT, Kittaneh AA, Laurene KR, & Kenne DR (2021). Increases in risky drinking during the COVID-19 pandemic assessed via longitudinal cohort design: Associations with racial tensions, financial distress, psychological distress and virus-related fears. Alcohol and Alcoholism. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Leventhal AM, Strong DR, Kirkpatrick MG, Unger JB, Sussman S, Riggs NR, & Audrain-McGovern J (2015). Association of electronic cigarette use with initiation of combustible tobacco product smoking in early adolescence. JAMA, 314(7), 700–707. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lund S, Ellingrud K, Hancock B, Manyika J (2020). COVID-19 and jobs: Monitoring the US impact on people and places. McKinsey & Company; 2020. [Google Scholar]
- Mohr CD, Umemoto SK, Rounds TW, Bouleh P, & Arpin SN (2021). Drinking to cope in the COVID-19 era: An investigation among college students. Journal of Studies on Alcohol and Drugs, 82(2), 178–187. [PubMed] [Google Scholar]
- Muthén LK, & Muthén BO (1998). Mplus: The comprehensive modeling program for applied researchers: User’s guide. Muthén & Muthén. Los Angeles, CA. [Google Scholar]
- Neighbors C, Lee CM, Lewis MA, Fossos N, & Larimer ME (2007). Are social norms the best predictor of outcomes among heavy-drinking college students?. Journal of Studies on Alcohol and Drugs, 68(4), 556–565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pollard MS, Tucker JS, & Green HD (2020). Changes in adult alcohol use and consequences during the COVID-19 pandemic in the US. JAMA Network Open, 3, e2022942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Prestigiacomo CJ, Liu MA, Plawecki MH, & Cyders MA (2021). Early impact of the US COVID-19 pandemic on drinking motives and alcohol use. Substance Use & Misuse, 1–4. [DOI] [PubMed] [Google Scholar]
- Radloff LS (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401. [Google Scholar]
- Sahu P (2020). Closure of universities due to coronavirus disease 2019 (COVID-19): Impact on education and mental health of students and academic staff. Cureus,12(4), e7541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schulenberg JE, Johnston LD, O'Malley PM, Bachman JG, Miech RA, & Patrick ME (2020). Monitoring the Future national survey results on drug use, 1975-2019: Volume II, college students and adults ages (pp. 19–60). Ann Arbor, MI: Institute for Social Research, The University of Michigan. [Google Scholar]
- Substance Abuse and Mental Health Services Administration. (2017). Key substance use and mental health indicators in the United States: Results from the 2016 National Survey on Drug Use and Health (HHS Publication No. SMA 17-5044, NSDUH Series H-52). Rockville, MD: Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration. Retrieved from https://www.samhsa.gov/data/ [Google Scholar]
- Wardell JD, Kempe T, Rapinda KK, Single A, Bilevicius E, Frohlich JR,& Keough MT (2020). Drinking to cope during COVID-19 pandemic: The role of external and internal factors in coping motive pathways to alcohol use, solitary drinking, and alcohol problems. Alcoholism: Clinical and Experimental Research, 44(10), 2073–2083. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.