MORNING DRINKING MEDIATION BETWEEN INSOMNIA AND DEPRESSION WITHIN INDIVIDUALS WHO DRINK: A MEDIATION ANALYSIS THROUGH THE FIRST YEAR OF THE COVID-19 PANDEMIC

Abstract

During the COVID-19 pandemic, many individuals increased their alcohol consumption to cope with stress. Excessive intake or drinking at inappropriate times can be important features of hazardous alcohol use. It is also known that alcohol can impair quality sleep. However, a potential relationship between aspects of hazardous alcohol use and insomnia during the pandemic has not been investigated. We hypothesized that there would be a significant correlation between morning drinking and insomnia. However, we also hypothesized that this relationship would be mediated by depression. A total of 13,313 individuals across the United States were surveyed (M=37.19 years old; SD ± 12.375). The online survey included self-report measures of depression, insomnia symptoms, and alcohol use. For this study, participants were asked specifically about their frequency of taking a drink first thing in the morning, which was assessed with an item from the Alcohol Use Disorders Identification Test (AUDIT). Insomnia was measured with the Insomnia Severity Index (ISI). Participants also completed the PHQ-9 Depression Scale. Morning drinking scores correlated with insomnia (r=.254, p<.001) and depression (r=.344, p<.001). A mediation model revealed a total effect of insomnia on morning drinking when depression was not accounted for (E=.0303, p<.0001). After including PHQ-9 Depression as a mediator, the direct effect was still significant (E=.0032, p=.0406), suggesting partial mediation. On the other hand, when insomnia was the mediator variable, the total effect and direct effect had no significant change (“total effect” E=.0414 p<.0001) (“direct effect” E=.0391 p<.0001), suggesting no mediating influence. These results suggest that the association between morning drinking and insomnia was partially mediated by depression during the pandemic. When depression was the independent variable and insomnia was the mediated variable, there was no longer efficiency in the mediation model. These results suggest that observed associations between insomnia and morning drinking during the pandemic were partially mediated by depression.