Globally, anxiety disorders are the most common psychiatric disorders, with an estimated 301 million individuals experiencing one or more anxiety disorders in their lifetime [1]. Alcohol use is also highly prevalent, with an estimated 32.5% of people aged 15 years or older globally being current drinkers [2]. Among drinkers, high rates of risky drinking and alcohol use disorders are seen. A large nationally representative survey of 36,309 adults in the United States found 12.9% of individuals drink at high-risk levels, defined as 4+ standard drinks for women and 5+ standard drinks for men, at least weekly [3]. Additionally, among those who drink, 17.5% experienced an alcohol use disorder in the past year [3]. Comorbidity between mental health and substance use concerns is common, most notably between anxiety and alcohol use. Indeed, among individuals with an anxiety disorder, it is estimated that 20%-40% will also experience an alcohol use disorder in their lifetime [4]. When anxiety and alcohol use co-occur, symptom severity can be worse, and treatment adherence and outcomes are typically poorer compared to individuals with an anxiety or alcohol disorder alone [5,6].

Several theories have been proposed to explain the comorbid, bidirectional relationship between anxiety and alcohol use. Arguably, the most common is the self-medication model [7,8] which suggests individuals drink alcohol to alleviate symptoms of anxiety or general negative affect, in turn leading to a reliance on alcohol over time and the development of a subsequent alcohol use disorder. The use of alcohol for self-medication also aligns with other theories, including the tension-reduction theory [9] and the stress response dampening model [10]. Conversely, there is the opposite causal explanation that alcohol misuse and alcohol use disorders promote anxiety disorders, known as the substance-induced anxiety model or kindling or stress hypothesis [6,11,12]. Following prolonged periods of heavy alcohol consumption, withdrawal or cessation can lead to adaptations in the brain and wider nervous system, which can induce or worsen anxiety [6,11]. In addition to evidence of causal, bidirectional pathways between anxiety and alcohol, there is also support for a mutual-maintenance model also referred to as the “vicious cycle of comorbidity” [13]. This model demonstrates how the biopsychosocial outcomes of one disorder (eg, anxiety) will often serve to maintain or exacerbate the other disorder (eg, alcohol use disorder). An example of such a relationship can be seen when an individual with social anxiety who is self-medicating with alcohol develops a reliance on alcohol and over time experiences alcohol-related consequences like decreased productivity, interpersonal issues, and anxiety-induced withdrawal. These consequences exacerbate the initial anxiety and result in further drinking, thus maintaining the problematic feed-forward cycle [6,13]. Together, these theoretical frameworks suggest a complex bidirectional relationship between anxiety and alcohol use, with longitudinal and prospective studies confirming that experiencing an anxiety disorder or an alcohol use disorder can increase the risk of developing the other [14,15].

Previous critical and systematic reviews have established a strong body of evidence for a relationship between anxiety and alcohol use [12,16]. These reviews have typically focused on the temporal sequencing of the anxiety-alcohol relationship [16,17]; comorbidity between a specific anxiety disorder, for example, social anxiety disorder, and alcohol use [5,18-21]; the treatment of comorbid anxiety and alcohol use disorders [22-25]; or all of the above [12,26]. Yet, despite substantial existing research, the specific mechanisms driving anxiety and alcohol comorbidity are still poorly understood. The proposed systematic review will be the first to synthesize and clarify the specific mediators and moderators driving the relationship between anxiety and alcohol use.

Given the bidirectional nature of comorbid anxiety and alcohol use, the planned review will investigate the relationship in both directions. Figures 1 and 2 are path diagrams visually demonstrating the relationships that will be examined. Figure 1 illustrates anxiety as the exposure/predictor variable (X) and alcohol use as the outcome variable (Y), while Figure 2 illustrates the relationship in the opposite direction, where alcohol use is the exposure/predictor (X) and anxiety is the outcome (Y). As illustrated in Figures 1 and 2, a mediator or moderator is a third variable that affects the relationship between the exposure and outcome of interest [27]. Investigating mediators is warranted to better understand causal mechanisms that account for a relationship between X and Y [28]. In short, by identifying mediators, the proposed review will highlight factors explaining how or why the association between anxiety and alcohol, or alcohol and anxiety, occurs. If modifiable, these mediators can then be targeted in prevention, early intervention, and treatment approaches. To get a complete understanding of the anxiety and alcohol relationship, investigating the role of moderating variables is also necessary. The synthesis of moderators will highlight factors that affect the direction or strength of the relationship between X and Y [27]. As such, moderators can help understand when the relationship between anxiety and alcohol or alcohol and anxiety will occur. Given the breadth of possible mediating and moderating mechanisms studies can investigate, the review may identify mechanisms that can act as both a mediator and a moderator in the anxiety-alcohol relationship. An example of such a factor is positive alcohol use expectancies, with previous research identifying it as both a mediator and moderator in the relationships of interest [29-31].

To our knowledge, this will be the first study to systematically identify and synthesize research on mediators and moderators of the comorbid anxiety and alcohol relationship. The outcomes of the review will specifically answer the following questions:

This systematic review is currently underway. In September 2022, initial database searches identified 7330 unique studies for title and abstract screening. It is anticipated that the review, including data extraction and synthesis, will be completed by December 2023. Upon completion, results will be reported in line with PRISMA and disseminated in a peer-reviewed journal. Before submission, searches will be rerun to capture any additional articles that have been published since the initial database searches.

This paper summarizes the protocol for a systematic review and prospective meta-analysis on mediators and moderators in the co-occurring anxiety and alcohol relationship. The review will incorporate studies from a wide spectrum of anxiety and alcohol classifications, including from anxiety symptoms to anxiety disorders and from alcohol use to alcohol use disorders. The biopsychosocial model will be used to synthesize identified factors to determine the distinct roles of biological, psychological, and social mechanisms.

Existing studies often examine a single mediator or moderator, despite evidence of multiple mechanisms contributing to the alcohol and anxiety relationship [12]. In light of this, this review will be the first to consolidate the literature to date on mediators and moderators driving this common comorbidity. Identified mediators and moderators can then be used as targets to develop and refine prevention, early intervention, and treatment approaches for comorbid anxiety and alcohol use concerns. In some cases, the identified mediating and moderating factors may not be modifiable in interventions; nonetheless, synthesizing this evidence will help advance the theoretical basis underlying this common comorbidity. As this review will focus on both the pathway from anxiety to alcohol use and the pathway from alcohol use to anxiety, findings may suggest different prevention and intervention approaches are needed depending on the temporal sequencing of the two disorders. One limitation, however, of the planned review is the inclusion of cohort and cross-sectional studies, which will limit the ability to draw causal conclusions for all results. The inclusion of these study designs is warranted nonetheless, as preliminary searches indicated most mediation and moderation studies for anxiety and alcohol were not longitudinal.

The findings of this review will offer critical insights into the conceptualization of comorbid anxiety and alcohol use, providing a foundation for researchers, clinicians, and policy makers to enhance outcomes in addressing this complex clinical manifestation. These results have the potential to drive advancements in understanding, prevention, and treatment strategies, thereby positively impacting individuals experiencing anxiety and alcohol use concerns.