Nearly 60% of college students use alcohol and 30% binge drink monthly. This is alarming given that heavy alcohol use is linked to serious detrimental outcomes. Despite various prevention and intervention strategies, heavy alcohol use has remained relatively stable over the past decade. Individual differences in stress response connote risk for alcohol use disorder. Anxiety sensitivity (AS) and intolerance of uncertainty (IU) are two key cognitive vulnerabilities that can hinder resilience by amplifying stress responses and promoting maladaptive coping strategies, such as alcohol use. Effective stress management is a cornerstone of resilience. The Intervention for Managing Psychological Responding to Overwhelming Emotions (IMPROVE) targets AS and IU, key barriers to resilience, by modifying cognitive processes that amplify stress and negative affect. In this study, undergraduate students who engage in heavy drinking behaviors and experience elevated anxiety symptoms will be randomized to IMPROVE or a control health promotion intervention (N=20 per arm). All participants will complete daily ecological momentary assessments (EMA) delivered to participants' mobile phones to capture real-world alcohol use before, during, and after the intervention. The investigators will evaluate the feasibility and acceptability of IMPROVE (Aim 1). The investigators will also include a multimodal battery of self-report and objective lab-based measures of AS and IU involving startle eyeblink potentiation and event-related potentials via electromyography (EMG) and electroencephalography (EEG). This will allow the investigators to examine whether IMPROVE changes IU and AS, and to assess if changes in these targets are associated with changes in alcohol use (Aim 2).

SIGNIFICANCE AND BACKGROUND A.1. What is the overarching problem? Heavy alcohol use is prevalent on college campuses across the US. According to recent data, nearly 60% of college students use alcohol, and 30% binge drink monthly (NSDUH, 2022). These rates are alarming given that heavy alcohol use is associated with academic difficulties, injuries, violence, suicide, risky sexual behaviors, and chronic and infectious diseases (CDC, 2021). Each year, over 1,500 college students die from alcohol-related unintentional injuries. The risky drinking patterns that are developed during college can persist for years after graduation (Arria et al., 2016). Although a mixture of prevention and intervention efforts have been adopted by universities to lower college drinking rates, heavy alcohol use in students has remained relatively stable over the past decade. Novel, innovative approaches are urgently needed to ameliorate heavy drinking and associated harmful consequences on college campuses.

A.2. Stress and problematic drinking. Converging lines of research indicate that individual differences in stress responding connote risk for the development of alcohol use disorder (AUD) across the developmental spectrum. Broadly, chronic and acute stressors increase AUD risk, frequency of heavy drinking, and worsen treatment outcomes. In college students, stress is a robust proximal antecedent to alcohol use (Aldridge-Gerry et al., 2011; Park et al., 2002). Studies utilizing ecological momentary assessments (EMA) have demonstrated that students are more likely to drink alcohol on days they endorse more perceived stress (CITE). Meanwhile, in the laboratory, exposure to acute psychological stress increases single-session alcohol consumption (CITE).

These studies are notable because stress is incredibly common amongst undergraduates. Nearly 80% of college students report ongoing moderate stress and an additional 10-12% report severe stress (Saleh et al., 2017). Perceived stress is a risk factor for heavy drinking as well as a variety of other mental and physical health conditions (CITE). Managing stress is therefore critical to subjective and objective well-being.

A.3. How can we target stress? Stress is ubiquitous and it is nearly impossible to reduce stress exposure without structural changes at the family, community, and university levels. Therefore, stress itself may not be the most viable target for changing trajectories of problematic alcohol use. A more practical and effective strategy is to intervene on the cognitive individual difference factors that shape responses to perceived stress. Indeed, this concept forms the basis of the Catastrophizing, Anxiety, Negative Urgency, and Expectancy (CANUE) model of alcohol use16. CANUE is a testable framework that was developed by Ferguson and colleagues to facilitate the development of behavioral and psychological interventions that target the processes that contribute to stress-related substance use. CANUE recognizes that alcohol use is often motivated by distress because of alcohol's acute stress-dampening effects. As attempts to escape stress are acutely reinforced, they become entrenched and alcohol use emerges as a primary coping strategy17. These processes set the stage for the onset of AUD. What is unique about the CANUE model is that there are several modifiable individual difference factors that are theorized to play an important role in self-medicating perceived stress with alcohol.

Examples of theoretically relevant moderators include distress-related attitudes and beliefs such as anxiety sensitivity (AS) and intolerance of uncertainty (IU). AS is defined as the tendency to interpret anxiety-related bodily sensations as indicative of impending danger or harm33 whereas IU is the propensity to respond to uncertain events and situations negatively34. AS and IU are implicated early in the temporal unfolding of the stress-to-alcohol use chain of events and influence the intensity of the negative emotional state that stems from perceived distress. In other words, IU and AS are involved in amplifying the intensity of negative affect in response to stressful stimuli and independently and synergistically interact to drive the negative reinforcement processes underlying stress-related alcohol use and onset of AUD.

A.4. Resilience versus Risk. AS and IU develop early in life and are shaped by genetic, familial, and environmental influences. Most importantly, they significantly affect the long-term course of mental health symptoms. Longitudinal studies in adolescents and young adults have shown that both AS and IU predict increases in heavy drinking and onset of AUD diagnoses (DeMartini & Carey, 2011; MORE). In college students, specifically, both MPI Gorka and Allan have demonstrated that AS/IU influence subsequent alcohol use behaviors (Gorka et el., 2023; MORE). While AS and IU are often labeled as 'risk factors,' they are dimensional constructs that can range from adaptive to maladaptive. High levels of AS/IU may increase risk, whereas low levels may promote resilience. Resilience is a complex concept that represents an individual's capacity to cope with stress and overcome adversity. Effective stress management is central to the definition of resilience (Luthar et al., 2000). Accordingly, several existing resilience programs for youth focus on building coping strategies for managing acute stressors. These initial studies demonstrated that resilience training can lower perceived stress and improve mental health outcomes; however, enthusiasm for stress-focused resilience training has recently plateaued. This shift is driven by the modest effect sizes observed in early trials, concerns about the long-term sustainability of training-related benefits, and increasing awareness that many stress-focused resilience trainings lack a clear theoretical foundation.

A.5. How to optimize resilience programs for stress? To enhance the efficacy of resilience programs, it is essential to integrate a theoretically grounded approach that addresses the cognitive vulnerabilities associated with stress, such as AS and IU. By targeting these modifiable factors, interventions can be tailored to individuals' specific cognitive profiles, enhancing their capacity to manage stress effectively and reducing reliance on maladaptive coping mechanisms like alcohol use.

A6 Our pilot studies have utilized brief, modular CBT interventions targeting AS and IU, focusing on psychoeducation, cognitive restructuring, and exposure exercises. This approach has shown promise in modifying cognitive biases associated with stress-related alcohol use. Data-driven adjustments have enhanced treatment fidelity and effectiveness. Future work will refine and evaluate these strategies for heavy alcohol users in the current study.

A8 IMPROVE targets AS and IU using self-report and objective measures, including EEG and startle potentiation. These measures will assess intervention impact on stress and alcohol use. This study will be the first to compare changes in subjective and objective responses to a CBT-based intervention, informing future precision medicine approaches.

Inclusion Criteria:

• report elevated psychological distress (i.e., Kessler Psychological Distress Scale scores >12)
• engage in heavy drinking behaviors (i.e., 15 drinks per week for biological males and 8 drinks per week for biological females)
• age 18 or older
• can read and comprehend English
• has access to a smartphone

Exclusion Criteria:

• limited mental competency/inability to give informed consent
• current comorbid moderate to severe substance use disorder other than nicotine

All de-identified demographic and clinical data collected as part of this project will be available as raw individual level data for sharing with external researchers working at an institution with a Federalwide Assurance (FWA) for the Protection of Human Subjects. Data will therefore be available for secondary analytic purposes. Names and institutions of persons either given or denied access to data will be tracked by our Administrative Core and will be available upon request from the sponsor. Data will not be available for data sharing until after this project is complete and the primary outcomes are published.

Data requests can be submitted starting 9 months after article publication and the data will be made accessible for at least the next two years. Extensions will be considered on a case-by-case basis.

Access to trial individual participant data (IPD) can be requested by qualified researchers engaging in independent scientific research, and will be provided following review and approval of a research proposal and Statistical Analysis Plan (SAP) and execution of a Data Sharing Agreement (DSA).

Raters completing the outcome assessments will be blind to condition. Participants are told they are receiving an intervention to help cope with distress but not which of the two interventions is the active and which is the control.