Alcohol and Anxiety: The Self-Medication Hypothesis and What the Evidence Shows

Using alcohol to calm anxiety is so common it’s practically a cultural norm. And alcohol does reduce anxiety — in the moment. Understanding what happens after that moment, and why the anxiety-alcohol cycle is so difficult to break, requires looking at both the pharmacology and the longitudinal evidence.

The acute effect is real

Alcohol enhances GABA transmission and reduces glutamate activity, producing anxiolysis (anxiety reduction), muscle relaxation and sedation. The subjective experience of feeling calmer after a drink has a genuine neurochemical basis [6]. This is the same receptor system targeted by benzodiazepines, and the acute anti-anxiety effect is similar.

The problem is what happens next.

The self-medication hypothesis

Khantzian proposed that people use substances to self-treat emotional distress they can’t otherwise manage [2]. For anxiety, this means: people with anxiety disorders are more likely to drink because alcohol provides temporary relief from symptoms they find unbearable.

The evidence partially supports this:

• Anxiety disorders and alcohol use disorders co-occur at rates of 20–40% in clinical samples, far above what chance would predict [1].
• Prospective studies show that anxiety disorders in adolescence and early adulthood predict later alcohol problems [3].
• People with anxiety disorders who drink often report anxiety reduction as their primary motivation.
• Social anxiety disorder has a particularly strong association with alcohol use, likely because alcohol reduces social inhibition.

But self-medication is not the complete story.

The bidirectional problem

The relationship goes both ways. Heavy drinking also causes anxiety through several mechanisms:

Withdrawal. Chronic alcohol use causes neuroadaptation — GABA receptors downregulate and glutamate activity increases. When alcohol levels drop, the resulting imbalance produces anxiety, tremor, agitation and insomnia. This withdrawal-related anxiety can be more severe than the original problem.

Kindling. Repeated cycles of drinking and withdrawal may progressively sensitize the anxiety response, making each withdrawal episode worse — a process similar to what happens with benzodiazepine withdrawal.

Sleep disruption. Alcohol disrupts sleep architecture, particularly REM sleep. Poor sleep worsens anxiety through the mechanisms described in sleep and anxiety research.

Cognitive impairment. Chronic heavy drinking impairs prefrontal cortex function, potentially reducing the brain’s capacity to regulate emotional responses.

Prospective data confirms this second direction: heavy drinking predicts later anxiety disorders even after controlling for baseline anxiety [5]. The relationship is genuinely bidirectional — anxiety leads to drinking, and drinking leads to more anxiety, creating a self-reinforcing cycle [3].

Distinguishing cause from consequence

This makes the correlation-causation problem particularly acute. When someone presents with both anxiety and alcohol problems, several scenarios are possible:

1. Pre-existing anxiety led to self-medication with alcohol
2. Heavy drinking caused or worsened the anxiety
3. Both arose from shared risk factors (genetics, adverse childhood experiences, personality traits)
4. All of the above, in a reinforcing cycle

In clinical practice, all four scenarios likely contribute in different proportions for different people. The practical implication is that treating one problem without addressing the other is often insufficient.

Treatment implications

Reviews of treatment for comorbid anxiety and alcohol use disorders consistently recommend integrated treatment [4]. This means addressing both problems simultaneously rather than sequentially (e.g., “get sober first, then we’ll treat the anxiety”). Sequential approaches often fail because untreated anxiety drives relapse to drinking, and continued drinking prevents anxiety treatment from working.

CBT approaches that combine anxiety-focused techniques (exposure, cognitive restructuring) with alcohol-focused techniques (motivational enhancement, relapse prevention) show promise, but the evidence base is smaller than you might expect given how common the comorbidity is.

Questions to ask about alcohol-anxiety claims

1. Does the study distinguish pre-existing anxiety from alcohol-induced anxiety?
2. Is the evidence prospective (following people over time) or cross-sectional?
3. Were both anxiety and alcohol use measured with validated instruments?
4. Was the causal direction tested, or just assumed?
5. Does the treatment evidence address both conditions together?

What remains uncertain

Whether moderate alcohol use has any anxiolytic benefit that outweighs its risks for people with anxiety disorders is contested. Whether specific anxiety disorders (social anxiety vs GAD vs panic) have different relationships with alcohol is not well mapped. And whether the increasing availability of non-alcoholic alternatives or cannabis products changes self-medication patterns and their consequences is too recent to study rigorously.

Alcohol and your anxiety. If you’re using alcohol to manage anxiety, you’re not alone, but you’re likely making the anxiety worse over time. Effective treatments exist for both anxiety and alcohol problems, ideally addressed together. If anxiety or alcohol use is affecting your daily life, talk to a GP or mental health professional. If you are in crisis or thinking about harming yourself, contact emergency services or a crisis line now: in the US, call or text 988; in the UK and Ireland, Samaritans on 116 123.