Alcohol Use Disorder, Anxiety, Depression, and the Metabolic Problem We May Be Missing
One of the most persistent stories therapists tell our clients and their families about alcoholism is also one of the most psychologically satisfying: people drink because they are depressed, anxious, traumatized, lonely, or otherwise suffering. Alcoholism is a disease because alcohol becomes the medication. Treat the underlying emotional disorder, the reasoning goes, and the need for alcohol should diminish.
Sometimes that story is true.
But some of the best longitudinal research we have suggests that we may frequently have the sequence backward.
This distinction matters clinically. If depression and anxiety are consequences—or amplifiers—of alcohol use disorder rather than its original cause, treating them primarily as independent psychiatric illnesses can leave the central pathology relatively untouched. And newer research suggests another variable belongs in the picture as well: metabolic dysfunction.
The Harvard Study That Complicated the Self-Medication Story
Beginning in the late 1930s and early 1940s, researchers followed two very different groups of men: 268 Harvard undergraduates and 456 socially disadvantaged young men from Boston. The research eventually became part of the Harvard Study of Adult Development, with psychiatrist George Vaillant examining the natural history of alcoholism across decades.
The great advantage of this work was prospective observation. Rather than asking people already suffering from alcoholism to reconstruct whether they had once been anxious, depressed, unhappy, or traumatized, researchers possessed information about participants before alcoholism developed.
That changed the picture.
Depression was certainly associated with alcoholism. In Vaillant’s Harvard cohort, severe depression was about five times more common among men with alcoholism than among those without it. But prevalence is not causation.
Among the 14 Harvard men who eventually experienced both alcoholism and major depression, only four developed depression before alcoholism. Vaillant reported similar findings for generalized anxiety disorder. Moreover, earlier clinical studies cited in his analysis found that sustained abstinence frequently reduced depressive symptoms.
His conclusion was provocative: in many patients, depression and anxiety appeared not to be the illnesses from which alcoholism emerged, but conditions that developed after alcoholism was established.
In other words, the depression may sometimes be downstream.
That conclusion should not be universalized. Vaillant’s original cohorts were male, overwhelmingly white, and drawn from unusual socioeconomic extremes. Subsequent epidemiological research has also demonstrated bidirectionality. A Harvard School of Public Health study of 14,480 people, for example, found that alcohol dependence increased the subsequent risk of major depression, while depression—particularly among women—could also increase subsequent risk for alcohol dependence.
The modern picture is therefore not “alcohol always causes depression.”
It is something more clinically useful:
Comorbidity does not tell us which disorder came first.
Then Metabolism Entered the Picture
A 2026 study from the National Institute on Alcohol Abuse and Alcoholism adds a particularly important dimension.
Alexandra Wagner and colleagues examined 1,220 participants enrolled in an NIH natural-history protocol. They divided participants into four groups: healthy controls, people with metabolic dysfunction alone, people with alcohol use disorder without metabolic dysfunction, and people with both AUD and metabolic dysfunction.
The last group—the researchers termed it “metAUD”—stood out.
People with both conditions had the greatest abnormalities in measures associated with liver injury, inflammation, and fibrosis. More importantly for mental-health clinicians, they also carried the greatest psychiatric burden. Within the AUD population, increasing metabolic dysfunction was associated with greater anxiety-disorder prevalence as well as worsening biological markers.
This study does not establish that metabolic dysfunction causes the psychiatric symptoms. It was observational and cross-sectional. But the clustering is difficult to dismiss.
Other research makes the metabolic connection increasingly plausible. Longitudinal work from the Maastricht Study found that higher fasting glucose, post-load glucose, and HbA1c predicted subsequent depressive symptoms over four years. Research involving metabolic syndrome, insulin resistance, depression, and anxiety has likewise demonstrated substantial overlap, although the direction of causality varies between studies.
What begins to emerge is not a neat causal line but a system.
Alcohol disrupts sleep, glucose regulation, hepatic function, inflammatory signaling, nutritional status, autonomic regulation, and brain function. Metabolic dysfunction independently affects many of those same systems. Depression and anxiety exist inside this biological environment rather than floating above it as purely psychological phenomena.
The question therefore changes.
Instead of asking only:
“What emotional pain is this person treating with alcohol?”
we might also ask:
“What has chronic alcohol exposure done to the organism in which this anxiety and depression are now occurring?”
Why AA Keeps Appearing in the Longitudinal Data
There is another interesting feature of Vaillant’s work.
Across decades of observation, formal treatment did not consistently predict long-term recovery. Alcoholics Anonymous was the conspicuous exception. Vaillant found AA at least as effective as clinic treatment in helping initiate stable abstinence and particularly important in maintaining it over time.
That finding was once easy to dismiss because people who remain in AA may differ substantially from people who do not.
But later evidence strengthened the case.
A 2020 Cochrane systematic review examined 27 studies involving more than 10,000 participants. It concluded that professionally delivered Twelve-Step Facilitation designed to increase AA involvement produced higher rates of continuous abstinence than several other established interventions, including CBT, while generally performing at least as well on other drinking-related outcomes.
The important point is not that psychotherapy fails or that everyone with AUD should attend AA.
It is that alcoholism behaves like a chronic disorder.
A weekly insight into childhood attachment may be psychologically meaningful while being biologically outmatched by a substance used every evening. AA, whatever one thinks of its language or philosophy, introduces something psychotherapy often cannot reproduce: an enduring behavioral structure organized around not drinking.
Vaillant noticed that decades ago.
The Clinical Error of Premature Psychological Explanation
Psychotherapists are trained to search for meaning. That is usually an asset.
But it can also become a form of diagnostic seduction.
A person drinks heavily and reports anxiety, shame, loneliness, disrupted relationships, low motivation, and depression. Every one of those experiences may have a genuine psychological history. Yet they are also entirely compatible with the physiological and psychosocial consequences of prolonged alcohol misuse.
If we immediately explain the drinking through the depression, we may inadvertently use a downstream symptom to explain the process producing it.
The same problem arises when every relapse is interpreted symbolically while sleep deprivation, glucose dysregulation, obesity, liver dysfunction, medication effects, nutritional deficiencies, and alcohol exposure themselves remain peripheral to the formulation.
Psychological meaning and biological causation are not competitors.
A person can drink because alcohol temporarily regulates an intolerable emotional state and simultaneously become more anxious and depressed because continued drinking progressively disrupts the biological systems regulating mood.
Both things can be true.
A More Complete Formulation
The emerging picture of alcohol use disorder is therefore less linear than either traditional psychiatry or traditional psychotherapy sometimes assumes.
There may be developmental vulnerability.
There may be trauma.
There may be anxiety or depression that clearly precedes alcohol misuse.
There may be genetic predisposition.
There may be learned reinforcement: alcohol works quickly, and therefore the brain learns to use it again.
But once persistent alcohol misuse becomes established, it becomes a causal agent in its own right.
And when substantial metabolic dysfunction accompanies it, the psychiatric and physical burden appears to intensify further.
For clinicians, this means that chronology matters. Metabolic health matters. Sleep matters. The trajectory of alcohol consumption matters. And the response of psychiatric symptoms to sustained abstinence matters.
We should be cautious about assuming that every depressed alcoholic is drinking because he or she is depressed.
Sometimes the more revealing question is the reverse:
How much of the person we are treating today has been shaped by the alcohol itself—and by the metabolic system in which that alcohol has been operating?
That is not a reduction of psychotherapy to biology.
It is an expansion of what we mean by psychological formulation.
References
Gilman, S. E., & Abraham, H. D. (2001). A longitudinal study of the order of onset of alcohol dependence and major depression. Drug and Alcohol Dependence, 63(3), 277–286. doi:10.1016/S0376-8716(00)00216-7.
Geraets, A. F. J., et al. (2020). The association of hyperglycaemia and insulin resistance with incident depressive symptoms over 4 years of follow-up: The Maastricht Study. Diabetologia, 63, 2315–2328.
Kelly, J. F., Humphreys, K., & Ferri, M. (2020). Alcoholics Anonymous and other 12-step programs for alcohol use disorder. Cochrane Database of Systematic Reviews, CD012880. doi:10.1002/14651858.CD012880.pub2.
Vaillant, G. E. (2003). A 60-year follow-up of alcoholic men. Addiction, 98(8), 1043–1051. doi:10.1046/j.1360-0443.2003.00422.x.
Vaillant, G. E., & Hiller-Sturmhöfel, S. (1996). The Natural History of Alcoholism. Alcohol Health & Research World.
Wagner, A. C., Jung, J., Reitz, J., et al. (2026). Alcohol Use Disorder With Metabolic Dysfunction Is Associated With Adverse Health Impacts in a United States Clinical Setting. Addiction Biology, 31(3), e70128. doi:10.1111/adb.70128.
