From the article: "A higher frequency of alcohol consumption was surprisingly associated with reduced frequency of depressed mood in people with depression. Th

From the article: "A higher frequency of alcohol consumption was surprisingly associated with reduced frequency of depressed mood in people with depression. This may potentially be due to the self-medicating use of alcohol by those with depression to manage their mood." So they explicitly consider your point for alcohol consumption. But they don't seem to consider it for sleep, screen time, or exercise. To complicate things, I find alcohol negatively impacts sleep and exercise, so the hypothesed self-medication with booze could score multiple hits.