For approximately three decades, one drink a day was presented as not just harmless but potentially beneficial — particularly for heart health. The "J-curve" hypothesis held that light drinkers had better cardiovascular outcomes than abstainers. Entire wine industries built marketing campaigns around it. Doctors gave their blessing. Now the evidence has shifted, and understanding why matters if you're using "just one drink" as your sustainable compromise.
The J-curve collapsed for a specific reason: the comparison group was contaminated. Most studies comparing "light drinkers" to "abstainers" included, in the abstainer category, former drinkers who had quit because of illness. When researchers removed these "sick quitters" from the abstainer group and compared light drinkers only to lifetime abstainers, the cardiovascular benefit largely disappeared. The 2018 Lancet Global Burden of Disease study — analyzing 195 countries and 28 million people — concluded that the safest level of alcohol consumption is zero, when all health outcomes are considered together.
What one drink a day actually does. For cancer risk: a consistent, dose-dependent relationship exists between alcohol and at least seven cancers. Even one drink per day modestly elevates risk for breast, colorectal, and esophageal cancer. For heart: the residual data shows some protective effect specifically for ischemic heart disease at very low doses — but this is offset by increased risk for stroke, cardiomyopathy, and atrial fibrillation, especially for women. Net cardiovascular effect at one drink per day: approximately neutral to slightly negative, depending on the study.
For sleep: one drink within three hours of bedtime measurably suppresses REM sleep even at low doses. For anxiety: the anxiolytic effect of one drink is real, but the corresponding rebound in cortisol and excitatory neurotransmitters several hours later is also real. For liver: one drink per day is below the fibrosis-risk threshold for most people, though this depends heavily on body weight, sex, and genetic metabolism variants.
Is one drink a day problematic in a clinical sense? Not inherently — it does not meet AUD criteria by itself. But for some people, one drink a day is the floor of a pattern that escalates. For others, it is a genuine steady-state. The difference is whether the "one drink" rule holds reliably over time, or whether it functions as an entry point.
The honest summary: one drink a day is not going to kill you, but the evidence no longer supports calling it healthy. It is a personal choice with modest, real costs, particularly for cancer risk and sleep quality. If you've moved from daily drinking back to one drink a day as a harm-reduction measure, that is a meaningful improvement. If you've been using "it's just one drink" to avoid looking at a pattern, the question worth asking is whether the one-drink rule actually holds — most days, most weeks, most months.