We tested the polarization hypothesis, which postulates that during times of crises, such as the COVID-19 pandemic, alcohol consumption increases among the heaviest drinkers but decreases among most other drinkers, resulting in an overall decrease in consumption among the population. We posited the increase in heavy drinking would lead to increases in 100% alcohol-attributable (AA) mortality. Furthermore, based on the high level of alcohol consumption in the Baltic countries compared to other European countries, we predicted that the increases in AA mortality would be more pronounced in these countries. Data for 100% AA deaths were obtained from the World Health Organization for the period 2010 to 2022, and standardized to the regional age distribution for 2010. Parametric and non-parametric tests were used to test the study hypotheses. The mean age-standardized 100% AA mortality rate increased during the COVID-19 pandemic in the 19 European countries with data by 3.12 deaths per 100,000 population, constituting an 18% increase (tested against no change: t = 2.52; df = 18; p = 0.021). This increase was higher in the Baltic countries (mean difference = 13.41 deaths per 100,000 population; standard deviation (SD) = 7.44; 46% increase) than for other European countries (mean difference = 1.19; SD = 1.55; 8% increase). The increases in 100% AA mortality were associated with decreases in the level of alcohol consumption in the majority of countries. As predicted, 100% AA mortality increased in 19 European countries during the COVID-19 pandemic, with the Baltic countries seeing a higher increase. Renewed alcohol control policy efforts should be considered.