Evaluating the impact of COVID-19 public health measures on alcohol-related harms

PROJECT SUMMARY During the coronavirus disease 2019 (COVID-19) pandemic, countries all over the world implemented various public health measures (e.g., complete lockdowns, occupancy limitations, and travel restrictions) in an effort to limit the spread of the virus. Such measures inherently impacted alcohol use, and, as such, likely had an impact on alcohol-attributable harms. The proposed Supplement will evaluate the effects of COVID-19 public health measures on alcohol consumption and on select alcohol-attributable disease and mortality outcomes in Lithuania, a high-income member state of the European Union. We postulate that the pandemic and its associated public health measures impacted alcohol consumption and alcohol-attributable harms in the following ways: consumption is hypothesized to polarize - i.e., people who had been drinking heavily before the pandemic (heavy drinking will be defined a 40+ grams of pure alcohol per day for women, and 60+ g of pure alcohol per day for men) will have, on average, increased their alcohol consumption, whereas the rest of the population will have, on average, decreased their consumption. We will also test mediating mechanisms of availability and affordability for decreases, and stress for increases. In terms of alcohol-attributable harm, we selected the following key consequences of alcohol consumption and theoretically derived several hypotheses impacted by confinement, social distancing and other public health measures taken during the pandemic to reduce infection risk: suicide attempts and deaths by suicide, domestic violence, non-domestic violence, traffic collisions and injuries, and liver cirrhosis. In short, we postulate that suicide attempts and suicide, domestic violence and liver cirrhosis hospitalizations and mortality increased during lockdowns, in part fueled by the polarization of drinking, while traffic collisions and injury as well as non-domestic violence decreased; the latter two because of less traffic and fewer situations where such violence would typically occur, overall. However, we also postulate that the proportion of alcohol-attributable traffic collisions and injury to all such collisions and injury will increase. If our hypotheses are supported, we will be able to improve future calculations of alcohol- attributable burden and disease and mortality, further moving away from simple attributable-fraction models based on level of exposure and outcome (as currently practiced by the World Health Organization and the Global Burden of Disease study) towards more complex models which specify country-specific boundary conditions. But as indicated above, the analyses planned for this Supplement will not only improve the theoretical knowledge in alcohol epidemiology, but also specify necessary boundary conditions for alcohol control policy implementation, not only during times of future pandemics, but economic and other crises as well. Lithuania offers a unique opportunity to test these hypotheses due to the richness of its data (e.g., five large-scale national general population surveys from 2019-21).