PROJECT SUMMARY
The COVID-19 pandemic has led to drastic declines in life expectancy in the United States (US),
disproportionately affecting individuals with low socioeconomic status (SES) as well as Black and Hispanic
populations. Mortality from non-COVID-related causes has also shown meaningful increases that contributed
to the losses in life expectancy, including a striking 27% increase in age-adjusted mortality rates from alcohol-
related causes from 2019 to 2020.
The current application requests a supplement to the parent grant (R01AA028009) to account for the social,
demographic, economic, and behavioral changes induced by the COVID-19 pandemic. Under the parent grant,
a microsimulation of the adult (18+) general population is being developed to model years of potential life lost
(YLL) from alcohol-attributable causes of death between 2000 and 2017 with a forward modeling time horizon
until 2028. This supplement aims to augment the microsimulation model by incorporating the role and effects of
the COVID-19 pandemic. Specifically, the role of alcohol use in excess mortality during the COVID-19
pandemic will be investigated among different population subgroups. Changes in psychological distress and
alcohol-related policies during the pandemic will be investigated as potential drivers of changes in alcohol use.
Current population-based data on alcohol use, psychological distress, alcohol-related policy changes,
mortality, and population statistics will be used. Linear spline multilevel models, stratified by sex, will be used to
investigate trends in alcohol use, conditional on SES and race/ethnicity. Age, individual-level psychological
distress, and state-level alcohol-related policies, along with their interaction with year, will be introduced as
covariates. Cause-specific contributions to changes in life expectancy will be calculated using a decomposition
method. Transition probabilities between i) drinking patterns and ii) levels of education before and during the
COVID-19 pandemic will be estimated using continuous-time Markov models. Findings will be integrated into
an augmented microsimulation model. YLL from nine selected causes of death including alcohol use disorders,
intentional and unintentional injuries, liver disease and cirrhosis, diabetes mellitus, and several cardiovascular
diseases will be modeled. Counterfactual scenarios will be used to estimate i) excess alcohol-attributable YLL
by population subgroup during the COVID-19 pandemic compared to a scenario without pandemic; ii) potential
long-term effects on alcohol-attributable mortality assuming sustained compared to temporary changes in
transition probabilities; and iii) alcohol-attributable YLL under different scenarios of long- and short-term
implementation of restrictive and permissive alcohol-related policies during the pandemic and beyond.
The findings will provide unique insights into the impact of the COVID-19 pandemic on alcohol use and
alcohol-attributable mortality in different population subgroups. Furthermore, the augmented microsimulation
model will constitute a robust foundation for policy and intervention modeling performed under the parent grant.
