Alcohol use and binge drinking are increasing among older adults, as is the need for tailored, age-specific
strategies to identify and treat unhealthy alcohol use. Current alcohol guidelines and risk thresholds for binge
drinking were not developed for older adults, and it is not known whether alcohol use consistent with these
guidelines is safe for older adults, particularly women. It is also not known whether any level of alcohol use is
safe for older adults who use cannabis or other psychoactive drugs (opioids or benzodiazepines) or have
functional impairment or comorbidity. Alcohol use, alone or combined with other drugs, is a risk factor for falls,
injuries, overdose, hospitalization, and death, but risk thresholds for older adults have not been established.
To address the gap in the literature on the risks of alcohol and polysubstance use among older adults, we
propose to construct a cohort of patients 65 and older with a recent clinical alcohol screen in a Veterans Health
Administration (VA) facility using merged national VA and Medicare data. We will stratify our sampling frame
by clinical alcohol screening score and the presence of a cannabis note in the health record (both available
from VA data) and will recruit a cohort of 4500 older patients (30% women). We will conduct a telephone
interview to assess exposure history (e.g., past-month alcohol use) and will quantify the association between
alcohol use (category of use or binge drinking) and adverse events (hospitalization or all-cause mortality, and
as secondary outcomes, falls, injuries, overdose, and hospitalization for cardiac, renal, hepatic, or mental
health conditions). We will also examine how the effects of alcohol differ for older patients who use cannabis.
Finally, we will explore whether the effects of alcohol are greater among higher risk older adults (women, older
adults with functional impairment/comorbidity, or who use other psychoactive drugs such as opioids or
benzodiazepines.
The goals of this proposal are to examine: 1) whether, after accounting for baseline differences in patient
characteristics, higher levels of alcohol use are associated with increased risk for adverse event; 2) whether
binge drinking is associated with increased risk for adverse events; and 3) whether associations between
alcohol use and adverse events vary for patients who use cannabis. We will also assess the effect of alcohol
use on adverse events among high-risk patients. We will use propensity score methods to control confounding.
There is currently insufficient evidence to establish safe drinking guidelines for older adults. This project will
develop the foundational evidence base needed to create better alcohol guidance, improve alcohol
assessment tools, and develop more patient-centered risk communication strategies for older adults.