(093) Associations between gender, race/ethnicity and age and the initiation of chronic high-risk medication in US older adults

Background: High-risk medication use is associated with an increased risk of adverse events in older adults.

Objectives: To study the associations between age, gender, and race/ethnicity, and new chronic use of high-risk medications in older adults.

Methods: In this retrospective cohort study, we identified older adults aged ≥65 years enrolled in a national health insurer who started a high-risk medication between 2017-2022 across 16 high-risk medication classes. We measured age, gender, and race/ethnicity directly from enrollment files. The primary outcome was the onset of chronic high-risk medication use after initiation (days’ supply of ≥90 days and ≥2 fills in the respective medication class within 180 days). We used generalized estimating equations to estimate the association between the sociodemographic classifications and outcomes (using White older adults, Female gender, and age 65-74 years as referent). The analyses were adjusted for sociodemographic and clinical patient characteristics measured during the 365-day baseline period. We added three-way interaction terms for race/ethnicity, gender, and age to the model (race/ethnicity#age category#gender) to explore whether the relationship between race/ethnicity, age, and gender and the outcome varies across different subgroups of race/ethnicity, age, and gender. We repeated the analyses by medication class using logistic regression.

Results: Across 2,751,069 patients (mean age: 74 years [SD=7], 72% White, 60% Female), 15% became new chronic users of ≥1 high-risk medication (n=406,075). Compared to White older adults, Asian (OR=0.66, 95% CI 0.64-0.69), Black (OR=0.84, 95% CI 0.82-0.85), and Hispanic (OR=0.77, 95% CI 0.77-0.79) older adults were at lower odds of becoming chronic users. Men had higher odds compared to women (OR=1.11, 95% CI1.10-1.12). We found inconclusive evidence regarding age (≥75 years: OR=1.00, 95% CI 0.99-1.01). The joint presence of specific age, gender, and race/ethnicity characteristics increased the odds of becoming a new chronic user (e.g., Hispanic # Male # ≥75 years: OR=1.06, 95% CI 1.03-1.09, Black # Female # ≥75 years: OR=1.05, 95% CI 1.02-1.07). We observed some differences across some medication classes, like benzodiazepines and other pain medications for which non-White older adults had higher odds of becoming new chronic users.

Conclusions: We found that the odds of becoming new chronic users of high-risk medications varied across individual sociodemographic groups and medication classes. These findings suggest the importance of individualizing medication optimization approaches.