Assessing Associations of Three Types of Higher-Risk Medication Use and Mortality in Older Adults

Higher-risk medication use-including regimens with polypharmacy, drug-drug interactions (DDIs), and potentially inappropriate medications (PIMs)-has been linked to adverse outcomes in older adults. However, studies establishing associations between higher-risk medication use and mortality have lacked measures of organ function and considered types of higher-risk medication exposures in isolation, allowing potential confounding and overestimation of harm.

We conducted a prospective cohort study of adults aged ≥ 65 years who took part in the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2016, with mortality follow-up through 2019. Our primary exposures were three types of higher-risk medication use: polypharmacy, DDIs, and PIMs. In primary analyses, these were defined, respectively, as follows: being on ≥ 5 prescription medications (polypharmacy); using two medications with ≥ 1 major DDI; and using ≥ 1 prescription PIM. Exposures were determined by medication review during the baseline survey. The primary outcome was all-cause mortality. Multivariable Cox regression models estimated mortality risk associated with each type of higher-risk medication use, sequentially adjusting for sociodemographic factors and comorbidities, organ function and health behaviors, and patterns of higher-risk medication use.

Among 7828 eligible participants (representing 26.5 million adults), 54.3% experienced at least one type of higher-risk medication use: 40.0% had polypharmacy, 11.4% had a DDI, and 37.6% used a PIM. In fully adjusted models including all higher-risk medication exposures, only polypharmacy remained significantly associated with increased mortality (hazard ratio [HR] 1.38; 95% CI 1.26-1.51). Sensitivity and subgroup analyses by age and health status yielded generally consistent findings.

When considering multiple types of higher-risk medication exposure, only polypharmacy was consistently and independently associated with increased mortality among older adults. These results highlight the harms of polypharmacy specifically and could help further target deprescribing interventions to the most high-risk patients.