Background: Polypharmacy, typically defined as five or more concurrent medications, is strongly associated with mortality. One mechanism through which this relationship is thought to operate is the increasing likelihood of exposure to potentially inappropriate medications (PIMs), which have been extensively studied among adults aged 65 years or older. Beers criteria list PIMs that are potentially inappropriate in most older adults. There is a paucity of studies that include middle-aged adults, although polypharmacy and Beers PIMs are common in this age group.
Objectives: To assess the contribution of PIMs to the association between medication count and risk of mortality among middle-aged patients, overall and by sex, race, and ethnicity.
Methods: Using data from the Department of Veterans Affairs (VA) – the largest integrated healthcare system in the US – we performed an observational cohort study of patients aged 41 to 64 from October 1, 2008 and September 30, 2017. Baseline was considered the first dispensing of a chronic medication, defined as continuous use of ≥90 days. PIMs were defined by the 2015 Beers criteria. Patients were excluded if they had ≥1 PIM or ≥5 chronic medications in baseline period to obtain incident exposures. Patients were followed until death, 5 years, or study end. Hazard ratios (HR) and 95% confidence intervals (CI) were estimated using time-updated Cox proportional hazards models. Nested models were performed separately: (a) unadjusted; (b) adjusted for age, sex, race, ethnicity, baseline year, smoking and alcohol consumption, and a validated measure of physiologic frailty; (c) additionally adjusting for PIM count.
Results: Of 733,728 included patients, 92% were men, 65% White, 21% Black, 6% Hispanic, and mean age (SD) was 56 years (5.7). Polypharmacy was common (11.2% on 5-9 medications and 1.2% on ≥10 medications), and the overall mortality rate was 11.2 deaths per 1000 person-years. Overall, HRs associated with 5-9 and ≥10 chronic medications were 1.16 [1.11 – 1.20] and 1.50 [1.38 – 1.62], respectively (model b). Additionally adjusting for PIMs (model b vs. c) attenuated these HRs (1.13 [1.09 – 1.18] for 5-9 medications and 1.41 [1.29 – 1.54] for ≥10 medications). Findings were similar in subgroup analyses by sex, race, and ethnicity.
Conclusions: Polypharmacy, particularly those on ≥10 concurrent medications, was strongly associated with risk of mortality. Although Beers PIMs are common in middle-aged adults, they modestly explained this risk. Further potential mechanisms, including drug-drug and drug-gene interactions and accumulated medication toxicity, should be explored.
