Background: Polypharmacy, the use of multiple medications by one individual, may be associated with adverse health outcomes including poor cognition. However, it remains unclear whether a longitudinal relationship exists.

Objectives: To investigate the association between polypharmacy and 3-year cognitive ability in older adults.

Methods: A longitudinal cohort study of older adults 65 years and older, residing in the community, who participated in waves 12 (2012), 13 (2013) and 16 (2016) of the Household Income and Labour Dynamics (HILDA) Survey was conducted. Polypharmacy was defined as the regular use of 5 or more prescription medications. Cognitive ability was assessed using backwards digit span test (BDS), 25-item version of the National Adult Reading Test (NART-25) and symbol-digit modalities test (SDM). Linear regression was used to test the longitudinal association between polypharmacy and cognitive test scores at 3 years. All analyses were adjusted for age, sex, education, comorbidities, socioeconomic and lifestyle factors, and baseline cognitive test scores.

Results: A total of 2141 participants (mean age 72.9 years, 54.4% female) were included in the study sample. Polypharmacy was present in 27.3%. After adjusting for potential confounders, polypharmacy was negatively associated with cognitive ability at 3 years: BDS: −0.067 (95% CI = −0.353 to −0.051), NART-25: −0.071 (95% CI = −1.428 to −0.294), SDM: −0.073 (95% CI = −2.960 to −0.696).

Conclusion: Polypharmacy was associated with poorer cognitive ability at 3 years, even after adjusting for comorbidities and other confounders. Future research should consider the long-term impact of polypharmacy on cognitive ability, and identify strategies to optimise medication use and cognition in older adults.