Variation over time in the association between polypharmacy and mortality in the older population

Background: In the older population, multimorbidity is common and best practice may invite the prescription of multiple medications. Objective: To examine the association between polypharmacy, defined as the concurrent use of five or more prescription or nonprescription medications, and mortality in 12 423 participants aged <65 years representative of the older population of England and Wales. Methods: Data on self-reportedmedication use, disability and health conditions were collected at baseline in 19913. The cohort was followed for up to 18 years with full mortality notification. Cox proportional hazards regression adjusted for age, baseline institutionalization, smoking, disability and health conditions was used to investigate polypharmacy and mortality, stratified by sex. Various methods for modelling the time-varying effect of polypharmacy are presented. Results: A strong independent association between polypharmacy and mortality existed in the short-term (first 2 years) for bothmen andwomen. This association remained, although attenuated, in the medium-long term (218 years of followup) for women, but became non-significant in the longer term for men. Conclusions: It remains unclear whether polypharmacy is a marker for poor health or is an independent risk factor for mortality. However, polypharmacy strongly predicts adverse outcomes, and multiple medication use should therefore be carefully monitored in the older population. © 2011 Adis Data Information BV. All rights reserved.