Polypharmacy Status as an Indicator of Mortality in an Elderly Population

Increased use of drugs has raised concern about the risks of polypharmacy in elderly populations. Adverse outcomes, such as hospitalizations and falls, have been shown to be associated with polypharmacy. So far, little information is available on the association between polypharmacy status and mortality.

To assess whether polypharmacy (six to nine drugs) or excessive polypharmacy (ten or more drugs) could be indicators of mortality in elderly persons.

This was a population-based cohort study conducted between 1998 and 2003 with mortality follow-up through to 2007. The data in this study were derived from the population-based Kuopio 75+ Study, which involved elderly persons aged ≥75 years living in the city of Kuopio, Finland. The initial sample (sample frame n=4518, random sample n=700) was drawn from the population register. For the purpose of this study, two separate analyses were carried out. In the first phase, participants (aged ≥75 years, n=601) were followed from 1998 (baseline) to 2002. In the second phase, survivors (aged ≥80 years, n=339) were followed from 2003 to 2007. Current medications were determined from drug containers and prescriptions during interviews conducted by a trained nurse. The Kaplan-Meier method and Cox proportional hazards regression were used to examine the association between polypharmacy status and mortality.

In the first phase, 28% (n=167) belonged to the excessive polypharmacy group, 33% (n=200) to the polypharmacy group, and the remaining 39% (n=234) to the non-polypharmacy (0–5 drugs) group. The corresponding figures in the second phase were 28% (n=95), 39% (n=132) and 33% (n=112), respectively. The mortality rate was 37% in the first phase and 40% in the second phase. In both phases, the survival curves showed a significant difference in all-cause mortality between the three polypharmacy groups. In the first phase, the univariate model showed an association between excessive polypharmacy and mortality (hazard ratio [HR] 2.53, 95% CI 1.83, 3.48); however, after adjustment for demographics and other variables measuring functional and cognitive status, this association did not remain statistically significant (HR 1.28, 95% CI 0.86, 1.91). In the second phase, the association between excessive polypharmacy and mortality (HR 2.23, 95% CI 1.21, 4.12) remained significant after adjustments. Age, male sex and dependency according to the Instrumental Activities of Daily Living screening instrument were associated with mortality in both phases.

This study points to the importance of excessive polypharmacy as an indicator for mortality in elderly persons. This association needs to be confirmed following adjustment for co-morbidities.