Longitudinal associations of polypharmacy and frailty with major cardiovascular events and mortality among more than half a million middle-aged participants of the UK Biobank

Abstract

Background

Studies of the associations of polypharmacy and frailty with adverse health outcomes in middle-aged adults are limited. Furthermore, a potentially stronger association of polypharmacy with adverse health outcomes in frail than in non-frail adults is of interest.

Objective

To evaluate associations of frailty (assessed using a frailty index) and polypharmacy (defined as taking five or more drugs) with major cardiovascular events, cancer incidence, all-cause, cardiovascular disease-specific, and cancer-specific mortality.

Methods

Cox proportional hazards regression models were used to analyze 501,548 participants of the UK Biobank cohort study aged 40–69 years who were followed up for an average of 12 years.

Results

The prevalence of pre-frailty and frailty were 43.2 % and 2.3 %, respectively, and that of polypharmacy was 18.3 %. Although strongly associated with each other, frailty and polypharmacy were independently, statistically significantly associated with major cardiovascular events, cardiovascular disease-specific, and all-cause mortality. In addition, the hazard ratios of polypharmacy were stronger among (pre-)frail than non-frail study participants. No profound associations with cancer incidence and cancer mortality were observed. No sex and age differences were observed.

Conclusions

This large cohort study showed that polypharmacy and frailty are independent risk factors for major cardiovascular events, cardiovascular disease-specific and all-cause mortality in both middle-aged (40–64 years) and older people (≥ 65 years). In addition, the hazard ratios of polypharmacy were stronger among (pre-)frail than non-frail study participants. This underlines the need to avoid polypharmacy as far as possible not only in older but also in middle-aged subjects (40–64 years), especially if they are pre-frail or frail.