Which better predicts mortality among older men, a prostate cancer (PCa) diagnosis or vulnerability on the Vulnerable Elders Survey (VES-13)? A retrospective cohort study.

Abstract

Older men with a prostate cancer (PCa) diagnosis face competing mortality risks. Little is known about the prevalence of vulnerability and predictors of mortality in this population compared to men without a PCa diagnosis. We examined the predictive utility of the Vulnerable Elders Survey (VES-13) for mortality in older men with a PCa diagnosis as compared to controls. Men aged ≥65years from an urban geriatrics clinic completed the VES-13 between 2003 and 2008. Each patient with a PCa diagnosis was matched by age to five controls, resulting in 59 patients with a PCa diagnosis and 318 controls. Cox proportional hazard models were used to determine the association of a PCa diagnosis and vulnerability on the VES-13 with mortality. The mean age for men with a PCa diagnosis and controls was 77.9years and 76.1years, respectively. Of those with a PCa diagnosis, 74.6% had no active disease or a rising PSA only. Regardless of PCa diagnosis, vulnerable individuals on the VES-13 were more likely to die during the study period (VES-13≥3: HR=4.46, p<0.01; VES13≥6: HR=3.77, p<0.01). Men with a PCa diagnosis were not more likely to die compared to age-matched controls (VES-13≥3: HR=1.14, p=0.59; VES13≥6: HR=1.06, p=0.83). Vulnerability for men with a PCa diagnosis was more predictive of mortality. Therefore, the assessment of vulnerability is important for establishing goals of care.