Skeletal-related events and mortality among older men with advanced prostate cancer.

OBJECTIVE: Skeletal-related events (SREs) are defined as a cluster of events including clinical diagnoses and treatment. Using claims data, the burden of SREs as a group has been reported among patients with cancer. We investigate the mortality impact of subcomponents of SREs, a topic that has received limited attention among older men. MATERIALS AND METHODS: We analyzed prostate cancer (PCa) and all-cause mortality among men diagnosed with metastatic PCa from 2000 to 2007 using Surveillance, Epidemiology, and End Results data linked with 1999-2009 Medicare data. We created three measures of pathological fracture (PF), spinal cord compression (SCC), and bone surgery (BS) that differed in the use of claims-based bone metastasis information. We reported covariate-adjusted hazard ratios (HRs) using the full sample and a propensity score-matched sample (PSMS). RESULTS: Application of inclusion/exclusion criteria resulted in 7062 men in the full sample (1776 in the PSMS). PCa-specific (all-cause mortality) was 54% (80%) at a median follow-up of 609days. SRE prevalence ranged from 9.7% to 17.1% across the measures. In a PCa mortality model, the HR associated with an SRE ranged from 1.07 (0.98-1.16) to 1.31 (1.18-1.45). The HRs for SCC and PF were statistically significant and positively associated with PCa-specific mortality. The results for BS depended on the measure. Results for SCC and BS, but not for PF, were preserved using a PSMS. CONCLUSIONS: The relationship between SREs and mortality among older men with metastatic PCa was driven by SCC and depended on the definition used to measure SREs.