Cost and Management of Fractures in Medicare Beneficiaries with Non-Metastatic Prostate Cancer Treated with Androgen Deprivation Therapy

Abstract

Background: Androgen deprivation therapy (ADT) in prostate cancer (PCa) is associated with increased fracture risk and costs. There is limited information regarding the use of guideline-recommended bone mineral density (BMD) testing and bisphosphonate therapy. This study examined patient-specific (e.g., comorbidity) and non-patient-specific (e.g., physician specialist visits) factors associated with BMD testing, bisphosphonate therapy, fracture risk, and costs. Methods: This retrospective cohort analysis used linked Surveillance, Epidemiology, and End Results & Medicare data on men aged 65+ with non-metastatic PCa and treated with ADT. Cox proportional hazards models examined the association between baseline Charlson Comorbidity Index (CCI) and fracture, accounting for death as a competing risk. Partitioned weighted least squares regression models quantified the 5-year incremental cost of fractures. Multivariable logistic regression models quantified the effect of specialist visits on BMD testing and bisphosphonate use. Results: Of 30,904 men, 6,779 (22%) had a fracture during 5-years of follow-up. Among men aged 66-74 years, the sub-hazard ratio (SHR) (95% CI) for fracture was 1.17 (1.05-1.29) for CCI=1 and 1.69 (1.51-1.90) for CCI=2+ (reference: CCI=0); CCI was not associated with fracture risk among the oldest men aged 85+ years. The mean (95% CI) incremental cost of fractures was $31,800 ($31,709-$31,891), $34,320 ($34,152-$34,488), and $46,678 ($46,534-$46,822) among patients with CCI=0, CCI=1, and CCI=2+, respectively. Following ADT initiation, 17.9% received BMD testing and 9.6% received bisphosphonate. Compared to men who saw a urologist and primary care physician, bisphosphonate receipt was more likely among men with a urologist and medical oncologist involved in their care (AOR=1.89; 95% CI=1.38-2.57), and less likely among men who saw a urologist and radiation oncologist (AOR=0.63; 95% CI=0.42-0.95). There were no statistically significant associations between specialist visits and BMD testing. Conclusions: Patients with greater comorbidity burden had increased fracture risk and higher costs associated with fractures. These patients could benefit from regular monitoring of BMD and bisphosphonate therapy, which occurred in approximately 2 in 10 men. Involving certain types of specialists (e.g., medical oncologists) in the management of men with PCa could be beneficial for bone health management.