In a phase 3 trial of denosumab vs zoledronic acid in patients (n=1776) with bone metastases and solid tumors or multiple myeloma, denosumab was superior to zoledronic acid for the primary end point of prevention of skeletal-related events. There was no difference in overall survival between the two groups; however, an ad hoc overall survival analysis in the multiple myeloma subset of patients (n=180) favored zoledronic acid (hazard ratio (HR) 2.26; 95% confidence interval (CI) 1.13-4.50; P=0.014). In the present analysis, we found imbalances between the groups with respect to baseline risk characteristics. HRs with two-sided 95% CIs were estimated using the Cox model. After adjustment in a covariate analysis, the CI crossed unity (HR 1.86; 95% CI 0.90-3.84; P=0.0954). Furthermore, we found a higher rate of early withdrawals for the reasons of lost to follow-up and withdrawal of consent in the zoledronic acid group; after accounting for these, the HR was 1.31 (95% CI 0.80-2.15; P=0.278). In conclusion, the survival results in multiple myeloma patients in this trial were confounded and will eventually be resolved by an ongoing phase 3 trial.