OBJECTIVES:  Hazard ratios (HRs) are commonly preferred as inputs to calculate clinical benefit across oncology value frameworks. However, these data are not consistently reported in publications, and individual patient data are not readily available. Therefore, we present a practical method for estimating the HR of survival data. METHODS:  We selected a published phase 3, multicenter trial (NCT00482833) in which investigators compared all-trans retinoic acid (ATRA) plus chemotherapy with ATRA plus arsenic trioxide (ATO) in the treatment of patients with low-to-intermediate risk acute promyelocytic leukemia (APL). The median follow-up was 34.4 months. At the time of the study publication, only 2-year overall survival (OS) and event free survival (EFS) were reported; HRs for both OS and EFS were not reported, and median OS and EFS had not been reached. To derive the HRs, we first obtained the number of patients at risk for each time point (12-month intervals until 48 months). Then, using a graphic tool, we extracted survival probabilities from the published Kaplan-Meier (K-M) curves for each treatment arm. Patient counts with censoring at each time point were not reported; thus we assumed a constant censoring rate within each 12-month interval. RESULTS: Using the K-M curve from the published trial, the calculated HRs of OS and EFS were 0.21 and 0.23, respectively, suggesting patients with ATRA combined with ATO and chemotherapy had better survival outcomes compared with patients receiving ATRA plus chemotherapy. This result is consistent with reported outcomes from a recent meta-analysis, estimating the HRs of OS and EFS to be 0.44 and 0.38, which included data from the same published trial. CONCLUSIONS:  Despite potential biases associated with relying on published data to derive median survival, the present analysis presents a practical method to estimate key survival inputs for application in oncology value frameworks.