The hazard rate function plays a fundamental role in survival analysis.Its statistical inference methods have been systemically and extensively studied.When does the hazard rate reach a particular warning level? This isa basic question of interest to the investigator but largely left to be explored in practice.We define a level set of hazard rate to addressthis issue and propose a kernel smoothing estimator for such a level set.In terms of the Hausdorff distance, we establish the consistency, convergence rateand asymptotic distribution of the level set estimator.The validity of the proposed confidence set, based on the bootstrapping method, forthe level set of hazard rate is theoretically justified.We conduct comprehensive simulation studies to assess the finite-sample performance ofthe proposed method, which is further illustrated with a lung cancer study.