This study examines 38 significant real-world earthquake events that occurred in the Chinese Mainland from 2018 to 2022,employing three distinct models to estimate the death toll associated with each event. These estimates are then compared against the actual fatalities to evaluate the models' accuracy. The models assessed include two intensity-based estimation models and one earthquake damage matrix model. The findings reveal that,among the cases reviewed,the earthquake damage matrix model demonstrates the highest precision in evaluating the number of deaths caused by earthquakes over the past five years. This underscores the importance of selecting the appropriate model based on the specific circumstances of an earthquake. Furthermore,the study highlights that the preconditions under which an earthquake results in fatalities are crucial for model selection. The post-earthquake evaluation is not a one-time task but an iterative process of refinement within a constrained timeframe. This continuous correction is essential for enhancing the accuracy and reliability of death toll assessments following earthquake events.