<p>Rapid estimation of earthquake consequences is crucial for emergency response. Integrated loss assessment methods, which estimate both building loss and fatalities, offer a more comprehensive impact evaluation and are thus widely applied. However, existing methods are either overly simplified or rely on assumptions that lack effective validation with actual data, leading to insufficient accuracy. Based on previous research in building seismic damage uncertainty and correlation modeling, this study developed a new integrated earthquake loss estimation method. The method treats building damage as a multi-dimensional correlated random variable and uses sampling techniques to simulate possible damage scenarios. It then establishes a nonlinear relationship between building damage extents and fatality rates, enabling an integrated estimation of both building loss and fatalities. Additionally, using Kriging interpolation improves the efficiency of handling multi-dimensional correlated random variables. The method was applied to estimate the residential building losses and fatalities caused by building damage in the M_w7.4 Hualien earthquake in Taiwan on April 3, 2024. The results indicate that the residential building losses are likely to range from 0.1 to 1.0 billion. All fatalities due to building damage occurred in Hualien County, with a 74.34% probability that fewer than 10 people perished. Compared to other existing methods, the integrated method’s estimates are more aligned with actual reports, both in terms of residential loss and fatality estimation, proving the method’s accuracy. This study represents the first application of the developed integrated earthquake loss estimation method, contributing new perspectives to earthquake risk quantification.</p>