How vital signs change at the end of life is a frequently asked question by health-care professionals, patients, and families, and research groups have attempted to identify specific changes and indicators associated with impending death. This study proposed a new method for detecting characteristic changes in vital signs associated with patient’s death based on time-series data measured by a pressuresensing mat. The pressure-sensing mat is less invasive and can automatically log heart and respiratory rates with high temporal resolutions. In the proposed method, the probability density functions (PDFs) of vital signs are estimated based on the large amount of data, and temporal changes in the PDFs are evaluated using an information measure, the Jenson-Shannon divergence (JSD). The proposed method was used to detect changes in vital signs of homecare patients to demonstrate its applicability. By tracking the temporal changes in the PDFs and JSD of vital signs, the specific changes associated with death can be detected. The proposed method can be a promising approach in quantitative and simplified prediction of impending death in homecare patients.