The purpose of this study was to use survival analysis to examine the influence of living arrangements and health care utilizations on total mortality among the middle aged and elderly in Taiwan. Apanel data design was applied to data to investigate whether living arrangements and health care utilizations were associated significantly to survival rates in the middle aged and elderly. Subject data was obtained from the Survey of Health and Living Status of the Middle Aged and Elderly in Taiwan, a study conducted in 1996 that encompassed observations on 2,462 individuals 50 years of age and older. Survey data was linked to 1996-2003 national death registry data. A Cox proportional hazard model was used to determine the relationship between mortality rate and age, gender, living arrangements and health care conditions. Principal empirical results confirmed that, after controlling for potentially confounding variables, the relatively younger elderly had a higher survival rate during the period 1996 to 2003. Females also had a longer life span than males in the same period. In addition, when other factors were excluded, this paper highlights that the middle aged and elderly living with their spouse and children enjoyed a significantly higher survival rate than those who lived alone. Furthermore, the findings showed that purchase of medicine (Chinese and/or Western) was positively correlated with survival rate and a lower survival rate for those with longer hospital stays and higher numbers of clinic visits. Findings suggest that living arrangements and health care utilization variables have strong and significant influences on the mortality rates for the middle aged and elderly. Research results should be useful in developing welfare strategies targeting living arrangement and health care services for the target group.