Abstract Objective: Readmission is a big part of health care expenditure and recent studies suggested that hospital readmissions can be applied as an important indicator of quality of care within health care system. Furthermore elderly population usually costs the large amount health care expenses and is the main group in readmission. Moreover readmission is usually attributed to chronic diseases. Nevertheless, evaluations of hospital readmissions under universal health care coverage areas were not well-studied in Taiwan. Therefore this study aims to explore the associations between initial hospitalizations and probability of hospital readmissions in details. Research method: Patients aged 65 or older with primary clinical diagnosis of asthma or COPD based on ICD-9-CM for hospital admissions and readmissions in Taiwan.National claims of these two diseases were collected and analyzed from year 2000 to 2004. Population-based descriptive analyses of related health care utilizations were estimated. Multivariate logistic regressions were conducted to predict the probability of hospital readmissions. Controlled variables included patient factors, medical institutions¡¦ characteristics, urbanizations, and air quality indicators. Result: Among asthma and COPD elderly patients, more health care utilizations were observed in the hospital readmissions than initial admissions. Multivariate logistic regressions indicated that age, gender, disease severity, hospital characteristics, and air quality were significant predictors of the probability of hospital readmission. Meanwhile, age, disease severity, and hospital characteristics also significantly affected the time interval between initial admission and readmission. In addition, longer length of stay in the initial admission will significantly shorten the time interval between initial admission and readmission (P<0.001). Conclusion: For Asthma and COPD elderly patients, longer length of stay in the initial admission will significantly shorten the time interval between initial admission and readmission and have higher probability of hospital readmission. This study provides the evidence of reducing the health care expenditure by controlling readmission rate. With more understandings of factors affecting hospital readmissions, we can improve the health care delivery and reduce unplanned readmissions in the future. Key words: Asthma, COPD, hospital readmission, health care utilization, length of stay, admission fees