Joint Model for Survival and Longitudinal Data － A Case Study on Taiwan AIDS Data

• Main Authors: Yi-shan Shih, 施伊珊
• Other Authors: 曾議寬
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/99985856413118889895

碩士 === 國立中央大學 === 統計研究所 === 95 === Over the past decade, much research have been done on Highly Active Antiretroviral Therapy (HAART), which has been proven to be able to postpone morbidity and lower mortality rate. This research collects data from 141 AIDS patients in Taiwan, which are separated into two groups, HAART and non-HAART. The research objective is to investigate the effect HAART and the CD4 cells to lifetime of AIDS patients. Research methods include graphic techniques and joint-modeling approach. Graphic techniques allow us to understand basic information from data, and the relationship between CD4 cells and mortality rate with the use of HARRT. Biased estimation usually occurs in the longitudinal data with intermittent measurement schedule. Joint modeling approach can correct this bias. Furthermore, the estimates derived by joint model have been proved to be consistent and efficient. Our findings are consistent with previous studies. The use of HARRT enables the increase of CD4 cells and extension of patients’ lifetime.