雞尾酒療法是目前治療愛滋病最有效的療法，衛生署自 1997 年 4 月開始提供病患雞尾酒療法之治療藥物，至今已逾十年之久。台灣關於雞尾酒療法對於愛滋病的影響使用數學模型來分析的例子仍很少見，因此本篇針對台灣愛滋病病患的資料，使用邊際模型 (marginal model) 及脆弱模型 (frailty model) 這兩種多維度存活分析的方法來分析這資料，以便探討使用雞尾酒療法對於 HIV 檢驗成陽性到愛滋病病發，及病發後到死亡這兩段時間療效之差異以及 CD4 / CD8 比值高低的影響。

Highly Active Anti-Retroviral Therapy, or HAART, is highly beneficial to many HIV-infected individuals. The Department of Health in Taiwan has began to provide the treatment
of HAART for the AIDS patients since April, 1997. However, so far in Taiwan, there are very few cases using mathematical models to analyse the efficacy of HAART to HIV patients. Conseqently, to investigate the problem, we use marginal model and frailty model, two methods of multivariate survival data analysis. We want to find the different effect of HAART in two time periods, HIV infection to onset of AIDS and onset of AIDS to death and the different effects of CD4 / CD8 ratio.

[1] Andersen, P. K. and Gill, R. D. (1982). Cox’s regression model for counting processes:
A large sample study. Annals of Statistics 10, 1100-1120.
[2] Cai, J. and Prentice, R. L. (1995). Estimating equations for hazard ratio parameters
based on correlated failure time data. Biometrika 82, 151–164.
[3] Chen, A. P. (2007). Ordered Bivariate Survival Time with Time Dependent Covariate
-Comparison of Marginal Method. NCU Thesis.
[4] Clayton, D. G. (1978). A model for association in bivariate life tables and its applica-
tion in epidemiological studies of chronic disease incidence. Biometrika 65, 141–151.
[5] Clayton, D. G. and Cuzick, J. (1985). Multivariate generalisations of the proportional
hazards model. Journal of the Royal Statistical Society, Series A, 148, 82–117.
[6] Cook, R. J. and Lawless, J. F. (2002). Analysis of repeated events. Statistical Methods
in Medical Research 11, 141-166.
[7] Cox, D. R. (1972). Regression models and life-tables (with Discussion). Journal of
the Royal Statistical Society:Series B, 34, 187-220.
[8] Crowder, M. (1989). A multivariate distribution with Weibull connections. J. R.
Statist. Soc. B, 51, 93-107.
[9] Fleming, T. R. and Harrington, D. P. (1991). Counting Processes and Survival Anal-
ysis. John Wiley and Sons, New York.
[10] Guo, G. and Rodr′ ıguez, G. (1992). Estimating a multivariate proportional hazards
model for clustered data using the EM algorithm. With an application to child sur-
vival in Guatemala. J. Am. Statist. Assoc. 87, 969-976.
[11] Hougaard, P. (1986). A class of multivariate failure time distributions. Biometrika
73, 671–678.
[12] Huber, P. J. (1967). The behaviour of maximum likelihood estimates under non-
standard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical
Statistics and Probability 1, 221–233.
[13] Klein, J. P. (1992). Semiparametirc estimation of random effects using the Cox model
based on the EM algorithm. Biometrika 48, 795-806.
[14] Lawless, J. F. and Nadeau, C. (1995). Some simple robust methods for the analysis
of recurrent events. Technometrics 37, 158-168.
[15] Lin, D. Y. and Wei, L. J. (1989). The robust inference for the Cox proportional
hazard model. Journal of the American Statistical Association 84, 1074-1078.
[16] Lin, D. Y. (1994). Cox regression analysis of multivariate failure time data: the
marginal approach. Statistics in Medicine 13, 2233-2247.
[17] Lin, D. Y., Wei, L. J., Yang, I. and Ying, Z. (2000). Semiparametric regression for the
mean and rate functions of recurrent events. Journal of the Royal Statistical Society,
Series B, 62, 711–730.
[18] MaGilchrist, C. A. and Aisbtt, C. W. (1991). Regression with frailty in survival
analysis. Biometrics 47, 461–466.
[19] Nielsen, G. G., Gill, R. D., Andersen, R. K. and Sørensen, T. I. A. (1992). A count-
ing process approach to maximum likelihood estimation in frailty models. Scand. J.
Statist. 19, 25-43.
[20] Oakes, D. (1982). A model for association in bivariate survival data. Journal of the
Royal Statistical Society, Series B, 44, 414–422.
[21] Oakes, D. (1992). Frailty models for multiple event times. Lifetime Data Anal. 4,
209–228.
[22] Pepe, M. S. and Cai, J. (1993). Some graphical displays and marginal regression anal-
yses for recurrent failure times and time-dependent covariates. Journal of American
Statistical Association 88, 811–820.
[23] Prentice, P. L., Williams, B. J. and Peterson, A. V. (1981). On the regression analysis
of multivariate failure time data. Biometrika 68, 373-379.
[24] Therneau, T. M. and Grambsch, P. M. (eds) (2000). Modeling survival data:extending
the Cox Model. Springer, New York.
[25] Vaupel, J. W., Manton, K. G. and Stallard, E. (1979). The impact of heterogeneity
in individual frailty on the dynamics of mortality. Demography, 16:439–454.
[26] Wei, L. J., Lin, D. Y., and Weissfeld, L. (1989). Regression analysis of multivariate
incomplete failure time data by modeling marginal distributions. Journal of American
Statistical Association 84, 1065-1073.
[27] Wulfsohn, M. S., Lin, D. Y., and Tsiatis, A. A. (1997). A Joint Model for Survival
and Longitudinal Data Measured with Error. Biometrics 53, 330-339.