The aim of this application is to develop statistical methods to make appropriate and efficient use of the data collected in epidemiologic and clinical AIDS studies. Two projects involving markers of disease progression and their relationship to clinical endpoints will be performed. They both utilize flexible and data driven methods, enabling more complete conclusions to be made about disease markers and clinical endpoints. In the first project the Investigator and his associates will develop, test, and apply a very general and comprehensive model for analyzing jointly longitudinal marker data and failure time data. The model incorporates (i) a smooth function for the marker population average, (ii) a flexible stochastic process determining marker progression, (iii) a time-dependent proportional hazards model for the endpoint with non-parametrically specified baseline hazard function, (iv) measurement error in the marker, (v) additional covariates influencing both the progression of the marker and the hazard of the event, and (vi) the ability to handle unbalanced and unequally spaced observations. This model will allow a proper assessment of the importance of markers and of other covariates in clinical and epidemiological studies. The model will be fit using Markov Chain Monte Carlo methods. The second project is to develop methods that will enable randomized clinical trials to be completed more quickly. The researchers will develop, test, and apply methods that recover information from censored observations using time-dependent markers to make efficient inference about the clinical endpoint. Two approaches will be developed, both utilizing multiple imputation of future clinical events followed by analysis of the augmented data. The first is an extension of the joint model in the first project, to forecast and thus impute future events. The second approach is to apply censored data regression techniques which allow for missing values to impute multiply for each censored case the event time. Non-parametric adaptations will be developed. The methods in both projects will be evaluated and compared to other approaches on simulated, clinical trial and Multicenter AIDS Cohort Study (MACS) data. The methodology has wide applicability to both AIDS and non-AIDS studies.

Agency
National Institute of Health (NIH)
Institute
National Institute of Allergy and Infectious Diseases (NIAID)
Type
Research Project (R01)
Project #
5R01AI029196-10
Application #
6052979
Study Section
AIDS and Related Research Study Section 2 (ARRB)
Program Officer
Dixon, Dennis O
Project Start
1990-04-01
Project End
2001-03-31
Budget Start
1999-04-01
Budget End
2001-03-31
Support Year
10
Fiscal Year
1999
Total Cost
Indirect Cost
Name
University of Michigan Ann Arbor
Department
Biostatistics & Other Math Sci
Type
Schools of Public Health
DUNS #
791277940
City
Ann Arbor
State
MI
Country
United States
Zip Code
48109
Hsu, Chiu-Hsieh; Taylor, Jeremy M G; Murray, Susan et al. (2007) Multiple imputation for interval censored data with auxiliary variables. Stat Med 26:769-81
Faucett, Cheryl L; Schenker, Nathaniel; Taylor, Jeremy M G (2002) Survival analysis using auxiliary variables via multiple imputation, with application to AIDS clinical trial data. Biometrics 58:37-47
Taylor, Jeremy M G; Wang, Yan (2002) Surrogate markers and joint models for longitudinal and survival data. Control Clin Trials 23:626-34
Liu, M; Taylor, J M; Belin, T R (2000) Multiple imputation and posterior simulation for multivariate missing data in longitudinal studies. Biometrics 56:1157-63
Taylor, J M; Wang, Y; Ahdieh, L et al. (2000) Causal pathways for CCR5 genotype and HIV progression. J Acquir Immune Defic Syndr 23:160-71
Roussanov, B V; Taylor, J M; Giorgi, J V (2000) Calculation and use of an HIV-1 disease progression score. AIDS 14:2715-22
Bycott, P; Taylor, J (1998) A comparison of smoothing techniques for CD4 data measured with error in a time-dependent Cox proportional hazards model. Stat Med 17:2061-77
Boscardin, W J; Taylor, J M; Law, N (1998) Longitudinal models for AIDS marker data. Stat Methods Med Res 7:13-27
Taylor, J M; Law, N (1998) Does the covariance structure matter in longitudinal modelling for the prediction of future CD4 counts? Stat Med 17:2381-94
Bycott, P W; Taylor, J M (1998) An evaluation of a measure of the proportion of the treatment effect explained by a surrogate marker. Control Clin Trials 19:555-68

Showing the most recent 10 out of 23 publications