In studies of cancer and AIDS, patients are monitored for clinical events and laboratory markers that are known to be associated with declining health and an increased risk of death. Understanding the pattern of disease progression is important for the clinical management of individual patients, as well as for the design and analysis of clinical trials for new therapies. The statistical analysis of disease progression is complicated by the fact that patients miss visits, resulting in incomplete information either on when an event occurred (interval censored data) or on the value of a clinical or laboratory measurement at some points in time. This issue is made more complex in the context of a study of a fatal disease, where it is important to appropriately handle mortality. This proposal is aimed at determining a useful approach to analyzing multiple outcomes of progression from a study with missing, truncated, and censored data. For this, we have four aims. 1.) We will develop methods for the analysis of multivariate failure time data, where the failures are subject to interval censoring. 2.) We will develop a method for the analysis of the failure time data from a cancer genetics study, where the participants are subject to truncation because the genetic test became available only recently (1996). 3.) We will develop a method for the analysis of screening data, for which the chance of being screened is dependent on the event (failure) of interest (informative censoring). 4.) We will develop methods of estimating the effect of longitudinal changes in a marker of progression on the hazard of a subsequent event indicating progression. These methods will be applied to data from clinical trials and observational studies in cancer (conducted at the MGH Cancer Center), in AIDS (conducted by the NIH sponsored AIDS Clinical Trials Group), and in cancer genetics (conducted by the NIH sponsored Cancer Genetics Network).
| Schoenfeld, David A; Rajicic, Natasa; Ficociello, Linda H et al. (2011) A test for the relationship between a time-varying marker and both recovery and progression with missing data. Stat Med 30:718-24 |
| Finkelstein, Dianne M; Wang, Rui; Ficociello, Linda H et al. (2010) A score test for association of a longitudinal marker and an event with missing data. Biometrics 66:726-32 |
| Rajicic, Natasa; Finkelstein, Dianne M; Schoenfeld, David A et al. (2009) Analysis of the relationship between longitudinal gene expressions and ordered categorical event data. Stat Med 28:2817-32 |
| Schoenfeld, David A; Hui Zheng; Finkelstein, Dianne M (2009) Bayesian design using adult data to augment pediatric trials. Clin Trials 6:297-304 |
| Matthews, Abigail G; Finkelstein, Dianne M; Betensky, Rebecca A (2008) Analysis of familial aggregation studies with complex ascertainment schemes. Stat Med 27:5076-92 |
| Kerber, Richard A; Amos, Christopher I; Yeap, Beow Y et al. (2008) Design considerations in a sib-pair study of linkage for susceptibility loci in cancer. BMC Med Genet 9:64 |
| Rao, Sowmya R; Schoenfeld, David A (2007) Survival methods. Circulation 115:109-13 |
| Matthews, Abigail G; Finkelstein, Dianne M; Betensky, Rebecca A (2007) Multivariate logistic regression for familial aggregation in age at disease onset. Lifetime Data Anal 13:191-209 |
| Zhang, Zhigang; Sun, Liuquan; Sun, Jianguo et al. (2007) Regression analysis of failure time data with informative interval censoring. Stat Med 26:2533-46 |
| Matthews, Abigail G; Betensky, Rebecca A; Anton-Culver, Hoda et al. (2006) Analysis of co-aggregation of cancer based on registry data. Community Genet 9:87-92 |
Showing the most recent 10 out of 23 publications
