Multi-state models for late e ffects in childhood cancer survivors | Kadriye Kaplan
| | Abstract | This thesis deals with statistical methods and their applications for describing diabetes-related morbidity and cause-specfii c mortality in the Nordic childhood cancer survivors. The purpose in the study is to analyze these outcomes in the survivors compared to the general population both separately and jointly by using multi-state models. The data is provided by a Nordic childhood cancer study, the Adult Life after Childhood Cancer in Scandinavia (ALiCCS). It encompasses information about 24936 childhood cancer survivors who are diagnosed during 1943 to 2008 in Denmark and Sweden and are matched with 124663 controls on gender and date of birth. In order to study statistical methods for analyzing the morbidity and mortality outcomes, the performing process is divided into two parts. In the first part of the analysis the standard two-state models are considered separately for each outcome whereas in the second part of the analysis more advanced multi-state models are constructed for describing the morbidity and mortality outcomes jointly. Both analyses are based on the ordinary as well as extended Cox models. The obtained results from the analyses are summarized and discussed. Both the standard two-state models and multi-state models have shown some similar results in the univariate analyses. These models have revealed that the childhood cancer survivors are associated with higher risk of experiencing both morbidity and mortality outcome when compared to the general population. In addition to this, multi-state analysis has shown that the childhood cancer survivors were more likely to die if they have developed diabetes than the other way around. Furthermore, it is found that the occurrence of diabetes has increased the risk of death in the study participants. | | Type | Master's thesis [Academic thesis] | | Year | 2012 | | Publisher | Technical University of Denmark, DTU Informatics, E-mail: reception@imm.dtu.dk | | Address | Asmussens Alle, Building 305, DK-2800 Kgs. Lyngby, Denmark | | Series | IMM-M.Sc.-2012-19 | | Note | Supervised by Professor Per Brockho ff, pbb@imm.dtu.dk, DTU Informatics, and co-supervised by Klaus Kaae Andersen | | Electronic version(s) | [pdf] | | Publication link | http://www.imm.dtu.dk/English.aspx | | BibTeX data | [bibtex] | | IMM Group(s) | Mathematical Statistics |
|