This program trains predoctoral and postdoctoral biostatisticians in statistical theory and methods as applied to clinical and preclinical cancer studies. Integrated collaboration between biostatisticians and oncologist provides trainees with (i) supervision by biostatisticians with extensive experience in developing and applying statistical methodology, (ii) interaction with cancer researchers, (iii) practical experience in consulting projects in cancer, (iv) mentoring trainees in their development of an independent academic career. The Department of Biostatistics in the University of Rochester Medical School administers the program, providing the critical link between the University of Rochester Cancer Center (URCC), where trainees gain their practical experience, and the Department of Statistics where they receive their education in statistical theory. Trainees (2 postdoctoral and 2 predoctoral) with a strong mathematical background and prior training in statistics and/or the biological sciences are individually matched to one of four (primary) biostatistical mentors and to one of four mentors from URCC. Pre-doctoral trainees will graduate with a doctorate in Statistics (Biostatistics option) after three to five years of study, depending on prior experience, postdoctoral trainees will complete their program in one or two years. All trainees undertake course work in statistical theory and methods, in research methods (including ethics in research) and in cancer biology. All trainees become involved in cancer studies, and all pursue independent research in biostatistical methodology applicable to cancer. Selection of pre-doctoral trainees is based on academic record, GRE scores, and recommendations. Postdoctoral trainees will be expected also to have demonstrated ability to undertake independent research. The Cancer Biostatistics Training Grant Committee, comprising both Biostatistical and Cancer mentors, meets three times a year to review progress. Applications are screened by the Program Director - all short-listed applications are subsequently assessed also by two other Biostatistical mentors and one URCC mentor. An outside advisory committee meets annually to review the program.
| Peng, F; Hall, W J (1996) Bayesian analysis of ROC curves using Markov-chain Monte Carlo methods. Med Decis Making 16:404-11 |
| Sun, Y; Sherman, M (1996) Some permutation tests for survival data. Biometrics 52:87-97 |
| Jacobs, R A; Tanner, M A; Peng, F (1996) Bayesian inference for hierarchical mixtures-of-experts with applications to regression and classification. Stat Methods Med Res 5:375-90 |
| Rapoport, A P; Rowe, J M; Kouides, P A et al. (1993) One hundred autotransplants for relapsed or refractory Hodgkin's disease and lymphoma: value of pretransplant disease status for predicting outcome. J Clin Oncol 11:2351-61 |
