The purpose of the Biostatistics Core is to provide design and analytic support to SPORE Projects, Developmental Projects, and other Cores. This core has three specific aims. 1. To provide biostatistical expertise for the planning, conduct, analysis, and reporting of laboratory, animal, clinical, and epidemiological studies for SPORE projects and developmental projects. The Biostatistics Core will work with SPORE researchers on protocol and experimental design, and will provide estimates of sample size and statistical power. The Core will advise researchers about optimal methods for monitoring ongoing studies. The Core will provide statistical analyses of completed experiments, and participate in manuscript preparation. 2. To provide or recommend supporting computational infrastructure, including computer databases, statistical computer packages, publicly available statistical programs and data sharing between computers or databases. Because an effective informatics infrastructure is critical to the functioning of the SPORE, the Biostatistics Core will work to assure that the storage and accessibility of data used and generated by the SPORE is efficient and effective. The Biostatistics Core will support efficient management of data resources by consulting on database design, helping to design efficient data collection methods and conducting audits of data quality. 3: To provide consulting and statistical education to SPORE researchers. In addition to formal collaborative efforts, the Biostatistics Core will be available for ongoing, informal consultation with researchers over the duration of the SPORE. In addition the Biostatistics Core will work with investigators within the DF/HCC Skin Cancer SPORE, other Skin Cancer SPOREs and Cooperative Groups to attempt to build a prognostic and predictive model for melanoma patients with a web-based statistical product that will help physicians define the prognosis and select the optimal therapy for individual patients with various stages of melanoma.
Showing the most recent 10 out of 132 publications
