2016 Annual Meeting
SC60-Practical Statistics for Pathologists: Case-based Instruction with Methods
Room CC 609, March 18 2016, 1:00pm to 4:30pm
SC60-Practical Statistics for Pathologists: Case-Based Instruction with Methods
Session Credits: 3 CME and 3 SAMs Faculty: Dan Milner, MD, Emily King Meserve, MD and Thing Rinda Soong, MD, PhD, Brigham and Womens Hospital, Boston, MA
There is a paucity within residency training (and even scientific research training) of formal statistical instruction such that residents and junior faculty who want to engage in pathology-related research do not have the skills required to read, interpret, or perform statistical testing on their data. Moreover, when they try to "self-learn" statistics for even the simplest projects, incorrect assumptions, insufficient knowledge, and incorrect application of certain tests leaves their data and results un-interpretable. Within academic institutions, there is a very heavy emphasis put on producing manuscripts dealing with basic clinical, surgical, radiological, and pathological data as measures of success or milestones of knowledge for resident trainees and junior faculty. Yet, these same programs and departments most often lack a statistical instruction series. Data sets themselves are becoming massive and require that a pathologist, who may provide only a small fraction of a data set or be attempting to integrate a large amount of non-pathological data, understand from start to finish the rules, requirements, and tools for data collection and analysis. This short course will provide example-based instruction in statistics relevant to Pathology practice by utilizing published studies to focus on a range of topics. Participants will be given the articles and reading guides prior to the session.
Upon completion of this educational activity, participants should be able to:
- Determine statistical tests necessary for pathology based studies
- Read and interpret with confidence the statistical portions of published papers
- Understand the importance of valid statistical testing when publishing in scientific literature.