Personalized Medicine in Benefit:Risk Assessment
*Scott Evans, Harvard University 


Methods for personalized/tailored medicine using within-patient analyses of benefits and risks will be discussed. The methods allow for high order interactions in defining subgroups, control error rates for global and subgroup-specific inference, and account for the censoring and competing risk issues associated with benefit and risk events. Incorporating patient preference and weighting outcomes will be discussed. Ideas for improved design, monitoring, analyses, and reporting of benefits and risks will be high-lighted.