Dr. Heller’s current research interests include the design and analysis of phase 2 clinical trials, the analysis of survival data, and the investigation of predictive accuracy measures. For the analysis of phase 2 studies, he has developed a test statistic that adjusts for patient risk when the patient population is heterogeneous. This approach is applicable in targeted therapy trials, where multiple patient populations are treated under a single protocol. In survival analysis, he has developed a measure of predictive accuracy, based on the concept of explained randomness. This metric, applied to risk models with survival data, is used to determine the adequacy of a risk classification model for clinical decision making. In addition, he is developing statistical methods to evaluate the improvement in risk classification measures due to the inclusion of new biomarkers. The statistical methods will provide a metric to decide whether a change in patient risk, due to the inclusion of new biomarkers into the risk model, is random variation or a true improvement in the accuracy of a risk classification model. Finally, Dr. Heller is involved in the design and analysis of laboratory and clinical studies emanating from the Departments of Medicine, Pediatrics, and Surgery.