Constantinos Koshiaris

I am a senior medical statistician interested in the development and validation of clinical prediction rules for both diagnosis and prognosis. My clinical areas of research include early diagnosis of cancer and particularly myeloma and cardiovascular disease prevention.

I am the principal statistician of the Stratified TreAtments Research (STAR) Group which uses data from routines electronic health care records to better understand the association between cardio-protective medication and harms. Our work focuses on the development of models which can identify which patients are at higher risk of experiencing side effects from their treatment.

Another area of research that I am working on is Monoclonal gammopathy of undetermined significance (MGUS). MGUS is an asymptomatic condition which is picked up incidentally when investigating for other conditions.  MGUS can eventually develop to myeloma but not much is known about the risk factors that might increase your risk of progression from one stage to the other. Currently I lead a research project where we use routines electronic health care records to identify the risk factors that might affect the risk of progression.

I currently supervise three DPhil students within the Department (Rik van der Veen, Ting Cai and Joseph Lee). I am a module coordinator for the Clinical perdition rules course run by the Oxford Postgraduate Programme in Evidence-Based Healthcare and also teach on other modules such as Big Data Epidemiology, Essential Medical Statistics, Introduction to Study design and Research methods and Statistical Computing with R and STATA.