This chapter discusses how medical illness can influence outcomes in standard randomized controlled trial (RCT) methodology. We review how medical illness can affect the occurrence, identification, and measurement of psychiatric disorders and how medical illness and environmental factors can affect psychiatric symptoms longitudinally over the course of an RCT. Finally, we argue that when outcomes for RCTs of validated psychological and behavioral treatments are substantially smaller among medical populations than among non-medical populations, the problem may lie in our ability to accurately identify the psychiatric disorder in that population, rather than in the intervention. We will discuss the implications of this for RCT design.