Heterogeneity of Treatment Effect — An Evolution in Subgroup Analysis

Heterogeneity of treatment effect is intuitive to the clinician at the bedside. When we recommend the best treatment for patients on the basis of their individual characteristics, we are applying the concept of heterogeneity of treatment effect. More formally, heterogeneity of treatment effect is defined as nonrandom variation in the benefit or harm of a treatment, in which the variation is associated with or attributable to patient characteristics.1 The article by Desai et al.2 in this issue of NEJM Evidence provides a contemporary example of exploring heterogeneity of treatment effect and offers an opportunity to understand how their approach fits within the larger research landscape.