One of the key challenges in interpreting data from patient-reported outcome measures (PROMs) is how to decide whether an effect is large enough to be clinically meaningful to the patient. Even when a threshold for clinical meaningfulness can be defined, it is sometimes unclear whether this threshold should be applied to the mean differences between treatment and control, or to an individual subject’s response, thus classifying each subject as a responder or nonresponder. In this presentation I will discuss the advantages and disadvantages of these two approaches, as well as a new approach, based on a clinical meaningfulness weighting function, that attempts to combine the advantages of both.