Problems with PROMs
Before discussing the kinds of changes and outlook that “new psychometric methods” suggest, it is helpful to understand the context in which they have arisen. Specifically, it is useful to examine some of the criticisms of PROMs that have led to the suggestion of new psychometric methods. The essence of the problem with PROMs can be summarized by a single deficiency that has multiple consequences; namely, PROMs lack a theory that provides a representation of the measurement interaction—the relationship between the construct and its instrument. This has been my argument throughout a number of papers, and, with regard to the general contours of my argument, I am not alone (McClimans 2010a, b, 2015; McClimans and Browne 2012). Consider Donna Lamping’s 2008 Presidential address to the International Society of Quality of Life Research, where she identified the need for a theoretical framework as one of three challenges facing the future of PROMs. Or take Jeremy Hobart et al. (2007 Lancet Neurology) where they lament the lack of explicit construct theories in their article criticizing the current state of PROMs. In a final example, Sonja Hunt, in her 1997 editorial for Quality of Life Research, argues that the surfeit of poorly designed measures suggests that we do not know what quality of life is (Hunt 1997). In what follows, I provide a brief overview of three consequences that result from this lack of theory: problems with validity, interpretability, and responsiveness.