Constructing and evaluating a validity argument for a performance outcome measure for clinical trials: An example using the Multi-luminance Mobility Test

Abstract

Background Clinical trials that evaluate new medical products often use clinical outcome assessments to measure how patients feel or function. Determining the evidentiary support needed for clinical outcome assessments is challenging but necessary to ensure scores from a clinical outcome assessment reflect the relevant aspects of patients’ health. Modern validity theory—from educational and psychological testing—addresses the challenge by requiring that investigators state key assumptions underlying the proposed use of a clinical outcome assessment and collect evidence for or against those assumptions. Methods This article describes the argument-based approach to validity using an example of a performance outcome measure—the Multi-luminance Mobility Test—designed to assess patients with inherited retinal dystrophy that causes progressive loss of night vision. For the proposed interpretation and use of a performance outcome measure to be reasonable, several key assumptions need to be plausible. I describe the assumptions along with examples of supporting evidence from the published literature to evaluate each assumption within the rationale. Results This article provides an example of a validity rationale to evaluate a clinical outcome assessment using the Multi-luminance Mobility Test as an example. Conclusion The demonstration illustrates the use of the argument-based approach to validity evaluation and the challenges in supporting parts of a validity rationale for clinical outcome assessments that measure how patients feel and function in a more indirect way. By making clinical outcome assessment validation practices consistent with modern validity theory, investigators, sponsors, and regulators should be able to communicate more clearly and direct resources more efficiently to support the creation of patient-centered endpoints in clinical trials.