Streamlining the Validation of Patient Reported Outcome (PRO) Measures in Drug Regulatory Processes

The context and setting in which patient-reported outcome (PRO) measures applied in drug development programmes are developed and applied are undergoing rapid changes. The emergence of social media, coupled with advancement of patient-centred methodologies, has led to more connected and empowered patients than before. Drug regulatory agencies are beginning to embrace a new role as collaborators in the drug development process, and new models for defining core outcomes in various disease areas are emerging. Technological advances and social changes have led to the wide use of personal gadgets with greater internet connectivity and computing power than most computers had a generation ago. To explore how progress in the measurement of PROs might be achieved amidst such changes, lessons were drawn from the PRO labelling claims reviewed by the US FDA following the publication of the PRO guidance in 2009. In addition, semi-structured interviews were carried out with seven experts working with drug regulatory agencies and in the pharmaceutical industry to identify major issues and possible solutions. The following recommendations are proposed: (1) that comprehensive hypotheses about the PRO concept be developed; (2) that PRO concepts/measures maintain substantive and clinical validity; (3) that the use of advanced psychometric methods be considered as the new gold standard in the development of PRO measures; (4) that psychometric properties of measures established through “validation by application” be enhanced; (5) that the generalizability of PRO measures be enhanced; (6) that the assessment of ability to detect change, and the interpretability of scores, be improved; and (7) that PRO guidelines by major regulatory authorities be aligned. These entail a greater transparency on the decisions taken during the PRO measure development process and greater understanding of the integration between qualitative aspects of the PRO measure and the score/measurement attributes.