PHP171 - Inclusion of real-world evidence in submission packages to health technology assessment bodies: What do current guidelines indicate?

Abstract

To describe and assess current recommendations by health technology assessment (HTA) bodies for incorporating real-world evidence (RWE) in submission packages. We reviewed the most recent pharmacoeconomic (PE) and submission guidelines and technical documentation, for HTA bodies identified through the ISPOR website. We limited the review to documents written in English. Within each document, we searched for explicit or implicit recommendations for incorporating RWE to inform PE models and submission packages. We defined RWE as evidence used for decision making that is not collected in randomized controlled trials (RCTs). We reviewed 18 PE and submission guidelines (3 American; 13 European; 2 Oceanian). In 13 guidelines, inclusion of RWE was recommended or considered. Among those, descriptions were provided regarding the relevance and suitability of RWE for: assessing the technology’s clinical effect and potential risks, inclusion as parameters in PE models, and external validation of such models. For guidance relating specifically to the description of clinical effect, eight deemed RWE acceptable under specific circumstances, with proper quality assessment, or as supplements to RCTs. Technical support documents from the United Kingdom provided guidance on RWE study design hierarchy, types of RWE to include in PE models, and on methods for estimating treatment effectiveness using RWE; such guidance was scarce in other submission guidelines. The remaining five guidelines were of limited scope and did not state their position regarding the inclusion of RWE. RWE can help capture the full value of a new technology and inform decision makers of the anticipated impact of being incorporated into real clinical practice. While the inclusion of RWE within HTA submission packages is addressed by most HTA bodies, the scope of the guidance is often limited or incomplete. Further work to clarify standards for generating and selecting relevant and robust RWE to be incorporated in submission packages is needed.