OBJECTIVES: With the growing interest in utilizing real-world evidence (RWE) for regulatory and reimbursement decision-making, manufacturers must navigate when and how to uilitize RWE in their evidence packages. Previous studies found that RWE is requested by HTA submission guidelines for cost, resource use, and adherence inputs in economic models; however, actual RWE adoption has not been estimated. This study quantifies the frequency of RWE use as inputs/assumptions in NICE and ICER cost-effectiveness models.

METHODS: Psoriasis HTAs from NICE and ICER published in 2018 - May 2019 were evaluated. Cost, resource use, and adherence inputs/assumptions and their sources were extracted; RWE sources were quantified. RWE sources was defined as observational data derived from claims, electronic health records, registries, or national cost data.

RESULTS: There were 28 model inputs/assumptions in the 5 Psoriasis HTAs evaluated. 64% (18/28) inputs/assumptions clearly utilized RWE, while 7 inputs/assumptions did not state the source and RWE use could not be determined. Both ICER (3) and NICE (4) inputs/assumptions lacked transparency. Even when RWE was evaluated, HTA agencies requested more updated and relevant data. For example, resource use for best supportive care in multiple NICE assessments was based on a 2010 observational study of data from 2003-2008. Inputs/assumptions sources differed between NICE and ICER assessments, but stayed consistent across NICE assessments.

CONCLUSIONS: Even in instances where RWE is accepted and preferred in HTAs, RWE is underutilized; the utility of RWE is not maximized in HTAs. Previous RWE submissions have been met with requests for updated data. While NICE assessments were consistent over time, more current data could have replaced older inputs/assumptions. There is an opportunity for manufacturers and HTA agencies to leverage recent RWE as inputs and to justify model assumptions in value propositions to payers.