Presentation (CTI)

OBJECTIVES: In Health Technology Assessment (HTA), multiple survival models may provide adequate fits to the observed data but provide heterogeneous predictions of survival at later time points. Structured expert elicitation (SEE) can be used to inform survival model selection. With the release of National Institute for Health and Care Excellence Technical Support Document 26 (TSD 26), it is envisioned that SEE methods for survival outcomes will feature more prominently within HTA. However, SEE can be resource and time intensive, this review explores the use of alternative remote approaches within SEE and discusses potential limitations.
METHODS: A targeted re-appraisal of the broader literature review presented in TSD 26 was conducted. Six studies were re-assessed, and data extracted relating to facilitator-led and remote components of the SEE. The findings were considered in the context of the recommendations in TSD 26.
RESULTS: Several elements of SEE appear to be conducted remotely, such as the review of the evidence dossier, whereas other elements, such as the provision of expert individual judgements appear to be conducted variably. Some studies provided training in a remote format as opposed to facilitator-led sessions, and concerns were raised regarding levels of expert understanding. Discussion between experts was highlighted as a key component of SEE, but in the case of remote formats, this was not always possible.
CONCLUSIONS: Although remote SEE exercises may appear to be less time and resource intensive, they may require multiple rounds of contact, as opposed to a single facilitated session. Additionally, exclusively remote formats may not fully comply with TSD 26 recommendations, raising concerns of the use of remotely provided judgements within HTA. In conclusion this work shows that some components of SEE can be conducted remotely in accordance with TSD 26, but these should be carefully chosen to ensure the provision of credible, contextual expert judgements.