ADDRESSING EVIDENCE GAPS THROUGH STRUCTURED EXPERT ELICITATION: INSIGHTS FROM RECENT NICE TECHNOLOGY APPRAISALS
Author(s)
Anurag Gupta, MSc1, Hugo Pedder, PhD2, Aman Yadav, MSc1, Kate Ren, PhD3.
1ConnectHEOR, Delhi, India, 2University of Bristol; ConnectHEOR Limited, Bristol, United Kingdom, 3ConnectHEOR, London, United Kingdom.
1ConnectHEOR, Delhi, India, 2University of Bristol; ConnectHEOR Limited, Bristol, United Kingdom, 3ConnectHEOR, London, United Kingdom.
OBJECTIVES: Survival extrapolation is often influential in HTA decision-making but is often associated with substantial uncertainty due to immature trial data. Expert judgement is typically needed to inform plausible long-term survival assumptions. NICE Technical Support Document (TSD) 26 provides structured guidance for obtaining systematic and reliable quantitative expert judgements on survival outcomes using structured expert elicitation (SEE). This study assessed the use of SEE in recent NICE technology appraisals (TAs) following publication of TSD 26 in March 2025.
METHODS: NICE TAs published between 1 June 2025, and 1 June 2026 were reviewed. Data were extracted on whether expert elicitation was conducted, whether SEE was used, whether SEE informed survival or non-survival outcomes, and the indication/disease area.
RESULTS: A total of 103 TAs were retrieved, of which 26 were terminated/withdrawn; 77 were reviewed. Expert elicitation was conducted in 10/77 (13%) TAs. Advisory board meetings to seek expert opinion were not considered formal expert elicitation. SEE was utilized in 6/10 (60%) TAs, whereas methodology of elicitation was not reported in 4/10 (40%) TAs. The 6 SEE appraisals spanned across five therapeutic areas: oncology, 2/6 (33.3%); and women's health, respiratory, cardiovascular, and dermatology, 1/6 each (16.7%). The methods used for SEE included SHELF (1/6; 16.7%), Delphi-panel (1/6; 16.7%) and not reported in remaining 4/6 (66.7%) TAs. SEE was used to inform survival extrapolation in 3/6 appraisals (50.0%) and to address other evidence gaps in the remaining 3/6 (50.0%).
CONCLUSIONS: SEE is increasingly being adopted in recent NICE TAs but its implementation and reporting in recent appraisals appears limited and inconsistent. Informal consultations through advisory boards may be vulnerable to bias, inconsistency, and limited transparency. NICE TSD 26 provides clear guidance for SEE and its recommendations should be adopted to improve the credibility, transparency, and robustness of long-term survival assumptions in HTA decision-making.
METHODS: NICE TAs published between 1 June 2025, and 1 June 2026 were reviewed. Data were extracted on whether expert elicitation was conducted, whether SEE was used, whether SEE informed survival or non-survival outcomes, and the indication/disease area.
RESULTS: A total of 103 TAs were retrieved, of which 26 were terminated/withdrawn; 77 were reviewed. Expert elicitation was conducted in 10/77 (13%) TAs. Advisory board meetings to seek expert opinion were not considered formal expert elicitation. SEE was utilized in 6/10 (60%) TAs, whereas methodology of elicitation was not reported in 4/10 (40%) TAs. The 6 SEE appraisals spanned across five therapeutic areas: oncology, 2/6 (33.3%); and women's health, respiratory, cardiovascular, and dermatology, 1/6 each (16.7%). The methods used for SEE included SHELF (1/6; 16.7%), Delphi-panel (1/6; 16.7%) and not reported in remaining 4/6 (66.7%) TAs. SEE was used to inform survival extrapolation in 3/6 appraisals (50.0%) and to address other evidence gaps in the remaining 3/6 (50.0%).
CONCLUSIONS: SEE is increasingly being adopted in recent NICE TAs but its implementation and reporting in recent appraisals appears limited and inconsistent. Informal consultations through advisory boards may be vulnerable to bias, inconsistency, and limited transparency. NICE TSD 26 provides clear guidance for SEE and its recommendations should be adopted to improve the credibility, transparency, and robustness of long-term survival assumptions in HTA decision-making.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
HTA392
Topic
Health Technology Assessment, Study Approaches
Topic Subcategory
Decision & Deliberative Processes
Disease
No Additional Disease & Conditions/Specialized Treatment Areas