ENSURING ESTIMAND AND POPULATION CONSISTENCY ACROSS HEALTH TECHNOLOGY ASSESSMENT DOSSIER COMPONENTS
Author(s)
Sylvaine Barbier, MSc1, Chloë Harden, PhD2, Catriona Crossan, PhD3, Mateusz Nikodem, PhD4;
1Putnam, Lyon, France, 2Putnam, London, United Kingdom, 3Putnam, Westport, Ireland, 4Putnam, Krakow, Poland
1Putnam, Lyon, France, 2Putnam, London, United Kingdom, 3Putnam, Westport, Ireland, 4Putnam, Krakow, Poland
OBJECTIVES: Health Technology Assessment (HTA) submissions integrate multiple interdependent evidence components, including systematic literature reviews (SLRs), indirect treatment comparisons (ITCs), cost-effectiveness (CE) models, and global value dossiers (GVDs). Failure to properly link these components, particularly regarding populations and outcomes definitions and estimands, may compromise internal validity and lead to misleading clinical and economic conclusions. This study aims to develop a practical guide to support internal consistency.
METHODS: A structured methodological assessment was conducted to identify critical “linking points” where misalignment commonly occurs. They were mapped across the evidence generation, with emphasis on alignment of target populations, estimands (including marginal versus conditional estimands), outcome definitions, treatment strategies, and time horizons. Identification of linking points was informed by HTA methodological guidance and expert review.
RESULTS: Key linking points requiring alignment were identified:- consistency between SLR PICOS criteria, populations parameters and outcomes used in CE models- alignment of ITC estimands with target decision population, including appropriate choice between marginal and conditional estimands- assessment of heterogeneity across trials populations included in the ITC and their transportability to the target population- coherence between relative treatment effects from ITCs and absolute event rates used in CE models- appropriate estimation of baseline risks, highlighting limitations of “average patient” approaches under non-linear link functions- alignment of treatment sequences and handling of treatment switching across evidence synthesis and CE models- consistency of endpoint and subgroup definitions across analyses.Failure to address these linking points may bias estimates of clinical and economic value.
CONCLUSIONS: Internal consistency is critical for generating valid, decision-relevant evidence. Misalignment of populations, outcomes, estimands, and treatments may introduce bias in the HTA conclusions, depending on the nature and direction of the mismatch. The proposed guide provides a structured framework to improve coherence and credibility of HTA submissions.
METHODS: A structured methodological assessment was conducted to identify critical “linking points” where misalignment commonly occurs. They were mapped across the evidence generation, with emphasis on alignment of target populations, estimands (including marginal versus conditional estimands), outcome definitions, treatment strategies, and time horizons. Identification of linking points was informed by HTA methodological guidance and expert review.
RESULTS: Key linking points requiring alignment were identified:- consistency between SLR PICOS criteria, populations parameters and outcomes used in CE models- alignment of ITC estimands with target decision population, including appropriate choice between marginal and conditional estimands- assessment of heterogeneity across trials populations included in the ITC and their transportability to the target population- coherence between relative treatment effects from ITCs and absolute event rates used in CE models- appropriate estimation of baseline risks, highlighting limitations of “average patient” approaches under non-linear link functions- alignment of treatment sequences and handling of treatment switching across evidence synthesis and CE models- consistency of endpoint and subgroup definitions across analyses.Failure to address these linking points may bias estimates of clinical and economic value.
CONCLUSIONS: Internal consistency is critical for generating valid, decision-relevant evidence. Misalignment of populations, outcomes, estimands, and treatments may introduce bias in the HTA conclusions, depending on the nature and direction of the mismatch. The proposed guide provides a structured framework to improve coherence and credibility of HTA submissions.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
HTA39
Topic
Health Technology Assessment
Disease
No Additional Disease & Conditions/Specialized Treatment Areas