ADAPTING RAPID REVIEW METHODS FOR HEOR DECISION-MAKING: A FIT-FOR-PURPOSE FRAMEWORK FOR EFFICIENT EVIDENCE SYNTHESIS
Author(s)
Kassandra Schaible1, Tracy Taylor, MSc2, Henry Ogden, PhD3, Apurva Pande, PhD4, Solange Gault, MD5, SOHAN NITIN DESHPANDE, MSc3.
1Thermo Fisher Scientific, Pittsburgh, PA, USA, 2Thermo Fisher Scientific, Ede, Netherlands, 3Thermo Fisher Scientific, London, United Kingdom, 4Thermo Fisher Scientific, Waltham, MA, USA, 5Thermo Fisher Scientific, Orlando, FL, USA.
1Thermo Fisher Scientific, Pittsburgh, PA, USA, 2Thermo Fisher Scientific, Ede, Netherlands, 3Thermo Fisher Scientific, London, United Kingdom, 4Thermo Fisher Scientific, Waltham, MA, USA, 5Thermo Fisher Scientific, Orlando, FL, USA.
OBJECTIVES: Existing guidelines provide methods for conducting rapid reviews but limited practical guidance for HEOR applications. General guidance on when rapid reviews are appropriate, how pragmatic approaches align with decision risk, and how AI can be safely incorporated to improve efficiency is needed. This work aimed to adapt the National Collaborating Centre for Methods and Tools (NCCMT) rapid review framework for HEOR applications and to identify appropriate use cases and AI-enabled efficiencies.
METHODS: The NCCMT Rapid Review Guidebook was used as the base framework and supplemented by other published methodological guidance. Core review steps were mapped to various HEOR decision contexts, from evidence landscaping to value propositions. Opportunities for AI assistance were identified across stages, including question refinement, search strategy development, citation deduplication, title and abstract prioritization, full-text triage, data extraction support, evidence table generation, and narrative synthesis drafting. Minimum safeguards for AI assistance in rapid reviews included human oversight, auditability, source verification, and transparent reporting of AI use.
RESULTS: The proposed framework includes seven domains: define the problem; refine research question(s); targeted searches; screening/prioritization of evidence; data extraction; streamlined synthesis; and reporting of limitations, uncertainty, and AI involvement. Rapid reviews are most useful for decisions that require fit-for-purpose evidence when comprehensive evidence synthesis is not required. Appropriate applications include early asset planning, evidence landscaping, payer evidence strategy, evidence gap analysis, review scoping, and determining whether a full systematic literature review is warranted. They are not recommended for regulatory-grade evidence submissions, formal comparative effectiveness conclusions, meta-analysis, or guideline development.
CONCLUSIONS: An NCCMT-based rapid review framework adapted for HEOR can improve consistency, transparency, and decision relevance by explicitly linking pragmatic approaches and appropriate AI use to decision context and evidentiary risk. Rapid reviews should complement, not replace, systematic reviews when comprehensive evidence synthesis is required.
METHODS: The NCCMT Rapid Review Guidebook was used as the base framework and supplemented by other published methodological guidance. Core review steps were mapped to various HEOR decision contexts, from evidence landscaping to value propositions. Opportunities for AI assistance were identified across stages, including question refinement, search strategy development, citation deduplication, title and abstract prioritization, full-text triage, data extraction support, evidence table generation, and narrative synthesis drafting. Minimum safeguards for AI assistance in rapid reviews included human oversight, auditability, source verification, and transparent reporting of AI use.
RESULTS: The proposed framework includes seven domains: define the problem; refine research question(s); targeted searches; screening/prioritization of evidence; data extraction; streamlined synthesis; and reporting of limitations, uncertainty, and AI involvement. Rapid reviews are most useful for decisions that require fit-for-purpose evidence when comprehensive evidence synthesis is not required. Appropriate applications include early asset planning, evidence landscaping, payer evidence strategy, evidence gap analysis, review scoping, and determining whether a full systematic literature review is warranted. They are not recommended for regulatory-grade evidence submissions, formal comparative effectiveness conclusions, meta-analysis, or guideline development.
CONCLUSIONS: An NCCMT-based rapid review framework adapted for HEOR can improve consistency, transparency, and decision relevance by explicitly linking pragmatic approaches and appropriate AI use to decision context and evidentiary risk. Rapid reviews should complement, not replace, systematic reviews when comprehensive evidence synthesis is required.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
SA64
Topic
Health Policy & Regulatory, Methodological & Statistical Research, Study Approaches
Topic Subcategory
Literature Review & Synthesis
Disease
No Additional Disease & Conditions/Specialized Treatment Areas