SYSTEMATIC REVIEW OF THE METHODOLOGY, MODEL STRUCTURE, AND DATA INPUTS OF HEALTH ECONOMIC MODELS EVALUATING SYSTEMIC THERAPIES IN ATOPIC DERMATITIS
Author(s)
Tessa Vernooij, MSc1, Cornelis Boersma, PhD2, Marlies de Graaf, dr.1, Marjolein de Bruin-Weller, prof. dr.1, Lisa Aniek de Jong, MSc, PharmD, PhD3.
1University Medical Center Utrecht, Utrecht, Netherlands, 2Health-Ecore, Zeist, Netherlands, 3University of Groningen, Groningen, Netherlands.
1University Medical Center Utrecht, Utrecht, Netherlands, 2Health-Ecore, Zeist, Netherlands, 3University of Groningen, Groningen, Netherlands.
OBJECTIVES: To provide an overview of health economic decision models evaluating systemic therapies in patients with AD, with a focus on methodological approaches, model structures, data sources, and key assumptions.
METHODS: This PRISMA-guided systematic review of PubMed (MEDLINE), Embase, the Cochrane Library, and Scopus (PROSPERO (CRD420261419329)) included structured health economic models evaluating systemic therapies in AD published from 2017 onwards, and was supplemented by manual searches of relevant HTA body websites. Data were extracted on the methodological characteristics of health economic models. Risk of bias was assessed by using the ECOBIAS checklist, a tool that identifies sources of bias in model-based economic evaluations.
RESULTS: Ten peer-reviewed publications and 12 HTA body reports met the predefined inclusion criteria, spanning 11 countries and 8 systematic therapies. A total of 20 out of 23 models shared a hybrid decision tree-Markov structure inherited from the first NICE dupilumab appraisal (2018), including an unvalidated long-term waning assumption replicated across 15 studies. Utility inputs varied substantially across studies, and flare disutility and treatment sequencing were absent in most models, though both emerged in the two most recent NICE appraisals, reflecting gradual methodological progress.
CONCLUSIONS: Most health economic models in AD share a common template with key unvalidated assumptions, particularly long-term treatment waning. Future models should empirically validate these assumptions, standardise utility measurement, and integrate real-world sequencing data to improve transferability and better reflect clinical practice.
METHODS: This PRISMA-guided systematic review of PubMed (MEDLINE), Embase, the Cochrane Library, and Scopus (PROSPERO (CRD420261419329)) included structured health economic models evaluating systemic therapies in AD published from 2017 onwards, and was supplemented by manual searches of relevant HTA body websites. Data were extracted on the methodological characteristics of health economic models. Risk of bias was assessed by using the ECOBIAS checklist, a tool that identifies sources of bias in model-based economic evaluations.
RESULTS: Ten peer-reviewed publications and 12 HTA body reports met the predefined inclusion criteria, spanning 11 countries and 8 systematic therapies. A total of 20 out of 23 models shared a hybrid decision tree-Markov structure inherited from the first NICE dupilumab appraisal (2018), including an unvalidated long-term waning assumption replicated across 15 studies. Utility inputs varied substantially across studies, and flare disutility and treatment sequencing were absent in most models, though both emerged in the two most recent NICE appraisals, reflecting gradual methodological progress.
CONCLUSIONS: Most health economic models in AD share a common template with key unvalidated assumptions, particularly long-term treatment waning. Future models should empirically validate these assumptions, standardise utility measurement, and integrate real-world sequencing data to improve transferability and better reflect clinical practice.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
SA110
Topic
Study Approaches
Topic Subcategory
Literature Review & Synthesis
Disease
No Additional Disease & Conditions/Specialized Treatment Areas