WHEN IS EXISTING PATIENT PREFERENCE INFORMATION FIT-FOR-REUSE? RECOMMENDATIONS FROM AN EDELPHI STUDY
Author(s)
Michael Bui, MSc1, Catharina G. M. Groothuis-Oudshoorn, PhD2, G. Ardine de Wit, PhD3, Byron Jones, PhD4, Conny Berlin, MSc5, Janine A. van Til, MSc, PhD2.
1PhD student, University of Twente, Enschede, Netherlands, 2University of Twente, Enschede, Netherlands, 3RIVM / Vrije Universiteit Amsterdam, 3720BA Bilthoven, Netherlands, 4Novartis, Basel, Switzerland, 5Novartis Pharma AG, Basel, Switzerland.
1PhD student, University of Twente, Enschede, Netherlands, 2University of Twente, Enschede, Netherlands, 3RIVM / Vrije Universiteit Amsterdam, 3720BA Bilthoven, Netherlands, 4Novartis, Basel, Switzerland, 5Novartis Pharma AG, Basel, Switzerland.
OBJECTIVES: Regulators increasingly endorse the reuse of patient preference information (PPI) to avoid unnecessary burden on patients during medical product development. However, guidance on when published PPI is fit-for-reuse is lacking. We conducted an electronic Delphi (eDelphi) study to develop content-valid and consensus-based guidance on this topic.
METHODS: A targeted literature review on factors influencing the transferability of evidence in health economics and outcomes research informed the draft guidance comprising 12 items categorized into 3 domains: methodological characteristics, population characteristics, and healthcare context. The guidance was iteratively refined through a 3-round eDelphi study involving 53 unique experts from academia, industry, consultancy, patient advocacy, regulatory agencies, and health technology assessment bodies. First, experts rated item clarity and essentiality, and suggested item revisions and additions through a survey. Second, focus groups were used to address the issues raised in the survey. Third, based on a final survey, items were retained if a significant majority of experts agreed on their essentiality according to a one-tailed exact binomial test (α=0.05), in accordance with Lawshe’s content validity framework.
RESULTS: After 3 eDelphi rounds, consensus was achieved on 9 items. Agreement was nearly unanimous for items assessing whether methodological and population characteristics of existing preference studies match the requirements of the new decision context. Methodological characteristics included items assessing the decision-relevance of attributes, levels, and framing of the preference instrument. Population characteristics included items assessing representativeness of the sample with respect to sociodemographic and clinical characteristics. Healthcare context items achieved consensus with lower agreement rates (68-76%), and were hypothesized to primarily apply to PPI applications concerning uptake prediction.
CONCLUSIONS: We provide a first step towards structured assessments of whether existing PPI can be reused, or whether a de novo study is warranted. Empirical applications of our guidance will provide opportunities to demonstrate its practical usefulness and further refine its development.
METHODS: A targeted literature review on factors influencing the transferability of evidence in health economics and outcomes research informed the draft guidance comprising 12 items categorized into 3 domains: methodological characteristics, population characteristics, and healthcare context. The guidance was iteratively refined through a 3-round eDelphi study involving 53 unique experts from academia, industry, consultancy, patient advocacy, regulatory agencies, and health technology assessment bodies. First, experts rated item clarity and essentiality, and suggested item revisions and additions through a survey. Second, focus groups were used to address the issues raised in the survey. Third, based on a final survey, items were retained if a significant majority of experts agreed on their essentiality according to a one-tailed exact binomial test (α=0.05), in accordance with Lawshe’s content validity framework.
RESULTS: After 3 eDelphi rounds, consensus was achieved on 9 items. Agreement was nearly unanimous for items assessing whether methodological and population characteristics of existing preference studies match the requirements of the new decision context. Methodological characteristics included items assessing the decision-relevance of attributes, levels, and framing of the preference instrument. Population characteristics included items assessing representativeness of the sample with respect to sociodemographic and clinical characteristics. Healthcare context items achieved consensus with lower agreement rates (68-76%), and were hypothesized to primarily apply to PPI applications concerning uptake prediction.
CONCLUSIONS: We provide a first step towards structured assessments of whether existing PPI can be reused, or whether a de novo study is warranted. Empirical applications of our guidance will provide opportunities to demonstrate its practical usefulness and further refine its development.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
PCR267
Topic
Methodological & Statistical Research, Patient-Centered Research
Disease
No Additional Disease & Conditions/Specialized Treatment Areas