EXPLORING THE IMPLICIT INCLUSION OF INFORMAL CARER SPILLOVERS IN HEALTH STATE VALUATION USING DISCRETE CHOICE EXPERIMENT
Author(s)
Jan Faller1, Cathy Mihalopoulos, PhD2, Gang Chen, MSc, PhD2, Brendan Mulhern, PhD3, David John Mott, PhD4, Richard Norman, BA, MSc, PhD5, Lidia Engel, PhD2.
1Melbourne, Australia, 2Monash University, Melbourne, Australia, 3Centre for Health Economics Research and Evaluation, Sydney, Australia, 4Office of Health Economics, London, United Kingdom, 5Curtin University, Perth, Australia.
1Melbourne, Australia, 2Monash University, Melbourne, Australia, 3Centre for Health Economics Research and Evaluation, Sydney, Australia, 4Office of Health Economics, London, United Kingdom, 5Curtin University, Perth, Australia.
OBJECTIVES: This study explored the extent to which informal carer spillover effects are considered in health state valuation using discrete choice experiment (DCE) with duration.
METHODS: A DCE was conducted in which participants completed 12 choice tasks, each involving a selection between two EQ-5D-5L health states with duration. Three survey versions were used: traditional-only DCE with duration, non-traditional-only DCE with duration, and combined DCE with duration (traditional then non-traditional). In the non-traditional versions, participants were explicitly told that the health states would require support from an informal carer. To assess whether preferences are similar between the traditional and non-traditional DCE with duration versions, indicating potential implicit incorporation of informal carer spillover effects, Swait-Louviere tests on conditional logit models were conducted, accounting for differences in choices between surveys.
RESULTS: A total of 2,089 members from the general Australian population participated in the study (traditional-only, n=516; non-traditional-only, n=517; and combined, n=1,056). About 45% of respondents completing the traditional version indicated considering the impact of the health states on others. Among those who did not consider impacts on other, 42% indicated that their responses would change if they were asked to consider others. In the non-traditional version, 36% indicated that their responses would change if the health states did not depend on informal care. Poolability was supported only between the traditional-only and non-traditional-only datasets, suggesting the potential implicit incorporation of informal care spillover in the traditional exercises. However, poolability was not supported within the combined arm, indicating public preferences in the traditional exercises would change if the same set of respondents were asked to revisit their choices and explicitly consider informal carer spillover.
CONCLUSIONS: Informal carer spillover effects are potentially partially captured in traditional valuation exercises, and that framing plays a meaningful role in shaping preferences.
METHODS: A DCE was conducted in which participants completed 12 choice tasks, each involving a selection between two EQ-5D-5L health states with duration. Three survey versions were used: traditional-only DCE with duration, non-traditional-only DCE with duration, and combined DCE with duration (traditional then non-traditional). In the non-traditional versions, participants were explicitly told that the health states would require support from an informal carer. To assess whether preferences are similar between the traditional and non-traditional DCE with duration versions, indicating potential implicit incorporation of informal carer spillover effects, Swait-Louviere tests on conditional logit models were conducted, accounting for differences in choices between surveys.
RESULTS: A total of 2,089 members from the general Australian population participated in the study (traditional-only, n=516; non-traditional-only, n=517; and combined, n=1,056). About 45% of respondents completing the traditional version indicated considering the impact of the health states on others. Among those who did not consider impacts on other, 42% indicated that their responses would change if they were asked to consider others. In the non-traditional version, 36% indicated that their responses would change if the health states did not depend on informal care. Poolability was supported only between the traditional-only and non-traditional-only datasets, suggesting the potential implicit incorporation of informal care spillover in the traditional exercises. However, poolability was not supported within the combined arm, indicating public preferences in the traditional exercises would change if the same set of respondents were asked to revisit their choices and explicitly consider informal carer spillover.
CONCLUSIONS: Informal carer spillover effects are potentially partially captured in traditional valuation exercises, and that framing plays a meaningful role in shaping preferences.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
PCR265
Topic
Economic Evaluation, Patient-Centered Research
Topic Subcategory
Health State Utilities
Disease
No Additional Disease & Conditions/Specialized Treatment Areas