OBJECTIVES: Fewer than 5% of UK health technology appraisals consider caregiver (dis)utility. It is unclear if participants completing traditional health state (HS) valuation tasks already implicitly consider the impact of the HS on others, e.g., caregivers, particularly for severe HS. This pilot study explored whether EQ-5D-5L HS valuations differ with the inclusion of caregiver information.

METHODS: Face-to-face composite time trade-off (cTTO) interviews were conducted with a convenience sample of people working at Roche Products Ltd, which included a ‘traditional’ (TE) and ‘combined’ (CE) cTTO exercise. The TE valued three HS: A (32312), B (55544) and C (55555). In the CE, participants valued the same HS but were informed the HS now required a caregiver. The caregiver’s HS varied depending on the participant’s HS (A with 11122; B with 11132; C with 11434). Participants provided qualitative feedback regarding their responses to both exercises. Primary analysis involved a paired t-test comparison of the utilities elicited in each exercise.

RESULTS: Thirty-three participants were enrolled (73% female, median age 45 years [range 24–51], 48% with experience of receiving/giving care for serious illness). Data suggest considerable heterogeneity in individual response trajectories from TE to CE in both directions, contributing to non-significant differences to the mean utilities: CE/TE means (mean difference 95% confidence interval; p-value) A: 0.72/0.79 (-0.17, 0.02; p=0.10), B: 0.07/0.0 (-0.06, 0.21; p=0.27) and C: -0.26/-0.19 (-0.23, 0.09; p=0.39). Following prompting 22/33 (67%) would have valued the HS differently in the CE had the caregiver been in full health.

CONCLUSIONS: Despite no significant differences in mean utilities, individual response trajectories varied greatly, suggesting that providing contextual information about caregivers could influence HS valuations in different directions. Larger studies to understand the full extent of this effect are required.