Africa’s first standard measure for child and adolescent health: The Nigerian EQ-5D-Y-3L value set using composite time trade-off and discrete choice experiments

Abstract

Nigeria has a rapidly growing child and adolescent population, yet no locally derived EQ-5D-Y-3 L utility weights (value set) exists for estimating quality-adjusted life years (QALYs) in paediatric economic evaluations. This study developed the first Nigerian EQ-5D-Y-3 L value set to support context-specific economic evaluation and evidence-informed priority setting for child and adolescent health interventions. A study was conducted across six Nigerian states using the EuroQol Portable Valuation Technology (EQ-PVT). A total of 2,700 adults were recruited. All respondents completed the DCE valuation tasks, while a quota-sampled subset of 540 respondents from the same sampling framework across the six states additionally completed the cTTO valuation tasks. Preferences for EQ-5D-Y-3 L health states were elicited through face-to-face interviewer-administered valuation tasks, in which adult respondents of the general population valued health states from the proxy perspective of a hypothetical 10-year-old child, using standardized EQ-VT protocols comprising composite time trade-off (cTTO) and discrete choice experiment (DCE) tasks. The best-fitting DCE model was a mixed logit model without constant, selected based on likelihood-based measures and convergence diagnostics. Latent DCE coefficients were subsequently anchored to the QALY scale by regressing mean cTTO disutilities on predicted DCE disutilities, consistent with established EQ-VT methodology. The anchored DCE coefficients were then used to generate utility values for all 243 EQ-5D-Y-3 L health states. Data from 2,679 respondents yielding 40,185 DCE choice tasks and 5400 cTTO observations were analysed. All EQ-5D-Y-3 L dimensions showed strong internal consistency and monotonic worsening across severity levels. Pain/discomfort Level 3 showed the largest utility decrement, followed by anxiety/depression Level 3, whereas looking after myself (self-care) had the smallest impact on utility. Utilities ranged from 1.000 for state 11,111 to 0.117 for 33,333. The mixed logit without a constant produced the best-performing mapping (RMSE = 0.1337; p < 0.001). This study provides the first Nigerian EQ-5D-Y-3 L value set and represents the first EQ-5D-Y-3 L valuation study in Africa conducted using the EQ-VT protocol. The resulting value set provides a methodologically rigorous foundation for QALY-based economic evaluation of child and adolescent health interventions. It will enable contextually relevant health economic analyses and support evidence-based priority setting in Nigeria and in similar low- and middle-income countries.