Abstract

Objectives

Ghana’s economic evaluation reference case recommends quality-adjusted life-years as an outcome measure for the conduct of cost-utility analysis. There is no Ghanaian value set available to be used in estimating quality-adjusted life-years. This study aimed to develop a value set for Ghana using the EQ-5D-5L instrument.

Methods

Face-to-face preference data were collected from 300 adults across 3 regions of Ghana using the adapted version of the EuroQol valuation technology (EQ-VT) standardized valuation protocol; with composite time-trade-off (cTTO) and discrete-choice experiments (DCEs) elicitation techniques. The cTTO and DCE data were modeled individually or together to provide complementary results on respondents’ utility preferences. Models explored were generalized least squares, Tobit, heteroskedastic, logit, and hybrid. The best-fitting model for the value set was selected based on its logical consistency, accounting for left-censored and heteroscedasticity data, and the statistical significance of parameters.

Results

The 300 interviews provided 4500 cTTO responses and 4200 DCE responses. The preferred model chosen for the Ghana value set was the Hybrid Tobit random effect heteroscedastic-constrained model. The predicted value for the worst attainable health state (55555) was −0.493 and the best health state (11112) was 0.969. The largest decrement was registered for level 5 mobility (0.369) followed by pain/discomfort (0.312), self-care (0.273), anxiety/depression (0.271), and usual activities (0.268).

Conclusions

This is the first Ghanaian EQ-5D-5L value set based on social preference derived from a nationally representative sample. The value set will play a key role in the use of economic evaluation studies to inform priority setting in Ghana where different health technologies can be compared.