Unravelling Elements of Value of Healthcare and Assessing their Importance Using Evidence from Two Discrete-Choice Experiments in England

Abstract

Background Health systems are moving towards value-based care, implementing new care models that allegedly aim beyond patient outcomes. Therefore, a policy and academic debate is underway regarding the definition of value in healthcare, the inclusion of costs in value metrics, and the importance of each value element. This study aimed to define healthcare value elements and assess their relative importance (RI) to the public in England. Method Using data from 26 semi-structured interviews and a literature review, and applying decision-theory axioms, we selected a comprehensive and applicable set of value-based elements. Their RI was determined using two discrete choice experiments (DCEs) based on Bayesian D-efficient DCE designs, with one DCE incorporating healthcare costs expressed as income tax rise. Respondent preferences were analysed using mixed logit models. Results Six value elements were identified: additional life-years, health-related quality of life, patient experience, target population size, equity, and cost. The DCE surveys were completed by 402 participants. All utility coefficients had the expected signs and were statistically significant ( p < 0.05). Additional life-years (25.3%; 95% confidence interval [CI] 22.5–28.6%) and patient experience (25.2%; 95% CI 21.6–28.9%) received the highest RI, followed by target population size (22.4%; 95% CI 19.1–25.6%) and quality of life (17.6%; 95% CI 15.0–20.3%). Equity had the lowest RI (9.6%; 95% CI 6.4–12.1%), decreasing by 8.8 percentage points with cost inclusion. A similar reduction was observed in the RI of quality of life when cost was included. Conclusion The public prioritizes value elements not captured by conventional metrics, such as quality-adjusted life-years. Although cost inclusion did not alter the preference ranking, its inclusion in the value metric warrants careful consideration. Supplementary Information The online version contains supplementary material available at 10.1007/s40273-024-01416-5.