A Taxonomy for Assessing Whether HRQoL Value Sets Are Obsolete.

Providing health-related quality of life (HRQoL) value sets to enable estimation of quality adjusted life years (QALYs) is important in facilitating economic evaluation and in supporting reliable decision-making about healthcare. However, as the field matures, many value sets across a range of HRQoL instruments are now old, based on potentially outdated valuation methodologies and preference data from samples that no longer represent the contemporary population. Having a clear strategy for identification and mitigation of obsolescence is important to ensure policy makers retain confidence in their country-specific value sets. In this Current Opinion, we develop a taxonomy of value set obsolescence. We then explore how the different types of obsolescence might be identified and how methodologists might work with local policymakers to address obsolescence and therefore ensure HRQoL instruments remain relevant for use. The taxonomy of obsolescence consists of four main areas: (1) the value set no longer aligns with current normative health technology assessment (HTA) requirements; (2) the methods used to generate it are no longer considered robust or adequately close to best practice; (3) the population composition has moved too far from the characteristics of the sample in which the original value set was derived; and (4) even after controlling for population differences, preferences are likely to have changed since the original data collection. Through identification of the type of obsolescence that applies in a particular setting, we then suggest a range of possible solutions to each, ranging from recommending particular sensitivity analyses, through reweighting of existing data to better account for population differences, to collecting new data for an updated value set. Obsolescence of existing value sets is driven by more than just time since data collection is often a matter of judgment rather than based on a clear definition. The taxonomy presented here provides a tool for assessing whether value sets are obsolete and what the appropriate response to this obsolescence should be. Working closely with local policymakers and involving discussions regarding the ongoing appropriateness of existing value sets should form an important part of future activities. This should include the consideration of updating value sets in contemporary populations using current best-practice methods. However, the benefits of updating value sets have to be balanced against the cost of doing so, including the challenges faced by policymakers when new values sets require a transition to new local decision-making processes.