Populating an economic model with health state utility values: moving towards better practice

Ara R, Brazier J

Background: When estimating health state utility values (HSUV) for multiple health conditions, the
alternative models used to combine these data can produce very different values. Results generated
using a baseline of perfect health are not comparable with those generated using a baseline adjusted for
not having the health condition taking into account age and gender. Despite this, there is no guidance
on the preferred techniques that should be used and very little research describing the effect on cost per
QALY results.
Methods: Using a cardiovascular disease (CVD) model and cost per QALY thresholds, we assess the consequence of using different baseline health state utility profiles (perfect health, individuals with no
history of CVD, general population) in conjunction with three models (minimum, additive,
multiplicative) frequently used to estimate proxy scores for multiple health conditions.
Results: Assuming a baseline of perfect health ignores the natural decline in quality of life associated
with co-morbidities, over-estimating the benefits of treatment to such an extent it could potentially
influence a threshold policy decision. The minimum model biases results in favour of younger aged
cohorts while the additive and multiplicative technique produced similar results.
Although further research in additional health conditions is required to support our findings, this pilot
study highlights the urgent need for analysts to conform to an agreed reference case and provides initial
recommendations for better practice. We demonstrate that in CVD, if data are not available from
individuals without the health condition, HSUVs from the general population provide a reasonable