A review of methods used to estimate mean preference-based utilities for comorbidities using published summary statistics

The prevalence of co-morbidities increases with age but preference-based utilities are generally obtained from cohorts who have a single condition. This can cause problems when populating health states in economic models which represent more than one condition. Analysts use the mean utility from the cohorts with the single conditions to estimate the mean utility for a cohort with co-morbidities. There is currently no consensus on which is the most appropriate method to combine these data and the different techniques can produce very different results.