Session (CTI)

Issue: A number of PRO instruments, designed to generate health state utilities, are now available for children as well as for adults. In principle, these allow HRQoL to be measured in age-appropriate ways 'from cradle to grave'. But can the estimates of QALYs produced from child and adult versions of these instruments be interpreted in the same way? Do transitions between age-specific measures of HRQoL require special attention in cost effectiveness models? Overview: Methods, instruments and evidence on child HRQoL have improved considerably in the last five years. There are now a number of generic measures of HRQoL developed to be age-appropriate, that are accompanied by utilities that allow estimation of QALYs. An ISPOR Taskforce on utilities for child HRQoL has recently concluded. However, a remaining challenge is understanding how the estimates of QALYs produced from child and adult HRQoL instruments compare. This is important, because many healthcare interventions in childhood generate survival gains and improvements in HRQoL that need to be modelled from childhood into adulthood. Systematic differences in how HRQoL is measured and valued in each age group could potentially cause discontinuities in HRQoL and QALYs, even when underlying health remains unchanged. In this issues panel, Nancy Devlin will present new evidence on how the measurement properties of the EQ-5D-5L and EQ-5D-Y-5L compare in data collected from adolescents who completed both; and how the characteristics of utilities for child and adult HRQoL instruments differ. Janine Vertsraete will present results showing how these differences in measurement and valuation between child and adult HRQoL instruments interact when analysing preference-weighted health state descriptions. Finally, Donna Rowen will consider how potential discontinuities in the measurement and valuation of HRQoL could, in principle, be addressed and what methods could be developed for use to support the transition between age-specific instruments in cost effectiveness modelling. The session will allow time for Q&A, led by moderator Louis Matza.