Development of a preference weighting system for the parent proxy child chronic cough quality-of-life (PC-QoL) instrument.

Background: Cough in children is the most common reason for seeking healthcare in Australia. When chronic, it is associated with decreased quality-of-life in parents/ carers and significant societal costs. Despite this, the spillover effect of chronic cough on parents/carers is seldom accounted for in economic evaluations of interventions. We aimed to develop a new method of estimating spillover health utility in this population in Australia.

Methods: We conducted a discrete choice experiment (DCE) on hypothetical health states based on the PC-QoL. These were analyzed using a garbage class multinomial logit model (GCL). We also obtained VAS scores for 6 health states and mapped them to the latent DCE utilities, rescaling them to the 0-1 health utility scale required to estimate QALYs.

Results: Five hundred and fifty participants broadly representative of Australian parents completed our survey. Parental concerns about their child being able to lead a normal life had the largest coefficients in the GCL. The resulting scoring algorithm had a minimum score of .21, and a maximum of 1 (full health) CONCLUSIONS: We have developed a new method of estimating spillover health utility values in parents of children with chronic cough in Australia. This study is also a use-case for the application of GCL in extracting respondent non-trading behavior, that may cause inaccurate preference estimates, and a VAS based anchoring methodological approach that could be iterated in further research. We have developed an R package and shiny app to allow the easy estimation of spillover utility scores from PC-QoL responses.