Abstract
BACKGROUND AND OBJECTIVES: 'Caregiver health spillovers' refer to the broader impacts of an individual's illness and interventions on informal caregivers' health and well-being. This study focuses on the spillover effects experienced by parental carers of children with coronavirus disease 2019 (COVID-19), aiming to compare the psychometric properties of the EQ-5D-5L and the experimental EQ Health and Wellbeing Short version (EQ-HWB-S) in capturing these effects. METHODS: A longitudinal study was conducted with 861 parental carers of children aged 0-18 years with COVID-19 and 231 parents of healthy children as the control group. The EQ-5D-5L and EQ-HWB-S were used to assess parental health and well-being. Analyses included known-groups validity (multivariable regression), test-retest reliability (Gwet's AC1, intraclass correlation coefficient) and responsiveness to health improvement (Glass' Δ effect size). RESULTS: Parents of infected children reported more problems than those of healthy controls. The EQ-HWB-S better discriminated between sub-groups defined by the child's COVID-19 presence, caring time and work impact. Test-retest reliability was fair to good for EQ-HWB-S dimensions (Gwet's AC1: 0.33-0.79), moderate to good for EQ-5D-5L (Gwet's AC1: 0.40-0.76), and good for index scores and EQ VAS (intraclass correlation coefficient: 0.70-0.77). Parental health and well-being improved as children recovered, with the EQ-5D-5L showing slightly higher responsiveness (effect size: 0.77-0.87) than EQ-HWB-S (effect size: 0.62-0.74). CONCLUSIONS: Both EQ-HWB-S and EQ-5D-5L are valid, reliable and responsive for measuring parental spillover effects related to a child's COVID-19 infection. EQ-HWB-S outperformed in distinguishing social and emotional impacts of caregiving, while EQ-5D-5L better captured physical health improvements. The choice between tools may depend on study objectives.