Do Family Spillover Effects of Illness Vary With the Age of the Ill Individual?

OBJECTIVES: To evaluate whether family spillover effects, as measured by health-related quality of life, vary with the age of an ill family member.

METHODS: Health-related quality of life associated with COVID-19 and RSV illness was measured using time-tradeoff questions to derive losses in quality-adjusted life years (QALYs). Respondents were asked to value spillover effects of a family member’s illness on themselves. Data were collected via two online surveys with nationally representative samples of US adults in 2021 and 2023. The first survey (n=915) collected data on spillover effects for an ill child; the second survey (n=1062) collected spillover for ill working-age adults (18-64 years) or older adults (65+ years). Health states included: outpatient and hospitalized COVID-19 or RSV, COVID-19 with an intensive care unit (ICU) stay, hospitalization for multisystem inflammatory syndrome in children (MIS-C), and long COVID-19. Unadjusted analyses reported mean and median QALY losses by health state and age of ill family member. Adjusted analyses used beta regression to estimate the effect of an ill individual's age on QALY losses, controlling for health state and respondent characteristics.

RESULTS: Unadjusted spillover QALY losses were larger for ill children and older adults compared with working-age adults for most health states. Mean QALY losses peaked for a child, older adult, and working-age adult at 0.081, 0.018, and 0.023 respectively. Adjusted analyses showed a significant association between a child as the ill family member and higher spillover QALY losses (p-value<0.05); no association for ill older adults. Higher QALY losses were also associated with increasing age, women, and higher income.

CONCLUSIONS: Age may be associated with the magnitude of family spillover effects from COVID-19 and RSV illnesses (i.e., higher for children than other family members) and this relationship may vary across illness severity and duration. Further research is warranted on variation in spillover effects by patient characteristics.