Mind the (inter-rater) gap. An investigation of self-reported versus proxy-reported assessments in the derivation of childhood utility values for economic evaluation: A systematic review.

PURPOSE

Evidence surrounding utilities for health states, derived either directly from the application of preference-based valuation methods or indirectly from the application of preference-based quality of life instruments, is increasingly being utilised to inform the cost-effectiveness of child health interventions. Proxy (parent or health professional) assessments are common in this area. This study sought to investigate the degree of convergence in childhood utilities generated directly or indirectly within dyad child and proxy assessments.

METHODS

A systematic literature review was conducted following PRISMA guidelines. A comprehensive literature search strategy conducted across six search engines (PubMed, Embase, Web of Science, PsychoINFO, EconLit, CINAHL and Cochrane Library). Original peer-reviewed articles that reported utilities derived directly or indirectly using simultaneous dyad child and proxy assessments were extracted. Mean and median utilities, correlation coefficients and levels of agreement were extracted, catalogued and assessed.

RESULTS

A total of 35 studies that reported utilities for two or more respondent types were identified. Of these, 29 studies reported dyad childhood self-report and proxy utilities whilst six studies reported levels of agreement and/or correlations only without documenting overall utilities. Proxy assessment was most often conducted by parents with the HUI3 representing the most commonly applied instrument across a range of health conditions. The utilities derived from child and parent proxy assessment were bidirectional with parental proxies tending to underestimate and health professional proxies tending to overestimate relative to child self-reports. Inter-rater agreement between child self-reports and parent-proxy reports were poorer for more subjective attributes (cognition, emotion and pain), relative to physical attributes (mobility, self-care, speech, vision) of health-related quality of life.

CONCLUSIONS

Childhood utilities derived from children or proxies are not interchangeable. The choice of self or proxy assessor may have potentially significant implications for economic evaluations of child health interventions.