Use of the PedsQL in childhood intermittent exotropia: estimates of feasibility, internal consistency reliability and parent-child agreement

Purpose To evaluate the PedsQLs performance in children with intermittent exotropia (X[T]) in terms of feasibility, internal consistency, floor-ceiling effects and levels of parent-child agreement. Methods Children with X(T) aged <12 years were recruited from 26 UK Hospital Eye Clinics/Orthoptic Departments. QOL was assessed using child (n = 166) and proxy (n = 392) versions of the PedsQLv4. Feasibility was assessed by percentage of missing responses; internal consistency by Cronbach's alpha and agreement by Bland-Altman plots and intraclass correlations. Analyses included age and gender comparisons. Results Missing response rates were no higher than 1.8%. Cronbach's alpha reached >= 0.70 on all but one parent-rated scale and on most child-rated Total, Psychosocial Summary and Social Functioning scales, but was <0.70 on most child-rated Physical, Emotional and School Functioning scales. On parent-rated scales, there were no floor effects; ceiling effects reached 27-56% in parents' Physical, Social and School Functioning. On child-rated scales, there were 0-1% floor effects and 0-28% ceiling effects. Parent-child agreement was fair to poor and varied by child's gender. Conclusions Proxy-rated PedsQLs demonstrated good internal consistency/feasibility in parents of children with X(T); child-rated reports appeared acceptable, although caution is advised regarding Physical, Emotional and School Functioning scales in younger children. Low-fair agreement between proxy and self-ratings is common in paediatric QOL assessment, reiterating the importance of obtaining both perspectives. We encourage future studies to explore the influence of child's age and gender, and the relationship of the proxy respondent.