Agreement Between Parent and Child Report of Health-Related Quality of Life: Impact of Time Postinjury

Abstract

Health-related quality of life (HRQL) is subjective concept and, therefore, should be captured directly from the patient. However, proxy reporting of HRQL is widespread, particularly in pediatric studies where children have been considered unreliable respondents. This study assessed the level of agreement between proxy (parent) and child reports of HRQL at key time points after injury. Thirty-seven seriously injured children aged 13 years to 16 years participated in this study. The Pediatric Quality of Life inventory was administered to the parent and child at 1 month, 6 months, and 12 months after injury by telephone interview. Agreement between child and parent responses was compared using Bland-Altman plots, and Pediatric Quality of Life inventory physical and psychosocial summary scales were compared using paired t tests or Wilcoxon signed-rank tests, respectively. At 1-month (psychosocial t = -4.6, p < 0.001; physical t = -6.5 p < 0.001) and 6-month (psychosocial z = -2.5, p = 0.01; physical z = -2.6, p = 0.01) postinjury there was a significant difference between the parent and child reports, with children rating their HRQL higher than their parents. At 12-months, there was no difference between the scores reported by parents and their children (psychosocial z = -0.3, p = 0.76; physical t = -0.7, p = 0.51). Agreement between parent and child ratings of HRQL improved with time postinjury. The findings have implications for the design of pediatric trauma outcomes studies and the routine collection of pediatric HRQL data. Parent and child reports should be considered separate but important information, particularly in the early stages following injury. Where collection of both is not feasible, parent or child report should be chosen, and interchangeable use of parent and child reports limited.