Background: Few studies have evaluated changes on parent-child agreement in HRQOL over time. The objectives of the study were to assess parent-child agreement on child's HRQOL in a 3-year longitudinal study, and to identify factors associated with possible disagreement.
Methods: A sample of Spanish children/adolescents aged 8-18 years and their parents both completed the KIDSCREEN-27 questionnaire. Data on age, gender, family socioeconomic status (SES), and mental health (Strengths and Difficulties Questionnaire, SDQ) was also collected at baseline (2003), and again after 3 years (2006). Changes in family composition were collected at follow-up. Agreement was assessed through intraclass correlation coefficient (ICC), and Bland and Altman plots. Generalizing Estimating Equation (GEE) models were built to analyze factors associated with parent-child disagreement.
Results: A total of 418 parent-child pairs were analyzed. At baseline the level of agreement on HRQOL was low to moderate and it was related to the level of HRQOL reported. Physical well-being at baseline showed the highest level of parent-child agreement (ICC=0.59; 0.53-0.65) while less "observable" dimensions presented lower levels of agreement, (i.e. Psychological well-being: ICC= 0.46; 0.38-0.53). Agreement parent-child was lower at follow-up. Some interactions were found between rater and child's age; with increasing age, child scored lower than parents on Parents relationships and Autonomy (Beta [B] -0.47; -0.71 / -0.23) and the KIDSCREEN-10 (-0.49; -0.73 /-0.25).
Conclusions: Parent-child agreement on child's HRQOL is moderate to low and tends to diminish with children age. Measuring HRQOL of children/adolescents mainly in healthy population samples might require direct self-assessments.