Abstract
<title>Abstract</title> <p> Purpose To explore the relationships between youth and caregiver scores on the Pediatric Quality of Life Inventory (PedsQL) in an intensive outpatient youth mental health services setting. Methods This is a secondary analysis of routinely collected PedsQL data. Informant discrepancies between youth self-report and caregiver proxy-report were described. Differences across domains scores, and gender impacts on discrepancies were examined using Intra-Class Correlations and repeated measures ANOVA. The associations between age and discrepancies were examined using Spearman’s rho. A hypothesis testing approach was used to assess the validity of interpreting discrepancies between scores. Results 205 youth-caregiver dyads met eligibility criteria. The mean age was 14.8 +/- 1.7 and participants identified as boys (25.9%), girls (65.4%), and gender diverse (8.6%). ICC point estimates indicated poor agreement for all domains and global scores (0.378–0.487). Average absolute discrepancies varied from 12.0-17.3 across domains. There was a significant main effect of domain (p = 0.018) on discrepancies, but post-hoc tests were not significant. Age was significantly negatively associated with emotional and social domain discrepancies ( <italic>ρ</italic> =-0.187,-0.123, respectively), and gender had no significant effect on discrepancies (p = 0.867). For construct validity, our hypotheses were supported by the data. Conclusion Youth and caregiver PedsQL scores show poor agreement. Informant discrepancies are highly variable across dyads, but do not show a bias towards youth or caregivers consistently under- or over-reporting quality of life. Age appears to influence discrepancies, but the impact of gender is unclear. The validity of interpreting informant discrepancies is supported; however, more research is needed to guide clinical interpretation. </p>