Anxiety disorder agreement among children with chronic physical illness and their parents

Abstract

Background: Assessment of child psychopathology using multiple informants provides a more comprehensive and accurate evaluation of child mental health; however, parent-child agreement is low-to-moderate in child psychiatry and tends to be lower for internalizing disorders. Children with chronic physical illness (CPI) are at an elevated risk of developing anxiety disorders, making accurate assessment especially important in this population. Despite this, longitudinal patterns and determinants of parent-child agreement in children with CPI remain underexplored. Objectives: The objectives of this thesis were to: (1) Estimate the magnitude of informant agreement for anxiety disorders on the Mini-International Neuropsychiatric Interview for Children and Adolescents (MINI-KID) between parents and children with CPI at baseline, 6, 12, 24, and 48 months, (2) Explore whether child sex moderates parent-child agreement, and (3) Identify sociodemographic and health factors associated with parent-child disagreement for anxiety disorders on the MINI-KID over time. Methods: Data for 119 dyads came from the Multimorbidity in Youth Across the Life-course (MY LIFE) study, a longitudinal study of children aged 2 to 16 years who had been diagnosed with a CPI and their primary caregiver. The prevalence-adjusted bias-adjusted kappa (PABAK) estimated the magnitude of agreement between parents and children with CPI at baseline, 6, 12, 24, and 48 months. Sex-stratified agreement analyses were conducted using the PABAK to investigate whether parent-child agreement was moderated by child sex. The method of variance estimates recovery (MOVER) was used to construct a confidence interval for the difference in κ estimates between male and female children at each timepoint. A generalized estimating equations model examined factors associated with parent-child disagreement over time. Results: Agreement ranged from fair to substantial over time (κ = 0.40-0.65). For male children, agreement was moderate to almost perfect (κ = 0.47-0.82), whereas for female children, fair to moderate agreement was observed (κ = 0.32-0.51). Moderation by child sex was only found at 6 and 48 months. Compared to baseline, time at 6 months (OR = 0.46, 95% CI = 0.23-0.91, p = 0.026) and 12 months (OR = 0.55, 95% CI = 0.31-0.97, p = 0.040) were associated with lower odds of disagreement. Female children were found to have significantly higher odds of disagreement compared to male children (OR = 2.04, 95% CI = 1.20-3.46, p = 0.008). Parents who were not partnered had lower odds of disagreement relative to partnered parents (OR = 0.27, 95% CI = 0.10-0.71, p = 0.008). Higher levels of parent psychopathology were also associated with increased odds of disagreement (OR = 1.15, 95% CI = 1.01-1.31, p = 0.032). Conclusion: Parent-child agreement ranged from low-to-substantial and varied over time. Moderation by child sex was only evident at 6 and 48 months. Predictors of parent-child disagreement may help identify dyads who may be at greater risk for informant discrepancies. Future research should examine the underlying mechanisms driving parent-child disagreement to inform targeted interventions that help strengthen agreement among parents and children with CPI.