Mark A Ferro1, Christiane Otto2, Ulrike Ravens-Sieberer2. 1. School of Public Health Sciences, University of Waterloo, 200 University Avenue West, Waterloo, Canada. mark.ferro@uwaterloo.ca. 2. Department of Child and Adolescent Psychiatry, Psychotherapy, and Psychosomatics, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Abstract
PURPOSE: This study examined whether the KIDSCREEN-27 was reliable and valid in young children 2-7 years with chronic physical illnesses which included estimating inter-domain correlations and internal consistency; measurement invariance testing; and, discriminant and convergent validity assessments. METHODS: Data come from the Multimorbidity in Children and Youth across the Life-course; a longitudinal study of individuals aged 2-16 years with physical illness. The parent-reported KIDSCREEN-27 was administered. Children (2-7 years; n = 106) were compared to adolescents (8-16 years; n = 157). Reliability was estimated using Cronbach α for internal consistency. Multiple group confirmatory factor analysis tested for measurement invariance. Cohen's d and Pearson coefficient were used to assess discriminant validity by sex and age. Convergent validity was tested using Pearson coefficients with the WHODAS 2.0 (child functioning/impairment). Multiple regression examined associations between multimorbidity (co-occurring physical and mental illness) and HRQL. RESULTS: Internal consistency reliabilities were α = 0.74-0.88 (children) and α = 0.77-0.88 (adolescents). Inter-domain correlations were relatively low (children: r = 0.18-0.59; adolescents: r = 0.30-0.62) indicating that each KIDSCREEN-27 domain was measuring a unique aspect of health-related quality of life. Measurement invariance was demonstrated (scalar level). Parameter estimates of the invariant models were similar for children and adolescents. Small, non-significant correlations were found for sex and age for children and adolescents. Medium, significant correlations were found for both groups between the KIDSCREEN-27 and WHODAS 2.0. Children and adolescents with multimorbidity had significantly lower physical well-being, psychological well-being, and school environment scores compare to those without multimorbidity. Regression coefficients were similar between groups. CONCLUSION: Findings provide evidence of adequate psychometrics for the KIDSCREEN-27 in young children with chronic physical illness.
PURPOSE: This study examined whether the KIDSCREEN-27 was reliable and valid in young children 2-7 years with chronic physical illnesses which included estimating inter-domain correlations and internal consistency; measurement invariance testing; and, discriminant and convergent validity assessments. METHODS: Data come from the Multimorbidity in Children and Youth across the Life-course; a longitudinal study of individuals aged 2-16 years with physical illness. The parent-reported KIDSCREEN-27 was administered. Children (2-7 years; n = 106) were compared to adolescents (8-16 years; n = 157). Reliability was estimated using Cronbach α for internal consistency. Multiple group confirmatory factor analysis tested for measurement invariance. Cohen's d and Pearson coefficient were used to assess discriminant validity by sex and age. Convergent validity was tested using Pearson coefficients with the WHODAS 2.0 (child functioning/impairment). Multiple regression examined associations between multimorbidity (co-occurring physical and mental illness) and HRQL. RESULTS: Internal consistency reliabilities were α = 0.74-0.88 (children) and α = 0.77-0.88 (adolescents). Inter-domain correlations were relatively low (children: r = 0.18-0.59; adolescents: r = 0.30-0.62) indicating that each KIDSCREEN-27 domain was measuring a unique aspect of health-related quality of life. Measurement invariance was demonstrated (scalar level). Parameter estimates of the invariant models were similar for children and adolescents. Small, non-significant correlations were found for sex and age for children and adolescents. Medium, significant correlations were found for both groups between the KIDSCREEN-27 and WHODAS 2.0. Children and adolescents with multimorbidity had significantly lower physical well-being, psychological well-being, and school environment scores compare to those without multimorbidity. Regression coefficients were similar between groups. CONCLUSION: Findings provide evidence of adequate psychometrics for the KIDSCREEN-27 in young children with chronic physical illness.
Authors: Louis S Matza; Donald L Patrick; Anne W Riley; John J Alexander; Luis Rajmil; Andreas M Pleil; Monika Bullinger Journal: Value Health Date: 2013-06 Impact factor: 5.725
Authors: Ryan J Van Lieshout; Mark A Ferro; Louis A Schmidt; Michael H Boyle; Saroj Saigal; Katherine M Morrison; Karen J Mathewson Journal: J Child Psychol Psychiatry Date: 2018-04-18 Impact factor: 8.982
Authors: Neuza Silva; Marco Pereira; Christiane Otto; Ulrike Ravens-Sieberer; Maria Cristina Canavarro; Monika Bullinger Journal: Qual Life Res Date: 2019-05-04 Impact factor: 4.147
Authors: Kelly M Kenzik; Sanjeev Y Tuli; Dennis A Revicki; Elizabeth A Shenkman; I-Chan Huang Journal: Med Decis Making Date: 2014-04-16 Impact factor: 2.583