Children and Adolescents Psychological Distress Scale During COVID-19 Pandemic: Validation of a Psychometric Instrument (CONFEADO Study)

Abstract

Aim and Object Purpose of the Study: In March 2020, the WHO declared a pandemic (COVID-19) due to the SARS-CoV-2 virus. In France, school closures and lockdowns were implemented. In this unprecedented context for French adolescents and children, the CONFEADO study surveyed children aged 9 to 18 years to assess their mental health, psychological distress, and resilience during and after the lockdown in relation to their living and housing conditions. To assess psychological distress, a psychometric tool (Children and Adolescent Psychological Distress Scale-CAPDS-10) was specifically designed for the research. This article presents the psychometric validity of the CAPDS-10. Methods: This cross-sectional study collected data from June 9 to September 14, 2020, from children and adolescents (9 to 18 years of age) via an online questionnaire after sending it to a large network of partners. Psychological distress, resilience, and trait anxiety were assessed using the CAPDS-10, the Child and Youth Resilience Measure (CYRM), and the State-Trait Anxiety Inventory for Children (STAIC). The CAPDS-10 measured perceived psychological distress in the most recent 2 weeks (primary endpoint). The predictive power of the CAPDS-10 was determined by statistical analysis. We proceeded to a confirmatory factor analysis to validate the scale at a clinical level. We carried out a psychometric validation with a step to verify the uni-dimensionality of the scale (PCA analysis) and the calculation of convergent and divergent validity, correlation coefficient between items and subscales, Cronbach's alpha for reliability, determination of a cut-off score for the AUROC index. Results: Three thousand and forty eight children and adolescents completed the CAPDS-10. Analysis confirmed a three-factor model (anxiety, depression, and aggressive behavior) (RMSEA = 0.072 [0.067; 0.077], CFI = 0.954), with a correlation coefficient between items >0.4. PCA analysis concluded that the scale is unidimensional. Reliability was satisfactory with Cronbach's alpha coefficients >0.7 (0.86). In addition, prediction was good with an AUROC index equal to 0.73 and a threshold score for severe distress greater than or equal to 19. Conclusion: The CAPDS-10 measures psychological distress over the most recent 2-week period with good psychometric qualities. It could be used in crisis or prevention contexts in the general population or in clinical settings.