Psychometric Validation of the Caregiver-Reported CDKL5 Clinical Severity Assessment and Developmental Questionnaire
Abstract number :
1.367
Submission category :
12. Genetics / 12A. Human Studies
Year :
2023
Submission ID :
430
Source :
www.aesnet.org
Presentation date :
12/2/2023 12:00:00 AM
Published date :
Authors :
Presenting Author: Scott Demarest, MD, MSCS – University of Colorado
Timothy Benke, MD, PhD – University of Colorado; Jenny Downs, PhD – University of Western Australia; Peter Jacoby, PhD – University of Western Australia; Helen Leonard, MD – University of Western Australia; Eric Marsh, MD, PhD – University of Pennsylvania; Heather Olson, MD – Boston Children's Hospital; Elia Pestana-Knight, MD – Cleveland Clinic; Dana Price, MD – NYU Langone Health; Raj rajaraman, MD – UCLA Mattel Children's Hospital; Jacinta Saldaris, PhD – University of Western Australia; Bernhard Suter, MD – Baylor College of Medicine; Gina VanderVeen, MS – University of Colorado; Judith Weisenberg, MD – Washington University of St. Louis
Rationale:
CDKL5 Deficiency Disorder (CDD) is a severe X-linked developmental and epileptic encephalopathy (DEE) with seizure onset in the first weeks of life. A minority of children with CDD attain independent walking and verbal speech. A novel caregiver-reported CDD clinical severity assessment (CCSA-Caregiver) and Developmental Questionnaire (CDQ-Caregiver) were developed and validated.
Methods:
The CCSA-Caregiver is an Observer Reported Outcome (ObsRO) with items that describe epilepsy, alertness, behaviors and feeding. It was developed using cognitive interviews and pilot testing (n=27 items). The CDQ is a ObsRO that describes gross motor, fine motor, expressive communication, receptive communication, and cognitive/social skills, developed using cognitive interviews and pilot testing. The CDQ-Caregiver comprises 66 items (Fine Motor 16, Gross Motor 18, Communication 24, Social/Cognitive 8) with branching logic to ensure that items are appropriate for the age and skills of the individual. Both measures were administered online to parents/caregivers of patients with CDD enrolled in the International CDKL5 Clinical Research Network (ICCRN) and International CDKL5 Disorder Database (ICDD) using REDCap (COMIRB 19-2756, RA/4/20/6198). Factor analysis, internal consistency, average variance extracted, divergent validity, test-retest reliability and known groups validity were evaluated and compared to predefined criteria. Items with poor performance were dropped or grouped into an alternative domain to improve model fit.
Results:
Factor analysis of 198 CCSA-Caregiver datasets (144 ICCRN, 54 ICDD) indicated that items loaded to four domains – Epilepsy, Alertness, Behavior, and Feeding (factor loadings >0.4). Cronbach’s alpha and Composite Reliability values for each of the identified domains were above the minimum threshold. As hypothesized for “known groups” validity, total scores were significantly lower, indicating milder severity, for females and individuals who could use words or walk. Confirmatory factor analysis of 212 CDQ-Caregiver datasets (146 ICCRN, 66 ICDD) with the domain composite scores (Gross Motor, Fine Motor, Expressive Communication, Communication, Cognitive/Social) identified a single Development domain with acceptably high ( >0.4) factor loadings for each composite score. The Cronbach’s alpha value was 0.92, composite reliability was 0.93 and the fit statistics were good. As hypothesized for “known groups” validity, scores were significantly lower, indicating milder severity, for individuals who were female or had no seizures. Measures were readministered after an interval of two to four weeks (ICCRN only – CCSA-Caregiver n=34, CDQ-Caregiver n=36) and there was good to excellent test-retest reliability for domain and total scores.
Conclusions:
This study supports excellent psychometric properties for these two novel observer reported measures (CCSA and CDQ) in a dataset of ~200 patients with CDD. These measures are likely to be valuable in a clinical trial setting for CDD as well as other DEEs.
Funding:
NIH/NINDS U01NS114312 (Benke/PD)
Genetics