Vineland-3 Growth Scale Values: Psychometric Properties for Clinical Trial Readiness in SCN2A.

Abstract

Purpose: The Vineland Adaptive Behavior Scales-3rd Edition (Vineland-3) is one of the most used measures of adaptive behavior among those with sodium channel protein type 2 subunit alpha related disorders (SCN2A-RDs). Several disease-modifying treatments are in early trials for SCN2A-RDs, and as such, clinical outcome assessments (COAs) are necessary. The Vineland-3 introduced growth scale values (GSVs), which are useful for measuring within-person change and thus may be useful in future clinical trials. The purpose of this study was to evaluate the psychometric properties of the Vineland-3 GSVs in SCN2A-RDs in preparation for future clinical trials. Methods: A sample of 65 individuals with SCN2A-RDs (mean = 108, SD = 76.0 months) was recruited for a clinical trial readiness study. The Vineland-3 Comprehensive Interview was administered by trained raters at regular intervals. Multiple psychometric properties were evaluated, including floor and ceiling effects, split-half internal consistency, test-retest reliability, and inter-rater reliability (on approximately 20% of all completions). Results: Floor effects were relatively infrequent on the GSV metric but occurred on all subdomains using the norm-referenced v-scale metric. Split-half and test-retest reliability were excellent for all subdomains (r_xx >0.95 and inter-class correlation coefficient [ICC] >0.90, respectively), except for coping, which still maintained adequate reliability (r_xx = 0.87, ICC = 0.65). Inter-rater reliability was also very strong, though it was more variable (α_kripp range 0.78-1.00). Conclusion: The Vineland-3 holds great potential as a COA in SCN2A-RDs; it exhibited very strong psychometric properties in this sample. This is a prerequisite level of evidence needed to demonstrate that a measure is fit-for-purpose for future clinical trials. While some reliability was high, some domains (e.g., domestic) still exhibited problems related to floor effects, which may suggest that they are less relevant to this population. Future studies should expand on this with mixed-methods research for prioritizing concepts of interest on the Vineland-3.