Validity and Reliability of the CVI Range for Clinical Research (CVI Range-CR): Baseline and One-Year Results

Purpose Cerebral/cortical visual impairment (CVI) is a leading cause of visual impairment in children, and significantly impacts both visual function and functional vision (the ability to use vision during real-life activities). We developed the CVI Range for Clinical Research (CVI-CR), a functional vision assessment that can be incorporated in CVI clinical trials. Herein we evaluate the psychometric properties, inter- and intra-rater reliability, and validity of the CVI Range-CR. Design Prospective, longitudinal cohort study Participants 40 children with CVI Methods Participants underwent CVI Range-CR assessments at baseline and one-year follow-up. All assessments were recorded. A trained in-person examiner scored in-person assessments; two expert remote graders scored recorded assessments. Visual acuity was evaluated by a pediatric neuro-ophthalmologist using the six-level visual behavior scale (VBS). Main Outcome Measures CVI Range-CR scores (ranging from 0 to 10) using the Across-CVI Characteristics (Rating 1) and Within-CVI Characteristics (Rating 2) methods of scoring were compared among graders using the intraclass correlation coefficient (ICC) to assess inter-rater reliability. A subset of videos was scored twice by the same remote examiner to assess intra-rater reliability using the ICC. CVI Range-CR scores were compared to VBS scores (Spearman’s correlation coefficient, ⍴) to evaluate validity. Cronbach’s ɑ was used to assess internal consistency of the 10 items on the Within-CVI Characteristics scale. Results The CVI Range-CR demonstrated high internal consistency (Cronbach’s ɑ=0.96), excellent intra-rater reliability (ICC=0.88-0.98), good-to-excellent inter-rater reliability (ICC=0.74-0.88), and strong correlation to clinical assessment of visual behavior as measured by VBS scores (Spearman’s ⍴=−0.76 to −0.86, p<0.0001). It was also sensitive to change; there were small but significant improvements in both Ratings 1 and 2 over one year (Rating 1 average: +0.35, p=0.002; Rating 2 average: +0.31, p=0.001). Conclusions The CVI Range-CR is a reliable and valid measure of functional vision in children with CVI. Because intra-rater reliability exceeds inter-rater reliability, longitudinal studies are recommended to use consistent graders over time. This could be accomplished by remote grading of recorded assessments via a centralized reading center.