Cerebral visual impairment remains one of the most underdiagnosed and poorly measured conditions affecting children's development, yet until now clinical trials have lacked a standardized, psychometrically validated tool to capture how vision functions in real-world settings. That gap may be closing. A prospective longitudinal cohort study evaluated the CVI Range for Clinical Research (CVI Range-CR), a structured functional vision assessment designed specifically for integration into CVI clinical trials. Forty children with confirmed CVI underwent assessments at baseline and again at one year, with scoring performed by both in-person examiners and remote expert graders reviewing recorded sessions.
The tool employs two distinct scoring methods — Across-CVI Characteristics and Within-CVI Characteristics — each producing scores on a 0-to-10 scale. Interrater reliability between trained graders was assessed using intraclass correlation coefficients, while intrarater reliability was tested by having the same remote examiner score a video subset twice. Validity was examined by correlating CVI Range-CR scores against a six-level Visual Behavior Scale administered by a pediatric neuro-ophthalmologist, using Spearman's correlation. Internal consistency was evaluated via McDonald's omega.
The significance of this work extends well beyond measurement science. CVI is now widely recognized as the leading cause of childhood visual impairment in high-income countries, yet it remains systematically overlooked because it does not originate in the eye itself but in cortical and subcortical visual processing pathways. Most existing visual acuity tests fail to capture the complex, context-dependent vision problems these children experience. A reliable, remotely scorable instrument could dramatically accelerate intervention research by enabling multicenter trials with consistent outcome measurement. The one-year longitudinal component also hints at the tool's sensitivity to change over time — a critical property for any clinical trial endpoint. Key limitations include the modest cohort of 40 participants and the absence of a healthy control comparison. Whether the CVI Range-CR can detect treatment-related change, beyond natural developmental progression, remains to be demonstrated in larger, controlled trials.