Diagnostic sensitivity and specificity of the teach CVI screening tool for identifying cerebral visual impairment in...

Cerebral Visual Impairment (CVI) is a brain-based visual disorder caused by damage to the visual processing areas of the brain, resulting in distinctive visual behaviors. Due to its unique presentation, medical professionals report challenges identifying CVI. This retrospective study explores the accuracy of the Teach CVI screening tool for detecting children with CVI. The Teach CVI has three versions targeting different motor abilities and age ranges. Through a retrospective review, 257 children were identified as having a Teach CVI screening tool completed during outpatient ophthalmology or occupational therapy visits. Data were analyzed for sensitivity, specificity, and Area Under the receiver operating characteristic Curve (AUC) for all three Teach versions. Logistic regression was used to explore the association between Teach CVI results and the diagnosis of CVI. The sensitivity and specificity of the Teach CVI varied across versions. Teach 1 demonstrated acceptable sensitivity and poor specificity (80 %; 29 %), Teach 2 and 3 both offered balanced levels of sensitivity and specificity (Teach 2: 64 %; 68 %; Teach 3: 72 %, 67 %, respectively). Both Teach 2 and 3 exhibited a significant association between screening results and CVI diagnosis (p < 0.05). Across all versions, AUC analyses indicated poor to marginally acceptable discriminative ability for identifying CVI. The Teach CVI screening tools demonstrated unique performance metrics for identifying CVI. Our sensitivity data demonstrated that Teach CVI, particularly versions 1 and 3, could add value when used with clinical observations and medical history to support clinicians in deciding when to refer a child for further CVI diagnostic evaluation. WHAT THIS PAPER ADDS: This is the first study to explore the performance and accuracy of the Teach CVI for screening children for traits and characteristics of Cerebral Visual Impairment (CVI). Teach CVI demonstrates varied levels of sensitivity and specificity for accurately identifying CVI in children; therefore, it would add value if utilized within a medical setting in conjunction with clinical observation and assessments to assist with identifying the presence of CVI. This paper highlights areas in which the Teach CVI demonstrates strengths and weaknesses, and potential considerations for implementing the Teach CVI in clinical care to screen for CVI.