Abstract
BACKGROUND: Vision impairment affects the accuracy of cognitive test outcomes, emphasizing the need for developing cognitive screening tools designed for visually impaired individuals, especially considering global aging trends. This study aimed to develop a modified, validated version of the Vision-Impairment version of Addenbrooke's Cognitive Examination III (VI-ACE-III) for Arabic-speaking elderly individuals with vision impairment in Egypt. In addition, the study aimed to assess the accuracy of VI-ACE-III in diagnosing dementia and mild cognitive impairment (MCI). METHODS: The VI-ACE-III was developed using large printed formats and verbal substitution of the vision-dependent items. One hundred and eighty participants aged ≥ 60, with varying degrees of vision impairment (including moderate, severe, and blindness), were divided into three equal groups: 60 individuals with MCI, 60 with mild to moderate dementia, and 60 with cognitively intact controls. Receiver operating characteristics (ROC) curves were plotted to assess the accuracy of the test screening. RESULTS: ROC analysis for dementia established an optimal cut-off point of 84 out of 115, demonstrating 100.0% sensitivity, 98.3% specificity, and an area under the curve (AUC) of 0.983, based on the comparison between the dementia and MCI groups. The analysis for MCI determined an optimal cut-off point of 94 out of 115, with 95.0% sensitivity, 96.7% specificity, and an AUC of 0.983 compared to controls. The VI-ACE-III demonstrated significant Cronbach's alpha values (α = 0.866, α = 0.771), indicating strong internal consistency within the dementia and MCI groups. CONCLUSIONS: The VI-ACE-III showed good sensitivity and specificity for assessing dementia and MCI in Arabic-speaking elderly individuals with visual impairment (VI) in Egypt. Regular screening and interventions are crucial for managing and preventing the deterioration of cognitive dysfunction and vision impairment in the elderly population.