The Oxford Cognitive Screen (OCS-AU): Sensitivity and Specificity of a Stroke-Specific Cognitive Screening Tool

Objective: The aim of this study was to evaluate the sensitivity and specificity of the Australian Oxford Cognitive Screen (OCS-AU) to detect post-stroke cognitive impairment within three months of stroke.Participants and setting: Stroke survivors (n=53) within 12 weeks of stroke were recruited from three states in Australia.Main measure: The OCS-AU.Other measures: The Montreal Cognitive Assessment (MoCA), and a comprehensive neuropsychological test battery.Design: A validation study was conducted to analyse the sensitivity and specificity of the OCS-AU in subacute stroke using a neuropsychological test battery as the reference standard. The MoCA was included for comparative purposes. Impairment was defined as failing any OCS-AU cognitive domain, scoring below 26 on the MoCA, or failing at least two domains on the neuropsychological test battery.Results: To detect impairment within individual cognitive domains, most OCS-AU scores had low sensitivity, ranging from 0.12 (Executive) to 0.92 (Spatial Attention). Specificity was higher, ranging from 0.80 (Spatial Attention) to 0.96 (Praxis). Regarding the detection of multi-domain cognitive impairments, MoCA scores showed high sensitivity (0.81) but low specificity (0.42), compared with OCS-AU lower sensitivity (0.70) but higher specificity (0.58).Conclusion: Our findings suggest that the OCS-AU has strong domain-level specificity, but may miss some individuals with mild to moderate memory and executive impairments, while the MoCA appears more sensitive to domain-general impairment, but may misclassify individuals without post-stroke cognitive impairment. Thus, the OCS-AU and MoCA could be utilised for different purposes, to leverage their strengths when addressing specific clinical needs.