Objective Cognitive screening tools are now recommended by national governing bodies to detect cognitive impairments following stroke and to prompt referral for further comprehensive assessment and rehabilitation. The primary aim of this review was to critically examine and integrate data across clinical and research domains to better understand Australian cognitive screening practices following stroke. Method Data from national clinical guidelines and audits, psychometric research, and clinical practice investigations were sourced, critically examined, and integrated. Results National Australian audit data suggest over two thirds of stroke units are routinely using screening tools to detect cognitive impairment. However, psychometric research suggests traditional cognitive screening tools, such as the Mini‐Mental State Examination, lack sensitivity to detect stroke‐related cognitive impairment. Furthermore, although more recently developed screeners, such as the Montreal Cognitive Examination, possess improved content validity, further modification, and/or supplemented assessment is required to improve their clinical utility. Of additional concern, even when cognitive impairments are detected during cognitive screening, very few stroke patients are referred for further comprehensive assessment as recommended within clinical practice guidelines. Conclusions Current evidence indicates cognitive screening tools, in their current form, do not perform well in stroke populations due to a variety of factors including poor content validity and lack of sensitivity. It appears that most Australian stroke patients with cognitive impairment are not receiving the assessment and rehabilitation services they require. Recommendations to adapt current screening tools, develop new stroke‐specific screening measures, and consider cognitive assessment protocols other than screening are discussed.