Table 4. Survey responses regarding stroke risk assessment in critically ill patients with new-onset atrial fibrillation.

Please tick all answers that reflect your views on stroke risk assessment in critically ill patients with new-onset atrial fibrillation:	Bar	Response	%
I do not use stroke risk scores routinely in critically ill patients with new onset AF to assess the need for anti-coagulation		246	67.96%
I regularly calculate a risk score (e.g., CHAD2, CHA2DS2-VASc) to assess the need for anti-coagulation		39	10.77%
Stroke risk scores inaccurately reflect the risk of embolic events in critically ill patients with new-onset atrial fibrillation due to prothrombotic changes associated with critical illness		112	30.94%
Stroke risk scores favour anti-coagulation despite a higher risk of bleeding in critical illness		73	20.17%
Modified risk scores should be developed for critically ill patients with new-onset atrial fibrillation		170	46.96%
Total		640	100.00%