Table 3.

Fixed attributes used in DCE and rationale for inclusion

Attribute	Rationale for inclusion	Fixed level	Rationale for level
Blood glucose level	Variable levels may result in diagnostic uncertainty	6 mmol/L	Average blood glucose level based on SITS data of treated patients
CT scan text description	To avoid skill/subjectivity around interpretation of scans	CT scan was conducted and is consistent with ischaemic stroke; it shows no haemorrhage or new ischaemic changes	Decided not to include image due to potential variability in CT image interpretation skill and subjectivity; difficulty finding scans to match multitude of various patient characteristics. Text description deemed most appropriate to remove diagnostic uncertainty
	To ensure confirmation of diagnosis of acute ischaemic stroke
Anticoagulation status	While it was deemed an influential attribute, only minority of stroke patients take an anticoagulant and therefore it was not included as variable attribute	patient is not on anticoagulation therapy	To avoid any issues surrounding INR levels that could complicate the decision to offer thrombolysis
Bleeding risk / recent surgery	Only relevant for a minority of patients. Challenging to operationalise variable and comparable levels in vignettes	no recent history of major bleeding
Diabetes	Not ranked as important in vignettes	no history of diabetes
	Included as fixed attribute for clinical validity
Patient consent/ family assent		assume either patient consent of family assent is available for treatment
Other / Comorbidities		there are no other attributes which would deter treatment	Due to difficulty defining fully and generating comparable and feasible levels of comorbidities. Potential overlap with pre-stroke cognitive and pre-stroke dependency status
Fixed attributes included post-pilot testing (Stage 4)
Handedness of patient		“All patients are right-handed”	To clarify and ensure the deficits will be interpreted consistently across all level of stroke severity (NIHSS)
Licenced dose bolus preparation time		“can be prepared for administration within 5 min”	Pilot testing revealed that participants would attribute in variable times in their decision-making so stating this will help control this potential error