Background: Optimal timing of early mobilisation following stroke, to maximise outcome, may vary according to patient characteristics. Assessments in ‘real world’ cohorts are lacking. Aim: To compare the characteristics of patients receiving Very Early Mobilisation (VEM, mobilised same day as stroke) to those receiving Early Mobilisation (EM, mobilised on the day after stroke) in relation to 90-day survival. Methods: Patient-level data from Queensland hospitals participating in the Australian Stroke Clinical Registry (2012-13) were linked with Queensland hospital data. Deaths obtained from national death data. Eligible patients: aged ≥ 18 years and received VEM/EM. Comorbidities were identified from ICD-10 codes (past five-years). Two multivariable, multi-level survival analyses were run. Model 1: VEM group only, model 2: EM group only. Factors associated with death were identified and effect sizes between models compared. Results: Overall, 1,221 patients (91% ischaemic stroke, 42% female, median age 76 years) were included, 34% received VEM. Within 90-days 8.8% had died (VEM: 5.7%, EM: 10.3%, unadjusted p = 0.007). VEM group had less severe stroke (able to walk on admission 55% vs 39%). Factors with the greatest difference in effect sizes between groups were atrial fibrillation (AF) (VEM: adjusted Hazard Ratio (aHR), 3.54, 95%CI, 1.24–10.10, EM: aHR, 1.52, 95%CI, 0.91–2.54) and intracerebral haemorrhage (ICH), (VEM: aHR, 3.21, 95%CI, 1.02–10.09; EM: aHR, 1.47, 95%CI, 0.89–2.41). Conclusion: Those with AF and ICH in the VEM group had a two-fold greater risk of death compared to the EM group. Delayed commencement of mobilisation in these types of patients is recommended.