Primary prophylaxis is the emerging standard treatment for boys with severe haemophilia. Tailored (escalating-dose) prophylaxis (EscDose), beginning at a low frequency and escalating with repeated bleeding may prevent arthropathy at a lower cost than standard prophylaxis (SP). From a societal perspective, we compared the incremental cost per joint-haemorrhage that is avoided and quality-adjusted-life-year (QALY) gained of SP and EscDose to on-demand (Demand) therapy in severe haemophilia A boys treated to age 6 using a decision analytic model. Costs included factor VIII (FVIII), professional visits and tests, central venous placement/complications, hospitalization, home programmes and parents' lost work-days. Resource utilization was estimated by surveying 17 Canadian clinics. The natural history of bleeding and other probabilities were determined from a longitudinal chart review (n = 24) and published literature. EscDose costs an additional $3192 per joint-haemorrhage that was avoided compared with Demand whereas SP costs an additional $9046 per joint-haemorrhage that was avoided compared with EscDose. Clinic costs and lost wages were reduced by 60-80% for EscDose and SP compared with Demand. EscDose attained more QALYs than SP and Demand on account of less bleeding than Demand and lower need for ports than SP. The incremental cost per QALY for EscDose vs. Demand was $542 938. EscDose was less expensive with similar QALYs compared to SP. Sensitivity analysis was performed on all probability- and cost-estimates, and showed the model was sensitive to the cost of FVIII and the SP and target joint utilities. In conclusion, prophylaxis will substantially improve clinical outcomes and quality of life compared to Demand treatment, but with substantial cost.