ABSTRACT
The article presents a new method for estimating the unconstrained factor VIII (FVIII) demand based on the principles of decision analysis. Epidemiology and treatment modalities were integrated into a model for unconstrained FVIII demand. Assumptions for each variable with impact on the unconstrained FVIII demand were defined and probability estimates for these variables were obtained from the literature and medical experts. The sensitivity of the unconstrained FVIII demand to each of the variables was determined, and the variables with the greatest impact were modelled probabilistically. The probability-weighted average for the unconstrained FVIII demand model was 6.9 units per capita with a 90% uncertainty interval of 2.7-13.6 units per capita. When compared with FVIII usage in countries, only Luxembourg's use of FVIII (7.7 units per capita) exceeded the probability-weighted average for the modelled unconstrained FVIII demand. As better information becomes available, revision of model variables is easily accomplished allowing for a more accurate and dynamic forecast of demand over time. More accurate modelling of the 'true' demand longitudinally should help prevent shortages of FVIII concentrates such as those that have occurred in the past. In addition, a more accurate forecast of FVIII demand will allow national health care policy makers to better allocate financial and other resources. Sufficient and consistent supply of FVIII concentrates and appropriate financing of haemophilia care will allow the clinical benefits of more aggressive treatment regimens such as prophylaxis to be realized.