The aim of the study was to investigate the feasibility of optimizing prophylactic dosing of factor VIII by the use of individual pharmacokinetic data. Twenty-one patients were enrolled in a randomized cross-over study on standard dosage regimens vs. dosing according to pharmacokinetic principles. The study period was 2×6 months. Using single-dose pharmacokinetic data for each patient, plasma factor VIII procoagulant activity ( C) curves following various doses and intervals were computer-simulated. From these calculations, a suitable dosage was chosen. C was also repeatedly measured during study periods. Trough levels of C, numbers of spontaneous joint bleedings and amounts of factor concentrate used during the two study periods were compared for each patient. There was a close correlation between predicted and measured values of C. As the half-lives of C in the patients varied from 7.8 to 18.3 h, it was obviously beneficial to base the dosage on individual pharmacokinetic data. Fourteen patients completed both study periods. Mean trough level of exogenous C was raised from 0.89 (SD 0.73) U dL(-1) during standard dosage to 2.2 (1.5) U dL(-1) during pharmacokinetic dosage. Concomitantly, mean 6-month consumption of factor VIII was decreased from 124 000 (SD 30 000) units to 84 000 (31 000) units. Numbers of reported bleedings were generally similar during both periods. The study demonstrates the usefulness of individual pharmacokinetics as a tool for cost-effective utilization of factor VIII in the prophylactic treatment of haemophilia A.