Factor VIII (F.VIII) and von Willebrand factor (VWF):Ag data collected by eight laboratories on a total of 336 obligatory carriers of hemophilia A and 137 normal women were used to answer several questions concerning the construction of linear discriminants for carrier detection. It was found: that a “universal” linear discriminant can be constructed which is suitable for use in all laboratories and is nearly as effective as laboratory-specific discriminants; that inclusion of age and ABO blood type data improved the efficacy of these discriminants; that substitution of alternative assays for F.VIII and VWF:Ag did not generally improve the efficacy of the discriminants over that obtained using the bioassay for F.VIII:C and Laurell's immunoassay for VWF:Ag; that linear discriminants were far more effective than discriminants based on the F.VIII:C/VWF:Ag ratio. A step-wise procedure is given which any laboratory may follow in using the universal discriminant for carrier detection.

Eight laboratories in six countries cooperated to clarify several issues concerning the phenotypes of heterozygous carriers of hemophilia “A.” Plasma levels of factor VIII (F.VIII:C, formerly VIII:C) and von Willebrand factor (VWF:Ag, formerly VIIIR:Ag) of carriers and normal women were determined by various “in-house” methods; a single lyophilized plasma standard was used for all assays. Analysis of the collated data from 336 carriers (296 obligatory carriers and 40 sporadic carriers) and 137 normal women showed that there was no difference in the F.VIII:C levels of “paternal” carriers (women who had obtained the abnormal gene from their fathers) and “maternal” carriers. Neither was there a difference in the VWF:Ag levels of normal women and either type of carrier. Age was found to have a significant effect on both F.VIII:C and VWF:Ag, values being higher at very young and very old ages, the minima occurring in the 25- to 30-year range. ABO blood type had a striking effect. Women of types A, B, and AB (designated non- O in the study), both normals and carriers, had significantly higher levels of both factors than did women of type O. Analysis by laboratories showed that differences in mean levels of both factors between laboratories were highly significant. It was concluded that age, ABO blood type, and laboratory variation should be taken into account in carrier detection.