I do so agree with you mollyredone. Sadly, in the UK, it is usual to have to perform a serological cross-match, using group O red cells if there is any hint of a mixed-field (even if the patient has, for example, been typed as group A for a decade, and a unit of group O blood was transfused the previous day). This pathetic attitude to the professional ethos of the laboratory personnel is insulting in the extreme, and, because the "mixed-field reactions" will be "reinforced" with each transfusion of group O red cells, if the patient requires frequent transfusions, the "discrepancy" will never be resolved, and precious group O units (in particular, group O, D Negative units) will be wasted unnecessarily.
Worse still is when the patient has an alloantibody, such as an anti-Fya, when group O, Fy(a-) blood has to be selected for the group A patient (in my example), when group A, Fy(a-) blood may be readily available, leaving fewer group O, Fy(a-) units readily available for a genuine group O patient, with anti-Fya, who may require it in an emergency.
Worse of all is when the patient has a complex mixture of common antibodies, or an antibody directed against a high=prevalence antigen, when there are only a certain number of group O units available in the country (and less donors, as numerous cryopreserved units may come from the same donor) and these units are wasted on patients who are patently not group O.
RANT OVER - FOR NOW!!!!!!!