Cost-effectiveness of selecting compatible donor blood in Rhc- end RhE negative females during and before reproductive age to prevent hemolytic disease in the newborn (HDN).

The aim is to decrease the incidence of allo-antibody-related HDN in the Dutch pregnant population. The presence of anti-RhE and anti-Rhc based antibodies ranks third (260 cases per year, cpy, 2003 OPZI-data) ) and fourth (125 cpy), after anti-K and anti-RhD, among cases of allo-antibodies in pregnant women. 66% of severe HDN induced by non-RhD alloantibodies are caused by anti-c or anti-E. We will investigate the cost-effectiveness of the introduction of a national bloodtransfusion policy consisting of matching donor blood for Rhc (and RhE) in Rhc- (or RhE-)negative female recipients (20% (and 71%),respectively) under 45 yrs. This national policy would apply to about 72000 transfusions/year. National consensus has been arrived at (CBO, oct. 2002) that all women under age 45 should receive Kell(K)-negative blood to prevent K- alloimmunization and subsequent HDN. The extension of this policy to Rhc- and RhE negative females was considered, yet postponed awaiting evidence from the on-going study (the socalled OPZI-study) on the Dutch screening program for Irregular Erytrocyte Antibodies (IEA) in pregnancy. OPZI is carried out by the applicants of this proposal but does not address the transfusion prevention potential; the coverage of OPZI exceeds 95% of IEA-cases. The study proposed here efficiently combines OPZI-data with newly collected data on an array of transfusion costs. OPZI will provide (1) the incidence Rhc- and RhE-immunizations in pregnancy, (2) direct costs (e.g. lab tests, echoscopy, invasive diagnostic tests, etc.), (3) the false negative rate of the screening test, (4) the burden of a positive screening test as well as (5) the actual risk of developing signs of hemolytic anaemia, requiring hospital stay and/or therapeutic intervention (a.o. risky fetal exchange transfusions). Newly collected data will include a.o. the costs of transfusing matched donor blood, including costs of [repeated] typing the recipient and the donor blood, and the indirect logistic costs for blood banks and user-hospitals. Additional analysis of OPZI data will derive the specific contribution of unmatched transfusion to incidence of IEA. Various transfusion policy models will be compared by modelling techniques with the average 2002/2003 pregnant population as the denominator of analysis. Outcome will be expressed in numbers of IEA-screen-positives (diagnostic burden gain) and Rhc/RhE-HDN (patient outcome gain) avoided. If the HDN difference is ignorable (as expected), then a cost-minimization analysis will be applied comparing (1) selective blood transfusion to (RhD-positive) women under 45 yrs (Rhc, RHE, K) with (1a) and without (1b) IEA screening, (2) selective blood transfusion to (RhD-positive) women in pregnancy or after delivery with (2a) or without (2b) IEA screening (3) blood transfusion as usual (K-matched cf. CBO) with IEA-screening. Regarding data collection the time schedule is dependent on the provision and creation of transfusion-related datasets of current business as usual. All data apply to the national population (no sampling). For rare events (like severe HDN) sensitivity analysis will be applied. No power calculations are thought to be applicable here.