Safety of the Canadian blood supply in 1980-85: using a paediatric cohort for risk assessment of human immunodeficiency virus (HIV) infection

The risk of HIV from transfusions in Canada in the period 1980-85 was estim ated, using the information from a transfused paediatric cohort. Children w ho were transfused between January 1980 and November 1985 at a tertiary car e paediatric hospital were contacted by letter. With this notification, HIV testing for recipients was recommended. HIV testing histories were obtaine d. The number tested for HIV was estimated from the questionnaire responses and from data matching with the HIV-testing laboratory. Cases of HIV infec tion were identified through multiple sources. In this cohort, 11 028 child ren were transfused a mean of 21 units. Of the 10 220 living recipients, th e estimated proportion tested for HIV was 86% to 91%. Thirty-one cases of H IV infection were identified, representing 0.28% of the cohort but 0.34% of those expected to have been tested. The estimated HIV incidence per 1000 u nits transfused ranged from 0.028 [95% CI 0.0007, 0.155] in 1980 to 0.445 [ 95% CI 0.2592, 0.712] in 1985. This suggests that the risk of HIV from tran sfusions in Canada continued to rise until the implementation of HIV testin g of donors in November 1985.