Avoided deaths as an indicator of improvement of childhood cancer treatment.

The aim of this study was to estimate how many deaths have been avoided because of the substantial improvement in childhood cancer survival both in Piedmont (2769 incident cases) and in Italy during 1970-1999. For each time period of diagnosis, the number of avoided deaths within 5 years after diagnosis was estimated as the difference between the observed number of deaths and the number of deaths that would have been observed if the patients had experienced the same mortality as in the reference period of 1970-1974. The national estimate of the number of avoided deaths was calculated by applying Piedmont cumulative mortality to the expected number of incident cases in Italy. An increase in the number of avoided deaths from 103 (95% confidence interval: 65-140) in 1975-1979 to 239 (95% confidence interval: 209-268) in 1995-1999 was observed in Piedmont. In Italy, the number of avoided deaths was 3880 in 1995-1999, with respect to 1970-1974. Results of these analyses provide an effective way to quantify the improvement in the quality of care for children with childhood cancer.