[Life expectancy as an indicator of outcome].

Usually survival studies evaluate the percentage of subjects alive or disease-free at a given point in time (cumulative survival percentage). These studies require a lengthy period both for follow up and for the collection of an exhaustive number of cases. In addition, for cancer types with a sharp gradient of cumulative mortality, the estimate may be unstable. On the database of children with Acute Lymphoblastic Leukaemia (ALL) collected by the Childhood Cancer Registry of Piedmont (CCRP) in 1979-1991, we have estimated an alternative time-dimensional index to estimate both patients' life expectancy of patient and number of years gained to death. These estimates have been compared in subsets of the database corresponding to different periods of diagnosis and efficacy of treatment. Life expectancy has been evaluated as the area under the survival curve, on the assumption that after a number of years since diagnosis, survivors have the same mortality rates of the general population. Clinicians are invited to take into consideration the approach described in the present note.