Number of children dying or last seen (censored) during each year of follow-up.

*All children in the cohort were aged 12 years or older at the time of ascertainment of deaths in 2007, thus losses from follow-up at younger ages were due to death or censoring alive on the date last seen (as recorded in hospital case notes). Between 12 and 15 years there are fewer children under follow-up (‘at risk’) in the older age groups as many children had not reached these ages and this reduced the precision of survival estimates after 12 years.
Number of children dying or last seen (censored) during each year of follow-up.