Survival
Interpreting figures about the chances of survival is a difficult process.
People often quote figures from different stages in the labour and delivery process. It is important to know exactly how they are derived, for example, at such low gestations many babies does not survive labour and a further proportion do poorly after birth and die in the delivery room.
Hence survival looks better if it is based on the number of babies admitted to a neonatal unit than if it is based on the number of babies who are alive, and in turn this looks better than if it is based on information on babies alive at the start of labour. Whereas for mature babies this makes little difference, for babies of extremely low gestational age it makes a huge amount of difference.
A second problem with interpreting figures is to know what the original population was. In specialist hospitals survival figures are often slightly better than figures for the whole population. At first sight this might be thought to be because the staff in the specialist hospital have more experience and thereby achieve better results but equally it might be because the women who are referred there have been selected as those with the very best chance of survival.
Survival in EPICure 1 has been thoroughly analysed and published as a full paper (click here for details). EPICure 2 data are still being analysed at present and in due course the numbers of surviving babies will change very little. As for EPICure 1, survival was only worked out based on babies admitted to neonatal units. The graph on the right shows the changes in survival for admitted babies between the two studies. When the analysis of the EPICure 2 study is complete we will post on the website survival for all liveborn babies and for those alive at the onset of labour.
You still need to be very careful in reading these figures as - firstly the number of babies surviving at 22 weeks was 2 in 1995, and 4 in the 2006 - only a tiny proportion of the babies born at that week - all we can reliably say is that at this gestation survival remains a rare event. Secondly the number of babies born and surviving at 23 weeks was also small and the change may have happened by chance. Another study from the Trent region of England has shown no increase in survival over the same period at either of these gestations (Field, D et al 2008).
However, at 24 and 25 weeks we can see very significant increases which are real improvements in survival at these gestations.
As with the 1995 study, in 2006 being a girl and the mother having steroid injections before birth were really important factors in helping babies to survive but this does not mean that boys whose mothers have not had steroids before birth do not survive, just that they do so less often.
It is also important to understand that the chances of survival improve steadily as the baby grows older on the neonatal unit and that most deaths occur in the first few days after birth. This can be seen in the graph survival after birth in the first study.