Background and aims A report suggested the UK neonatal mortality rate (NMR) in 2010 was the second highest of 22 Western European Countries having been the ninth lowest in 1990.1 There are a number of known differences in data categorisation that could explain some of these differences.2 Excluding babies below a certain gestational age or birthweight is frequently done in other countries but not in the UK. CMACE information on all UK neonatal deaths was therefore analysed to determine the effect of different cut-offs.

Methods Data on all neonatal deaths across the UK were collected by CMACE with Office of National Statistics data utilised to obtain missing and denominator data. Deaths of babies born before different gestational ages were sequentially removed from the cohort and the NMR recalculated. In a separate analysis birthweight cut-offs were applied in the same way.

Results In 2008, there were 734 342 livebirths, 2394 neonatal deaths and a NMR of 3.3 deaths per 1000 livebirths. Applying gestational and birthweight cut offs substantially reduced the calculated NMR bringing it into line with other countries. 22, 24 and 25+5 week cut offs removed 256, 694 and 985 deaths respectively lowering the calculated NMR to 2.9, 2.3 and 1.9. Similarly 500 g and 1000 g birthweight cut-offs removed 412, and 1220 deaths respectively lowering the calculated NMR to 2.7 and 1.6.

Conclusions Cut-offs applied to neonatal deaths significantly alter calculated NMRs. A standardised approach to gestational age and birthweight is required to enable accurate comparisons between different countries.