Doubts were cast yesterday on the quality of maternity services after the Office for National Statistics revealed a sharp rise in the number of stillbirths over the past two years.
The ONS said it would work with the National Institute for Clinical Excellence (Nice) to find the reasons why stillbirths in England and Wales rose from 3,159 in 2001 to 3,372 in 2002 and 3,585 last year.
The number of babies dying in the first week of life also rose last year for the first time on record, increasing from 1,620 to 1,734. Most of this was due to more babies dying within 24 hours of birth.
The figures ran against the trend of improvements in safety for mothers and children that has persisted for many decades.
The ONS said it could not yet explain the figures. Its confidential inquiry with Nice into maternal death and child health would seek to establish whether the deterioration was due to social or clinical factors.
After the stillbirth total increased in 2002, the ONS looked to see if it was linked to a trend towards more multiple births - associated with older women becoming pregnant, sometimes after fertility treatment. There was no evidence to support that theory.
It found single mothers were nine times more likely to have a stillbirth, but could not establish the reason. There were also more stillbirths in spring and summer, when there had been a reduction in previous years.
The inquiry will try to establish whether more women are becoming pregnant from high risk groups, or whether there are clinical factors such as shortage of qualified midwives.
The definition of a stillbirth changed in 1993 to cover babies born dead from the 24th week of pregnancy. Those born earlier are counted as miscarriages. Previously the threshold had been 28 weeks.
The ONS said: "This is the first time we have observed an increase in deaths in the first week of life. It could be a blip or the beginning of a trend." It will examine health records to find if the deaths are linked to prematurity or other factors.
Other figures showed the sudden infant death rate in England and Wales decreased by 13% between 2002 and 2003, from 0.32 per 1,000 live births to 0.28.
It was highest for babies of mothers aged under 20 at the time of the child's birth (0.52 per 1,000 live births). The rate fell with increasing age of mother.
Birmingham and the Black Country had the highest infant mortality rate among the strategic health authorities in England, at 8.6 deaths per 1,000 live births, while Dorset and Somerset had the lowest at 3.3 deaths per 1,000 live births.
