SOME hospitals are failing to investigate the deaths of stillborn babies or those who die in the first month of life, according to a new report.

Experts behind a major study into Britain's high stillbirth rate said some hospitals carry out "cursory" investigations into whether lessons can be learned about the care of mothers and babies, while others fail to record data.

While the overall stillbirth and neonatal death rate has fallen slightly, the team found that many parts of the UK still have death rates that are too high. The experts said these hospitals should carry out thorough reviews of each case.

Even when factors such as the age of the mother and social deprivation were taken into account, networks covering large parts of central, north-west and north-east England had a stillbirth rate and death rate in the first 28 days of life that was more than 10% higher than the UK average.

These included networks covering Lancashire and south Cumbria, Yorkshire and the Humber, Southern West Midlands, Staffordshire, Shropshire and the Black Country and Trent.

Overall, death rates varied across the UK, from 4.9 deaths to 7.1 per 1,000 births. This variation cannot solely be explained by factors such as poverty, mother's age, multiple birth and ethnicity, the experts said.

Some NHS trusts also had poor levels of completed data for key areas such as antenatal care, the experts found, while there were also wide variations in how some deaths were recorded, such as for heart defects.

David Field, professor of neonatal medicine at the University of Leicester, and one of the authors of the report - called MBRRACE-UK - said hospitals needed to carry out proper investigations.

He said: "If you look at different trusts, some do it in fantastic detail and some do it in a cursory fashion. In some trusts, it's done by two people, in others it's being done by a team by who really pull it apart."

He said reviews included looking at factors that may have influenced a baby's death, such as whether a woman should have been classed as high risk instead of low risk.

He said trusts needed to be "hyper-critical" to see whether things could have been done better, adding: "You can't just say all the deaths are inevitable".

While in many cases, babies might not have been able to be saved, he said "you want to know about those deaths where things might have been done differently".

Dr Brad Manktelow, senior lecturer at the University of Leicester, said some trusts were "very, very good" at investigating deaths.

"Some, we suspect, are less good at investigating individual deaths and the care that was received," he added.

According to the report, there were 4,633 stillbirths and deaths in the first 28 days of life (for babies born at 24 weeks' gestation and over) in 2014 - almost six deaths for every 1,000 births - down slightly on the 4,722 in 2013.

The stillbirth rate and death rate in the first month of life is now 5.9 deaths per 1,000 births in the UK, which is still "significantly higher than rates reported in other high-income European countries", the report said.

Almost 50% of stillbirths and just over 5% of deaths in the first month are classed as having an unknown cause.

More than 90% of families were offered a post-mortem to help establish what happened but even this varied according to the hospital.

In 45% of trusts and health boards this offer was made for all deaths, but 22% of organisations offered a post-mortem for less than 90% of deaths.

Overall, parents consented to a post-mortem around 40% of the time.