Babies are dying because of 'poor clinical management' and the failure of some GPs to detect serious illness, an influential study has found.

The deaths of about 60 babies a year could be averted through better care, the Confidential Enquiry into Stillbirths and Deaths in Infancy has concluded.

It calls today for a review of the training requirements of all doctors with responsibility for babies' emergency care.

CESDI director Mary Macintosh told HSJ: 'We are not doctor-bashing. It is our job to highlight issues where the professions can improve.'

CESDI's fifth annual report includes a major study of sudden unexpected infant deaths.

A total of 456 deaths were identified in data from five regions between 1993 and 1996. Eighty per cent were classed as sudden infant death syndrome, or 'cot death'.

The remaining 20 per cent - 93 cases - had a specific cause. CESDI looked in detail at 67 of these 'explained' cases and found 'suboptimal' care in half.

Extrapolating data nationally, it says about 60 deaths in England and Wales could be averted each year.

The highest proportion of 'explained' deaths occurred in the first month of life. These babies were more likely to come from disadvantaged families and have mothers who smoked during pregnancy.

While circumstances varied, the report adds: 'Perhaps the most striking finding was that half of these babies were ill enough to have needed medical attention in the 24 hours before they died.'

It adds: 'This confirms the Enquiry conclusion that half of these deaths might have been avoided if professionals or carers had behaved differently.'

A combination of faults rather than a single error may have led to death in most cases. But the most common single deficiency - cited in 14 cases - was GPs' failure to recognise the severity of illness.

The report notes that at present, GP training does not always include hospital-based paediatric experience. This should be rectified, it says, in a review of GP training, paediatrics and child health.

But the report also found failings in hospital-based care and 'poor clinical management'. Three babies died after being assessed in hospital by a junior doctor and discharged without the opinion of a more experienced doctor.

Paediatric and casualty teams should ensure that a baby is always assessed by a doctor of sufficient experience before considering discharge, the report says.

CESDI also found that carers were as likely as professionals to have contributed to deaths, highlighting the need for targeting support services to vulnerable groups of people.

Dr Macintosh said the findings should be put in context: 'We are always going to see the worst cases by definition.'

CESDI also reviewed 22 cases of planned home birth associated with the death of the baby. It identified 'substandard care' in 77 per cent of cases.

But Dr Macintosh stressed that this percentage tallied with previous studies of hospital deliveries.

CESDI found failings in recognising foetal distress and highlighted the need for back-up and transfer arrangements.

The report says: 'We are unable to assess absolute or relative risks between planned home birth and hospital delivery in this study.

'However, the difficulty in estimating these risks should not discourage attempts to acquire information on the safety of planned home births.'

5th Annual Report, CESDI, 188 Baker Street, London NW1 5SD, pounds 6.