The standard practice in maternity units of examining a newborn baby twice is a waste of resources and should be stopped. One examination is sufficient to detect any abnormalities, researchers have recommended.

Most babies born in hospital are examined within 24 hours of birth and again a few days later. The first is to detect abnormalities that may require early action, and the second is to pick up any that become apparent later, such as heart defects.

But researchers from the Aberdeen Maternity Hospital have discovered that very few abnormalities were picked up from the second examination that had not been detected first time round, and those that were did not need any special intervention.

Of the 4,835 babies who were screened just the once, 8.3 congenital defects per 100 babies were detected, compared with 9.9 defects discovered in the 4,877 babies who were screened twice. Yet the difference between the two groups in those needing active management was just 0.2 per cent, suggesting the defects were very minor.

In an accompanying article, medical statistician Jonathan Deeks from the Institute of Heath Sciences in Oxford accuses the researchers of bad science. He argues that the mothers were not truly randomly selected into either group for examination, but this was due to the haphazard nature of bed allocation in hospital (BMJ, 1999; 318: 627-32).