We are indebted to our distinguished colleagues in Australia for pointing out this ambiguity. We will be indicating in our definitive EPDS book, soon to be published by Gaskell (Reference Cox and HoldenCox & Holden, 2003), that the scale from the first validation study as published in 1987 contains the correct and original wording.

The differences between ‘being’ and ‘feeling’, ‘anxious or worried’ and ‘worried and anxious’ are not only semantic. Perhaps committed EPDS advocates, like your correspondents, will test their hypothesis that these word changes may affect the total EPDS score. We doubt it, but a local grant-giving body might support an ambitious master's student.

The EPDS is not, of course, a precise measuring-rod of feelings, but its total score has been shown to provide a remarkably accurate indication of the likelihood of clinical depression in many cultures and countries.

Our new book, Perinatal Mental Health: A Guide to the Edinburgh Postnatal Depression Scale (EPDS), is our definitive and final attempt to ensure that the EPDS is used as frequently as appropriate; and misused — never!