In this issue a paper by Yadegarfar and McNamee1 reports data on mortality from ischaemic heart disease (IHD) among male industrial workers employed in a company producing and reprocessing nuclear fuel (see page 10.1136/oem.2006.030627).1 The risk of dying from IHD was slightly higher among shift workers than day workers (adjusted OR 1.10).

Ten years ago a similar study was carried out at the same factory.2 That study showed reduced mortality in shift workers (adjusted OR 0.90). It is not clear if the present paper has included all subjects from the initial cohort. For the sake of future meta-analyses it would be valuable to have this information. The inclusion of subjects started on 1 January 1950 in both studies, but in the first study inclusions continued until 1992 while in the second until 1998.

Results from the first study were surprising, as there was no plausible explanation as to why shift work should protect against …