The quality and stability of essential drugs in rural Zimbabwe: controlled longitudinal study.

distress. A study carried out as part of the NHS workforce survey found no difference in the levels of distress between responders and non-responders, suggesting that non-response was not systematically related to psychological health. Our results may not be generalisable to non-teaching hospitals and might be confounded by the fact that consultants and house officers were selected from different trusts. In fact, one hospital was a source of both consultants and house officers for this study, and the occupational group differences at this hospital remained significant. Some of these data are three years old. The continued upheaval in the health service and further reductions in junior house doctors’ working hours mean that the differences in psychological distress between senior hospital doctors and their most junior colleagues may now be more pronounced. One approach to addressing this problem could be to tackle specific sources of occupational stress in different groups. For example, our results suggest consultants might benefit from a reduction in their workload, while house officers might benefit if they had greater autonomy. We would like to thank the consultants and house officers. Contributors: NK coordinated and collected data for the house officer study, helped analyse the data, and wrote the initial draft of the paper. CB coordinated the consultant study and commented on drafts of the paper. CS helped to extract and analyse the data, and commented on drafts of the paper. NK and CB will act as guarantors for the paper. Funding: The NHS workforce survey was funded under the mental health programme of the NHS executive (Northern and Yorkshire). Conflict of interest: None.