BACKGROUND: Many public health researchers conducting studies in resource-constrained settings have experienced negative 'rumours' about their work; in some cases they have been reported to create serious challenges and derail studies. However, what may appear superficially as 'gossip' or 'rumours' can also be regarded and understood as metaphors which represent local concerns. For researchers unaccustomed to having concerns expressed from participants in this manner, possible reactions can be to be unduly perturbed or conversely dismissive.This paper represents a retrospective examination of a malnutrition study conducted by an international team of researchers in Zambia, Southern Africa. The fears of mothers whose children were involved in the study and some of the concerns which were expressed as rumours are also presented. This paper argues that there is an underlying logic to these anxieties and to dismiss them simply as 'rumours' or 'gossip' would be to overlook the historic and socio-economic factors which have contributed to their production. METHODS: Qualitative interviews were conducted with the mothers whose children were involved in the study and with the research nurses. Twenty five face-to-face interviews and 2 focus group discussions (FGDs) were conducted with mothers. In addition, face-to-face interviews were conducted with research nurses participating in the trial. RESULTS: A prominent anxiety expressed as rumours by the mothers whose children were involved in the study was that recruitment into the trial was an indicator that the child was HIV-infected. Other anxieties included that the trial was a disguise for witchcraft or Satanism and that the children's body parts would be removed and sold. In addition, the liquid, milk-based food given to the children to improve their nutrition was suspected of being insufficiently nutritious, thus worsening their condition.The form which these anxieties took, such as rumours related to the stealing of body parts and other anxieties about a stigmatised condition, provide an insight into the historical, socio-economic and cultural influences in such settings. CONCLUSIONS: Employing strategies to understand local concerns should accompany research aims to achieve optimal success. The concerns raised by the participants we interviewed are not unique to this study. They are produced in countries where the historic, socio-economic and cultural settings communicate anxieties in this format. By examining this study we have shown that by contextualizing these 'rumours', the concerns they express can be constructively addressed and in turn result in the successful conduct of research aims.