The publication process itself was the major cause of publication bias in genetic epidemiology

Objectives: How do scientists in the field of genetic epidemiology see the problem of publication bias, what are the possible solutions and what particular pressures are they under that might either promote or prevent publication and other biases? Study Design and Setting: An exploratory study of beliefs and practices among scientists working in the field of genetic epidemiology. A purposive sample of senior, genetic scientists was selected (N = 6) and informants were interviewed face to face. Results: There was some consensus that the issue of nonreplication might be particularly problematic in genetic epidemiology, and that publication bias could contribute to this. The informants suggested that the problem lay mainly with the publication process. Publication of negative results was seen as important but fraught with difficulties. Conclusion: Possible solutions included education of editors, and reviewers, dissemination of negative findings through Web sites or accessible data archives and pooling of data. However, none of these were perceived to be straightforward given the current pressures on the research industry. (C) 2006 Elsevier Inc. All rights reserved.