Open Access News

News from the open access movement


Tuesday, March 30, 2004

Open access can make the research record less biased

The Wellcome Trust has released a new report, Public Health Sciences: Challenges and Opportunities, March 2004. It focuses on solving public health problems, not models of scientific publication. But it points out a natural connection (p. 26): "At the heart of the public health science research process is the need to distinguish true and causative associations from those that arise from chance, bias and confounding. The failure to publish and report negative findings (which is driven by funding, scientific publishing and media considerations) accentuate positive research findings. The unrepresentative nature of many positive findings in public health science results in a flawed research record, the accumulation of non-replicated findings, the duplication of research and a misleading picture being conveyed to the public. Researchers, public policy makers and commercial and non-commercial funders must recognize the responsibility to release all research findings for the public record, whether negative or positive. Some progress is being made in the area of clinical trials, where registration of trials is becoming an obligation and the systematic review of previous work is now a necessary prelude to further trials. Similar measures must be taken in observational epidemiology (including genetic epidemiology) and in public health interventions. Open access publishing may facilitate the introduction of systems approaches to dealing with this issue." (Thanks to Gary Price.)