LAUSR

‘For any given research area, one cannot tell how many studies have been conducted but never reported.’ This formed the basis of what Rosenthal coined the ‘file drawer problem’ in 1979;1 that is, journals are filled with the few studies conducted that show statistically significant results, while the vast majority of studies, which indicate the null hypotheses, remain unaccounted for and unpublished, lying in the locked file drawers of researchers. Now often referred to as publication bias, the implications are far reaching. The absence of negative results from the published literature can lead to the unnecessary replication of work by others, wasted funding and resources and, perhaps most importantly, a weaker evidence base on which to inform decision-making.2 In veterinary medicine, the need for the complete data, including negative results, is obvious, having all of the evidence available can inform veterinary practice and lead to better health and welfare outcomes for animals. Yet, despite concerns over the implications of publication bias being reported nearly 40 years ago, many data remain hidden. Albeit now stored in different forms – the computer cloud instead of the file drawer – studies continue to show that papers which report statistically significantly findings are more likely to be published and …