An analysis of head and neck cancer trials identified two factors that strongly correlated with publication bias.Clinical trials that did not yield statistically significant results were 11 times less ...