In an editorial, the authors of a previously published study examining industry bias in drug-drug comparisons of statins say that the authors of the network meta-analysis incorrectly argue that their finding differs from the previously published study. They note that like the current meta-analysis, their study also found no differences in the results of industry and non-industry sponsored studies comparing statins with statins or other drugs. However, they also analysed a pre-specified subset of studies, all of which were industry sponsored, to determine whether the results favoured the drug of the company sponsoring the study. This was not done in the meta-analysis. They found that the results of head-to-head comparisons favoured the product made by the sponsor of the study so for head-to-head comparisons, it is not industry sponsorship per se, but the company sponsoring the study that is associated with the bias. They postulated that this favouritism was accomplished by testing the sponsor’s drug at a higher dose than the competitor’s drug; and they add that it’s not just about the dose: the administration route of the drug or the coding and analysis of outcomes can also result in comparisons that favour one product over another. They note that although the network meta-analysis shows that the effects of statins are dose related, it does not investigate whether the studies funded by a particular company tested non-equivalent doses of the company’s product compared with a competitor’s product, or examine other differences in study design among the industry funded studies. They acknowledge that network meta-analyses are a valuable tool for making indirect comparisons of drugs, but they can be difficult to interpret because of the complexity of their methods. The wide confidence intervals often shown for the relatively small number of indirect comparisons that can be generated make it difficult to detect meaningful differences. They call for further empirical research on the association of industry funding with research outcomes in order to understand possible mechanisms for the observed industry bias in the context of different types of test drug or study design. In the meantime, they advise that the potential for industry bias should not be ignored.