As the BMJ is one of a few journals which requires authors to declare funding and competing interests, the researchers used 159 trials published in the journal between 1997 and 2001 as the basis for their study. Each study was examined for a link between any competing interests and the author's conclusions.
For the purpose of the trial the author's conclusion was defined as 'the interpretation of extent to which overall results favoured experimental intervention'.
Competing interests were defined as anything which could influence professional judgement. Funding from profit organisations was considered to be a financial competing interest and was analysed separately from the other competing interests, which included personal, academic, and political influences.
Authors' conclusions were not significantly different in trials without competing interests, trials with other competing interests or trials funded jointly by profit and non-profit organisations.
However in both pharmacological and non-pharmocological trials funded by profit organisations, the author's conclusions were positively associated with financial competing interests - a significant proportion of author conclusions in these trials favoured experimental intervention.
Due to the BMJ's policy of requiring authors to report competing interests, it is possible that some authors choose not to publish in the journal. If this is the case the researchers conclude that the study may actually underestimate the extent of association between competing interests and author's conclusions.