The authors defined ghost writing as "the failure to name, as an author, an individual who has made substantial contributions to an article".
44 trials were identified and it was found that in 75% of the papers there was evidence of ghost authorship. This rose to 91% when papers were included that acknowledged a contribution but still did not name the individual as an author.
The authors conclude that ghost authoring is common and may undermine accountability and transparency.
Action: This article raises an interesting dilemma: How much should we trust a paper if we don't really know who wrote it?