ELECTIVE PUBLICATION OF STUDies with statistically significant results has received widespread recognition. 1 In contrast, selective reporting of favorable outcomes within published studies has not undergone comparable empirical investigation. The existence of outcome reporting bias has been widely suspected for years, 2-12 but direct evidence is limited to case reports that have low generalizability [13][14][15] and may themselves be subject to publication bias.Our study had 3 goals: (1) to determine the prevalence of incomplete outcome reporting in published reports of randomized trials; (2) to assess the association between outcome reporting and statistical significance; and (3) to evaluate the consistency between primary outcomes specified in trial protocols and those defined in the published articles. METHODSIn February 2003, we identified protocols and protocol amendments for randomized trials by reviewing paper files from clinical studies approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg, Denmark, in 1994-1995. This period was

BackgroundGhost authorship, the failure to name, as an author, an individual who has made substantial contributions to an article, may result in lack of accountability. The prevalence and nature of ghost authorship in industry-initiated randomised trials is not known.Methods and FindingsWe conducted a cohort study comparing protocols and corresponding publications for industry-initiated trials approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg in 1994–1995. We defined ghost authorship as present if individuals who wrote the trial protocol, performed the statistical analyses, or wrote the manuscript, were not listed as authors of the publication, or as members of a study group or writing committee, or in an acknowledgment. We identified 44 industry-initiated trials. We did not find any trial protocol or publication that stated explicitly that the clinical study report or the manuscript was to be written or was written by the clinical investigators, and none of the protocols stated that clinical investigators were to be involved with data analysis. We found evidence of ghost authorship for 33 trials (75%; 95% confidence interval 60%–87%). The prevalence of ghost authorship was increased to 91% (40 of 44 articles; 95% confidence interval 78%–98%) when we included cases where a person qualifying for authorship was acknowledged rather than appearing as an author. In 31 trials, the ghost authors we identified were statisticians. It is likely that we have overlooked some ghost authors, as we had very limited information to identify the possible omission of other individuals who would have qualified as authors.ConclusionsGhost authorship in industry-initiated trials is very common. Its prevalence could be considerably reduced, and transparency improved, if existing guidelines were followed, and if protocols were publicly available.

Blinding is rarely tested. Test methods vary, and the reporting of tests, and test results, is incomplete. There is a considerable methodological uncertainty how best to assess blinding, and an urgent need for improved methodology and improved reporting.

The blinding status of key trial persons was incompletely reported in most randomized clinical trials. Unreported blinding may be frequent, but one of five 'double blind' trials did not blind either patients, treatment providers or data collectors. Authors, referees, and journal editors could improve the completeness of reporting of blinding, eg, by adhering to the CONSORT statement. It is inappropriate to presume blinding of key trial persons based only on the ambiguous term 'double blind'.