Article type
Year
Abstract
Background: Bias in industry-sponsored trials is common and interpretation of results may be distorted in favour of the sponsors' products. Therefore many industry-sponsored trials are coordinated by seemingly independent steering committees. However, this may not prevent the sponsors' influence on reporting of results.
Objectives: To investigate the sponsors' influence on reporting of results in industry-sponsored trials.
Methods: We included a sample of industry-sponsored trials published in The Lancet in 2008-2009 and corresponding trial protocols provided by the editors. For each protocol and published paper, we extracted information on coordination of publication, sponsors' publication rights and involvement of medical writing assistance. We compared the information from protocols with the published papers.
Results: We identified 169papers of randomised trials and included 69(41%) industry-sponsored trials. An additional 12(7%) fully industry-funded trials with trial conduct apparently independent of the sponsor were analysed separately. In 24 out of the 69 trials (35%) the sponsor or a hired contract research organization were involved in coordinating writing of the manuscript, in 10 (14%) the sponsor was not involved and in 35 (51%) it was not described. In 64 trials (93%), the sponsor had influence over publication of the results through co-authorship or an explicitly stated right to approve or comment on the paper, and 37 trials (54%) had evidence of medical writing assistance supported by the sponsor. For the 12 apparently independently conducted trials 4 described the sponsors' involvement in reporting of results, without this being stated in the published papers. Conclusions: The sponsors are usually involved in reporting of results in industry-sponsored trials. Even for fully industry-funded trials that appear to be conducted independently, the sponsors are also sometimes explicitly involved in the reporting of results.
Objectives: To investigate the sponsors' influence on reporting of results in industry-sponsored trials.
Methods: We included a sample of industry-sponsored trials published in The Lancet in 2008-2009 and corresponding trial protocols provided by the editors. For each protocol and published paper, we extracted information on coordination of publication, sponsors' publication rights and involvement of medical writing assistance. We compared the information from protocols with the published papers.
Results: We identified 169papers of randomised trials and included 69(41%) industry-sponsored trials. An additional 12(7%) fully industry-funded trials with trial conduct apparently independent of the sponsor were analysed separately. In 24 out of the 69 trials (35%) the sponsor or a hired contract research organization were involved in coordinating writing of the manuscript, in 10 (14%) the sponsor was not involved and in 35 (51%) it was not described. In 64 trials (93%), the sponsor had influence over publication of the results through co-authorship or an explicitly stated right to approve or comment on the paper, and 37 trials (54%) had evidence of medical writing assistance supported by the sponsor. For the 12 apparently independently conducted trials 4 described the sponsors' involvement in reporting of results, without this being stated in the published papers. Conclusions: The sponsors are usually involved in reporting of results in industry-sponsored trials. Even for fully industry-funded trials that appear to be conducted independently, the sponsors are also sometimes explicitly involved in the reporting of results.