Are investigators' access to trial data and rights to publish restricted and are potential trial participants informed about this? A comparison of trial protocols and informed consent materials.

OBJECTIVES

To determine to which degree industry partners in randomised clinical trials own the data and can constrain publication rights of academic investigators.

METHODS

Cohort study of trial protocols, publication agreements and other documents obtained through Freedom of Information requests, for a sample of 42 trials with industry involvement approved by ethics committees in Denmark. The main outcome measures used were: proportion of trials where data was owned by the industry partner, where the investigators right to publish were constrained and if this was mentioned in informed consent documents, and where the industry partner could review data while the trial was ongoing and stop the trial early.

RESULTS

The industry partner owned all data in 20 trials (48%) and in 16 trials (38%) it was unclear. Publication constraints were described for 30 trials (71%) and this was not communicated to trial participants in informed consent documents in any of the trials. In eight trials (19%) the industry partner could review data during the trial, for 20 trials (48%) it was unclear. The industry partner could stop the trial early without any specific reason in 23 trials (55%).

CONCLUSIONS

Publication constraints are common, and data is often owned by industry partners. This is rarely communicated to trial participants. Such constraints might contribute to problems with selective outcome reporting. Patients should be fully informed about these aspects of trial conduct.