Box 86 Structure of a typical data monitoring and ethics committee meeting

♦ The committee meet in private to review the data they have, to discuss the progress of the trial, and to identify any questions they need to ask of the trial management group

♦ Members of the trial management group may be called upon to discuss matters relating to the conduct of the trial such as the rate of intake and the quality of the data, but not the confidential interim analyses

♦ The committee and the trial management group discuss the progress of other related trials being conducted by other groups

♦ The trial statistician is called in to present the analyses and respond to any questions the committee may have

♦ The committee discuss in private their collective view on the progress of the trial and their recommendations about whether the intake should be continued or terminated, or the protocol modified

♦ The committee convey their conclusions to the trial statistician and the principal investigator, with their reasons, and draft reports to the sponsor, the trial management group, and the trial steering committee analyses. This could include information on the progress of similar trials conducted by other groups. Thus, the meeting typically has the structure shown in Box 8.6.

Reporting and making recommendations to the trial steering committee After each meeting, the committee should make a confidential report to the trial steering committee, taking care that this does not reveal confidential interim results. Their report should make recommendations on the continuation of the trial. If they recommend stopping the intake prematurely or making a major modification to the protocol, it must be made clear whom they should make such recommendations to, and the process by which a decision is reached. In the model we propose here, it is for the trial steering committee to decide whether to accept the data monitoring and ethics committee's recommendation.

Recommendations for major changes to the conduct of the trial should not be made lightly. If the reasons for undertaking the trial are still valid, it is important that the findings are as reliable as possible. The committee must appreciate the inherent unreliability of interim analyses based on immature data and avoid recommending too hasty termination of a trial (see Chapter 9). Nevertheless, they should not hesitate to recommend major changes if these are necessary for preserving the safety of patients or if the results of the primary outcome measures are already sufficiently convincing that it would be inappropriate and unethical to continue the intake.

Considering requests for release of interim findings Some trials groups present interim findings, even on primary outcome measures, at national or international conferences, sometimes even before the intake has been completed. This is an inappropriate and undesirable practice: it may jeopardize further recruitment or lead to unjustified changes in practice; and it throws into question the information available at the time the trial was planned and the statistical calculations on which the size of the trial was based. Interim findings based on immature data are inherently unreliable and conclusions based on them may well turn out to be incorrect. Moreover, groups planning or undertaking similar trials may find themselves subjected to considerable pressure to abandon their trials on the basis that the primary question has already been answered.

The data monitoring and ethics committee should therefore view with extreme caution any requests for the release of interim data. Nevertheless, it may be necessary to make collaborating clinicians aware of differences in toxicity, for example, if to do so could protect patients from avoidable risk. The committee should always make the reasons for their recommendations clear.

Responding to requests for additional support If the trial management group feels that it may need to apply for additional support to complete the trial satisfactorily, the data monitoring and ethics committee maybe asked for their recommendations on the matter. This may be difficult because they know the interim results. In response to such a request, they should therefore confine their comments to reporting on the progress of the trial, the quality of the data, and the likelihood that, with additional support, the trial would be completed successfully and would provide results that are still likely to be scientifically important and clinically relevant in the revised timescale. In their report, they should provide evidence to support their recommendations but not reveal the results of interim analyses. They should include and comment on relevant information that maybe available from similar trials conducted by other groups.

0 0

Post a comment