Statistics Collaborative - Design and analysis for biomedical research

CapabilitiesCapabilities

Data Monitoring Committees (DMCs)

For several decades, the National Institutes of Health (NIH) has used independent advisory boards, whose members include physicians, statisticians, and others, to monitor the safety of participants in clinical trials. These boards go by various names, such as Data Monitoring Committees (DMCs), Data and Safety Monitoring Boards (DSMBs), or Independent Data Monitoring Committees (IDMCs). Throughout this website we use the acronym DMC, except when we refer to a specific committee that may be known by a different name.

The Food and Drug Administration (FDA) has recently established guidelines for the use of such boards to review the safety and efficacy of many ongoing trials: http://www.fda.gov/cber/gdlns/clindatmon.htm

Statistics Collaborative, Inc. (SCI) and its staff have many years of experience both serving on DMCs and acting as the independent statistical group reporting to DMCs. SCI's internal operating procedures conform to the FDA's guidelines for DMC reporting.

SCI takes an active role when we report to a DMC. We provide the DMC with clearly written reports that assist the members in assessing the risks and benefits of the interventions being studied. We tailor each report to the study at hand. We consider the data being collected, the disease under study, and the emerging pattern of risks present in the study or suspected for the class of agent.

Our tables and text are readily interpretable by clinicians and statisticians. We aim for completeness and succinctness. We come to meetings prepared to answer questions from the DMC on the report and on the trial.

SCI takes pride in the many compliments we have received from DMC members and sponsors over the years on the content and clarity of our DMC reports.

Liaison between the sponsor and the DMC

Because of the highly confidential nature of DMC reports and decisions, SCI takes exquisite care in all its interactions with DMCs. We bind DMC reports and number them to maintain accountability of confidential information. We serve as an intermediary facilitating communication between the sponsor and the DMC members. Our goal is to maintain the integrity of the trial; the safety of participants; and the interests of the sponsor, the investigators, the Institutional Review Boards and Ethics Committees, the FDA, and, often, the European Medicines Agency (EMEA). We are practiced in securely transferring study materials, managing DMC requests while keeping the sponsor blinded, and managing communications among the sponsor, the data provider, and the DMC.

Programming for interim reports

The methods appropriate for programming clean, locked data are often not ideally suited to programming for DMC reports. At SCI we have had many years of experience programming with interim data that are neither fully clean nor complete. We use defensive programming techniques to anticipate possible variations in future transfers of interim data. We work with the data provider to choose a course of action that will effectively present a clear snapshot of the interim data. Our thorough interim data checks have prompted some clients to ask us to perform additional checks to assist with the quality-control processes they use for their databases.

We provide a full range of services for DMCs:

Management and operations

  • Developing a DMC charter
  • Recommending experts to sit on a DMC
  • Managing DMC meetings: scheduling, planning logistics of meetings, creating agendas, preparing minutes
  • Facilitating the sponsor's communication with the DMC
  • Facilitating the DMC's recommendation to the sponsor
  • Archiving DMC materials for the sponsor until the study is unblinded

Statistics

  • Developing an interim analysis plan, including statistical methods and table shells
  • Developing statistical stopping guidelines for efficacy, safety, and futility
  • Constructing adaptive designs

Programming

  • Programming analysis datasets with interim data
  • Creating presentations requested by the DMC

Reporting

  • Preparing and presenting blinded or unblinded safety and efficacy reports
  • Preparing blinded sponsor reports

Examples of publications for trials with which SCI was involved:

  • Wittes J, Barrett-Connor E, Braunwald E, Chesney M, Cohen HJ, Demets D, Dunn L, Dwyer J, Heaney RP, Vogel V, Walters L, Yusuf S. Monitoring the randomized trials of the Women's Health Initiative: the experience of the Data and Safety Monitoring Board. Clinical Trials 2007; 4:218-234.
  • Cairns JA, Wittes J, Wyse DG, Pogue J, Gent M, Hirsh J, Marler J, Pritchett EL. Monitoring the ACTIVE-W trial: some issues in monitoring a noninferiority trial. Am Heart J 2008; 155:33-41.
  • Wittes J, Palensky J, Asner D, Julian D, Boissel J, Furberg C, Kulbertus H, Pocock S, Roniker B. Experience collecting interim data on mortality: an example from the RALES study. Current Controlled Trials in Cardiovascular Medicine 2001; 2:59-62.