Data Monitoring Committee (DSMB) Charter Template

Data Monitoring Committee Charter: [Trial short title]

Charter for the independent data monitoring committee (DMC, also called a data and safety monitoring board, DSMB, or independent data monitoring committee, IDMC) of a clinical trial: its role, membership, independence, meetings, reports, stopping guidelines and how its recommendations are communicated. The structure follows the charter proposed by the DAMOCLES Study Group (Lancet 2005;365:711-722) and supports SPIRIT 2025 items 28a and 28b on data monitoring. Replace every [bracketed] value and every row marked Example. This is a documentation format, not clinical, legal, statistical or regulatory advice; check the current official texts and consult your sponsor, statistician and ethics committee.

Document control

Approvals

Revision history

Contents

1. Introduction and scope
2. Trial summary
3. Role and responsibilities
4. Membership and independence
5. Relationships and data flow
6. Meetings
7. Reports and confidentiality
8. Interim analyses and stopping guidelines
9. Decision making and recommendations
10. After the trial
11. Agreement by members
12. Glossary and references
13. Standard and licence note

1. Introduction and scope

This charter describes how the data monitoring committee for [trial title] ([registry ID]) works. It covers the committee's purpose, membership, meetings, the data it receives, the guidelines it uses and how it communicates with the trial steering committee (TSC) and the sponsor.

The trial has a DMC because [reason, for example: the primary outcome includes death and hospitalisation, the intervention is new to this population, and the protocol plans two interim analyses of efficacy].

2. Trial summary

3. Role and responsibilities

The DMC's primary responsibility is to protect the safety and interests of trial participants, while also safeguarding the integrity and credibility of the trial. In this trial the DMC will:

• review accumulating safety data at the intervals in section 6;
• review the planned interim analyses of efficacy in section 8;
• review trial conduct data that bear on the reliability of the results: recruitment, adherence, follow-up completeness, data quality and protocol deviations;
• consider relevant external evidence, for example results from other trials of the same intervention, that the TSC or sponsor passes on;
• recommend to the TSC whether the trial should continue unchanged, continue with changes, pause recruitment or stop.

The DMC will not: design or amend the protocol, manage recruitment, see individual participant data beyond what it needs, or communicate interim results to anyone outside the closed session.

4. Membership and independence

Quorum: [the chair or a nominated deputy, the statistician and at least one other clinician]. If a member leaves, the TSC appoints a replacement with the DMC chair's agreement. Payment: [members receive an honorarium of a fixed amount per meeting, not linked to any trial outcome].

5. Relationships and data flow

6. Meetings

Each meeting has an open session (DMC with the trial statistician, chief investigator and sponsor representatives, discussing blinded and pooled data, conduct and external evidence) and a closed session (DMC members and the independent reporting statistician only, discussing data by arm). Meetings may be held by video call. The DMC may ask for an extra closed session at any time.

7. Reports and confidentiality

Reports are sent [7] days before each meeting through [secure transfer method]. Closed reports and closed minutes are kept by the independent reporting statistician until the trial is unblinded, then filed in the trial master file. Members do not discuss interim data with anyone outside the closed session and keep reports confidential until the main results are public.

8. Interim analyses and stopping guidelines

The protocol plans two interim analyses of the primary outcome (section 6). The Example below uses a Haybittle-Peto approach.

With such a strict interim threshold the overall type I error stays close to the nominal 0.05 (about 0.051 for this design by simulation, Example); the trial statistician confirms the exact figure in the SAP. Alternatives include alpha spending functions (Lan-DeMets with O'Brien-Fleming-type boundaries) and formal futility boundaries; name the method and software in the SAP. Safety guidelines are usually descriptive, because a trial may need to stop for harm on less evidence than for benefit.

9. Decision making and recommendations

After each meeting the DMC recommends one of:

1. continue the trial unchanged;
2. continue with a change (for example to eligibility, monitoring or consent information), with reasons;
3. pause recruitment or treatment pending further data;
4. stop the trial, or one arm, for clear benefit, harm or futility.

The DMC aims for consensus; if it cannot reach consensus, the recommendation follows a majority vote of members and records the minority view. The chair sends a signed recommendation letter to the TSC chair within [5] working days, or within [24 hours] for an urgent safety recommendation. The letter states the recommendation and, where it is not "continue unchanged", the reasons in terms that do not unblind the reader beyond what is needed. If the TSC or sponsor does not follow a recommendation to stop or change the trial, the TSC chair explains why in writing and the DMC may [ask to meet the sponsor or, if participants are at risk, raise its concern with the ethics committee].

10. After the trial

The DMC continues until [the final analysis is complete / the last participant completes follow-up]. Members receive the main results before publication. Publications name the DMC members (with their agreement) and state the number of meetings and any recommendations that changed the trial. DMC members are not authors of the main results paper unless the TSC agrees it in advance.

Related templates: science/trial-protocol (data monitoring section, items 28a and 28b), academic/stats-analysis-plan (interim analysis methods), science/case-report-form (safety data collected), science/data-management-plan and science/clinical-trial-report.

11. Agreement by members

12. Glossary and references

13. Standard and licence note

Structure aligned with the DAMOCLES Study Group's proposed charter for data monitoring committees (Lancet 2005;365:711-722) and with SPIRIT 2025 items 28a and 28b on data monitoring, cited by name and item number only. The text is original and reproduces no wording from these sources. The US FDA draft guidance (February 2024) and the EMA guideline (effective 1 January 2006) are named as dated examples; check the current official versions and your national requirements before use, because the FDA guidance was still a draft when this template was checked in October 2026. Stopping boundaries and error rates shown are Example values to be replaced by your statistician's calculations. This is a documentation format, not clinical, legal, statistical or regulatory advice; consult your sponsor, statistician and research ethics committee. Template licence: MIT.