Data Safety

What is in a Data and Safety Monitoring Plan?

A study team will be prompted to answer questions about a DSMP to the IRB when the study exposes subjects to greater than minimal risk. A DSMP, depending on the risks, size, and complexity of the research, may account for:

• Types of data and events, for example:
  • Data quality
  • Subject recruitment, accrual, and retention
  • Adverse events
  • Procedures for the protection of subject privacy
  • Information from outside sources that may impact subject safety (related studies, scientific reports, therapeutic development)
• Roles and responsibilities for gathering, evaluating, and monitoring data
• Frequency of review of data and events (no less than annually)
• Reporting of adverse events and unanticipated problems

Because the risks inherent to research vary greatly from study to study, data and safety monitoring is reviewed within the context of the study. A DSMP should take into account the risks within the study and provide an adequate plan to review study data in order to minimize those risks. Additionally, monitoring may at times need to account for stopping rules should an overwhelmingly positive response be noted.

Resources are available to investigators to construct a data and safety monitoring plan by contacting the CTSI Bioethics and Subject Advocacy Program (BSAP).

Who monitors data and safety?

The party performing data and safety monitoring should vary depending on the potential risks, complexity, and nature of the study. In many situations, various groups or individuals may share responsibility for different aspects of study monitoring and oversight. If a study is multi-site, an entity should be identified to communicate between sites and IRBs.

Internal to Study Team

• Principal investigator/study team: Generally appropriate for studies that involve a small number of subjects at a single site and the interventions are minimal risk or when there is a low risk-benefit assessment. Close, continuous monitoring is expected.
• Sponsor: The sponsor often completes the monitoring when it is a multi-site trial with a low risk-benefit assessment.
• Safety/Medical monitor: Often affiliated with a contract research organization (CRO)
  • Safety/medical monitors may act alone in monitoring or may act as a supplemental monitor to other individuals or entities
• Internal monitoring is often suitable for Phase 1 and early Phase 2 trials. Early studies are frequently not randomized or controlled so accumulating results are known to the investigators and sponsor. Internal monitors often review the data during frequent conferences due to study design. Issues regarding statistical interpretation or confidentiality of interim data are therefore generally less relevant in this setting.

External to Study Team

• Institutional/departmental monitor: Independent affiliation from the research/researchers but possesses relevant expertise
  • May be suitable when research is occurring at a single site and contains a low risk-benefit assessment
  • Appropriate to include additional expertise or objectivity from the investigators involved in the research
• DSMB/DSMC: A group of individuals with relevant expertise that reviews accumulating study data on a regular basis
  • Generally established for large, randomized, multi-site studies that evaluate treatments intended to prolong life or reduce risk of a major adverse health outcome
  • Commonly recommended for late Phase 2 and Phase 3 clinical trial
  • Required by FDA for emergency research where informed consent is waived (21 CFR 50.24(a)(7)(iv))
  • Required by NIH for multi-site clinical trials involving interventions that entail potential risk to the subjects (NIH Policy)

When is a DSMB necessary?

The following types of studies are required to have a DSMB appointed:

• NIH-sponsored Phase 3 clinical trials (as well as some Phase 1 and 2)
• FDA-sponsored planned emergency research
• Other studies when required by the IRB

Studies that fall under the following categories are most likely to have a DSMB:

• Large, multi-site randomized studies evaluating treatments intended to prolong life or reduce risk of a major adverse health outcome
• Controlled trials comparing rates of mortality or major morbidity

The FDA recommends the use of a DSMB when a sponsored study includes:

• A study endpoint which might ethically require termination of the study at interim analysis, where a highly favorable or unfavorable finding is made
• A particular safety concern, such as administration of treatment by an invasive method (done only for research purposes)
• A subject group that includes a fragile population such as children, pregnant women, the elderly, or the terminally ill or those with diminished capacity
• A subject group at elevated risk of death or other serious consequences, even with a study objective that addresses a lesser endpoint

When is a DSMB unnecessary?

The FDA cautions that DSMBs add administrative complexity and require additional resources, so they are not recommended for all clinical trials. (FDA Guidance)

DSMBs are generally not needed for:

• Trials at early stages of product development; or
• Trials addressing lesser outcomes, such as relief of symptoms, unless the trial population is at elevated risk of more severe outcomes.
• Multi-site studies or late-phase studies using an approved drug for a new indication may not require an independent monitor such as a DSMB. The IRB may request justification for not using an independent monitor, such as when there is a low risk-benefit assessment.

DSMBs are not needed for:

• Social/Behavioral/Educational Studies
• Phase 1 studies. Generally, a DSMB is not required in Phase 1 studies because they are designed to include frequent internal monitoring. These trials typically advance faster than late-phase trials, so the monitoring plan must be commensurate with that frequency of monitoring.

Data and safety monitoring information is collected in the IRB application for NIH-funded clinical trials and greater than minimal risk research.

Greater than Minimal Risk Research and NIH-funded Clinical Trials

For studies that involve greater than minimal risk to subjects or are NIH-funded clinical trials, the initial IRB submission must provide a plan addressing several key elements by either attaching a charter or providing these details in the IRB application form:

• Who is responsible for the monitoring of the data, their expertise, and if the individual or committee is independent from the sponsor and/or the PI of the study.
• What data will be reviewed: at minimum, this should include adverse event data. This may also include, data quality, subject recruitment, accrual, retention, outcome data, assessment of scientific reports or therapeutic development, results of related studies that may impact subject safety, and procedures designed to protect the privacy of subjects.
  • There may be other elements to be monitored, and those should also be included as a part of the initial submission.
• How often the individuals or committee responsible for the DSMP will review the data, including what documentation will be maintained to demonstrate that the safety review was conducted. Keep in mind that this information will be reviewed at the time of study renewal. The Board will want to see that the individuals/committee actually met at the intervals and associated documentation was maintained as was indicated in the DSMP in the initial study submission.

Renewal/Ongoing Review

Study teams provide a summary of the findings from data and safety monitoring at time of renewal (if renewal is required) or time of study closure. If the DSMP includes a DSMB, the study team should provide meeting minutes or a report from the DSMB. Generally, the DSMB report should include:

• Confirmation that the monitoring entity reviewed adverse events, subject withdrawal, complaints, and any interim findings
• The date of the review
• The monitoring entity's assessment of the information that was reviewed. If the monitoring entity has not identified any issues and recommends continuation of the study, this recommendation should be included in the report as well

The criteria for approval, and accreditation standards, require that the research plan includes adequate provisions for monitoring the data collected to ensure the safety of subjects.

For studies that are greater than minimal risk or are NIH-funded clinical trials, the IRB likely will expect that all of the elements listed above (under What is in a Data and Safety Monitoring Plan?) are being monitored as part of DSMP. In addition, the IRB will evaluate who is responsible for the data and safety monitoring and whether the frequency/timing of the monitoring is appropriate. Finally, the IRB will review any planned statistical analyses, pre-planned stopping rules, actions to be taken upon specific events, and/or endpoints.

Renewal

The IRB has the responsibility to ensure that results from data and safety monitoring justify the continuation of IRB approval of the research study. If the plan outlined in the IRB application indicates that a DSMB will meet regularly, the IRB requires the renewal submission to include all reports for DSMB meetings that occurred since the last renewal. At the time of renewal, the IRB has the authority to request changes to the monitoring plan.

• AAHRPP Standards
  • Element I.8.C
  • Element II.3.B
• IU HRPP Documents
  • Policies
    • IRB Review Process
• Regulatory References
  
  • 21 CFR 56.111(a)(6)
  • 45 CFR 46.111(a)(6)
  • DOD Instruction 3216.02 – Protection of Human Subjects and Adherence to Ethical Standards in DOD-Supported Research
  • FDA Guidance for Clinical Trial Sponsors: Establishment and Operation of Clinical Trial Data Monitoring Committees
  • NIH Policy for Data and Safety Monitoring
    • Further Guidance on a Data and Safety Monitoring for Phase 1 and Phase 2 Trials
    • Guidance on Reporting Adverse Events to Institutional Review Boards for NIH-Supported Multicenter Clinical Trials
  • OHRP Guidance on IRB Continuing Review of Research
  • VHA Directive 1200.05 – Requirements for the Protection of Human Subjects in Research