Data and Safety Monitoring Guidance

What is data and safety monitoring?

Data and safety monitoring involves reviewing accumulated outcome data from an ongoing research study to protect the safety of research participants and the validity and scientific merit of the trial.

What is the purpose of a Data and Safety Monitoring Plan (DSMP)?

A written Data and Safety Monitoring Plan (DSMP) formalizes the monitoring process and assigns responsibility for monitoring study-specific events on a pre-defined schedule. All research proposals that present greater than minimal risk to subjects should include a DSMP. Often, the plan will be included in the protocol. If not, the study submission should include a DSMP in the project summary or as an addendum using the DSMP template.

Interventional research refers to any prospective study involving human participants, which is designed to answer specific questions about the effect or impact of a particular biomedical or behavioral intervention or to answer specific questions about human physiology.

Minimal risk means the probability and magnitude of harm and/or discomfort anticipated in the research is not greater than ordinarily encountered in daily life; or during the performance of routine physical or psychological examinations or tests (45 CFR 46.102(i) and 21 CFR 56.102 (i)).

What are the key elements of a DSMP?

A Data Safety Monitoring Plan (DSMP) should describe:

  • The type of data or events to be captured under the monitoring plan.
  • The name(s) of the person(s) who will be responsible for monitoring the data collected (i.e., study accruals, protocol deviations, protocol violations, unanticipated problems, adverse events) and their respective roles (e.g. investigators, research sponsor).
  • The frequency of assessments/analysis of data or events captured by the monitoring plan (e.g., a periodic time interval or after a specific number of participants are enrolled).
  • The time frame for reporting unanticipated problems, adverse events, protocol deviations, and protocol violations.
  • Definition of specific or stopping rules dictating when some action is required. Stopping rules are predetermined guidelines used to determine that the study should be altered or stopped based on review of study-related events occurring during the conduct of the study, specific study endpoints, and the decisions prompted. For example, studies may be stopped when there is greater than expected morbidity or mortality rate or when the experimental arm in a comparison study is shown to be better or worse statistically than the standard care arm.
  • Procedures and time frames for communicating outcomes of monitoring reviews to the Institutional Review Board (IRB), study sponsor, and/or indicated entities (e.g., NIH, FDA).
  • Plans to monitor adherence to the IRB-approved protocol and assure the validity and integrity of data.

What influences the breadth and depth of a DSMP?

DSMPs vary depending on the:

  • Nature, size, and complexity of the research protocol.
  • Expected risks of the research.
  • Type of participant population being studied.

The nature and complexity of DSMPs fall along a continuum. If the study involves a small number of participants, is conducted at only one site, and/or the range of possible study events having a important impact on risks and benefits to participants is narrow, the principal investigator or a group of investigators could perform the monitoring functions. Alternatively, if a study is larger, more complex, and higher risk, a monitor/monitoring group or an independent Data and Safety Monitoring Board (DSMB) may be required.

What are appropriate monitoring methods/entities?

Depending upon the size, complexity or inherent risk of the protocol monitoring can be conducted by the investigators, experts in the field of study, consultants (such as biostatisticians), and other specialists, as needed.

Investigator Review

For many studies, the principal investigator will be the monitoring entity. This is appropriate when the study involves a small number of subjects, the study is conducted only at one site, and the range of possible study events that could have an important impact on the risks and benefits of to research participants is narrow. In such cases, continuous monitoring of events by the investigator and prompt reporting of unanticipated problems to the IRB, and FDA or others as appropriate may be adequate.

The issue of possible conflicts of interest must be taken into account, especially if the investigator assumes the role of the monitor. Use of an independent monitor can address the need for an unbiased review. Independent review can be achieved through a range of solutions.

The investigator can enlist a physician or other appropriate expert who is independent of the study and available in real time to review and recommend appropriate action regarding adverse events and other safety issues. The name and qualifications of the expert should be stated to the IRB at the time of review.

This is a small group (2–3 members) of independent investigators and biostatisticians who review data from a particular study. The independent investigator may be recruited from UT Southwestern or an outside institution. The biostatistician should be independent to avoid unblinding the study.

Data Safety Monitoring Board (DSMB)

A Data Safety Monitoring Board (DSMB) is a group of scientists, physicians, statisticians, and others that collect and analyze data during the course of a research study to monitor for adverse events and other trends that would warrant modification or termination of the trial or notification of subjects about new information that might affect their willingness to continue in the trial.

In general, a DSMB may be the most appropriate way to monitor data and safety for studies that involve:

  • Large numbers of subjects where risk may be better assessed through statistical comparisons of treatment groups;
  • Blinded study treatment groups where the validity and integrity of the study may be adversely affected by having an individual or group associated with the design and conduct of the study break the blind;
  • Multiple clinical sites where there is a need for investigators to submit reports of adverse events to a central reporting entity, such as a coordinating center or statistical center, responsible for preparing timely summary reports of adverse events for distribution among the clinical sites, and to the IRBs;
  • High risk interventions where death or severe disability is a major risk of research participation; and/or
  • Controlled trials with mortality or major morbidity as a primary or secondary endpoint where increased morbidity or mortality may be better assessed through statistical comparisons among treatment groups.

Note: The Principal Investigator is responsible for oversight of all aspects of the trial, including safety. The inclusion of other reviewers or safety monitoring mechanisms does not relieve the investigator's responsibility.

Information and Assistance

As always, the IRB staff are available to provide assistance or answer questions.

Additional specialized assistance with independent data monitoring is available through the Department of Clinical Sciences. For more information or to request assistance, please contact the CIRC Services Office at 214-648-5005 or circ@utsouthwestern.edu.

NIH Policy requires data and safety monitoring for all NIH-unded clinical trials. The divisions within the Institute have each established policies, procedures, guidelines etc. to implement and comply with the general NIH policy. (See guidance for Data Monitoring Plans and Data Monitoring Committees – NIH and NCI policies).

Sponsors of FDA-regulated research are responsible for safety reporting to the IRB, to the study sponsors, and to the FDA (particularly in the case of investigator-initiated INDs or IDEs). These guidelines incorporate Sponsor-directed safety planning. (See guidance Data Monitoring Committees – FDA March 2006 "Guidance for Clinical Trial Sponsors,")

Return to Top