Determination of Study Risk and Data and Safety Monitoringfor Research
Supported by the Johns Hopkins University General Clinical Research Center Grant # M01 RR00052  
December 2005

Determination of risk should include a consideration of both the interventions being performed and the study population.  Risk assessments must take into account special circumstances that are unique to the study such as disclosures of HIV status, results of genetic tests, or surgical procedures.  Levels of risk will also vary depending on the characteristics of individual subjects and the experience of investigators with the interventions proposed. ⁽¹⁾  An initial assessment of risk should be made by the principal investigator. ⁽²⁾   The GAC will consider that assessment and assign a level of monitoring appropriate for the protocol at the time of initial review. ⁽²⁾

Risk shall be defined as:  The probability of harm or injury (physical, psychological, social, or economic) occurring as a result of participation in a research study.  Both the probability and magnitude of possible harm may vary from minimal to significant. ⁽³⁾  Federal regulations define only "minimal risk."  “Significant risk” has no single accepted definition, and is not categorized, for example, as “high, medium or low”. 

Minimal Risk – [45 CFR 46.102(i)] - means that the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests. ⁽⁴⁾  Risks include all harms, discomforts, indignities, embarrassments, and potential breaches of privacy and confidentiality associated with the research. ⁽¹⁾

Minimization of Risk

It is the duty of the investigator and everyone involved in clinical research to ensure that risks are minimized in all research. Thus, even in research studies that have risks deemed minimal, or a minor increase over minimal, every attempt should be made to minimize risks.  For example, procedures should only be performed by appropriate personnel and protocols should include, when appropriate, specific rules setting limits on the number of attempts at a procedure or the length of time for completion of a questionnaire. In addition, appropriate methods should be used to orient the [research subject] to the research and decrease potential anxiety and discomfort, and explicit plans should be developed to protect subjects from breaches of privacy and confidentiality. ⁽¹⁾

Data and Safety Monitoring Plans (DSMP)

ALL protocols conducted on the GCRC must have a GAC-approved data and safety monitoring plan. ⁽²⁾  A DSMP should accomplish the following:

• identify the investigator’s assessment of risk,
• establish the overall framework for data and safety monitoring,
• describe the entity that will be responsible for monitoring subject safety,
• establish the frequency of monitoring,
• describe the mechanisms for reporting adverse events to the IRB, FDA, GCRC and NIH in accordance with IND or IDE regulations
• include the name of the individual responsible for adverse event reporting

Investigators should be aware that various regulatory bodies use different language to address the level of risk.  For example, the NCRR requires that all protocols that place participants at significant risk have an independent Data and Safety Monitoring Board (DSMB), ⁽²⁾ while NIH requires the establishment of a DSMB for multi-site clinical trials involving interventions that entail potential risk to the participants, and generally for Phase III clinical trials. ^{(6, 7)}

Data and Safety Monitoring Boards (DSMB)

NCRR requires that all protocols that place participants at significant risk have an independent Data and Safety Monitoring Board (DSMB). ⁽²⁾

NIH requires the establishment of a DSMB for multi-site clinical trials involving interventions that entail potential risk to the participants, and generally for Phase III clinical trials. ^{(6, 7)}

The main objective of a DSMB is to ensure the safety of research participants.  DSMBs accomplish this by assisting the investigators in:

• evaluating interim results and making decisions about continuing, modifying, or terminating a study if adverse effects are serious enough, ⁽⁸⁾
• reporting credible study results, ⁽⁸⁾
• conducting the study ethically and above reproach, ⁽⁸⁾
• ending the study when the data are sufficiently favorable to one treatment. ^{(3, 8)}

Protocols that typically require a DSMB or an individual independent monitor include:

1. Protocols that generate blinded/randomized data, ⁽⁹⁾
2. Multi-site protocols presenting more than minimal risk to subjects, ⁽⁹⁾
3. Protocols using gene transfer or gene therapy methodology, ⁽⁹⁾
4. Phase I or Phase II trials that are high risk, involve vulnerable populations, or are blinded, ⁽¹⁰⁾
5. Protocols that the sponsoring IC believes require special scrutiny because of high public interest or public perception of risk, ⁽⁹⁾
6. Controlled trials with mortality or major morbidity as a primary or secondary endpoint or in settings where trial participants may be at elevated risk of such outcomes even if the study intervention addresses lesser outcomes such as relief of symptoms.  ⁽¹¹⁾

• If either the IRB or GAC decides that a study should have a DSMB, one will be required prior to study initiation in the GCRC. ⁽²⁾
• For multicenter studies, the GAC will consider the DSMB proposed; whether it will be single site or study-wide should be specified by the investigator and the GAC.
• The constitution/charter (or list of the responsibilities, meeting frequency, etc.) and membership of any DSMB, as detailed in the NIH Guidance on Data and Safety Monitoring, must be described. ^{(2, 5)}

 References: