Component Analysis: Analyzing Benefits and Risks of Individual Interventions to Ensure Appropriate Research Participant Protections
A major element of any IRB review is the examination of potential benefits and risks to study participants. In studies involving multiple interventions and/or placebo, applying component analysis can help IRB members understand the different levels of risk associated with each intervention.
Component analysis differs from an overall risk assessment, in which a study’s combined interventions and procedures are evaluated as a whole. In 1978, the National Commission stated, “To determine the overall acceptability of the research, the risk and anticipated benefit of activities described in a protocol must be evaluated individually as well as collectively.” Component analysis recognizes that each intervention may have different risks and may or may not offer direct benefit to study participants.
One way IRBs commonly apply component analysis is in evaluating research involving vulnerable subject populations like children, pregnant women, prisoners, and others. FDA and OHRP regulations require that research involving vulnerable populations include additional protections to ensure those vulnerable participants are not exposed to undue risk. In studies with multiple possible interventions or arms, component analysis can help uncover the greatest possible risk a participant could encounter.
Let’s dive a little deeper and examine how component analysis may be applied to research involving children. FDA and OHRP regulations for pediatric research define four categories of approvable pediatric research, using the criteria of minimal risk and potential for direct benefit to determine the review process and possible additional requirements.
In 2013, the FDA stated that in pediatric research, a placebo cannot be evaluated to offer the prospect of direct benefit to participants. So for pediatric studies involving multiple interventions and/or placebo, each procedure should be evaluated individually to determine the highest possible level of risk study participants may encounter.
Here is an example to illustrate:
Researchers plan to conduct a blinded, randomized, placebo-controlled study of an investigational oral flu preventative drug in healthy children 12-16 years old. The protocol requires a physical exam, administration of the investigational product (IP) or placebo, two nasopharyngeal swabs, and a flu symptom questionnaire. Component analysis asks the IRB to individually examine of the study’s possible interventions: participants will either receive the investigational product or a placebo.
- Children receiving the IP will be exposed to research involving greater than minimal risk, as the nasopharyngeal swabs go past the nares and present risk slightly greater than what the average child would experience in daily life. Additionally, the IP presents the potential for direct benefit to the participants.
- The IP arm of this study meets the requirements for approval category 2. Only one parent’s consent is required for this category.
- Children receiving placebo will also be exposed to research involving greater than minimal risk because of the nasopharyngeal swab procedure. However, placebo cannot present the potential for direct benefit, though the study may yield generalizable knowledge about the participants’ disorder or condition.
- The placebo arm of this study meets the requirements for approval category 3. Both parents’ consents are required for this category.
- Because this is a blinded, placebo-controlled study, researchers won’t know which participants will receive the IP versus the placebo. In this example, the IRB applies approval category 3 to the entire study and requires both parents’ consents be obtained.
The regulations require IRBs to ensure that risks to study participants are minimized, and “risks to subjects are reasonable in relation to anticipated benefits, if any” (45 CFR 46.111[a] ). By using component analysis, IRBs help to ensure that study participants are appropriately protected no matter what the study design.