Exempt research and the IRB's residual obligations

Ross Woods, 2023, with thanks to Dαwiε ναn Vυυrεn

Let's start with an example:

John wants to do research using a survey. He writes a research proposal and fills in a form for ethical approval. He ensures that subjects will not be at risk of any harm. He will also conceal subjects’ identities, and withhold any information that could be used to identify them.

The research ethics committee (the IRB) assesses the proposal and agrees that subjects will not be at risk. The committee notices that the category is survey research, which is one of the categories eligible for exemption from the US legal requirements (Part 46 Protection of Human Subjects).

It then decides that John’s proposal is exempt from Part 46 requirements, and issues an exemption letter. The IRB has done its job according to law.

However, the committee is still responsible to make sure that John complies with ethical guidelines. What does it still have to do?

In other words, if the IRB grants an exemption under the regulations, what ethical obligations should it still impose on researchers? These may be termed “residual obligations,” because they remain after the legal requirement has been met.

Some background on exemptions

The US Part 46 Protection of Human Subjects allows exemptions from the regulations for the following kinds of research:

“Research, conducted in established or commonly accepted educational settings, that specifically involves normal educational practices that are not likely to adversely impact students' opportunity to learn required educational content or the assessment of educators who provide instruction.” §46.104 (d)(1)
“Research that only includes interactions involving educational tests (cognitive, diagnostic, aptitude, achievement) …” §46.104 (d)(2)
“Research that only includes interactions involving

… survey procedures …” §46.104 (d)(2)
… interview procedures …” §46.104 (d)(2)
…observation of public behavior (including visual or auditory recording) …” §46.104 (d)(2)

Research involving benign behavioral interventions in conjunction with the collection of information from an adult subject through verbal or written responses (including data entry) or audiovisual recording if the subject prospectively agrees to the intervention and information collection §46.104 (d)(3)
Secondary research for which consent is not required §46.104 (d)(4)

Despite minor differences, they all tend to require no risk of harm to subjects and that data be de-identified.

After meeting legal requirements for exemption, the matter is then completely at the discretion of the IRB, which must decide on the ethical requirements it will require of researchers.

Putting aside several other considerations

The first two of the exemptions listed above refer to education. We can put aside several considerations regarding the question: Could a school or college collect information and then release it to the researcher?

Breach of privacy. The institution (e.g. a school or college) may collect information and then release it to the researcher if it is de-identfied and participants cannot be re-identified. If so, there is no breach of privacy, especially if data is agglomerated into statistical form. However, the ownership or right to use the data transfers to the researcher only after it is de-identfied.
Pool of private information. The data is not a pool of private information for which broad consent is required, because:

It is not a repository of data to be used for secondary research.
It contains no private information.
Cf. §46.102, 46.116(a)(1)-(4), (a)(6), and (d).

Residual obligations

Even though IRB has met its legal obligations, it must still address the basic principles of ethical research with human subjects, which are:

The researchers must inform subjects about the research project:

Participation is voluntary.
Subjects have the right to withdraw.
Information must be in a form that subjects can understand.
Researchers may not deceive subjects.
Get subjects’ permission.

Minimize the risk of harm OR Favorable benefit for amount of risk to both subjects and to parties other than subjects.*
Protect private information.

*One of the most serious weaknesses in Part 46 Protection of Human Subjects, is that it does not consider harm to parties other than subjects. These typically include other people and might include other organizations.

The following tables suggests how these principles might be applied:

Principle	Application
The researchers must inform subjects about the research project.	This is quite easy to do as there is no risk of harm or identification. In a school, it might be enough to give a class announcement (for adults), or (for minors) a note home to parents.
Participation is voluntary.	In a school, voluntary has a different meaning in a classroom that includes the whole class. It more likely means that students or parents can opt out rather an opt in. See also below regarding the question of perceived coercion.
Subjects have the right to withdraw.	This is not an impediment. If the research is normal class activities and records, it means that some records (written or video) will be excluded from the research data.
Information must be in a form that subjects can understand.	This is not an impediment.
Researchers may not deceive subjects.	In some cases, telling subjects of the actual purpose of the research makes them act differently, making the research invalid. In this case, researchers should give subjects an honest, broader explanation of the research but not reveal the details of their specific interests. (Cf. §46.104 (d)(3)(iii))
Get subjects’ permission.	Tacit approval is sufficient. An approval form is not required because there is no risk to students. (Cf. §46.117(c)(1)(ii)) However, an opt out form is appropriate.
Minimize the risk of harm OR Favorable benefit for amount of risk.	A condition of the exemption is no risk of harm. However, the IRB should take steps to ensure there is also no risk of harm to other parties.
Protect private information.	Information is de-identfied such that participants cannot be re-identified; this is a condition of the exemption. In a school example, names and unique characteristics are removed when school data transfers to research data. Video images of class members who are not subjects are pixelated.

The question of percieved coercion

Ginting (2022, p. 80) compares two earlier studies on students' views of possible forms of coercion:

Item	Yes	Neutral	No
1. Do you object to being recruited in class for participation?^a	1%	3%	96%
2. Did you feel coerced or forced into participating?^a	2%	1%	97%
3. Is the giving of extra credit for participation coercive to you?^a	47%	14%	39%
4. Do you object to being given extra credit for participation?^a	3%	1%	96%
5. Is a professor encouraging students to volunteer to participate in their research projects as subjects unethical?^b	34.8%	1.7%	63.5%
6. Is having a student be a research participant as part of a course requirement (with no alternative) unethical?^b	71%	1.5%	27.2%

^aLeak, 1981.^bKeith-Spiegal et al. 1993.

It is quite possible that students vary in perceptions over time and between cultures, but at least a score of over 90% seems quite conclusive. Anyway, some tentative conclusions seem to be:

Students do not object to being recruited in class for participation and do not feel coerced. (Q. 1 & 2)
Students also do not object to being given extra credit for participation (Q. 4) but nearly half thought it was coercive. (Q. 3)
Most students thought it was ethical for a professor to encourage students to become voluntary research subjects (Q. 5), but most thought it was unethical to make it a course requirement with no alternative. (Q. 6)

Ginting also points out that IRBs view that compensation (whether financial or otherwise) exerts undue influence, that is, “excessive persuasion that causes another person to act or refrain from acting by overcoming that person's free will and results in inequity.” In other words, “compensation may not be the main factor that encourages or influences the decision to participate in research.” (p.81) This still allows for some level of compensation that is not coercive.

________________
Ginting, Daniel. 2022. Ke Pertimbangan Etis Penelitian: Ke Lika-liku Pendidikan dan Penelitian dalam Ilmu Sosial Humaniora di Era new Normal. (Malang Indonesia: Media Nusa Creative).