1. New Human Subjects Rule
This presentation was prepared for the use of the Carleton College community. It was delivered on Jan. 11, 2018.
AN OVERVIEW

2. IRB Review: The Old Rule
Three categories: Exempt, Expedited, Full
Never exempt: Children, Prisoners, Pregnant women
Approval good for one year only
Multi-institutional research? Storing and maintaining data for secondary use?
Is oral history covered?
Old rule:
Most projects at Carleton are “exempt,” meaning they fall into a category that the Rule defines as low-risk research not needing IRB oversight.
Specifically, the IRB won’t require you to get informed consent. Under the old rule, no research on children could be exempt.
Some projects are low-risk but don’t fall into an “exempt” category. These get “expedited” review: approval by one IRB member is sufficient.
A very few projects are high-risk and need “full review” (the whole board meets and reviews it). All research on prisoners falls into this category.
There were some other questions that the old rule simply didn’t address.

3. New rule New definition of research & new exempt categories
“Exempt with limited review”
Some research on minors is exempt (and pregnant women no longer a vulnerable category)
Expedited research need not reapply after 1 year
Multi-institutional research, storing data, oral history addressed
Plus some stuff on consent
The new rule was finalized in January of 2017 and set to go into effect in January of HHS recently proposed to change the rule to delay implementation until but that proposal has to go through a long process before it is finalized, so we’re going ahead with implementation in Jan 2018, as the new rule directs.

4. Carleton Policy vs. Federal Rule
All persons involved in conducting research have an obligation to respect the dignity and integrity of the persons beings studied, including their right not to be the subject of potentially harmful research.
Where possible, potential subjects should be provided the opportunity and means to decide freely whether to participate.
Researchers who promise confidentiality are responsible for maintaining it and for informing subjects of the limits of their capacity to meet that responsibility.
Research procedures should minimize the risk of harm and maximize the possible benefits to the subject and to society.
Subjects should be selected for reasons directly related to the problem being studied, not because of their easy availability, their compromised position, or their manipulability.
IRB implements the federal rule, but it also implements Carleton policy--which might be more demanding. We may require consent where the federal rule doesn’t:
E.g. survey or interview research that might retraumatize the subject. We treat that as requiring at least informed consent; the new rule does not. It doesn’t recognize this sort of psychological harm.

5. Multi-institutional Research
One IRB To Rule Them
(unless you’re doing research in
another country)
Our policy: one IRB must be designated as the IRB of record, which IRB must agree in writing to be responsible for approving and ensuring compliance with the federal regulations. This agreement and documentation of the IRB’s approval must be received by the Carleton IRB before research may begin.
Note: THIS does not apply to research in foreign countries; Carleton IRB must approve it and will ask that a comparable institution in the other country does as well.
Note: The rules don’t specify who gets to choose which IRB will be in charge (although sometimes federal granting agencies will specify that). Our assumption is that the researchers can choose which institution to apply to.

6. “Not research”
Research: Systematic investigation designed to contribute to generalizable knowledge.
The following are not research:
“Scholarly and journalistic activities (e.g., oral history, journalism, biography, literary criticism, legal research and historical scholarship)…that focus directly on the specific individuals about whom the information is collected.”
This new definition of “research” is not a big change, it just make the exclusion of oral history etc. explicit. This may still leave some gray areas. Case studies are an example; some look like social science research, but others are more journalistic.

7. Exempt categories The major ones for us are:
Cat 1: Normal educational research
Cat. 2: Low-risk social research*
Cat. 3: Behavioral experiments*
Cat. 4 & 8: Secondary research on (identifiable data or biospecimens)
Cat. 7: Collecting (identifiable data or biospecimens) for secondary use*
*Requires “limited review” if you’re collecting data that are both sensitive and identifiable
Exempt means you do not need to get informed consent.
*These categories are “exempt with limited review” if you are collecting sensitive data that will remain identifiable. “Sensitive” means disclosure could damage subject’s financial standing, employability, educational attainment, or reputation or make them subject to civil or criminal liability. “Limited review” means making sure the privacy and confidentiality of data is protected.

8. 1: Educational research
Research conducted in established or commonly accepted educational settings, involving normal educational practices that are not likely to impact students’ opportunity to learn required educational content.  
Note: We are proposing additional safeguards if a professor wants to study her or his own students. We’ll discuss that next.
This category generally refers to “scholarship of teaching and learning” research, such as evaluating the effectiveness of a new pedagogic method. Note: This exemption applies to research on minors, too!
However, the rule does not offer any clarification as to what is a “commonly accepted education setting.” Commentators on the rule from universities noted that this phrase is hard to interpret--is it only research in classrooms? What about off-campus programs, e.g.? However, it is clear from the discussion that the intent of this rule is to exempt normal pedagogical research.

9. Pedagogical Research Using One’s Own Students—Carleton policy
When making adjustments to a class, professors often wish to study the learning outcomes of their own students. This is both admirable and a hallmark of sound pedagogy. For the most part, these activities are considered to be exempt from IRB oversight (note that it is the IRB that must determine “exempt” status).  That said, a pillar of human subjects research is voluntary, uncoerced, informed consent of the participants. Due to the inherent power dynamics between a professor and the students currently in their classes, it is the opinion of Carleton’s IRB that students cannot fully consent. In order to protect our students, and to preserve their right to opt out of human subjects research, we ask the following of our faculty:

10. Pedagogial Research: Continued
Please enlist a faculty partner, who is not in any way associated with the class in question, to handle the consent forms and administer the activities or surveys in question. This faculty partner will hold all documents, files, or outcomes related to the study in their possession until grades have been assigned and the course is complete. In this way, the professor of the course being studied cannot know which students do and do not consent to participate in the research until a time when this knowledge cannot affect the grades of any students. Knowing this allows the students more freedom to opt out of the study without negative impacts. We require that their consent be obtained before the study, which necessitates that we designate these studies not “Exempt,” but rather “Limited Review.” 

11. 2: Low-risk social research
Research that only includes interactions involving educational tests (cognitive, diagnostic, aptitude, achievement), survey procedures, or observations  of public behavior…. If at least one of the following criteria is met:
Identity of subjects can’t be readily ascertained (data aren’t identifiable), or
Disclosure wouldn’t put subjects at risk (data aren’t sensitive), or
The data are sensitive and identifiable but: IRB conducts a limited review to make the determination that there are adequate provisions to protect the privacy of subject and confidentiality are protected
See IRB website for more info on protecting your data
For this category “educational tests” refers to things like IQ tests--unlike the first category, it covers research that takes place outside the classroom.
Note: “data isn’t identifiable” means that it’s been recorded in such a way that someone couldn’t identify the subjects from your notes. (*You* might know their identity, but your data won’t reveal them to others). The IRB FAQs #8 addresses how to protect your data.

12. Low-risk social research on minors?
Exempt:
Surveys & interviews, if data are not sensitive or not identifiable;
Educational tests & observations of public behavior if data are not sensitive or not identifiable AND you’re not interacting directly with minors
Not exempt:
You’re doing educational tests or observations of public behavior and interacting directly with minors; or
You’re collecting sensitive, identifiable data
This sort of research can be expedited, but you will need parental consent
Some research in category 2 is exempt even when subjects are minors, but there are a few exceptions.

13. 3: Behavioral experiments
Research involving benign behavioral interventions in conjunction with the collection of information from an adult subject or audiovisual recording if (1) subject agrees to intervention, (2) there’s no deceit involved (or the subject agrees to be deceived) and at least one of these conditions is met:
Identity of subjects can’t be readily ascertained (data isn’t identifiable), or
Disclosure wouldn’t put subjects at risk (data isn’t sensitive), or
The data are sensitive and identifiable but: IRB conducts a limited review to make the determination that there are adequate provisions to protect the privacy of subject and confidentiality of data
Benign: defined as brief in duration, harmless, painless, not physically invasive, not likely to have adverse impacts, not offensive or embarrassing.
What is “not physically invasive”? This term isn’t defined, but the discussion of the rule makes clear that use of medical interventions (medical tests, procedures, devices) is not covered by this exemption. But putting people in a noisy environment is covered.
The discussion also explicitly says that “withholding the purpose of the research” counts as deception. (It does not say that withholding the hypothesis is deception.)

14. 4 & 8: Secondary research
Secondary research using potentially  identifiable private information or biospecimens:  
Cat. 4: Fully exempt if:
The data are publicly available, or
The information is recorded in manner that subject’s identity isn’t readily ascertained and investigator won’t contact or re-identify subjects
Cat. 8: If data or biospecimens will remain identifiable, the IRB will review to ensure that:
Subjects consented to this sort of use; and
Privacy and confidentiality of data is protected.
The category 4 exemption applies to research on minors, but cat. 8 doesn’t.
Note: The OHRP promises to develop a list of technologies that could be used to re- identify biological specimens. Again, we address how to de-identify data in our FAQ question 8.

15. 7: What if you’re collecting & storing data for secondary use?
This is covered by category 7:
Storage and maintenance for secondary use of identifiable data or identifiable biospecimens is technically exempt, but not really. IRB will review for
“Broad consent” from subjects;
Privacy and confidentiality of data is protected.
“Broad consent”? What’s that? We’ll get to that in a minute….
This “exemption” does apply to research on minors, as long as their rights are protected (e.g. you got parental consent).

16. Participant Observation?
Probably not exempt (unless it’s limited to public observation & interviews only)
Please look at the guidance on the IRB website on how we review participant observation research

17. Wait, what about 5 & 6?
Cat. 5: evaluation of federal public service or benefit projects, if they are approved by federal dept. head
Cat. 6: Taste test, food quality studies
We rarely have projects that fall into these categories.

18. Summary
FULLY EXEMPT: most low-risk research collecting info that isn’t sensitive or isn’t identifiable
LIMITED REVIEW: most low-risk research collecting info that is sensitive and identifiable.
MIGHT NOT BE EXEMPT, EVEN IF LOW-RISK: Research on minors, participant observation, deceptive experiments
ALWAYS FULL REVIEW: More than minimal risk, research on prisoners

19. Consent
Key Points

20. Why Consent?
The aim of requiring consent is to make sure participation is voluntary 
Subjects of special concern:
children
prisoners
persons with impaired decision making abilities, and
persons who are economically or educationally disadvantaged
(for us): students of the Principal Investigator, who may feel coerced
If your research is exempt, that means it’s exempt from the informed consent requirements. However, you might still want to get consent. Our campus policy is: All research on human subjects conducted by Carleton faculty, students and staff, at Carleton or at other institutions and research sites, must conform to these ethical principles.

21. Informed Consent All consents form should:
Begin with a concise, focused presentation of the key information needed to make this decision;
Present the information in a way that facilitates comprehension of why one might and might not want to participate; and
Contain no exculpatory language (waiving of legal rights).
The new rule is trying to discourage the long, detailed, hard-to-read consent forms that have become common (especially in medical research).

22. Include only if relevant:
An “informed consent” form should also include:
An explanation of the research, duration and procedures subjects will experience;
Reasonably foreseeable risks to subjects;
Reasonably foreseeable benefits to subject or others of the research;
The extent to which subjects’ info will remain confidential;
Contact info for researcher & IRB;
A statement that participation is voluntary and subject can discontinue participation at any time without loss of any benefits to which the subject is entitled.
If you’re collecting identifiable information, you also have to explain whether it may be de-identified, saved and used for additional research in the future without getting subject’s consent.
Nevertheless, the rule still gives us a long list of stuff to cover. We will read “if relevant” into this list. E.g.: Risks and benefits may not be relevant to your projects. You don’t have to make up something to say about these nonexistent risks and benefits. Just give your subjects the relevant info.

23. Broad consent
When you’re collecting data or specimens to be stored for secondary use by other folks, you may need to secure broad consent. Explain:
the kind of research that may be conducted with the data or biospecimens;
whether you’ll be sharing this stuff with other institutions, and what types of researchers or institutions might use it;
how long it will be maintained and used; and
whether the subjects will receive reports on any research using this stuff.
This is new. Broad consent forms should also be simple and easy to read.

24. Again, only if relevant:
You should also:
describe foreseeable risks to subject from the research;
describe any benefits to the subject or others from the research;
describe how confidentiality of the records identifying the subject will be maintained; and
tell subjects whom to contact about storage & use of the stuff, and whom to contact about any harms resulting from the research.
Also, if relevant (this applies only to biospecimens):
whether the biospecimens will be used for commercial profit and
whether the research will or might include whole genome sequencing
Again, only if relevant.

25. Waiving consent Informed consent can be waived by the IRB if
the research involves no more than minimal risk,
the subjects’ rights and welfare won’t be affected,
the research couldn’t be carried out without the waiver, or
the subjects will be given pertinent information after they participate (when appropriate).
Broad consent:
If subject was asked to consent to storage or private identifiable information and refused, the IRB cannot waive consent (although it can waive documentation of consent in appropriate circumstances).
No big change here for most research. On our campus, waiving consent is usually called for only experiments involving deception or in research involving interaction with minors (for example, participant observation studies in communities that include children.) But note that broad consent can’t be waived.

26. Waiving documentation of consent
The IRB can waive documentation of informed and broad consent for minimal risk research for any of these reasons:
To secure privacy:  If the consent form is the only record linking the subject to the research, and the principal risk of harm from the research is breach of confidentiality.  In this case, you must ask the subject whether they want documentation linking the subject with the research.  The subject’s wishes will govern.
Written consent isn’t the norm:  The research involves no procedures for which written consent is normally required outside the research context.  
The subjects don’t do forms:  the subjects are members of a distinct cultural group in which signing forms is not the norm.  In this case, you must use an alternative way to document informed consent (like noting it in your field notes).
Note: waiving documentation of consent is not the same as waiving consent.  Also, note that the IRB might require that you present a written explanation of the research even if you aren’t having subjects sign a form.

27. Additional resources:
CITI resources on the new rule (link to CITI is on IRB website)

28. This presentation was prepared by:
Kim Smith, with support from:
Kathleen Galotti
Dave Huyck
Paula Lackie
Christopher Tassava
Dannette Demann
Charlotte Whited