Special Population Requirements

Main Page Content

Changes with Final Rule

With the update of the Common Rule set to go into effect on July 19, 2018,  there will be modifications and revisions involving vulnerable populations.

The Final Rule’s preamble states that the possibility of coercion or undue influence could affect the ability
to make an informed decision about participating in research. Therefore, the vulnerability of the subjects
in research studies should be considered as a function of the possibility of coercion or undue influence.
The preamble states that this type of vulnerability alone should be the IRB focus of concern in determinations about vulnerable populations.

The preamble also notes that the assessment of the equitable selection of subjects (46.111[a][3]) should
include factors like societal marginalization or discrimination. Likewise, the preamble discusses that
the criterion at 46.111(a)(1) includes risks that some might term “vulnerabilities,” which are not covered
by the regulatory term.

NEW: Pregnant women and physically disabled individuals will no longer included as examples of populations that are potentially vulnerable to coercion or undue influence. 

NEW: The Final Rule uses the term “individuals with impaired decision-making ability” to replace the term “mentally disabled persons.” 

NEW: Prisoners can be used as research participants in exempt or expedited research if the research propulation only involves prisoners and is not aimed specifically at them.  

NEW: For research conducted with non-English-speaking and/or international populations, signed consent can be waived if it is not the cultural norm for the subjects to sign such documents, as long as:

  • The research is no more than minimal risk and
  • An alternative documentation mechanism is used

If the special population or issue you are studying is not addressed below, please check the IRB FAQ or contact the IRB office.


Classrooms and students

Illegal behavior

Inclusion of women and minorities in study populations

Non-English-speaking populations and international research projects



Consent/assent issues

Parental / Guardian Consent

The informed consent must clearly explain what is required of child as well as what is asked of the parent. The document should also clearly state if there is to be future contact with the family or if there will be any observations or interactions with the child while the parent is not present.

Child Assent

Since children cannot give legal consent to participate in research, we speak of parents giving consent, but their children give their assent, where appropriate. Children can only be approached for assent after parental consent has been given. The assent is usually a scripted interaction, though with older minors it can be a document. The assent must be in age-appropriate language, explaining that their parents have consented to have them participate, but they do not have to participate if they don’t want to. If appropriate, the assent should state that the research study is not school and will not affect grades or teacher interaction. The assent should state if they will miss any activities from the regular school day, and how that will be made up. The script should also clearly state what the child is being asked to do, where, for how long, and with whom, and the nature of any compensation; explain that they can stop participating at any time and still receive part of that compensation for their time; and the extent that their responses will be confidential.

Group Assent

The IRB discourages group assenting procedures (asking children in a classroom setting to raise their hand if they do/do not wish to participate). This creates social pressure that may sway judgment. If at possible, it is preferable to ask children in a one-on-one setting or devise a system of passing out papers and having the child decide and covertly mark the paper individually.

Opt-Out Consents

The IRB does not approve “opt out” “consent” procedures. IRB does not recognize “passive consent” as such; you must request a waiver of (parental) consent.

Waiver or Parental Consent

The IRB can waive parental consent under current regulations (under 45CFR, both 46.408(b) / 46.116(d)), if the study is minimal risk, not practicable without the waiver, and waiver won’t affect the participants’ risk or rights, and if appropriate efforts are made to inform the parents about the work. Arguments that these conditions are met should be included in the protocol submission. The IRB tends to be very conservative in approving this unless the survey is benign and anonymous. If this consent method is sought, the researchers much also submit a letter from the school indicating their approval for the procedure in the absence of parental consent.

Asking children to evaluate their peers or family
  • Some studies are interested in peer social interactions, resulting in a need to have identified children evaluate other identified children, for example to correlate what one says about oneself to others’ views of them. This creates some difficulty, however, because the the named classmates are participants in the research, as the regulations [45 CFR 46.102(f)(2)] define human subject as a person “about whom an investigator obtains identifiable private information.” The consequence is that researchers will need to get parental consent and child assent both for children who are evaluating others and those classmates who are being evaluated.
  • If the evaluation of others is intended only to inform you about the views of the respondents, then there may be no need to obtain/maintain identifiers for the classmates. If that’s the case, the procedures to keep any information about those who are being evaluated anonymous should be made clear in the protocol and consent forms.
  • See Special Populations: Classrooms and Students in this section for more information about classroom studies, as well as Classroom and School-based Research elsewhere on the IRB website.
  • When researchers wish to ask children about their family environment or the behaviors that they witness from their family members, the family members are similarly participants in the study. In such cases, there may be significant risks to the family members if information about financial status, emotional well-being, illegal behaviors or other sensitive topics were disclosed. Their consent must be sought, or the researcher needs to explain why their consent should be waived.
Asking parents / guardians / teachers to report on the behavior of children
  • As when children are asked to report on the behavior of peers and parents (see above), assent may be needed from the true subjects of the study, in this case, the children, when parents are being asked to report about their child’s behavior. This requirement is particularly important when the child can be identified, and the topics are sensitive ones. The IRB will consider the particular circumstances to determine if assent is needed in all such third-party interviews.
Mandatory reporting of abuse
  • If the researcher or their staff obtains credible evidence that a child being evaluated for research purposes shows signs of abuse, it is mandated by state law that the researcher must report that information to the New York Statewide Central Register of Child Abuse and Maltreatment (sometimes referred to as the State Central Register or SCR). Research that, because of the topics explored (for example, parental discipline) or because of the population to be studied (for example, children with mood disorders or other at-risk groups), is more likely to uncover evidence of abuse, the consent and/or assent scripts should state explicitly that any evidence of abuse must be reported to a state hotline.

Click here for more information about reporting, or call 1-800-342-3720.

Asking children for information about behavior unknown to parents / illegal behavior
  • Some studies ask children about behavior unknown to the parents; this often concerns sensitive areas such as drug use, sexual practices, or criminal behaviors. It is not necessary that this information be shared with the parents, and often it is critical that it be withheld from parents. However, researchers must let parents and children know in the informed consent that they will ask about certain behaviors, and that the parents will not be able to find out their children’s answers.
  • If this information could be brought into a court of law or otherwise disclosed, the data should have potential identifiers as soon as possible, either when collected, or when coded, so that it cannot be connected to participants. (These include direct identifiers like social security numbers, driver’s license numbers, etc., as well as indirect information that could uniquely identify a person, such as a combination of demographic and descriptive information). A Certificate of Confidentiality may provide further protection from such disclosure. For more information, see Special Populations: Illegal Behavior.
Children who are not under their parents’ care (wards of the state, foster children, etc.)
  • Consent and assent must still be obtained when studying children who are not under their parents’ care.
  • It is the researchers’ responsibility to determine who is legally responsible for the participant (who could consent to have medical treatment done to the children?).
  • Information about who is legal guardian for the group of interest, and the source for this information, should be included in the protocol.
  • It is important to note that children may have been removed from the care of their parents for a reason that makes them particularly vulnerable, such as abuse/neglect. How this vulnerable population will be protected must also be addressed in the protocol.
Children with compromised mental/communicative capacities
  • See the section on Special Populations: Compromised Mental/Communication Capacity.

Sample assent scripts can be found on the Forms page of the IRB website.

Classrooms and Students

The college classroom is the site of many research activities, both when students are being trained in the research procedures and methods in the behavioral and social sciences, and when investigators are interested in research questions where students are the appropriate population.

Class Projects
  • Class projects do not need to be reviewed by the IRB if they are class projects within The College at Brockport classes, conducted only for instructional purposes, and there are no plans to use the research to “add to generalized knowledge,” such as presenting the results at a conference, publishing them, posting them to a publicly available website, etc.
  • Any class studies where there is a possibility that the results may be presented/published outside of the class should be reviewed and approved by the IRB and have proper consent/assent procedures.
  • Class projects may involve participants other than the students in that class; nonetheless, if the purpose of the project is strictly instructional, it does not qualify as “research” for regulatory purposes, and does not need to be reviewed by the IRB. Of course, professional and academic ethical guidelines should be followed for such activities, and the instructor is responsible for ethical oversight of their students’ projects.
Consenting minors (including college students who are younger than 18 years)
  • See section on Special Populations: Children
  • College students younger than 18 are minors, and parental consent as well as the student’s assent is necessary. When studying a college student population it is the responsibility of the researcher to ask the age of the college student and NOT TEST or COLLECT DATA FROM students who are younger than 18 years.
  • If a college classroom population is being tested that includes students younger than 18 years, and extra credit is being offered as compensation, then an equivalent alternative non-research opportunity must be available to those who are under 18.

If Observing Students’ Behavior In A Classroom Setting Or Having Participants Fill Out Surveys In A Classroom Setting

  • All consents and assents must include what activities students may do during testing time if they do not wish to participate in the classroom testing.
  • If participants you are observing in the classroom might be stigmatized by being singled out for observation, appropriate precautions must be made so that the other students will not know who is being observed and why. Similar precautions must be taken if select students are being removed from the classroom for some special status or if surveys are being filled out by a whole class that only will be appropriate for a subset of participants to fully fill out. Details must be provided in the protocol about how this stigma will be avoided.
  • Given the sensitive nature of some surveys, participants may be afraid that their answers will be viewed by other students or tracked back to them. Having the survey be anonymous and providing each participant with an unmarked closeable envelope in which the surveys can be placed prior to handing them to the researcher or advisor may decrease this risk. Providing adequate spacing between students during data collection may be essential to ensure confidentiality in these situations.

Offering extra credit as compensation for participation

  • If compensation is extra credit, those participants who do not wish to participate in research must be given a time/effort equivalent academic extra credit opportunity so that they can feel free not to participate, but still get benefits available to those who can/want to participate in research. This alternative activity must be described in detail in the protocol submission as well as in the consent form/assent script.
  • Extra credit for participation should be calculated such that any risk of coercion via “undue compensation” is minimized.

Accessing academic records

  • Explicit permission must be obtained to access confidential academic records. The researcher must detail in the consent/parent consent to access exactly what records will be seen and when. If the participant is a minor, but child is old enough to understand what these records are and how their release may affect them, this aspect of the research must also be assented to by the children.
  • Records cannot be accessed unless it is expressly consented to, even if some member of the research group would otherwise have access to these records for non-research purposes.

Students taken out of the classroom setting for testing

  • If any participants are going to be taken away from classroom activities for this research, the consent forms and assent script must include information about what classroom activities will be missed by the participants when they are removed from the classroom for testing. If students are to miss some classroom material, it must be detailed how it is that they will make up this material.
  • If only a subset of students will be removed from the class to be tested (for example, those with reading difficulties), it must be detailed in the protocol what procedures will be used to prevent these participants from being stigmatized by the other students who notice that the child is being removed because they are of a special population.

Evaluating Educational Programs/Educational Tools

  • It’s not uncommon for the IRB to receive protocols that have a mix of what’s called “program evaluation/quality assurance” aspects, such as examining a standard part of the curriculum or new method of educating. This may or may not require IRB approval. It is for the IRB to decide if the project is exempt from research. See Proposal Guidance for more information about this determination.

Classroom Observation

  • Common issues and considerations for protocols that observe students in classroom settings include the following:
    • Public school classrooms are not considered public spaces and therefore do not fall under Federal Exemption #2.
    • Will data be collected for all students? Are some singled out, because of teacher selection or other criteria?
    • Is consent necessary for all students about whom data is recorded, even if identifiers are not recorded?
    • Teacher consent may also be necessary.
  • Each of these issues must be addressed by researchers submitting a protocol of this nature.

Wanting to test students in a class where a researcher is teacher/supervisor/teacher’s aide, etc.

  • Though it may be convenient, there are many reasons NOT to test one’s own class. In fact, unless there is a compelling reason other than convenience to test one’s own class, it is highly encouraged to test another teacher’s class to prevent some of the problems of potential coercion. The idea of “coercion” is that students may feel obliged to participate, even if there is no direct “compulsion” to do so. This is a specific instance of a general problem that occurs when researchers are in some supervisory capacity regarding their targeted participants. Often two professors will swap classes so to not be testing their own students, who may feel pressured to participate.
  • If for some reason it is essential to test own students, it sometimes can be done. However, the reason why it must be one’s own class must be argued in the protocol. When this is allowed, special procedures must be in place to prevent coercion. Generally, this is managed by preventing instructor/supervisor knowledge of who has and has not consented to participate in the study.
  • For the same reason, we ask investigators collecting data from their own classes that they be blind to who has been / will be participating until the final exams are graded. Specifically, the researchers must explain how they will be blind to who has agreed to participate until after grades are finalized – or alternatively, how grades are determined “mechanistically” (e.g., objective tests) and the instructor is not making any judgments regarding grades. There is more of a problem if there is a “subjective” component to grading that could be influenced, even slightly, by the instructor’s knowledge of who has and has not agreed to participate. The procedure for doing this must be detailed in both the protocol and the consent forms.
  • There are often ways to arrange things to avoid the potential for coerciveness, such as by having everyone do some activity or exercise as part of the regular class. After final grades are determined, students and/or parents/guardians can be asked if their data can be used for research purposes. In other cases, when this is not practicable, the instructor can have the research component conducted by an assistant, and the instructor remains blind as to who has and has not participated. It is also helpful if it can be confirmed that there is no “subjective” component to the grade that could be so influenced.

Use of materials that have already been collected for non-research purposes

  • These documents, such as class assignments, papers, etc., can be examined for research. However, these materials need to be those that were gathered for purposes other than for the research study.
  • The researcher must explain that prior purpose in the protocol submission and who is offering the data and how it will be transferred to the researcher.
  • If at all possible, the data need to be processed before they are given to the researcher such that all identifiers have been removed. An identifier is a marker or set of markers that could be used to identify a specific participant.

Use of Banner ID for Identifiers with Students at The College at Brockport

  • Banner ID number can be used as identifiers for research and may be useful as thus if extra credit is being assigned as compensation for participation. However, collection of Banner ID is an identifier and should not be stored on data that is being kept confidential.
  • Researchers may never ask for participants’ password.

Opt-Out (Passive) Consent In Schools/Classroom Settings

  • The College at Brockport IRB does not approve “opt out” consent forms. The IRB doesn’t recognize “passive consent” as such; rather, primary investigators need to request a waiver of consent.
  • Federal regulations allow for waiver of parental consent/consent for minimal risk research under appropriate conditions, i.e., that the research has potential benefit but could not be practicably carried out without the waiver.
  • If the research is to take place in a P-12 school, there needs to be a letter of institutional support (Form H) submitted along with the protocol indicating school administrators’ approval for the procedure in the absence of parental consent.

Compromised Cognitive and/or Communication Capacity

Consent/Assent Issues

Two crucial issues when a researcher wishes to collect data from cognitively challenged participants are (a) legal consent and (b) communication about the research with the participant.

  • It is the researcher’s responsibility to determine if the participant themselves can legally consent. If not, it is the researcher’s responsibility to find out who can legally consent, and seek that person’s signed consent.
  • Regardless of whether the participants may legally consent for themselves, if it is judged that the participant is capable of understanding what is being asked of them, then it is the researcher’s responsibility to make a consent/assent process that will be optimally understood by the participant.
  • As with any consent/assent process, the protocol must detail how all efforts will be made to have the participant understand who the researchers are and how they can withdraw their consent if desired.
  • Also, in cases of those who are communicatively challenged, the protocol needs to detail how the researcher will determine that the participant wants to stop participating. The researcher may, for example, include consulting with the participants’ regular caregivers on how the participant might express when they wish to withdraw from an activity.

 Safety issues

  • Some special populations may pose a risk to themselves, those around them, or those working on the study. This is especially true of violent or self-injurious individuals.
  • It is imperative that the protocol describe the experience and credentials of those working with the population, so that the IRB can judge if the study will be carried out by those who have the skills to minimize the risk of injury, and to respond effectively to potentially harmful events.
  • There are also certain behavioral interventions with cognitively challenged violent individuals where the researchers must first determine what is rewarding unwanted behavior (commonly called functional analysis). If this unwanted behavior is violent to the self or others, the protocol must include details about how the functional analysis will be set up so to decrease risk of injury, when the functional analysis will be stopped, and when/how intervention will take place to stop self-harm or harm of others.
  • All of this information should also be included in consent documents. 

Illegal Behavior

General concerns when collecting data about illegal behavior

  • Any research that is gathering data on participants who are engaging in illegal behavior puts those participants at risk of being discovered and perpetrator and facing legal repercussions. This is most certainly true if the data collected details illegal behavior. However, it is also true if the data collected is not about the illegal behavior per se, but that the population of interest is those who engage in illegal behavior.
  • A number of illegal behaviors are often not recognized as such by researchers, but these behaviors raise the same concerns as any other illegal behavior. These include: underage drinking, underage smoking, and sexual behavior that, because of the age of the participants, is illegal.
  • Even though the researchers and The Co do not condone any illegal behavior, it is the researchers’ and IRB’s responsibility to determine how to study this population to offer them and/or society some benefit, while protecting these participants from the risk of negative repercussions of participation in research.
  • Generally, participants can be protected from later having the data collected used against them when subpoenaed in court by one of two strategies discussed further on this page:
    • Having no identifiers linked with the data
    • Obtaining a federal exception from having data obtainable by courts of law, a Certificate of Confidentiality

Truly anonymous data collection

One method for managing data that is collected from participants who are engaging in illegal behavior is to collect the data completely anonymously. This is a case where the requirement for a signature on the informed consent document can be waived by the board. No information that could identify the participants, directly or indirectly, should be recorded. This data includes but is not limited to: photographs, video, voice, or unique personal information (such as name, birthday, address, etc.). Depending on the population being studied, certain demographic information, alone or in combination with other facts about a person, could easily lead to their identification (for example, a sex offender’s race, age, and specific crime, in a particular county).

Use of non-identifying codes to track anonymous participants across time/sessions

  • In some cases, a group of participants are examined at multiple points, and the research design necessitates tracking of participants from one time point to the next.
  • In these cases, the IRB suggests having participants generate an identifying number that they can remember but that you could not trace them back to. Subject-generated codes can be unique, easily and reliably generated by participants, and usually involve some combination of parts of family first names, letters of hometowns, etc.

Certificate of Confidentiality

  • Some research designs necessitate collection of identifying information. In order to decrease the risk to participants, there is a federal mechanism for preventing the data from being subpoenaed by a court of law. This mechanism is a Certificate of Confidentiality, which may be provided by NIH.
  • It is the researchers’ responsibility to apply for this certificate and to then furnish it to the IRB board along with their protocol. If the researchers wish, they may first submit their protocol and their intent to apply for a Certificate of Confidentiality to the board.
  • For more information on a certificate of confidentiality, click here. https://humansubjects.nih.gov/coc/index Please read the information carefully and note that there may be a substantial wait to obtaining this certificate. The delay should be inquired about to the granting agency. The IRB has no knowledge or control over the granting of these certificates.

Inclusion of Women and Minorities in Study Populations

In the IRB’s analysis of the balance between the risks and benefits within a protocol, an evaluation of the quality of the research is a key factor. In order for research to have the optimal societal benefit, it must sample a population representing the population at large so that the findings are valuable as generalizable to the population at large.

NIH has specific recommendations concerning the proper representation of women and minorities in research studies:

  • Special emphasis should be placed on the need for inclusion of minorities and women in studies of diseases, disorders, and conditions which disproportionately affect them.
  • If one gender and/or minorities are excluded or are inadequately represented in this research, particularly in proposed population-based studies, a clear compelling rationale for exclusion or inadequate representation should be provided.
  • Composition of study population must be described in terms of gender and racial/ethnic group, together with a rationale for its choice.
  • Gender and racial/ethnic issues should be addressed when developing a research protocol and sample size appropriate for the scientific objectives of the study.
  • Protocols must employ a study design with gender and/or minority representation (by age distribution, risk factors, incidence/prevalence, etc.) appropriate to the scientific objectives of the research.
  • It is not a requirement for the study design to provide statistical power to answer the questions posed for men and women and racial/ethnic groups separately, however, whenever there are scientific reasons to anticipate differences between men and women, and racial/ethnic groups, with regard to the hypothesis under investigation you should include an evaluation of these gender and minority group differences in the proposed study.
  • If adequate inclusion of one gender and/or minorities is impossible or inappropriate with respect to the purpose of the research because of the health of the subjects, or other reasons, or if they are the only study population available, or if there is a disproportionate representation of one gender or minority/majority group, the rationale for the study population must be well explained and justified.

Non-English-Speaking Populations and International Research Projects

A wide range of protocols involve potential participants whose primary language is not English, or who do not speak English at all.

Informed Consent for Non-English Speaking Populations

  • It is the responsibility of the principal investigator to insure that the information provided during the consent process is comprehensible to the potential participants. If there are non-English-speaking participants to be recruited, the consent documents—including both written consent forms and verbal consent and assent (for children) script—should be submitted both in English and in the language of the target population. The investigator should document the qualifications of the translator, whether themselves or another person, in that language. Are they native speakers or the language? If not, what sort of training have they had? Since the goal of these scripts is effective communication, it’s important that the translator have extensive conversational experience in the target language, and not just “academic” training.
  • It is helpful, particularly for protocols that may involve more than minimal risk, to have the consent documents “back-translated” into English, so we can assess whether essential information has been accurately communicated in the target language.
  • One element of informed consent which may need modification with participants in studies taking place in other countries is the contact information for the Primary Investigator or the Brockport IRB. Participants may not have access to or experience with phones or email, for example; or it may be impracticable to have them phone an international number to contact a supervisor or the IRB office. In such cases, the PI should explain how participants will be provided with appropriate contacts for any questions about the research, or their rights as participants. In some instances it would be helpful to have a member of the local community act as a liaison in this capacity.
  • If the research itself involves direct interactions with participants, it is important that the individuals who will be interacting with participants in these studies be fluent in that language, whether it is merely to be ready to answer questions that might arise during the consent phase, or to conduct in-depth interviews. The investigator should explain in the protocol who will be interacting with participants, and again describe their qualifications and experience in that language and culture.
  • Processes for transporting data from the international location to The College at Brockport must be specified.
  • Specific processes for assuring confidentiality of all data must be specified, particularly if the analysis will occur away from the College at Brockport.

Literacy and informed consent

  • Particularly in certain international settings, potential participants may be illiterate. Obviously, neither a written informed consent document nor signed documentation of consent makes sense in such cases. Verbal consent scripts, with a waiver of documentation of informed consent (i.e. signed consent) may be proposed under these circumstances.

Review of research activities by other countries and communities

  • In addition to review by The College at Brockport IRB, research projects to be conducted in other countries may require approval by various agencies from those countries. This may involve anything from informal approval from the local community, to more formal review by regional or national bodies analogous to the IRB. It is the responsibility of the investigator to determine the appropriate mechanisms for such local review and consent, and to include a description of those mechanisms, and a statement of how they will be implemented before the start of any data collection.
  • The OHRP recently established a database of contacts for exploring the requirements for international research in many countries. If your research involves work in one of those countries, this is a good place to begin learning about human research protections in those communities. Although currently the bulk of the focus of those protections concern biomedical issues such as drugs and biological specimens, there is a growing interest aspects of research protections more appropriate to social and behavioral science research activities.

Cultural considerations in research and informed consent

  • It is not uncommon for expectations and concepts about research to be different among different cultures and communities. This may be especially true among non-English-speaking populations in foreign countries.
  • These cultural differences can mean a reluctance to sign any form, which can be seen as a potential threat in cultures where signatures are not the norm for economic or contractual social agreements. It is appropriate in such cases for the researcher to request a waiver of the requirement for signed documentation of consent, and obtain oral consent for participation.
  • Similarly, participant attributes that may define your target population—or that are disclosed as part of your research—such as a disorder, disease, ethnic or religious identification, sexual orientation, and the like, may be benign in our culture, but stigmatizing and potentially very harmful in another culture. In such cases, any additional steps that will be taken to protect the privacy of individuals or the confidentiality of this information, both during recruitment and in other phases of the research, should be explicitly described.
  • In general, you should explain in the protocol (a) how you will be able to identify such cross-cultural issues in your work (have you become familiar with the conventions or that culture? Do you have a local collaborator who is sensitive to these kinds of issues?); (b) any potential for such problems in your project; and (c) the steps that will been taken to minimize any additional risks that these may create.

A helpful resource for thinking about these kinds of issues is the American Anthropological Association’s statement about ethnographic research and IRBs.


 Why are prisoners a “special population?”

One of the three subparts of the federal code that focus on particular populations of subjects that may be especially vulnerable as research participants concern prisoners (Subpart C, 45 CFR 46.301-306). In the past, prisoners were sometimes seen as a convenient population for conducting all kinds of research where their ability to refuse to participate was compromised. The regulations are intended to afford these individuals additional protections because of this potential and historical vulnerability.

Scope of the definition of “prisoner”

46.303(c) defines “prisoner” as

[A]ny individual involuntarily confined or detained in a penal institution. The term is intended to encompass individuals sentenced to such an institution under a criminal or civil statute, individuals detained in other facilities by virtue of statutes or commitment procedures which provide alternatives to criminal prosecution or incarceration in a penal institution, and individuals detained pending arraignment, trial, or sentencing.

For purposes of the IRB, this would include, individuals held in a prison, county penitentiary, county or city jail, court, detention facility, hospital prison ward, police lockup, halfway house, juvenile secure and/or non-secure detention facility, and juvenile residential placement facility, as well as those committed to a host of alternative settings, such as day reporting facilities, youth courts and drug courts.

The “residential” nature of the detention is critical; for example, someone ordered to a residential substance-abuse treatment program as an alternative to prison would be considered a prisoner, but those ordered to a similar treatment program but residing in the community would not be considered prisoners. Similarly, probationers are not considered prisoners, even if they are required to wear monitoring devices (although OHRP recognizes there may be special circumstances in that case that must be considered). Residents of a psychiatric facility who were ordered there as an alternative to imprisonment would be prisoners, but those committed involuntarily through civil action because of the risk they pose to themselves or others, would not be prisoners.

Special protections for prisoners

Research involving prisoners as subjects is constrained in several ways:

  • The IRB review process must include a prisoner or prisoner representative. (46.304(b)). Regulations state: At least one member of the IRB must be a prisoner, or a prisoner representative with appropriate background and experience to serve in that capacity, except that where a particular research project is reviewed by more than one IRB, only one IRB need satisfy this requirement.
  • In the absence of choosing someone who is a prisoner or has been a prisoner, the IRB should choose a prisoner representative who has a close working knowledge, understanding and appreciation of prison conditions from the perspective of the prisoner.
  • Special attention is paid to the compensation, or any “advantages” being offered for participation to avoid potentially coercive incentives, such as access to special facilities, medical care, special consideration for parole, or earnings (46.305)(a)(2).
  • The selection of prisoners should be “fair to all prisoners” and, unless otherwise approved, be done on a random basis from among potential participants. Selection and special privileges have been issues, for example, in recent “faith-based” rehabilitation programs.
  • Approvable research topics is restricted to those that directly concern prisoners; these are specified as (a) the causes, effects, and processes of incarceration, (b) prisoners as incarcerated persons, or prisons “as institutional structures;” (c) those that concern “conditions particularly affecting prisoners,” such as drug addiction or sexual assaults; or (d) those concerning practices that may improve the health or well-being of the prisoners. [46.306]. The first two of these categories must be minimal risk to be approved.
  • The exemptions that otherwise could be applied to protocols, thus waiving the review process, do not apply to studies involving prisoners (see 46.101(i), footnote 1).

Federal regulations require additional safeguards for research involving prisoners because of the constraints of incarceration may affect the prisoners' ability to make a truly voluntary and uncoerced decision to participate. Most research involving prisoners requires a full board review at The College at Brockport. 

Expedited review of studies involving prisoners

Research with prisoners as subjects does not necessarily have to be reviewed by the convened Board, but can be given expedited review. However, given the additional protections and constraints on such research, it is much more likely that the full Board will be asked to consider the protocol.

Permitted research with prisoners

According to federal regulations, research can be conducted using prisoners, only if the proposed research involves the following: 

  • "A study of possible causes, effects, and processes of incarceration, and of criminal behavior"
  • "Study of prisons as institutional structures or of prisoners as incarcerated persons"
  • "Research on conditions particularly affecting prisoners as a class (for example vaccine trial and other research on hepatitis which is much more prevalent in prisons than elsewhere; and research on social and psychological problems such as alcoholism, drug addiction, and sexual assaults)" or
  • "Research on practices, both innovative and accepted, which have the intent and reasonable probability of improving the health or well-being of the subject"

Review procedures for research with prisoners
Federal regulations require that the IRB consider several additional factors beyond its normal review process when the research involves prisoners as participants. For example, the IRB must be satisfied that: 

  • "any possible advantages to the prisoner through his or her participation in the research, when compared to the general living conditions, medical care, quality of food, amenities and opportunities of earnings in the prison, are not of such a magnitude that his or her ability to weigh the risks of the research against the value of such advantages in the limited choice environment of the prison is impaired"
  • "The risks involved in the research are commensurate with the risks that would be accepted by non-prisoner volunteers"
  • "Procedures for the selection of participants within the prison are fair to all prisoners and immune from arbitrary intervention by prison authorities or other prisoners"
  • "The information is presented in language which is understandable to the subject population"
    "Adequate assurances exists that the parole boards will not take into account a prisoner's participation in the research in making decisions regarding parole" and
  • "Adequate provision has been made for any follow-up examination or care of participants, given varying lengths of prisoners' sentences"

Issues Related to Consent

In a closed institution such as a prison, there may be extraordinary interpersonal pressures that intrude into the decision regarding whether or not to participate as a subject in research. This may be particularly evident in group situations and classroom environments. Whenever possible, prisoners should be given the opportunity to reflect on the decision to participate in private.

On occasion, research will be situated in a prison classroom setting assuring the structured program segment for the day. A prisoner who elects not to participate in such research should be offered an alternative program for the time in question to minimize coercion. 

Some prisoners may feel they will lose privileges or be punished if they choose not to participate in research; others may hope for favorable treatment or early release if they do participate. Prisoners must be assured that they will be neither punished nor rewarded for their participation, and that they can discontinue their participation at any time without an institutional penalty.

Many adult prisoners are deficient readers, many have an incomplete formal education, and many speak English poorly or not at all. Researchers must use necessary measures to assure that these populations clearly understand the nature of the research and its potential risks.

Issues Related to Confidentiality

Special care should be taken to avoid requesting information in a group setting that could jeopardize the safety of individual prisoners. In the collection of research data, special care should be taken to assure that confidential materials do not come into possession of prison administrators, guards and correctional officers or other prisoners.

Prisoners are much more likely than other populations to be associated with sensitive data. This could include for example, involvement in illegal activity, and HIV/AIDS. Appropriate safeguards are necessary regarding the collection, storage and destruction of such information.

Researchers must be aware that research into certain topical areas within the institutional setting can be potentially dangerous for participants. For example, the mere act of interviewing a prisoner about sensitive topics such as gang activity, contraband, and prison prostitution may inadvertently label the participant as an informant. Great care must be taken to balance the research against protection of the prisoner as a participant.

The risk of suicide is an ever-present concern in the penal environment. The researcher must assure that debriefing is available whenever the subject is questioned about sensitive topics that could evoke self-injury once the prisoner has returned to the privacy of his or her cell.

For more information about prisoner research, see the regulations regarding prisoners at the OHRP website, as well as the OHRP Prisoner Research FAQ.

Last Updated 9/18/18

Close mobile navigation