NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on Ethical Considerations for Revisions to DHHS Regulations for Protection of Prisoners Involved in Research; Gostin LO, Vanchieri C, Pope A, editors. Ethical Considerations for Research Involving Prisoners. Washington (DC): National Academies Press (US); 2007.

Cover of Ethical Considerations for Research Involving Prisoners

Ethical Considerations for Research Involving Prisoners.

Show details

4Defining Prisoners and Correctional Settings

This chapter provides the committee’s recommendation for a new definition of the term prisoner, which considers the contexts, or “places,” relevant to research with prisoners. The goal of this definition is to expand the reach of the regulatory procedures and oversight mechanisms recommended in this report to the fuller population of individuals who are under restricted liberty and, therefore, face potentially greater risks than the general population when participating in research. It identifies the personal interests that may be violated because of research participants’ status as prisoners1 and the settings in which protections against such potential violations are required. As a point of departure, this chapter briefly reviews the ethical foundations that underpin research regulations and current regulatory language relevant to prisoner settings and notes the relevant ethical principles that led the committee to expand the definition of the term prisoner for purposes of protecting those involved in research.


The ethical foundations of research protections in the United States are based on two key ethical considerations identified by the National Com mission for the Protection of Human Subjects of Biomedical and Behavioral Research (NCPHSBBR, 1976). They are respect for persons and justice. The principle of respect for persons invokes the protection of individuals’ autonomy and personal dignity and requires that informed and voluntary consent be obtained from subjects before their involvement in research. This basic principle is often difficult to implement in a correctional setting because of the power dynamics and inherent deprivations within such a setting, especially with respect to voluntariness. Privacy and confidentiality play central roles within the principle of respect for persons as well. The principle of justice concerns the fair treatment of persons and groups. In the context of research involving prisoners, justice requires that prisoners not bear a disproportionate share of the research burden without a commensurate share of benefit, and also that prisoners have the freedom to decide questions of research participation for themselves. Justice becomes particularly important to encouraging research on a system that disproportionately affects the disadvantaged and racial and ethnic minorities (Chapter 2).

The competence and freedom of a prisoner to make a choice as well as the reality of privacy protection through confidentiality can be hampered in any of the correctional settings that restrict liberty, whether by the correctional officers or other prisoners within the prison walls or by probation officers, for example, in the community. If, for instance, researchers plan to study the effectiveness of electronic monitoring as compared with parole supervision, a system of oversight should be in place to protect the persons involved in the study—who currently are not covered under Subpart C protections. An expanded definition of prisoner is offered in this chapter. A fuller description of the ethical framework followed by the committee is found in Chapter 5.


In Subpart C of 45 C.F.R. § 46.303(c), the term prisoner is defined as:

Any individual involuntarily confined or detained in a penal institution. The term is intended to encompass individuals sentenced to 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 [emphases added].

The current regulations clearly emphasize custodial confinement as a consequence of the state’s exercise of its power via the criminal justice system. The potential impact of confinement in highly controlled institutional settings on individual autonomy is explicitly recognized in other sections of the current regulations. For example, 45 C.F.R. § 46.305 (a)(2)– (7) includes the following regarding the issue of voluntariness to ensure that the subject’s participation in the research is not coerced:

  • any possible advantages accruing to the prisoner through his or her participation in research, when compared to the general living conditions, medical care, quality of food, amenities and opportunity for earnings in the prison are not of such magnitude that his or her ability to weigh the risks of the research against the value of such advantages in a limited choice environment of the prison is impaired;
  • the risks involved in the research are commensurate with risks that would be accepted by nonprisoner volunteers; and
  • adequate assurance exists that parole boards will not take into account participation in making parole decisions, and prisoners will be informed that participation will have no effect on parole.

Institutional review boards (IRBs), however, have received guidance from the Office for Human Research Protections (OHRP) that suggests the definition may include parolees but not probationers. There is clearly confusion as to the parameters of the definition as it stands today. The important issue, as noted by the commission is that “prisoners are, as a consequence of being prisoners, more subject to coerced choice and more readily available for the imposition of burdens which others will not willingly bear” (NCPHSBRR, 1976).


Although this committee believes that research in correctional institutional settings should be subject to federal regulations, it also believes that the present emphasis on custodial detention is too narrow and results in depriving many other justice-involved individuals of human subjects protections appropriate to prisoner research participants. Several tables in Chapter 2 provide details relevant to this issue. Table 2-1 provides a broad snapshot of the number of individuals within the correctional population and notes that, as of December 2003, only 2.1 million of the 7 million total correctional population were in prisons and jails (Bureau of Justice Statistics [BJS], 2004). The rest had restricted liberties but outside the razor wire, in programs such as those listed in Table 4-1. These 4.9 million individuals (up from 1 million in 1978) require special protections when participating in research as well. Table 4-2 on pages 106–107 illustrates the vast array of incarceration options within the state of California and details a long list of alternatives to incarceration offered, including community service, electronic monitoring, and probation.

TABLE 4-1. Alternatives to Incarceration That May Be Available to Offenders.


Alternatives to Incarceration That May Be Available to Offenders.

TABLE 4-2. Descriptions of Various Criminal Justice Agencies and Facilities in California.


Descriptions of Various Criminal Justice Agencies and Facilities in California.

According to the Bureau of Justice Statistics (BJS, 2001), among state parole discharges in 1999, 43 percent were returned to prison, a statistic relatively unchanged since 1990. This high rate of return to incarceration demonstrates that significant power dynamics continue for persons who are outside the prison walls but still under some form of community correction. Thus, the element of voluntariness in the informed consent process is conceptually very similar for persons incarcerated and those under some form of disposition alternative to incarceration.

A logical system of oversight would expand the definition of the term prisoner to include parole and probation. There is little logic in providing protections, as do the current regulations, to a person detained before trial and not yet convicted of a crime, but not to a person who has been convicted of a crime and is subject to incarceration because of violations of parole or probation conditions. The expanded definition is also supported by the findings of one researcher that policing whether conditions are violated (with more drug tests, more tracking of movement, and so on) is becoming more of a priority for parole officers than promoting reintegration (Petersilia, 2000).


This section articulates the committee’s revised definition of prisoner, which places an emphasis on liberty restrictions resulting from the interactions with the criminal justice system. These restrictions include, but are not limited to, custodial confinement. The aim is to expand the reach of regulations to protect prisoners and others with restricted liberty.

Such an individual may be ordered to reside in settings in which freedom of movement is restricted (e.g., precinct holding pen, jail, prison, halfway house, or prerelease center) or in the community under constraints ordered by the criminal justice system (e.g., probation, parole or house arrest, or drug court sentence).

Recommendation 4.1 Redefine the term prisoners to expand the reach of human subjects protections. The Department of Health and Human Services and other relevant agencies that write, implement, or enforce regulations pertaining to research with prisoners should expand the definition of the term prisoner to include all settings, whether a correctional institution or a community setting, in which a person’s liberty is restricted by the criminal justice system.

It is not custodial confinement alone that creates the potential for coercion and threatens an individual’s right to autonomous decision mak ing (consent or refusal to participate in research). Rather, it is restrictions of freedom or the imposition of other sanctions by an agent of the criminal justice system who observes, scrutinizes, supervises or monitors, and ultimately determines the imposition of punishments for an individual’s behavior. Persons on parole and others subject to limits on liberty or privacy are particularly vulnerable to the risk of reinstitutionalization based on decisions made by parole officers and other personnel of the criminal justice system. This threat to independence creates the potential for coercion and requires that prisoners be afforded additional levels of research protection as well. Note that persons who become prisoners while in a nonprisoner study must also be afforded special protections on entering a correctional setting.

A prisoner, for the purposes of this report, is any person whose liberty is restricted as a result of the interaction with the criminal justice system. Although the limitations on personal choice and control are perhaps most evident and oppressive in locked detention facilities (e.g., jails and prisons), the power differential between criminal justice agents and prisoners exists in many other contexts as well; the differences are a matter of degree. Individuals involved in a wide variety of community-based criminal justice programs, ranging from probation and parole to pre- or postadjudication diversions such as drug courts or mental health courts, are subject to coercion by the array of agents (e.g., parole or probation officers and diversion program counselors) who monitor the individuals’ compliance with program requirements; such agents may also invoke further sanctions for program failure or noncompliance. Thus, one can fall within the protections recommended by this committee by virtue of being ordered by the criminal justice system to reside in a confined setting or, if living in the community, by virtue of the restrictions on individual decision making imposed by some part of the criminal justice system. To be a prisoner, there must be some nexus or connection between the setting or restricted liberty of the person and the action of the criminal justice system. For all individuals who meet this definition, regulations that govern research with prisoners should be applied. The current statutory or constitutional distinctions between the civil and criminal processes, which are evolving, are not sufficiently clear to allow an easy determination of whether the proposed systems of protection should apply. For instance, juveniles confined under orders of family court (or the equivalent, such as a juvenile court) likely require additional protections, but an analysis of their needs is beyond the scope of this report. If, however, they are transferred from the original jurisdiction of the family court (or the equivalent, such as a juvenile court) to the jurisdiction of a state or federal criminal court, then they would fall under this definition. Other types of confinement on the civil/criminal frontier that our definition encompasses include commitment by reason of an acquittal on the grounds of insanity and commitment as a sexual offender (e.g., so-called criminal sexual psychopaths or violent sexual predators) under various state laws. However, those persons whose confinement is the result of other civil proceedings, such as those designed to protect a mentally ill person from harming self or others, would not be covered by our system of protections. There are substantial ethical issues involved in conducting research with these populations that go beyond the scope of this report. Parallel studies, such as the ones undertaken by this committee, may be needed to explore ethical issues of research involving these groups.


In the interest of clarity, the committee specifies an array of settings in which regulations governing research with prisoners should apply, as well as a brief listing of settings for which the recommended regulations are not considered necessary. Some cases may arise that are not specifically addressed, and these must be decided on a case-by-case basis. Further discussion of research gradation and level of risk, which determine the study design and monitoring safeguards required, is provided in Chapter 6.

When Proposed Regulations Should Apply

Settings and situations in which proposed regulations governing research with prisoners should apply are as follows:

  • state or federal prisons (including, e.g., camps, farms, boot camps);
  • any jail or detention facility (e.g., municipal, city, county, federal);
  • any community-based criminal justice supervision program such as bail or bond supervision, parole, or probation; community correctional reentry centers; work furlough programs; halfway houses; or similar programs;
  • any community-based alternative disposition program, including (but not limited to) restitution programs, community service programs, or participation in other activities deemed to constitute punishment or other mandatory activities resulting from being sentenced;
  • inpatient or outpatient psychiatric treatment settings if an individual is involuntarily committed as a result of a finding of not guilty by reason of insanity;
  • treatment settings for any person who, having served a sentence for a sex offense, is subsequently deemed to meet criteria for involuntary commitment under various criminal sexual psychopath or violent sexual predator (or equivalent) statutes in a civil proceeding;
  • placement or participation in a community residential drug, alcohol, or mental health treatment program, day program, or partial-day program if mandated by a criminal court as part of an order for conditional release or sentencing; and
  • participation in a drug court, mental health court, or other criminal justice specialty court that functions to divert arrested or convicted individuals into substance abuse or mental health service programs.

When Proposed Regulations Should Not Apply

Settings and situation in which proposed regulations governing research with prisoners would not apply are as follows:

  • individuals committed to involuntary inpatient psychiatric or substance abuse treatment, or to any form of mandated community treatment, by the order of a juvenile, civil, or probate court (as stated earlier in this chapter, juveniles likely need additional protections, but this committee was not equipped to address their specific needs);
  • individuals living in a noncustodial community setting who meet the above definition of prisoner but whose status as a prisoner is not relevant or related to their enrollment in a particular community-based research project. This means that the criminal justice agent or agency having supervisory jurisdiction over the individual was not involved in the plans for study enrollment, and the study itself does not include prisoner status as a criterion for participation. This exclusion permits prisoners living in the community to enroll in research that is open to any citizen in the community (e.g., hospital or medical school-based clinical trials, survey studies) without imposing the restrictions of the proposed regulations on those research entities.


If the subject’s liberty status changes through arrest or revocation of probation or parole, and the person is then confined in a custodial setting, do the provisions of Subpart C become applicable to the subject? The committee’s answer was yes, regardless of the nature of the research. Upon entering a custodial setting, that person becomes formally and in every way a prisoner, subject to the same constraints and concerns that the committee has expressed in this report, and therefore needs the same safeguards. There is no ethical justification for providing fewer safeguards for a new entrant into prison than those already in prison.

This does not mean, however, that study participation would be automatically terminated. It means that continued participation requires a new review of that prisoner’s participation in the study. The original IRB should review the impact of the correctional setting on the procedures, with input from the IRB affiliated with the correctional setting. It should be done in an expedient manner (e.g., within 30 to 45 days) to allow for continuation of study participation.

Continuity of care is an important issue, especially for treatment studies. If terminating prisoner participation would adversely affect the health of the subject, participation may continue until the IRB review takes place. The custodial official receiving the prisoner should be informed that the prisoner is enrolled in a research protocol, provided information on the protocol, and explained the potential risks of not allowing continued participation. The researcher would have an obligation to advocate for providing the appropriate care while seeking to comply with regulations applicable to that setting.

Certain social or behavioral studies that, for example, examine health and risk behaviors over several years, may see some participants move into correctional settings. If the researchers wish to continue study involvement for those individuals, review would be necessary to weigh the risks and benefits within the new setting. If the risks remain low, continuation may be approvable. However, if new risks are foreseen, but they are still low, the consent may need to be updated for the participant’s continued involvement.

This need for review may not be an issue for many jailed detainees who are often incarcerated for just a few days if that short period does not affect participation in the protocol. In a related, but different situation, if a prisoner is participating in a study within a correctional setting and is transitorily confined to disciplinary segregation, the researcher will have to evaluate whether or not the potential interruption in access during the period of segregation would preclude continued participation.

Research studies lose participants for many reasons. There may be occasions in which participation in the study may not continue for the research participant who becomes incarcerated. For example:

  • The IRB may determine that the risks outweigh potential benefits to remaining in the study.
  • By statute, some states do not permit biomedical/clinical trials research in prisons.
  • By Department of Corrections policy, some states do not permit biomedical/clinical trials research in prisons.


  1. BJS (Bureau of Justice Statistics) Trends in State Parole, 2000. 2001. [accessed December 23, 2005]. [Online]. Available:
  2. BJS. Probation and Parole in the United States, 2003. 2004. [accessed December 23, 2005]. [Online]. Available:
  3. John Howard Society of Alberta. Community Corrections. 1998. [accessed June 27, 2005]. [Online]. Available:
  4. NCPHSBBR (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research) Research Involving Prisoners. Washington, DC: U.S. Department of Health, Education, and Welfare; 1976. DHEW Pub. No. (OS) 76-131.
  5. Petersilia J. Challenges to prisoner reentry and parole in California. California Policy Research Center Brief Series. 2000. [accessed April 4, 2006]. [Online]. Available:



Using the place-centric term prisoner to define individuals also found outside of the typical prison setting may be confusing to some readers. This definition is aimed, however, at a systemic approach to oversight of research involving those subject to restricted liberty through the criminal justice system.

Copyright © 2007, National Academy of Sciences.
Bookshelf ID: NBK19878


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.1M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...