Overview of the problem: Suicide and self-harm in correctional facilities

Suicide is the leading cause of death in local jails, accounting for over one-third of jail deaths in 2013; it is less frequent in prison settings but still accounted for nearly 6 percent of the deaths of people in state custody in 2013.9 Although suicide rates in jail declined steadily from 129 deaths per 100,000 people in 1983 to 47 deaths per 100,000 in 2002, the rate of suicide increased by 12 percent between 2008 and 2013 and the suicide mortality rate is three times higher in jail populations than in either prison populations or in the community.10 There are also clear demographic trends in jail suicides: men incarcerated in jail are 50 percent more likely to commit suicide than women and incarcerated people who are white are far more likely to commit suicide than those who are black or Latino (six and three times more likely, respectively).11       

            Self-harm in correctional settings is a less well-understood phenomenon but nonetheless a serious health concern. Historically, research on its occurrence has been hindered by a lack of consistent terminology as well as ambiguity in defining what constitutes self-harm or self-injury. (See “Defining suicide and self-harm.”) No comprehensive national data exist on self-harm in correctional facilities, but recent research indicates that up to 15 percent of adults and up to 24 percent of young people engage in non-suicidal self-injury while in custody (defined as “deliberate, self-inflicted tissue damage without intent to die”); rates are even higher when the person has a mental health disorder (up to 61 percent).12 Surveillance data from the New York City jail system are also revealing; there were 2,514 acts of self-injury from 2007 to 2011, with the annual rate of self-injury increasing significantly during this time period.13

Defining suicide and self-harm

Although the definition of suicide is generally considered straightforward (defined by the Centers for Disease Control and Prevention as “death caused by self-directed injurious behavior with an intent to die as a result of the behavior”), appropriate definitions for various acts of self-harm are debated.a The term “self-harm” is commonly used to define a wide range of behaviors in which people deliberately inflict physical harm on themselves. Terminology for this behavior varies and may include terms such as “attempted suicide,” “deliberate self-harm,” “parasuicide,” “self-injury,” and “self-mutilation.”b  

One particularly difficult question to resolve is whether and how to include a person’s motivation or “intent” in definitions of self-harm. While some believe that definitions of self-harm should include those acts where someone has intent to kill oneself, others argue that the definition should only include acts undertaken without the intent to die.c Still others argue that self-harm can be defined without reference to the presence or absence of intent since intent itself is so difficult to assess and since intention may not be relevant to the physical damage that occurs.d For this report, Vera employs a broad definition of self-harm that includes all acts leading to direct and deliberate harm to oneself with or without intent to die. The terms self-harm and self-injury are used interchangeably and the inclusion or exclusion of intent is noted where appropriate. 

A full review of how suicide and self-harm may differ in terms of intent, function, and prevalence is beyond the scope of this report; however, recent literature suggests there are important differences in the people who engage in self-harm versus those who attempt or complete suicide while in custody. For example, rates of suicide and self-harm vary by gender, with men more likely to commit suicide in jail but with documented rates of self-harm significantly higher among women.e  Other research suggests that whereas most people who engage in suicide are engaging in an act that is intended to be life-ending, some people who engage in self-harm are employing the behavior as a coping mechanism to relieve psychological distress and are not intending to end their lives.f A better understanding of the differences between suicide and self-harm is thus important for practitioners, both because self-harm can be a risk factor for suicide and because these distinctions affect how a person should be assessed and treated.

            The reasons for elevated rates of suicide and self-harm in correctional facilities are myriad, ranging from the characteristics of the population, to the experience of incarceration, to the common features of the environment. Many people enter the correctional system with multiple risk factors for engaging in suicide or self-harming behavior, including having a serious mental illness and/or substance use problems, a history of trauma, and a history of self-harm, suicide attempts, and recent suicidal ideation.14 These individual risk factors, combined with environmental risk factors, such as the stress of the correctional environment and the trauma of arrest, place detained people at a particularly high risk for suicide and self-harm.

            In addition to the direct harm caused by these behaviors, staff, or other people who witness suicidal or self-harming acts, are at serious risk for experiencing psychological repercussions, such as post-traumatic stress disorder.15 They are also likely to experience burnout and apathy over time.16 Even having relatively small numbers of people who self-harm can present substantial challenges to institutions and the people who both work and are detained there, since self-injurious behaviors consume significant institutional resources through disrupted routines, higher staffing levels, and security risks.17 

Traditional responses to suicide and self-harm in correctional facilities

Best practices for suicide prevention and response for correctional systems do exist and a growing body of research establishes the essential components of a reasonable suicide prevention program in jails and prisons.18 These include initial and annual staff trainings, intake and on-going assessment, communication procedures, and housing that includes architectural and environmental safeguards—for example, buildings free of protrusions and designed to ensure the incarcerated person is maximally visible in any location, procedures for emergency response, appropriate mental health care, and multidisciplinary mortality reviews.19 Even so, prevention and review processes for incidents of suicide and self-harm in many correctional facilities lag in three specific ways, detailed below.

Lack of review processes

When a death in custody occurs, the National Commission on Correctional Health Care (NCCHC) has standards that stipulate a three-pronged review:

  • an administrative review (an assessment of the correctional and emergency response actions);
  • a clinical mortality review (an assessment of the clinical care provided and the circumstances leading up to a death); and
  • a psychological autopsy if the death is by suicide (a written reconstruction of the person’s life emphasizing factors that may have contributed to his or her death).

The goal of such a review is “to determine the appropriateness of clinical care; to ascertain whether changes to policies, procedures, or practices are warranted; and to identify issues that require further study."20 These mortality reviews share many features of sentinel event reviews but are limited in scope to incidents of death; a sentinel events framework calls for a broader scope of review for negative outcomes and could include serious, non-fatal incidents.

            However, even with standards for review in place, it is clear that many correctional facilities do not follow them. A national survey of jail suicide, conducted in 2005 and 2006 by the National Institute of Correction (NIC), found that the majority (63 percent) of jails did not conduct a mortality review following a jail suicide.21 Although no comparable research exists on the prevalence of reviews for incidents of self-harm, it is reasonable to assume that the majority of correctional facilities also do not regularly conduct formal reviews when such incidents occur. The lack of a system-wide, institutionalized response to these events inhibits an honest assessment of the “error” that, in turn, forecloses opportunities for staff and correctional leaders to learn from mistakes and prevent future incidents of suicide and self-harm.

Emphasis on individual versus system responsibility

The criminal justice system has traditionally employed a “retrospective, adversarial inspection model of quality control” characterized by assigning individual responsibility for errors.22 Such an approach may prove effective at providing isolated fixes to specific problems (i.e., eliminating “bad apples”), but it fails to account for the sometimes multiple and complex root causes of error that are related to underlying system problems and it is divorced from an ideal of continuous quality improvement—the idea that process-based and data-driven approaches can improve the quality of a process or service. A robust response to reducing the problem of suicide and self-harm in correctional facilities thus requires a bolder approach—one that encourages the reporting of incidents, prioritizes the identification of systemic causal factors, and looks to develop the capacity for forward-looking and shared accountability.

Inadequate staff training

Most correctional facilities have fairly minimal staff training on suicide and self-harm prevention. The NIC survey found that 62 percent of jails provide suicide prevention training to at least 90 percent of their correctional staff. However, the majority of these trainings were two hours or less and only 75 percent of those who held trainings did so annually.23 Furthermore, it is not clear whether correctional facilities have training on non-lethal self-harm, which is essential if correctional administrators are to address the important differences in the underlying mechanisms and motivations for suicide and self-harm (for example, the differences in the issue of “intent” described in “Defining suicide and self-harm” above).24 One study found that many prisons use suicide protocols to respond to acts of self-injury, a response that may be inappropriate if the differences in the motivations and risk factors for suicide and self-harm are not addressed in these protocols.25