Suicide Prevention in Correctional Facilities: Reflections and Next Steps

Suicide Prevention in Correctional Facilities: Reflections and Next Steps[1]


Lindsay M. Hayes


            Data from a recent national study of inmate suicides indicates that the suicide rate in county jails throughout the United States has steadily decreased. Despite this progress, the author argues that rather than developing and maintaining comprehensive policies and practices, policymakers and correctional administrators appear preoccupied with the notion that suicides can only be prevented when inmates are on suicide precautions. Measures such as closed-circuit television monitoring, suicide-resistant jail cells, safety smocks, and new technology are popular tools to keep certain inmates safe. There is more to suicide prevention than simply observing suicidal inmates and waiting for them to attempt suicide. The author argues that suicides are prevented and suicide rates reduced when correctional facilities provide a comprehensive array of programming that identifies suicidal inmates who are otherwise difficult to identify, ensures their safety on suicide precautions, and provides a continuity of care throughout confinement.


1) Introduction

A recent national study of inmate suicide found that suicides were evenly distributed from the first few days of confinement to over several months of confinement, many suicides occurred during waking hours, most inmates were not under the influence of drugs and/or alcohol at the time of death and many suicides occurred in close proximity to a court hearing (Hayes 2012). Perhaps the most significant finding was that the suicide rate in detention facilities throughout United States has been substantially reduced during the past 20 years, dropping from 107 county jails suicides per 100,000 inmates in 1986 to 38 suicides per 100,000 inmates in 2006 (Hayes 1989; 2012). There may be several explanations for this reduced suicide rate, including the fact that national studies of jail suicide conducted over this time period gave a face to this long-standing and often ignored public health issue, recurring research has been incorporated into suicide prevention training curricula, increased awareness about the problem of suicide among jail inmates is now reflected in national correctional standards that advocate comprehensive suicide prevention programming, and inmate suicide litigation has persuaded (or forced) counties and facility administrators to take corrective action in reducing the opportunity for future deaths. With a reduced suicide rate, the antiquated mindset that “inmate suicides cannot be prevented” is no longer true in many cases.

Despite this progress, policymakers and correctional officials, supported by medical and mental health administrators, continue to take the tunnel view that “suicide prevention” is simply the safe management of inmates while they are on suicide precautions. A review of many suicide prevention policies will find a disproportionate amount of narrative regarding the conditions of a suicide precautions, i.e., level observation, prohibited clothing items and other possessions, etc., and little description of other essential components to an effective program.

More times than not, correctional, medical, and mental health personnel do a fine job of safely managing inmates identified as suicidal and placed on precautions. After all, few inmates successfully commit suicide while on suicide precautions (Hayes 2012). When they do, the likely cause(s) were required cell checks not performed, the inmate was not housed in a “suicide-resistant” cell, and/or the level of observation was not commensurate with the level of suicide risk (i.e., an inmate at high risk for suicide placed on a lower level of observation).

The correctional field has long been obsessed with trying to thwart suicide attempts and manage suicidal inmates with technology and short-sighted responses. Back in 1986, I received correspondence from a police officer who fancied himself as the inventor of a system of placing a series of sensory strips on the floor and bed of the jail cell. The system operated under the principle of “weight off” in which an inmate confined to their cell, but not laying on their bunk or standing on the floor, would presumably be hanging from a ligature off the floor. With the weight off the floor, the sensory strips would trigger an alarm in the main control station of the jail. Although this young inventor obtained a patent,[2] his discovery literally never got off the ground presumably because many inmates were found to commit suicide by hanging in either the standing or sitting position on the floor (Hayes 1989; 2012). More recently, I received correspondence from a research professor who was looking to patent a device that an inmate would wear as an earpiece on suicide precautions to monitor their pulse and oxygen level. If the inmate’s vital signs were detected as being outside the normal range, an alarm would presumably go off and an emergency response would be called. Of course, if the inmate simply removed the earpiece, an alarm would presumably go off and an emergency response would also be called. Use of this pulse oximetry technology has been found in a psychiatric setting for restrained patients (Master & Wandless 2005).

Safety smocks and blankets, made of heavy nylon fabric that is very heavy and difficult to tear, have become standard issue for suicidal inmates in correctional facilities throughout the country. The mental health director of a large county jail once called to ask who sold the best safety smock on the market? A simple perusal of the Internet would find numerous vendors hawking “anti-suicide” products for the correctional field (e.g., a safety smock with breakaway sleeves “with the durability and interior softness….provides the most comfortable safety option for individuals in crisis,”[3] although most inmates clothed in these degrading and humiliating garments would disagree). A particular manufacturer once claimed to be in the final design stage of a line of anti-suicide underwear.[4]

Consistent with national correctional standards, inmates on suicide precautions are now required to be housed in “suicide-resistant” cells which contain tamper-proof light fixtures, smoke detectors, sprinkler heads, and ceiling/wall air vents that are protrusion-free. Fiberglass-molded bunks in these cells have rounded edges and no tie-off points. Clothing hooks are now collapsible and towel racks, sinks, radiator vents have been modified to reduce their use as anchoring devices for hanging (Atlas 1989). Corded telephones, an obvious suicide hazard, have been replaced with cordless telephones in many jurisdictions (Hayes 2003; Quinton & Dolinak 2003).

Used predominantly in jail and prison facilities that choose not to provide a constant observation option for inmates at high risk for suicide, closed-circuit television (CCTV) has become a popular, although deadly form of inmate supervision (i.e., the technology generally does not prevent a suicide unless staff are observing the monitor; it only records be suicide attempt in progress) (Hayes 2006). Similarly, the use of inmate companions to observe other inmates on suicide precautions has also become popular in some jurisdictions throughout the country struggling with overtime budgets, although national correctional standards advocate that their use should only be a supplement to, and not a substitute for, correctional officer monitoring (National Commission on Correctional Health Care 2008). In addition, mental health clinicians often develop contracts with suicidal inmates, seeking assurances that their patients will not engage in self-injurious behavior as a condition of discharge from suicide precautions. Correctional agencies might, in turn, request that each incoming inmate sign a standard letter as an apparent shield against liability. Of course, although there may be many positive therapeutic aspects to no-harm contracts, most experts agree that once an inmate comes acutely suicidal the written or verbal assurances are no longer sufficient to counter suicidal impulses (Garvey et al 2009).

Similar to the argument that use of CCTV or inmate companions can alleviate correctional staff responsibilities for suicide precautions, a research arm (National Institute of Justice) of the U.S. Department of Justice is current funding an evaluation of equipment that can measure an inmate’s heart and breathing rate, and body motions while on suicide precautions. A wall-mounted range controlled radar system, originally designed for home security motion detectors, measures subtle motions on the body’s surface caused by heart and lung activity. Alarms are activated when the system detects suspicious changes in heart rate, breathing rate or body motion that are typically found when an inmate is engaging in a suicide attempt. In a program bulletin, the National Institute of Justice acknowledged a declining inmate suicide rate, but complained that “it remains a troubling problem and traditional suicide watch requires dedicated staffing, taking officers away from other duties,”[5] suggesting that there are other more important duties than keeping inmates safe from themselves.

Suicide-resistant architecture and other environmental safeguards are critically important to ensuring the safety of individuals housed in correctional facilities and other settings (see, for example, Watts et al 2012). However, what inmate companions, CCTV, contracting for safety, range controlled radar systems, pulse oximetry, and anti-suicide products all have in common is the further separation of correctional, medical, and mental health personnel from the inmate that has already been identified as suicidal. These quick-fix responses also have little to do with the most important aspects of suicide prevention: how we identify the suicidal inmate that is not easily identifiable and, when found, how we subsequently know when they are no longer suicidal and can be safely discharged from suicide precautions.

When an inmate self-reports suicidal ideation, the system easily responds appropriately: they are placed on suicide precautions and the anti-suicide products kick in. What we continue to struggle with is the ability to prevent the suicide of an inmate who is not on suicide precautions. These are inmates that might not be easily identifiable as being at risk for self-harm. These are inmates that emphatically deny they are suicidal, they may even contract for safety, but their actions and history suggest otherwise. These are inmates with recent past histories of self-injurious behavior and poor coping skills who find themselves locked down in segregation or who recently received bad news during a court hearing, telephone call, or family/legal visit. These are inmates who are not on suicide precautions, but should be.

Kay Redfield Jamison, a prominent psychologist and author, best articulated the point by stating that if “suicidal patients were able or willing to articulate the severity of their suicidal thoughts and plans, little risk would exist” (Jamison 1999, at page 150). With this in mind, several guidelines for better identification and management of suicidal inmates are offered.

2) Assessment of Suicide Risk as an On-Going Process

The face of jail suicide victims has changed considerably during the past 20 years, and such change is no more apparent than in the time in custody before suicide. Recent research found that less than a quarter of all inmates who committed suicide were dead within the first 24 hours of confinement, and half were dead between two days if four months of confinement (Hayes 2012). This finding is contrasted with earlier research that found more than 50% of jail suicide victims were dead within the first 24 hours (Hayes 1989). The availability of better screening to identify suicide risk during the initial booking process, coupled with increased staff awareness and emphasize on the first few hours of confinement as a high risk period for suicide was probably responsible for this changing pattern.

As such, the assessment of suicide risk should not be viewed as a single opportunity at intake, but as an on-going process. Because an inmate may become suicidal at any point during confinement, suicide prevention should begin at the point of arrest or transfer to the correctional facility and continue until the inmate is released. We should be creating more opportunities to gather information, as well as periodically assess inmates at risk. So, for example, there should be a formalized process by which intake staff ask arresting or transporting officers whether the newly arrived inmate is at risk for suicide, as well as a determination as to whether the inmate had been on suicide precautions during a previous confinement in the facility. Once an inmate has been successfully managed on, and discharged from, suicide precautions, they should remain on a mental health caseload and assessed periodically until release pursuant to a thoughtful treatment plan.

Screening for suicide risk during the initial booking and intake process should be viewed as something similar to taking one’s temperature – it can identify a current fever, but not a future cold. The shelf life of current behavior that is observed and/or self-reported during intake screening is time-limited, and we often place far too much weight upon this initial data collection stage. Following an inmate suicide, it is not unusual for the mortality review process to focus exclusively upon whether the victim threatened suicide during the initial intake stage, a time period that could be far removed from the date of suicide. If the victim had answered in the negative to suicide risk during intake, there is often a sense of relief expressed by participants of the mortality review process, as well as a misguided conclusion that the death was not preventable. Although the intake screening form remains a valuable front-end prevention tool, the more important determination of suicide risk is the current behavior expressed and/or displayed by the inmate during their confinement. Most suicide prevention policies are heavy on explaining the intake screening process, but light on most of the other critical areas of identification.

In addition to early stages of confinement, many suicides occur in close proximity to a court hearing. We must begin to devise ways to be more attentive to this risk period. In one county jail, inmates who are on the mental health caseload and/or have a prior history of suicidal behavior, receive a brief mental health status screening after each court hearing (Hayes 2005). In another, inmates arrested for murder, domestic violence, or child molestation receive similar scrutiny. Some jurisdictions add a secondary layer of assessment for inmates charged in highly publicized cases.

A disproportionate number of inmate suicides take place in “special housing units” (e.g., disciplinary/administrative segregation) of the facility (Patterson & Hughes 2008; Way et al 2007). One effective prevention strategy is to create more interaction between inmates and correctional, medical and mental health personnel in these housing areas by: 1) increasing rounds of medical and/or mental health staff; 2) requiring regular follow-up of all inmates released from suicide precautions; 3) increasing rounds of correctional staff; 4) providing additional mental health screening to inmates admitted to disciplinary/administrative segregation; and 5) avoiding lockdown due to staff shortages (and the resulting limited access of medical and mental health personnel to the units).

Finally, other commentators have spoken about “clean” suicides, i.e., the suicides of inmates with no prior psychiatric history that develop the onset of mental illness during confinement (Hanson 2010). Few, if any, correctional policies address the identification of mental illness and suicide risk during confinement, but the need to.

3) Suicide Risk Despite Denial

Why would a suicidal inmate deny that they are suicidal? The most common reason might be they are committed to ending their life and do not want to be stopped. For others, however, they might be unable or unwilling to articulate their thoughts, or the lack of privacy offered when the questions are asked, or the manner in which the questions are asked, or fear of being ostracized by other inmates, or the perceived punitive aspects of suicide precautions.

We should not rely exclusively on the direct statements of an inmate who denies that they are suicidal and/or have a prior history of suicidal behavior, particularly when their behavior, actions and/or history suggest otherwise. Often, despite an inmate’s denial of suicidal ideation, their behavior, actions, and/or history speak louder than their words. Take, for example, the inmate who is on suicide precautions for attempting suicide the previous day. He is now naked in a cell with only a suicide smock, given finger foods, and on lockdown status. The mental health clinician approaches the cell and asks the inmate through the food slot (within hearing distance of others on the cellblock): “How are you feeling today? Still feeling suicidal? Can you contract for safety?” Will this inmate’s response be influenced by their current predicament? How would any of us respond?

It is not all that surprising that some preventable deaths often escape our detection. The booking area of any jail is traditionally both chaotic and noisy; an environment where staff feel pressure to process a high number of arrestees in a short period of time. Two key ingredients for identifying suicidal behavior – time and privacy – are at a minimum. The ability to carefully assess the potential for suicide by asking the inmate a series of questions, interpreting their responses (including gauging the truthfulness of their denial of suicide risk), and observing their behavior is greatly compromised by an impersonal environment that lends itself to something quite the opposite. As a result, the clearly suicidal behavior of many arrestees, as well as circumstances that may lend themselves to potential self-injury, are lost.

In another example, a suicidal inmate sent to the local hospital for an assessment may appear to be stable in front of an emergency room physician, even deny suicide risk, only to be discharged from hospital and returned to jail where they again revert to the same self-injurious behavior that prompted the initial referral. Given such a scenario, healthcare and correctional staff should not assume that the hospital was cognizant or even appreciative of this cyclical behavior. On the contrary, regardless of what the hospital clinician might have observed and/or recommended, as well as the inmate’s denial of risk, whenever healthcare and correctional staff hear an inmate verbalize a desire or intent to commit suicide, observe an inmate engaging in suicidal behavior or otherwise believe an inmate is at risk for suicide, they should take immediate steps to ensure that the inmate’s safety.

4) Meaningful Suicide Prevention Training

We must provide useful initial and annual suicide prevention training to all staff. While implementing suicide precautions for an inmate that verbally threatens suicide requires little training, identifying suicidal behavior of inmates unwilling and/or unable to articulate their feelings, or who deny suicidal ideation, requires both initial and annual training. Simply stated, correctional staff, as well as medical and mental health personnel, cannot detect, make an assessment, nor prevent a suicide for which they have little, if any, useful training.

All suicide prevention training must be meaningful, i.e., timely, long-lasting information that is reflective of our current knowledge base of the problem. Training should not be scheduled to simply comply with an accreditation standard. A workshop that is limited to an antiquated videotape or DVD, or the reciting of current policies and procedures, might demonstrate compliance (albeit wrongly) with an accreditation standard, but is not meaningful, nor helpful, to the goal of reducing inmate suicides. In addition, although webinar-based and/or e-learning question-answer formatted training have become popular cost-effective alternatives to traditional classroom training, such technology should be discouraged in this area. The topic of suicide prevention is one that is best provided in a live, interactive environment amongst correctional, mental health, and medical personnel. Suicide prevention is all about collaboration, and training that is reduced to an individual sitting alone and watching a DVD or webinar-based workshop or e-learning instruction on a desktop screen has questionable value. Without regular suicide prevention training, staff often make wrong and/or ill-informed decisions, demonstrate inaction, become complacent, or react contrary to standard correctional practice, thereby incurring unnecessary liability.

In reviewing a litigation case recently, I came across this rather interesting deposition testimony. When asked if he was ever provided with training on how to properly conduct intake screening, the un-trained booking officer replied that “I wasn’t trained as far as what specific questions to ask them.” When questioned whether he would ask an inmate if they were thinking about suicide, the officer replied, “I would not personally ….You don’t want to turn around to an inmate and say, ‘Do you feel like you’re suicidal’ and give him an idea. You have to, like, judge that person when they come in…. if he’s not suicidal and he’s just depressed, I’m not going to say, ‘are you suicidal,’ and give them the idea to kill himself….I don’t push the issue for the simple reason I don’t want to push it in his mind to do it.”

In another case, an untrained officer was asked how he would monitor an inmate for signs of depression (as recommended by the jail nurse) and he responded: “Depression is not a word that means much. You’re in jail, you should be a little depressed. That’s not an uncommon emotion. Okay, it’s not something we get really excited about.” In addition, when asked during the deposition whether he ever reviewed an inmate’s prior jail record to see if they had previously been identified at risk of suicide, the officer responded: “There’s no reason to be looking at the previous record….Should he always be treated as a suicidal person for the rest of his life? No.” These two officers might have reacted differently had they received appropriate training.

5) One Size Does Not Fit All

One size does not fit all and basic decisions regarding the management of a suicidal inmate should be based upon their individual clinical needs, not simply on the resources that are said to be available. For example, if an acutely suicidal inmate requires continuous, uninterrupted observation from staff, they should not be monitored only by CCTV simply because that is the option jail officials choose to offer. A clinician should never feel pressured, however subtle that pressure may be, to downgrade and/or discharge an inmate from suicide precautions because additional staff resources (such as overtime, post transfer, etc.) are required to maintain the desired level of observation. Although they would rarely admit it, clinicians have prematurely downgraded, discharged, and/or changed the management plan for a suicidal inmate based upon pressure from facility officials.

I was conducting an assessment of a county jail a few years ago and, while interviewing a mental health clinician, the telephone rang. It was the head nurse. Apparently a female detainee had just arrived into the clinic from booking. During booking, the intake nurse had scored the inmate high on a suicide risk screening form (for loss of relationship, psychiatric history, drug history, displaying signs of depression, anger, incoherence, and inability to focus). Now the head nurse was calling to ask the mental health clinician to assess and basically clear the detainee from the clinic. I followed the clinician to the clinic and came upon a female detainee sitting in a chair surrounded by the head nurse and several officers. The detainee was barely conscious, appeared incoherent, and should not even have received medical clearance into the facility without a thorough examination. In any event, the clinician tried to talk to the detainee, but it was pointless. She could not respond to any questions and had to be held up from falling off the chair. The clinician clearly could not conduct the assessment and told the head nurse that the detainee would need to be placed on suicide precautions until such time as she could be interviewed. Given her condition, the detainee would also need to be housed alone and, given the jail’s policy of requiring constant observation for suicidal inmates who were housed alone, an officer would need to be reassigned to conduct the supervision.

The clinician and I returned to her office. My interview continued until there was a knock on the door. It was the shift supervisor who wondered aloud how long it would take to assess the detainee. In other words, how long would an officer need to be assigned to provide constant observation? The clinician calmly responded that the assessment could not be conducted until the detainee became coherent and could understand the screening questions. The supervisor thought about it for a moment, glanced at me, then departed. The interview continued again for a few minutes until the telephone rang. This time it was the jail commander. I only heard half of the conversation, but it seemed to be of similar content to that of the shift supervisor. Again the clinician responded politely that the assessment would occur only when the detainee was coherent enough to understand the questions. The telephone conversation ended and my interview continued. Well, not really. Another knock on the door. This time it was the head nurse with some evidence. It seemed that, in reviewing the intake screening form, the head nurse noticed that one of the “no” answers had been scratched out and replaced with a “yes” response. Perhaps the intake nurse had made a mistake and the screening form score was really a “7,” not an “8.” If so, that would drop the inmate out of the high risk category, thus eliminating the need for constant observation. The clinician remained calm and told the head nurse that, “No, the scoring on the form would remain the same.” The inmate would remain on constant observation and hopefully be assessed the following day. The nurse reluctantly left and my interview continued without any further interruptions.

A subsequent review of the detainee’s prior records revealed that she had been in the jail approximately six months earlier and placed on suicide precautions following a suicide attempt. The clinician I interviewed had worked at the facility for approximately five years and was obviously experienced and not easily intimidated. There was one other mental health clinician at the facility. He had only started working there the previous week. The outcome of this case might have been different had an inexperienced clinician been confronted by the head nurse and supervisory jail staff.

6) Only Qualified Mental Health Professional Should Determine Non-Suicide Risk

By far the most difficult decision in the area of suicide precaution is the determination that an inmate is no longer suicidal and can be discharged from suicide precautions. That decision must always be made by a qualified mental health professional (QMHP) following a comprehensive suicide risk assessment. Why would we want such critical decisions to be made by anyone other than a licensed, masters-level or above clinician? According to national standards, a QMHP would include a psychiatrist, psychologist, psychiatric social worker, psychiatric nurse, and others by virtue of their education, credentials, and experience are permitted by law to evaluate and care for the mental health needs of patients (NCCHC 2008). These decisions must be respected by non-QMHP staff. Decisions by non-QMHPs that result in bad outcomes incur unnecessary liability.

Take, for example, the case of 27-year-old Phillip Hall.[6] He was arrested by local police for automobile theft and eluding police following a high-speed chase. During his subsequent arrest at a hotel, officers noticed he had self-inflicted lacerations on his left wrist. Mr. Hall was then transported to the emergency room of a local hospital for treatment of his self-inflicted injury and assessment as to whether his behavior warranted emergency psychiatric commitment. During a nursing assessment, Mr. Hall expressed continuing suicidal ideation, and complained of a recent breakup with his girlfriend, financial and legal problems (“patient is under arrest and broke”), appetite (“sometimes I don’t eat”) and sleep (“unable to sleep”) disturbances, and self-reported experiencing both bi-polar and post-traumatic stress disorders. In addition, the assessment noted that Mr. Hall had recently engaged in self-injurious behavior and had a prior suicide attempt (by drug overdose). He self-reported not having a support system and being impulsive. Based upon the assessment, Mr. Hall was placed on suicide precautions at the hospital.

Soon thereafter, Mr. Hall was assessed by the emergency room physician. According to this assessment, he presented with anxiety, depression, and suicidal thoughts. The physician, however, was reluctant to sign an emergency psychiatric commitment because this rural hospital could not provide the safety and security required to manage a suicidal inmate. The physician called the local county jail and received assurances the facility could provide a “safe and secure environment” for Mr. Hall. The physician wrote on the discharge summary form that “Phillip should be put in a cell, naked, with nothing in it.”

When Mr. Hall subsequently arrived at the jail, he was screened and then placed in an isolation cell with only a safety smock and blanket, and scheduled to be observed at 15-minute intervals. Although jail staff were well aware of his level of suicide risk, they inexplicably chose not to notify either medical or mental health personnel, although both were available for consultation. Less than six hours later, Mr. Hall was removed from suicide precautions by correctional officers after he denied being suicidal. According to one officer, Mr. Hall complained that he was having trouble sleeping because the lights in the isolation cell remained on during the safety precautions and also made the cell warm. “I thought Mr. Hall’s mood would improve and that he would be more comfortable if it were easier for him to sleep.” Another officer added that “when Mr. Hall was in isolation he was suicidal, but after he visited with me, he promised me that he would not harm himself.” Mr. Hall was subsequently observed by one of these officers to be “meticulously making his bed, which I thought significant because he was coming to terms with being incarcerated.” The following day, a correctional officer was conducting rounds of the cellblock, shined his flashlight into Mr. Hall’s cell, and observed that the inmate was hanging by a pair of tube socks that were tied to the holes in the top bunk. Phillip Hall was subsequently pronounced dead at the same emergency room he had been transported to a few days earlier.

7) Suicide Precautions Should Not Appear To Be Punitive

We must avoid creating barriers that discourage an inmate from accessing mental health services should they feel suicidal. Some inmates are reluctant to seek out mental health services because it might result in their transfer to mental health housing or because inmates on psychotropic medication or often denied institutional jobs or early release from custody (Hanson 2010). Often, certain management conditions of a facility’s suicide prevention policy appear punitive to an inmate (e.g., automatic clothing removal/issuance of safety garment, lockdown, limited visiting, telephone, and shower access, etc.), as well as excessive and unrelated to their level of suicide risk. As a result, an inmate who becomes suicidal and/or despondent during confinement may be reluctant to seek out mental health services, and even deny there is a problem (e.g., Phillip Hall above), if they know that loss of these and other basic amenities are an automatic outcome.

I was assessing suicide prevention precautions in a county jail recently. The sheriff and medical director boasted that the jail had not had a suicide in several years because they housed their suicidal inmates in small booking cages. Inmates called them “squirrel cages.” To me, the cage closely resembled a telephone booth made of sturdy chain-linked fencing, approximately 3 by 3 feet in diameter and 7 feet tall. It was not uncommon for an inmate to be placed in these cages for more than 24 hours. Several inmates who had been placed in these cages on suicide precautions were interviewed. Most confided that they were still experiencing suicidal ideation, but refused to self-report their ideation to staff for fear of being placed back in these cages. An extreme example? Yes; but, the point remains if we treat suicidal inmates with punitive or overly restrictive measures, we run the risk of creating barriers to mental health services.

Although we must safely manage inmates on suicide precautions, we also need to make better decisions in avoiding what appear to be punitive responses to self-injurious behavior. We often take these measures to discourage inmates we perceive as manipulative from threatening or engaging in self-harm, but at what cost? What if we are also discouraging suicidal inmates from coming forward? Most suicide prevention policies prohibit inmates from making telephone calls or having family visits. Why? Facility officials, with support from mental health personnel, argue that providing a suicidal inmate with telephone cells and family visits might increase suicidal ideation and even precipitate a suicide attempt. They might, but what better way to gauge an inmate’s reaction to negative news than by monitoring the reaction while on suicide precautions? At a minimum, inmates on suicide precautions should receive showers, access to telephone cells, legal and family visits, and other routine privileges we provide to non-suicidal inmates unless they are serving a disciplinary sanction and/or a clinical assessment determines otherwise.

8) Inmates Viewed As Manipulative Can Also Be Suicidal

Few issues challenge us more than that of inmates we perceive to be manipulative. It is not unusual for inmates to call attention to themselves by threatening suicide or even feigning an attempt in order to avoid a court appearance, or bolster an insanity defense; gain cell relocation; transfer to the local hospital; receive preferential staff treatment; or seek compassion from a previously unsympathetic spouse or other family member. Some inmates simply use manipulation as a survival technique.

Although the prevailing theory is that any inmate who would go to the extreme of threatening suicide or even engaging in self-injurious behavior is suffering from at least an emotional imbalance that requires special attention; too often we conclude that the inmate is simply attempting to manipulative their environment and, therefore, such behavior should be ignored and not reinforced through intervention. Too often, however, a feigned suicide attempt goes further than anticipated and results in death. Research has warned us that we should not assume that inmates who appear manipulative are not also suicidal, i.e., they are not necessarily members of mutually exclusive groups (Dear et al 2000).

Although there are no easy solutions to the management of manipulative inmates who threaten suicide or engage in self-injurious behavior for a perceived secondary gain, the critical issue is not how we label the behavior, but how we react to it. The reaction must include a multidisciplinary treatment plan.

9) Few Or No Inmates On Suicide Precautions Can Be A Red Flag

A lack of inmates on suicide precautions should not be interpreted as meaning there are not any currently suicidal inmates in your facility, nor a barometer of sound suicide prevention practices. Invariably I arrive at a correctional facility on the first day of an assessment to find that most of the inmates have been cleared from suicide precautions. It is as if the jurisdiction is promoting the idea that only a small number of inmates on suicide precautions is indicative of a solid suicide prevention program. It is not. In fact, the opposite is probably true. We cannot make the argument that our correctional systems are increasingly housing more mentally ill and/or other high risk individuals and then state there are not any suicidal inmates in our facility today. Correctional facilities contain suicidal inmates every day; the challenge is to find them. The goal should not be a “zero” number of inmates on suicide precautions; rather the goal should be to identify, manage and stabilize suicidal inmates in our custody.

Similarly, both healthcare and correctional professionals commonly use the terms “watch closely” or “keep an eye on him” when describing an inmate they are concerned about, but have not placed on suicide precautions. They might have received a telephone call from a family member of an inmate who said, “Please watch my husband or son closely, he might do something stupid.” And the response normally is, “Sure, we’ll watch him closely” or “We’ll keep an eye on him” or “Ma’am, this is a jail, we watch everyone closely.” Such casual responses rarely result in increased observation. If there is any concern that an inmate may be at risk for suicide, the inmate should be on suicide precautions or at least immediately referred to a mental health clinician for a thorough assessment.

10) Avoiding Obstacles to Prevention

In order for our suicide prevention efforts to be meaningful, we must avoid the obstacles to prevention. Experience has shown that negative attitudes often impede meaningful suicide prevention efforts. These obstacles to prevention often embody a state of mind that unconditionally implies that inmate suicides cannot be prevented. Popular slogans such as “If someone really wants to kill themselves there’s generally nothing you can do about it” and “We did everything we could to prevent this death, but he showed no signs of suicidal behavior” are often heard immediately after an inmate suicide and prior to any thoughtful and comprehensive review of the death. There are also some rather subtle (or not so subtle) obstacles to prevention that have been seen over the years. One jail system utilized a fee-for-service program for health care services. That was not unusual, but the Charge Sheet that listed the co-payments for each service was:

  • Nurse Call ¾ $10.00
  • Transportation Fee ¾ $25.00
  • Over-the Counter Medication ¾ $3.00
  • Physician Evaluation ¾ $60.00
  • Pregnancy Test ¾ $20.00
  • Request Release from Suicide Watch ¾ $10.00

This jail was later investigated by the Civil Rights Division of the U.S. Department of Justice and persuaded to stop charging inmates $10 to be discharged from suicide precautions. But in revising their policies, facility officials thought it a good idea to have every suicidal inmate assessed by a physician prior to discharge. Not a bad idea, but the physician evaluation then would cost the inmate $60.

Another example: A nurse once asked an inmate during intake screening if he felt suicidal. When the inmate answered “Yes,” the nurse quickly replied, “If you tell me you’re suicidal, we’re going to have to strip you of all your clothes and house you in a bare cell.” “Okay, then I’m not,” replied the inmate. A clear obstacle to prevention.

One final example: A 16-year-old youth sustained a traumatic suicide attempt at a California boot camp that left him severely brain damaged. It was a very tragic case. The young man had been committed to the boot camp by the court, a poor decision given his known history of both mental illness and suicidal behavior. The boot camp program was ill-equipped to handle him and, in fact, had exclusion criteria that should have kept him out of the facility. It lacked adequate intake screening, mental health services, and staff training. It lacked a suicide prevention policy. The family sued. Following a three-week trial, the jury returned a $4.6 million jury verdict against the county and several employees. But it was the response from the county administrator that was perhaps the most disturbing. He told a local newspaper reporter after the verdict that there would be no changes made at the boot camp: “Our defense is that we have adequate policies and procedures. I don’t really want to give any indication that we would do something that’s not consistent with what our defense was.”[7]

There are several ways to overcome these obstacles to prevention and the negative attitudes that drive them. We can speak to the dramatic reduction in the rate of suicide in correctional facilities throughout the country, particularly in jails. We can look at specific model suicide prevention programs that are currently in operation. I once asked a jail commander of one such program how his facility was able to maintain success despite budget pressures that caused low staffing levels, as well as other challenges, and he responded: “When you begin to use excuses to justify a bad outcome, whether it be low staffing levels, inadequate funding, physical plant concerns, etc., issues we struggle with each day, you lack the philosophy that even one death is not acceptable. If you are going to tolerate a few deaths in your jail system, then you’ve already lost the battle” (Hayes 1998, at page 6).

11) A Reasonable Suicide Prevention Program

Recent research has suggested that many jail suicides occur in facilities lacking comprehensive suicide prevention programs, with only 20% having written policies encompassing all the essential components (Hayes 2012). The essential components to a reasonable suicide prevention program have been discussed elsewhere (Metzner & Hayes 2012), and include training of all correctional, medical, and mental health staff on both an initial and annual basis; intake and on-going screening/assessment for suicide risk; procedures that encourage communication between outside entities and correctional facilities, as well as internally between and among facility staff and the suicidal inmate; suicide-resistant housing and restrictions (e.g., clothing, showers, telephone calls, visiting, etc.) commensurate with risk level; procedures for emergency response to a suicide attempt; reporting and notification of a suicide to the facility’s chain of command and family of suicide victim; and multidisciplinary examination of the inmate suicide through a mortality review. Correctional facility officials should not conclude that an inmate suicide was not preventable unless they have demonstrated that their facility initiated and maintained a comprehensive suicide prevention program.

12) Conclusion

Has progress ended? I think not. Edwin Shneidman, the father of contemporary suicidology, once noted that “Suicide is not a bizarre and incomprehensible act of self-destruction. Rather, suicidal people use a particular logic, a style of thinking that brings them to the conclusion that death is the only solution to their problems. The style can be readily seen, and there are steps we can take to stop suicide, if we know where to look” (Shneidman 1987, at page 58). Many preventable suicides remain within our reach if we know where to look.


Atlas, R. (1989). Reducing the opportunity for inmate suicide. Psychiatric Quarterly, 60, 161-171.

Dear, G., Thomson, D. & Hills, A. (2000). Self-harm in prison: Manipulators can also be suicide attempters, Criminal Justice and Behavior, 27, 160-175.

Garvey, K., Penn, J., Campbell, A., Esposito, C., & Spirito, A. (2009). Contracting for safety with patients: Clinical practice and forensic implication. Journal of the American Academy of Psychiatry and the Law, 37, 363-370.

Hanson, A. (2010). Correctional suicide: Has progress ended? The Journal of the American Academy of Psychiatry and the Law, 38, 6-10.

Hayes, L. (2012). National study of jail suicide: 20 years later. Journal of Correctional Health Care, 18, 233-245.

Hayes, L. (2006). Suicide prevention in correctional facilities: An overview. In M. Puisis (Ed.), Clinical Practice in Correctional Medicine (pp. 317-340). Philadelphia: Mosby Elsevier.

Hayes, L. (2005). Model suicide prevention programs: Part 1. Jail Suicide/Mental Health Update, 14 (1), 1-9.

Hayes L. (2003). A jail cell, two deaths, and a telephone cord. Jail Suicide/Mental Health Update, 11 (4), 1-8.

Hayes, L. (1998). Model suicide prevention programs: Part 3. Jail Suicide/Mental Health Update, 8 (1), 1-6.

Hayes, L. (1989). National study of jail suicide: Seven years later, Psychiatric Quarterly, 60, 7-29.

Jamison, K. (1999). Night Falls Fast: Understanding Suicide. New York: Knopf.

Masters, K. & Wandless, D. (2005). Use of pulse oximetry during restraint episodes. Psychiatric Services, 56, 1313-1314.

Metzner, J. & Hayes, L. (2012). Suicide prevention in jails and prisons. In R. Simon & R. Hales (Eds.), Textbook of Suicide Assessment and Management (pp. 389-403). Washington, D.C.: American Psychiatric Association.

National Commission on Correctional Health Care (2008). Standards for Health Services in Jails, Chicago: Author.

Patterson, R. & Hughes, K. (2008). Review of completed suicides in California department of corrections and rehabilitation, 1999 to 2004. Psychiatric Services, 59, 676-682.

Quinton, R. & Dolinak, D. (2003), Suicidal hangings in jail using telephone cords. Journal of Forensic Sciences, 48, 1151-1152.

Shneidman, E. (1987). At the point of no return. Psychology Today, March, 55-58.

Watts, B., Young-Xu, Y., Mills, P., Derosier, J., Kemp, J., Shiner, B., & Duncan, W. (2012), Examination of the effectiveness of the mental health environment of care checklist in reducing suicide on in-patient mental health units, Archives of General Psychiatry, 69, 588-592.

Way, B., Sawyer, D., Barboza, S., et al (2007). Inmate suicide and time spent in special disciplinary housing in New York state prison. Psychiatric Services, 58, 558-560.

[1]This article appeared in the International Journal of Law and Psychiatry 36 (2013) 188–194.



[4]Personal correspondence with author, March 31, 2008.


[6]In order to ensure complete confidentiality, the names of the facility, staff, and suicide victim have been changed. No other modifications have been made.

[7]Hanford Sentinel, “Jury awards former inmate $4.6 million,” April 6, 2007, page 2,