Patient Safety Tip of the Week

October 6, 2015

Suicide and Other Violent Inpatient Deaths



There has been a lot of attention in the Canadian healthcare system recently about suicide in hospital inpatients. The Toronto Star (Carville 2015) reported a sampling of half of Ontario hospitals found 96 inpatient suicide deaths since 2007 and a further 760 inpatients seriously harmed while attempting suicide. In addition, the Canadian Patient Safety Institute and Health Quality Ontario just published the first list of “never events” for Canadian hospitals (CPSI 2015) and it includes patient suicides or serious harm related to suicide attempts.


And, of course, Canada is not alone. The Joint Commission sentinel event statistics continue to show 80-90 suicides per year reported as sentinel events in the US over the past decade (Joint Commission 2015). While suicides numerically are more common in non-hospital settings, fatal suicide attempts are more common in inpatients than in outpatient or residential settings (NYSOMH 2013).


The Toronto Star article noted that suicides and attempts occurred in all hospital departments, from maternity to neuro-clinical, emergency, medical and psychiatry and that methods ranged from strangulation and suffocation to drowning, overdose and electrocution.


But not all violent deaths of inpatients are suicides. Others violent deaths may occur in patients who are confused or attempting to abscond from the hospital. But many of the root causes and contributing factors are common to both occurrences.


Our first column on preventing inpatient suicides was our January 6, 2009 Patient Safety Tip of the Week “Preventing Inpatient Suicides”. In that we discussed changes that were made, primarily on behavioral health units, after a New York State report (CQC 1989) and the first Joint Commission Sentinel Event report on inpatient suicides (Joint Commission 1998). Since hangings were the most frequent method of inpatient suicide in the New York State report, attention was directed to environmental factors including exposed pipes, showerheads, bars in showers and toilet stalls. The commission recommended environmental safeguards such as use of breakaway physical structures. Joint Commission also recommended that breakaway hardware actually be tested regularly (weight testing to ensure that hardware will, in fact, break away when sufficient weight is applied). Other environmental safeguards include keeping doors closed, and keeping sharps or other hazardous materials away from patients. The Joint Commission Sentinel Event Alert noted 75% of inpatient suicides were by hanging but also noted that 20% involved jumping from roofs or windows. So careful attention to access to such sites is important.


While the NYS report focused on inpatients in psychiatric facilities, the 1998 Joint Commission Sentinel Event Alert noted that of 27 suicides occurring in general hospitals, 12 occurred in med-surg units and one in the emergency room. In an excellent review, based on root cause analyses (RCA’s) of inpatient suicides and suicide attempts in the VA Hospital system, Mills and colleagues (Mills et al 2008) noted methods of suicide varied by location. Whereas hanging/asphyxiation, cutting, and fires occurred most often on the psychiatric units, overdoses, jumpings, stabbings, and ingestion of chemicals was more common on non-psychiatric units. They provide tables describing the types of anchor points in hangings, the materials used as nooses, the implements used in cuttings, and the locations for jumpings. They point out that, for a variety of reasons, it may be impossible to eliminate the materials used for nooses so they suggest a focus on eliminating anchor points. For example, they note that interior doors and cabinets can be removed or replaced with accordion doors that cannot be used as anchor points. And they recommend that things like door knobs, railings, faucets and hooks be eliminated or constructed so as to break away when weight is applied. They also note that most of the attempted overdoses took place on units other than psychiatric inpatient units (though many of these units were detox units, etc.) so careful assessment of security of medications is important on all units.


In 2010 The Joint Commission issued an update (The Joint Commission 2010) to its previous Sentinel Event Alert on preventing inpatient suicide and focused on suicide occurring outside of psychiatric units (see our December 2010 What’s New in the Patient Safety World column “Joint Commission Sentinel Event Alert on Suicide Risk Outside Psych Units”). That provided statistics about suicide on various units, places the suicides take place, and means used. They pointed out that there are 2 types of patients who commit suicide on med/surg units: (1) those admitted after a suicide attempt and (2) those with no known psychiatric disorder or known suicide intent. They detail the many risk factors for suicide plus the potential warning signs. Suicide on med/surg units also tends to occur earlier after admission than those occurring on behavioral health units.


They discuss many of the environmental items, usually not available on psychiatric units but readily available on med/surg units, that may be used in suicides. These include items such as bell cords, sheets, restraint belts, various types of tubing, bandages, etc. But key contributing factors are lack of screening, failure to identify risk or recognize warning signs, lack of appropriate training, poor communication, and lack of appropriate staffing for proper observation. They make a number of recommendations regarding screening for depression and suicide risk (both inpatient and emergency department) and watching for behaviors that are potential warning signs of impending suicide attempts. They then offer tips about engaging the patient and family or others capable of providing peer support. And they stress the importance of communication at all levels of care.


A coalition of healthcare organizations has put together a national “Zero Suicide” campaign. That campaign focuses on prevention of suicide in all parts of the healthcare continuum and stresses the important roles of leadership, cultural change, training, communication, and transitioning. But one of the key areas they stress is identification of at-risk patients. Historically, one of the problems has always been identifying those patients who are at risk for suicide. While there are lots of studies noting various risk factors for suicide, there were few validated tools for accurate risk assessment. That is no longer the case. One of the core elements of the Zero Suicide program is use of the Columbia-Suicide Severity Rating Scale (Posner 2011). The C-SSRS may be found online (see also our December 2011 What’s New in the Patient Safety World column “Columbia-Suicide Severity Rating Scale”).


One important consideration, since so many patients at risk for suicide enter the hospital via the emergency department, is that the initial risk assessment should begin in the emergency department (McBroom 2013). But suicide risk must not be considered as an isolated event in time. Suicide risk assessments are often incomplete or not done and the 1998 Joint Commission alert especially noted a dearth of suicide reassessments. Sound familiar? How often have we noted that other key assessments in healthcare, such as fall risk or DVT risk assessment, are done on admission but are not repeated even though clinical circumstances have changed during the hospital course? Another issue is that even though the reason for admission may have been related to a suicide attempt, the patient on admission may be unable to cooperate with a suicide risk assessment (eg. the patient may be comatose or obtunded because of a drug overdose). Sometimes many days pass where the patient physically would have been incapable of another suicide attempt but, ironically, as they begin to improve medically the suicide risk reappears.


Risk assessments should be done not only on admission, discharge, and all transitions of care but also repeated frequently as the clinical course demands. It is also particularly important to consider the risk of suicide in patients on leave or pass and several studies have also noted the association between such leaves and suicide (see our April 2, 2013 Patient Safety Tip of the Week “Absconding from Behavioral Health Services”).


Another critical factor we see over and over is that there may be inadequate training for those charged with close monitoring or observation of patients. This is especially the case on med-surg floors when patients are identified as being at high risk for suicide, other self harm, or wandering and elopement. A case from a year ago illustrates many of the problems encountered with such patients on non-behavioral health units (Darragh 2014). A patient with an “impulse control disorder” following head trauma was on continuous observation by a hospital security guard in an ICU when he entered the bathroom and locked the door. He then smashed double-paned locked windows and jumped to his death from the sixth floor. The guard had not received the same training that nurses who usually provide continuous observation would have had. Such would have required continued observation of the patient in the bathroom at least via a partially open door.


Surprisingly, many completed suicides occur in patients who were on close observation protocols. One study looked at human factors to improve observation practices (Janofsky 2009). Janofsky reviewed the literature on observation practices and did a FMEA (Failure Mode and Effects Analysis) at Johns Hopkins Hospital and found that the language used to describe observation practices and procedures had not been standardized and that the terminology and practice varied widely, even in the same health system and across shifts and units. Moreover, staff often tended not to follow observation policies or covertly modified them and some considered observation a low-level task that was not clinically useful. He noted Hopkins had one level of intermittent observation and 3 levels of continuous observation. In the intermittent level patients are monitored every 15 minutes. In the lower level of continuous observation one staff member may observe more than one patient but the patient must be kept in constant view. At 1:1 observation one staff member must remain in close proximity at all times, with no physical barriers between the patient and observer. But the highest level (“intensive’) requires staff remain within arm’s length of the patient at all times in an adequately lighted area. That includes observation during hygiene and toileting activities. They may use security personnel for protection with patients at high risk of violence but the security personnel do not perform observation.


The FMEA done by Janofsky showed critical failure modes caused by communication failures between physicians, nurses, and observers. Often nurses’ workflow issues interfered with regular communication with observers. Sometimes observers were not comfortable in communicating changes to nursing staff. Shift changes often interfered with communication. And both expectations and documentation requirements for the observers were often inconsistent. After feedback they developed a simple support list for the observers and nurses, and they formalized handoffs. They also found unclear physician and nurse decision making regarding when to start or stop constant observation.


We suspect many other organizations are having similar problems. The role and activities and expectations of the observers are often poorly defined whether the observation is for suicide risk or wandering risk. And without more structure to those there are also obviously difficulties in training observers.


And one other caveat: when we put a patient on 1:1 observation, whether for suicidal risk or because of delirium, we also need to remember it is not practical for any one individual to remain continuously vigilant for long periods of time nor should they be engaged in other activities. Tishler et al (Tishler 2009) recommend changing the observation person every two hours to avoid burnout.



And we’ve discussed some of the tools available to help identify risk factors for inpatient suicide that might be mitigated or avoided. In our January 6, 2009 Patient Safety Tip of the Week “Preventing Inpatient Suicides” we noted that the VA had developed a mental health environment-of-care checklist (MHEOCC). That checklist is available online on the VA Patient Safety website. And in our July 2012 What’s New in the Patient Safety World column “VA Checklist Reduces Suicide Risk” we discussed a study that demonstrated use of the MHEOCC significantly reduced the risk of inpatient suicide (Watts 2012).


Bathrooms are favored sites for patient suicides for several reasons. One obvious reason is that the patient is usually completely or partially out of sight of any observers. Another reason is that many bathrooms can be locked from the inside. The other is that they typically contain many loopable items. While most behavioral health units have removed locks and loopable items from such bathrooms, that is seldom the case in other areas of the hospital. Many of you will recall our March 16, 2010 Patient Safety Tip of the Week A Patient Safety Scavenger Hunt” where we challenged you to find a bathroom in your radiology suite or a supply closet on a med-surg unit. Both have numerous implements a suicidal patient could use to commit suicide and may be lockable from the inside. And in most cases the key to open the door from the outside is not readily available, allowing the patient ample time to commit the act.


Another potentially vulnerable situation is one in which an inpatient who is at risk for suicide or wandering (whether on a behavioral health unit or med/surg unit) is transported to another area of the hospital such as the Radiology suite. In our several columns on use of a “Ticket to Ride” checklist for intrahospital transports (see, for example, our August 25, 2015 Patient Safety Tip of the Week “Checklist for Intrahospital Transport”) we have emphasized inclusion of an item on the checklist to alert staff of potential suicide. If the patient is at risk, appropriately trained staff would need to accompany the patient and especially ensure he/she is not allowed into a locked bathroom.


Another potential vulnerability has to do with fire alarms. In one case a patient pulled a fire alarm which automatically unlocked doors on a behavioral health unit, allowing him to escape and jump to his death from a rooftop (Pfeiffer 2010). After we heard about that case we began to include inspection of stairwells and rooftop access points adjacent to behavioral health units in our patient safety walkrounds or environmental walkrounds.


While the focus of today’s column has been on prevention of inpatient suicides, it is critical we don’t lose focus on a potentially even more vulnerable period for suicide – the transition period following hospital discharge. Most such suicides occur within the first 72 hours following discharge or within the first 30 days.


Given the nationwide shortage of psychiatrists, getting timely followup arranged can be problematic. The “Zero Suicide” approach, as adopted by NYS Office of Mental Health (NYSOMH 2013), stresses the importance of the “warm handoff” and use of “bridger” staff. The latter are peer specialists who meet with patients either face-to-face or by phone prior to discharge and accompany the patients to their first outpatient appointment. They also ensure that additional appointments are scheduled and educate the patients about other support services.


But what about those patients who reject further treatment at discharge? A “caring letter” expressing concern and support sent to the patient every four months has been shown to reduce the completed suicide rate in a randomized controlled trial (NYSOMH 2013).



So what should your organization be doing to reduce the suicide risk? We recommend you consider the following:




Some of our prior columns on preventing hospital suicides:





See also our previous columns on wandering, eloping, and missing patients:








Carville O, Boyle T. How do suicides happen in hospitals? Toronto Star 2015; Sept. 27, 2015



CPSI (Canadian Patient Safety Institute)/Health Quality Ontario. Never Events for Hospital Care in Canada. September 2015



The Joint Commission. Summary Data of Sentinel Events Reviewed by The Joint Commission. Sentinel event statistics as of 7/9/2015.



New York State Office of Mental Health. Getting to the Goal. Suicide as a Never Event

in New York State. August 2013



CQC (New York State Commission on Quality of Care). Preventing Inpatient Suicides: An Analysis of 84 Suicides by Hanging in New York State Psychiatric Facilities (1980-1985). 1989



Joint Commission. Sentinel Event Alert Issue #7. Inpatient Suicides: Recommendations for Prevention. November 6, 1998



Mills PD, Watts BV, Miller S, Kemp J, Knox K. DeRosier JM, Bagian JP.

A Checklist to Identify Inpatient Suicide Hazards in Veterans Affairs Hospitals
Joint Commission Journal on Quality and Patient Safety. Volume 36, Number 2, February 2010 pp. 87-93(7)



The Joint Commission. Sentinel Event Alert Issue #46. A follow-up report on preventing suicide:  Focus on medical/surgical units and the emergency department. Sentinel Event Alert 2010; 46: 1-5 November 17, 2010



Zero Suicide in Heatlh and Behavioral Healthcare. website



Posner K, Brown GK, Stanley B, et al. The Columbia–Suicide Severity Rating Scale: Initial Validity and Internal Consistency Findings from Three Multisite Studies With Adolescents and Adults. Am J Psychiatry 2011; 168(12): 1266-1277

Published online November 8, 2011



C-SSRS scales for clinical practice.



McBroom S. Reducing Inpatient Suicide Risk In The Hospital Setting. Compass (McNeary, Inc.) 2013; 12(3): 1-4



Janofsky JS. Reducing Inpatient Suicide Risk: Using Human Factors Analysis to Improve Observation Practices. J Am Acad Psychiatry Law 2009. 37(1): 15-24



Darragh T. State: St. Luke's staff not properly trained to monitor man who jumped to death from hospital window. The Morning Call (Allentown, PA) August 6, 2014



Tishler CL, Reiss NS. Inpatient suicide: preventing a common sentinel event. General Hospital Psychiatry 2009; 31: 103-109



VA. Mental Health Environment of Care Checklist. 2015



Watts BV, Young-Xu Y, Mills PD, et al. Examination of the Effectiveness of the Mental Health Environment of Care Checklist in Reducing Suicide on Inpatient Mental Health Units. Arch Gen Psychiatry. 2012; 69(6): 588-592



Pfeiffer R. Man survives plunge from roof of hospital. Niagara Gazette 2010; October 17, 2010







Print “PDF version









Tip of the Week Archive


What’s New in the Patient Safety World Archive