A 38-year-old podiatrist, well known in his small community, was admitted through the emergency department to a local hospital. He was brought in by the police who discovered him in his crashed car; he had presumably driven the car off a cliff. The patient suffered multiple fractures and injuries for which he was treated in the emergency department. A sister, his closest relative, found a suicide note at his home. She lived approximately 2 hours away. The police were sure this was a serious, but miraculously failed suicide attempt. The patient was transferred to a locked, inpatient psychiatric service where a history was taken.

Much of his childhood history was unremarkable. He had graduated from high school, had completed podiatric medical school, and had set up a practice not far from where he grew up. He was well liked by his patients, had a pleasant manner, and was fairly popular in his community. It emerged during the hospitalization that the patient, who was an avid skier, had stolen expensive ski equipment from several stores, and this had come to the notice of local law enforcement. The patient had ongoing marital conflicts with his wife from whom he was separated. They had three young children who were now with his wife and with whom he had sporadic contact as she lived a few hours away with her parents. His solo practice had declined, and his business was on the brink of ruin.

In the hospital, the patient was placed on observation for ongoing suicidal ideation, severe depression, anhedonia, poor self-care, hopelessness, and tearfulness. Several medications were tried without much improvement. He was then given ECT [electroconvulsive therapy] with good results. Gradually, he began to improve, was more engageable, and attended groups but was not always forthcoming about his concerns. He continued to be anxious and did not always sleep well. The patient was transferred to an open ward with no privilege restrictions, and his treatment with a SNRI [serotonin-norepinephrine reuptake inhibitor] continued. His sister visited him on and off. The nurses tried to mobilize the patient toward establishing life goals, but he had a hard time focusing. At least three psychiatrists treated the patient, making notes that were hard to interpret in the chart. It was unclear from the chart review whether they were members of the same practice or whether they had spoken to one another about the patient.

The patient learned that the police had charged him; he was awaiting a hearing. He also heard that his wife now wanted a divorce and wanted full custody of their children. He spoke to the treatment team and asked the social worker to contact his sister. He told his sister that he wanted her to handle all his affairs and prepared a note to that effect. He remained on an open unit without observation but was still on 15-minute checks. Early one morning soon after he saw his sister, the patient was found hanging from a makeshift hook in the bathroom of his room. On the 15-minute checks the previous night, the nurse had thought he was in bed. He had placed pillows under the sheets, deceiving the nurse into thinking he was in bed. The family, colleagues, and staff were devastated. Clinicians asked themselves: "What did I miss? Could I have prevented this? Does this mean I am not a good psychiatrist? Will I be sued?

In the words of Robert I. Simon, M.D. "There are two kinds of psychiatrists: those who have had patients commit suicide and those who will" (1). Suicide is a high-risk, yet relatively low-frequency event that we are not good at predicting (2). There are more than 30,000 suicides per year in the United States (3). Of these, 5%—6% occur in hospitals (3), translating to nearly 1,800 inpatient suicides per year. Suicide had been the number one Joint Commission Sentinel Event in our nation in 2005 (4). Suicide as a hospital sentinel event was more common than operative and postoperative complications, more common than wrong-site surgeries, and more common than medication errors. It is now the fifth most frequently reported sentinel event by The Joint Commission (Sentinel Event Statistics 2004—2010). (5) Suicide is the number one cause of psychiatric malpractice settlements and verdicts (6). It was the fifth most frequently reported Sentinel Event by The Joint Commission (Sentinel Event Statistics 2004—2010).

Psychiatrists and organizations cannot expect to wait for the suicides or lawsuits to occur to realize opportunities for change and improvement. There will be mistakes despite good psychiatrists practicing within good systems of care. Barriers exist for individual clinicians within themselves, their treatment teams, their organizations, and mental health systems, creating failure modes affecting suicide risk assessment and risk reduction.

Failures can and do occur in systems of care. Failure modes are problems that might occur at various levels in the risk assessment, management, and treatment of a person at risk and result in a suicide. A failure mode and effect analysis (FMEA) is a prospective analysis of the entire system and process of suicide risk assessment and management that assesses where high-risk problems might occur. The final goal of an FMEA is to then mitigate or, if possible, eliminate the failure points to change the process, design, or system to support improved care and safety for the patient and improved success for the clinician. Failure modes can be extrapolated from the realm of common allegations of negligence, as summarized by Scott and colleagues' (6) (Table 1).

Failures of the system can be organized across a spectrum of eight areas, ranging from the quality of the suicide risk assessment to appropriate training and orientation. In each of the eight areas of failure modes, the subsequent solutions can be viewed through multiple lenses, including the scope of 1) the individual clinician, 2) the team, 3) the organization, and 4) the entire system. Table 2 summarizes the failure modes and some of the strategies to overcome barriers to effective suicide risk assessment and risk reduction (7—13).

Is "NO SI (suicidal ideation)" a good enough suicide risk assessment?

In the study of Busch et al. (3) of patients who committed suicide while inpatients or immediately after discharge, 78% denied suicidal ideation at their last communication. The greatest clinical root cause of inpatient suicide is a failure in clinical assessment (Figure 1). Reported suicides have increased by 20% from 2005 to 2008.

Clinicians are not able to predict suicide. However, psychiatric standard of care requires a suicide risk assessment (6). Suicide risk assessment should entail the following five elements:

Barriers to improved patient safety in regard to suicide also exist beyond the scale of the psychiatrist into the regions of the team, organization, and system (Table 2).

Evidence supports the fact that psychiatrists are not able to predict suicide (2). However, suicide risk assessment and risk reduction are attainable. The right skill, teamwork, and system support minimize the risk for the patient and physician. With the failure modes for suicide being addressed, the case of the podiatrist presented here might have had a different outcome. Awareness of the barriers to safe patient care is the first step toward improving suicide risk. Once identified, means to overcome these barriers can be put in place. Psychiatrists can be more confident in suicide risk assessment and reduction, thereby improving patient safety.