The poet A. Alvarez titled his book about suicide, “The Savage God” to evoke its lure and horror for its victims and their families. Suicide is one of the most upsetting paths to death for all touched by it. It has been the subject of much study, aiming to understand what drives it and how to diminish it. Changes in public health and education as well as patient care are understood to have contributed to the national incidence of suicide per 100,000 of population in the United States gradually but significantly decreasing from 13.2 in 1950 to 10.8 in 2003. A parallel improvement occurred in New York State where the suicide rate per 100,000 fell from 9.5 during the period from 1980 – 1985 to 6.2 in 2005, the 49th among the 50 states.
No matter the gains for the broad population, it is essential that once a person is hospitalized for mental illness their safety is assured. During June, 2009 the NYS Office of Mental Health released a report about inpatient suicide titled, “Incident Reports and Root Cause Analyses 2002-2008: What They Reveal About Suicides”. Their overview addresses inpatient suicide, suicides within 72 hours of discharge, and suicide while on pass or AWOL. During those years the average number of suicides of inpatients was 5 per year and for patients in the 72 hours post-discharge, AWOL and or on pass group the rate was about 12 per year. When presented with positive data such as this, the question arises as to how to further think about it. To consider data and draw conclusions from it, a frame of reference is necessary. OMH’s overview, draws on broad numbers about suicide and how it compares to other causes of death, to present a rather gray picture about inpatient suicide. We ask, is that perspective accurate?
The OMH report provides a limited frame of reference when it says that, “…. NYS inpatient mental health treatment facilities operate approximately 3,660,000 bed days each year.” The decreased numbers of suicides would be better understood if the report included data on total numbers of admissions or discharges, and on average lengths of stay. For instance, in New York State in 2004 there were 107,271 discharges from general hospital inpatient psychiatric units and the average length of stay (ALOS) was 14.81 days. By contrast, in 1990 the number of discharges 74,563 and the ALOS was 24.77 days. Additionally, state psychiatric centers discharge approximately 7,000 persons per year. Elsewhere, OMH has presented data on where patients are served: in one recent year, general hospitals served 69,939, state psychiatric facilities served 11,288 and private hospitals served 10,378 individuals. These additional numbers flesh out a picture of service delivery and underscore just how infrequent suicide is amongst people served in our mental health system. While we agree that every suicide is a tragedy, 17 deaths (5 inpatient and 12 post-discharge) per 114,000 discharges, a rate of 0.01%, is an achievement to be commended. Another study conducted on all discharges between 2002 and 2007 from a consortium of 6 prominent free-standing private psychiatric hospitals reported 3 inpatient suicides, all by hanging, among 153,552 discharges, a rate of 0.002%. This lower rate adds support to the conclusion that OMH could have more vigorously emphasized – suicide among inpatients has meaningfully declined. In a review of post-discharge suicides among the same cohort, the consortium has preliminarily identified 49 suicides in the first post-discharge month. We suggest that the immediate post-discharge period warrants more careful study and add that study of this higher-risk period should separate planned discharges from patients who were AWOL or on pass at the time of suicide.
The recent OMH review might have benefited by drawing on earlier work by another interested NYS agency. In May, 1989 the NYS Commission on Quality of Care for the Mentally Disabled (CQC) released a report, “Preventing Inpatient Suicide: An Analysis of 84 Suicides by Hanging In New York State Psychiatric Facilities (1980-1985)”. Their study focused on a subset of the 131 inpatient suicides during those years. The suicide rate of 48 per 100,000 inpatients was 0.05%. (For purposes of perspective, the odds that a person who auditions for American Idol will win is 1 in 103,000 or 0.001 %.) Thus, over the past 25 years the inpatient suicide rate in NYS psychiatric facilities has dropped from 48 per 100,000 to 5 per 100,000 – clearly, inpatients are far safer now, despite the risks during both eras being very low. A strength of the CQC work was that it studied over 50 variables allowing for data-based conclusions, which providers could incorporate into practice. It recognized, as most scholars have, the unreliability of predicting which persons were likely to attempt suicide. It directed attention to the value of making the environment safer and paid detailed attention to specific matters such as actual times of higher risk and human factors such as making sure safety orders are clearly written and implemented. (We suggest that CQC consider posting their study on their web site.) The OMH report might have been strengthened had it followed up on some important questions it raised and presented in a fuller context. As an example, the important question of increasingly shorter lengths of stay is raised but not addressed.
The OMH report based its recommendations on the results of root cause analyses. While this methodology is designed to identify system failures, it also is dependent on self-examination, introducing the risk of subjectivity and bias. To counter that risk OMH requests re-investigation when they feel an institution has overlooked something. We do not dispute the value of fostering rigorous scrutiny of processes, and we feel there are important and generalizable lessons to be learned from their summary of root cause analyses. However, it should be recalled that these lessons emanate from reviews of very rare occurrences, and they are unavoidably anecdotal and subjective. The OMH report highlights “communication” deficiencies and makes particular note of a parallel conclusion drawn in a 2005 Joint Commission report. While this certainly may be a contributory problem, the likelihood that similar communication occurred in tens of thousands of other cases where suicide was not the outcome, must not be overlooked. We must not return to the approach of believing we can predict who among a high-risk population is most likely to attempt suicide. Identifying faulty communication as something to eradicate stops significantly short of outlining processes and practices based on data, which if adhered to could further improve outcomes. Once we understood the frequency and means of suicide by hanging, including where and how it occurred within the hospital, significant environmental changes were made and the rates dropped – this was a major achievement. How to systematically lower post-discharge suicide rates is far more difficult to imagine; yet real data on frequency compels us to address this more sizeable challenge. As we attempt to meet this challenge, in an age of limited resources, it will be important to weigh the benefits of purchasing and building increasingly sophisticated environmental safety elements against the costs of enhancing the skills of staff who provide programs and aftercare.
In conclusion, the OMH report makes clear that more than 2 decades of effort have made our inpatient units far safer. These gains are the result of concerted and collaborative efforts among governmental agencies, hospitals, and the professional teams providing direct care to our patients. While no system should “rest on its laurels,” the mental health care system in NYS has done a remarkable job of driving down the number and probability of inpatient suicides. Persons admitted to our hospitals have every reason to believe they are in a safe place. All who worked to realize these goals should feel justifiable pride in what has been accomplished.
Barry B. Perlman, M.D. is Director, Dept. of Psychiatry, Saint Joseph’s Medical Center, Yonkers, New York. Virginia L. Susman, M.D. is Associate Medical Director & Site Director, New York Presbyterian Hospital, Westchester Division.