INTRODUCTION
Intuition indicates that heavy objects fall faster than light objects, and that all suicide is the result of mental disorder. A derivative article of faith (in some quarters) is that when a suicide occurs, safeguards put in place by society have failed and there is a health professional to be brought to book.
To explore the relationship between suicide, mental disorder and attributing responsibility it is necessary to first look at some interlocking components: health, mental disorder, personality disorder, medicalization and mental health problems. A synthesis can then be attempted.
This topic is of relevance to those working in psychiatric intensive care units as, in spite of best efforts, suicide does occur in these especially careful facilities. Those employed to scrutinize the work of health professionals find such events particularly perplexing, and there is an inclination to particular criticism of intensive care unit suicides. Also, there are important medical/legal/political/ethical issues raised when psychiatric intensive care units are used for the management of a troublesome behavior (which may include suicide) which is not the product of mental disorder.
HEALTH
Health is the super concept under which mental health and related topics cluster. The World Health Organization (1948) defined health as: ‘a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity’. Thus, health is synonymous with well-being. Dictionary definitions of well-being make reference to health, ‘a contented state of being happy, healthy and prosperous’ (Princeton’s Wordnet, 2009). When the unhelpful circularity is discarded, well-being (and health) becomes ‘a contented state of being happy … and prosperous’. Recent definitions of health such as ‘The overall condition of an organism at a given time’ (American Heritage Dictionary, 2003), reflect the six decade old WHO concept.
As health is ‘not merely the absence of disease’, but is a matter of ‘the overall condition of the organism’, and includes being ‘happy … and prosperous’, it is not merely a matter for doctors and others who treat, but for every agency which affects the lives of people, from education, employment, transport, wealth distribution, security and safety, to child care and internet dating services.
Health departments are, in general, misnamed, because they only provide small components of health treatment. Health departments began as hospitals and were focal points for the treatment of disease and injury. The predominant workers were doctors and others who treated people. As the concept of health broadened, some additional responsibilities were awarded to them. There has been greater emphasis in the training of health professionals in prevention and public health matters. But, by the time most people present to health professionals, maladies are well established; effective prevention is a matter for public education and is better commenced at the cradle than the clinic.
MENTAL DISORDER
The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revised (DSM-IV-TR; American Psychiatric Association, 2000) is the most influential diagnostic instrument currently available. It states: ‘… no definition adequately specifies precise boundaries for the concept of mental disorder’ (p xxx). In the absence of a definition, DSM-IV-TR provides a description (p xxxi) which begins, ‘… each of the mental disorders is conceptualized as a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is associated with present distress or disability…’. This description employs vague undefined terms including ‘clinically significant’, ‘psychological syndrome’ and ‘distress’ and is unable to differentiate mental disorders from normal human experiences such as guilt and grief.
Although ‘precise boundaries’ between mental disorder and no mental disorder are lacking, in clinical practice normative judgments are still required. With respect to patients located in these boundary regions, judgments are rationally informed, but not rationally determined. Thus, different responsible third parties of the pre-mortem period may reach different conclusions regarding the presence or otherwise of mental disorder.
PERSONALITY DISORDER
Personality can be defined as the organization of the individual that determines her/his unique response to the environment (Allport, Reference Allport1938). This resonates with experience. We can fairly accurately predict the reactions of people we know well to environmental events. Personality disorder is defined in the DSM-IV-TR (p 685) as ‘an enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual’s culture, is pervasive and inflexible … and leads to distress and impairment’.
There are two reasons for giving personality disorder special mention in this paper: personality disorder is 1) the diagnosis which is most difficult to distinguish from normal, and 2) frequently associated with suicidal behavior.
Personality disorder is a topic of contrasting views. Some authorities believe that the so-called personality disorder simply represents variants of normal (like extreme tallness and shortness) and should not be classified as a mental disorder (Schneider, Reference Schneider1950). Unlike mental disorders such as schizophrenia and bipolar disorder, which generally come in episodes, personality disorder is diagnosed when the individual habitually responds to the environment in a more extreme manner than people without personality disorder (e.g., with undue anger, seduction or avoidance). The difference between those with personality disorder and those without is quantitative not qualitative. Given that the validity of the diagnostic category can be questioned, disagreement on the diagnosis in individual cases is to be expected.
Personality has to do with characteristic long term responses. Personality disorder is diagnosed when the individual has long used a restricted number of extreme or maladaptive responses. Suicidal behavior and completion are maladaptive responses (with possible exceptions, such as in the case of people terminal illness). Accordingly, suicidal behavior and completion is common in personality disorder (Cheng et al., Reference Cheng, Mann and Chan1997).
BORDERLINE PERSONALITY DISORDER
Borderline personality disorder deserves special mention as it carries a 10% lifetime risk of suicide (Plakun et al., Reference Plakun, Burkhardt and Muller1985). This disorder is characterized by a pattern of unstable personal relationships, self-image and affects, and marked impulsivity. Personality disorders are chronic conditions, however, there may be acute episodes, and people with borderline personality have frequent broken relationships and episodes of anger and distress which may result in self-injury or suicide.
For these people, lengthy inpatient periods in psychiatric facilities are at best useless and at worst, damaging; they remove individuals from the real world in which they must learn to function, and delay the development of a sense of personal responsibility. However, brief hospitalization may be helpful during crisis periods and allow the settling of acute episodes of distress (Krawitz & Watson, Reference Krawitz and Watson2000). Wyder (2004) reported that of those who attempt suicide, in 79% the impulse has passed within 12 hours.
The management of patients with borderline personality disorder is legally perilous for doctors because of the lack of understanding in the community of the chronic risk of suicide and the optimal treatment outlined in the above paragraphs (Gutheil, Reference Gutheil1985). However, there are some informed jurisdictions. The Ministry of Health (New Zealand) Guidelines (1998) state: ‘In order to achieve therapeutic gain, it is sometimes necessary to take risks. A strategy of total risk avoidance, could lead to excessively restricted management, which may in itself be damaging to the individual’.
MEDICALIZATION AND MENTAL HEALTH PROBLEMS
Medicalization is the process by which non-medical problems are reclassified as medical problems. That is, non-medical problems (such as the normal response to loss) become called disorders, with the naturally following expectation that a medical treatment is appropriate (Zola, Reference Zola1972). Van Praag (2000) described medicalization as the process by which ‘normal’ human behavior and experience is ‘re-badged’ as a series of medical conditions, and Chodoff (2002) has warned against the process of ‘Medicalization of the human condition’. Critics of medicalization claim that shyness has been re-badged as social phobia and promiscuity as sexual addiction. While Zola (1972) laid total blame for medicalization at the door of the medical profession, other drivers are now recognized and include drug companies, special interest groups and patients themselves (Conrad, Reference Conrad2007).
Major depressive disorder is relevant to suicide. In their book, Horwitz & Wakefield (2007) drew attention to the fact that the DSM-IV-TR takes no account of the context (except for bereavement) in which individuals find themselves, and consequently normal distress can be, and is, diagnosed as a disorder. Accordingly, when suicide follows a loss of some type, the normal human reaction of distress will be retrospectively diagnosed as a disorder. Suicide has been widely medicalized by well meaning community education, which recommends that if you are having thoughts of suicide, go not to your family and friends, the clergyman or a social worker, but to the doctor.
Until 1992, when the first National Mental Health Strategy (Australian Health Ministers, 1992) was implemented, mental health services in Australia dealt predominantly with mental disorders, that is, disorders meeting standardized criteria (such as the DSM-IV-TR). At this time a new category of ‘mental health problem’ was invented. ‘A mental health problem also [like a mental disorder] interferes with how a person thinks, feels, and behaves, but to a lesser extent than a mental illness … Mental health problems are more common and include the mental ill health that can be experienced temporarily as a reaction to the stresses of life’ (National Mental Health Strategy, 2009).
The motivation behind the invention of this category of ‘mental health problems’ is unclear. It appears to have been to make the temporary ‘reactions to the stresses of life’ the responsibility of health services. This is medicalization by government. It raises many issues which remain to be addressed in the scientific and philosophical literature.
NOTION: ALL SUICIDE IS THE RESULT OF MENTAL DISORDER
The view commonly expressed by some suicide researchers (and accepted by members of the public) is that 98% of suicide completers had a mental disorder (Bertolote et al., Reference Bertolote, Fleischmann, De Leo and Wasserman2004). Some researchers go further and conclude that even those suicide completers reported as having no evidence of psychiatric disorder, ‘probably have an underlying psychiatric process’, which investigations ‘failed to detect’ (Ernst et al., Reference Ernst, Lalovic, Lesage, Seguin, Tousignant and Turecki2004).
Reports of high rates of mental disorder among suicide completers are the results of ‘psychological autopsies’. All the evidence which can be gathered from records and witnesses regarding the thinking and behavior of the deceased is considered by researchers, and conclusions are drawn. However, psychological autopsies are retrospective studies; for good reasons, retrospective studies are accorded low scientific status in other areas of psychiatry or psychology. Those who consider the evidence may be invested in the outcomes, and although we aspire to objectivity, the avoidance of bias may be difficult (Selkin & Loya, Reference Selkin and Loya1979). There are various types of psychological autopsy (Scott et al., Reference Scott, Swartz and Warburton2006), leading to the belief that one study can not be compared to another (Abondo et al., Reference Abondo, Masson, Le Gueut and Milleet2008). Concerns have been raised about validity and reliability (Ogloff & Otto, Reference Ogloff and Otto1993; Biffl, Reference Biffl1996; Hawton et al., Reference Hawton, Appleby, Platt, Foster, Cooper and Malmberg1998; Werlang & Botega, Reference Werlang and Botega2003). And finally, ‘the vast majority have used ill-defined instruments’ (Pouliot & De Leo, Reference Pouliot and De Leo2006). Taken together, the evidence casts doubt on the findings of psychological autopsies.
NOTION: NOT ALL SUICIDE IS THE RESULT OF MENTAL DISORDER
There is no doubt that people who have a mental disorder are more likely to complete suicide than people who do not have a mental disorder. However, the pioneer suicide researcher Erwin Stengel (1964) estimated mental disorder was present in only 37% of those who complete suicide. Recent work (Wang & Stora, Reference Wang and Stora2009) found evidence of psychiatric or drug disorders in 61% of suicides in the Faroe Islands, and work from China found a ‘startlingly’ low rate of mental disorder among completers (Law & Liu, Reference Law and Liu2008). Research in India also found a low rate of mental disorder, but a high rate of ‘stresses stemming from social practices and perceptions’ among completers (Vijayakumar et al., Reference Vijayakumar, John, Pirkis and Whiteford2005; Jacob, Reference Jacob2008, Bastia & Kar, Reference Bastia and Kar2009). Of women from Goa who attempted or completed suicide, Maselko & Patel (2009) found mental disorder in 37% of cases, but significantly higher levels of exposure to violence and recent hunger.
Work by the current authors provides evidence of completed suicide by individuals who have not suffered mental disorder (Cheah et al., Reference Cheah, Schmitt and Pridmore2008; Pridmore & McArthur, Reference Pridmore and McArthur2008, Reference Pridmore and McArthur2009; Pridmore, Reference Pridmore2009).
Thus, the evidence is compelling that suicide may occur in people who do not have a mental disorder. For completeness, it is important to note that suicide is influenced by societal factors (Durkheim, Reference Durkheim1951), education level, socioeconomic status, gender, age and level of intoxication, and that genetics accounts for 30–55% of suicide risk (Voracek & Loibl, Reference Voracek and Loibl2007).
SYNTHESIS
A view widely held in the community is that suicide is always the result of mental disorder. The currency of this view is in part due to the fact that the average person has trouble comprehending or is unsettled by the notion of suicide, and the explanation that those who complete suicide are mentally disordered is convenient and comforting (‘I’m not mentally disordered so I don’t have to worry that I might complete suicide’). This view is also influenced by the findings of scientific studies called psychological autopsies. These studies are conducted by responsible researchers, but as discussed above, there is reason to read them with a degree of skepticism.
People with mental disorder are more likely to complete suicide than those who do not have mental disorder. For example, the life-time risk of suicide of people with major depression is 3.4% to 15% (Blair-West & Mellsop, Reference Blair-West and Mellsop2001). But this does not make the case that all those who complete suicide have a mental disorder.
Not all suicide is the result of mental disorder, and this can be substantiated by recent population (Wang & Stora, Reference Wang and Stora2009; Law & Liu, Reference Law and Liu2008; Jacob, Reference Jacob2008, Bastia & Kar, Reference Bastia and Kar2009; Maselko & Patel, Reference Maselko and Patel2009) and individual (Cheah et al., Reference Cheah, Schmitt and Pridmore2008; Pridmore & McArthur Reference Pridmore and McArthur2008, Reference Pridmore and McArthur2009; Pridmore Reference Pridmore2009) studies.
The lack of clear definitions contributes to current misunderstanding. The WHO definition makes health synonymous with well being (World Health Organization, 1948). In many medicine/legal arenas, this has been interpreted to mean that health departments are responsible for helping those who suffer life stressors, such as bullying, loss of hierarchy or shame. The leading mental health authority (DSM-IV-TR) states that no satisfactory definition of mental disorder is currently available. The DSM-IV-TR definition of personality disorder (a category in which events are followed by maladaptive responses) is imprecise, for example, with the words ‘behavior deviates markedly from the expectations of the individual’s culture’.
Medicalization is the process by which non-medical issues are re-classified as medical disorders, implying a responsibility for mental health workers. Following the 1992 Australian Health Ministers Conference, a new category of ‘mental health problem’ was invented. ‘Mental health problem’ has never been satisfactorily defined but ‘includes the mental ill health that can be experienced temporarily as a reaction to the stresses of life’.
Mental health workers in every state are guided by legislation on the topic of involuntary hospitalization. In general they are trained to assist those with mental disorder as defined by the DSM-IV-TR, and less prepared to become involved in the management of temporary ‘reactions to the stresses of life’. There is little medical/psychological literature to inform on the best ways to deal with stresses such as lost cheques and eviction, and to repeatedly ‘rescue’ people from normal life experiences may have a deleterious effect on their psychological development in the long term. Finally, public resources (hospital beds and clinician time) are always limited, and there is prioritization depending on the estimated level of need.
Given the absence of clear definitions of health and mental disorder, and the invention of the category of ‘mental health problem’, almost all of us at any point in our lives could be cast as the responsibility of mental health services.
Suicide is not the exclusive domain of mental disorder and should not be the exclusive domain of the health system. The highest incidence of suicide in the world at the moment is among the Inuit of Greenland, the First Nations of the USA and the Aboriginal People of Australia. This is understood to be the result of the destruction of the culture and personal support structure of these people, and sociologists (Durkheim, Reference Durkheim1951) have been demonstrating the importance of social factors in suicide for over 100 years. Accordingly, it is appropriate to reconsider the current practice of managing all people with suicidal thinking as if they are sick and in need of medical treatment. To reduce suicide rates a cultural change will be necessary, with an emphasis on increased educational, employment and housing opportunities, equality and fairness, and innovative approaches to the building of more integrated and supportive communities.
A busy general hospital Emergency Medicine department is not the place to manage social disintegration and disappointment. In many instances it is not mental health professionals who are the most appropriate, but a range of mature caring people who bring a range of life experiences to the task.
There is evidence that at times a period of asylum to allow a crisis to settle can be beneficial. Thus, an alternative suicide management system will require some form or residential arrangement. If the individual does not have a mental disorder, he/she should not reside on a psychiatric ward designed for those who are mentally disordered. However, if the current system is to be maintained and mental health professionals are to continue to be held responsible for those who suicide, more hospital beds and greater ease of compulsory detention will be required.
Returning to the status quo, like homicide, suicide is the killing of a human by a human. In homicide, the person who kills may be found not guilty (not responsible) if the Rule in McNaughton be satisfied. Should McNaughton not be satisfied, the presence of a mental disorder may be a mitigating factor, suggesting a reduced level of responsibility. In self-killing, it is rare for the individual not to understand the nature of the act. When a person does not understand the nature of the act, it cannot be suicide, which by definition is the intentional killing of oneself.
However, the presence of mental disorder may reduce the level of responsibility.
Mental health workers in all jurisdictions are guided by a mental health act which allows for compulsory admission to a psychiatric facility. In most legislation there must be a mental disorder and a danger to the individual or others. In most legislation, mental disorder is a psychotic disorder, with evidence in the form of impaired mood, thinking, perception or behavior. Most legislation has been written to expressly exclude people with a personality disorder, and protect the liberty people who are simply different.
Difficult situation arise for clinicians. One of these is when an individual who is assessed as being at suicidal risk refuses to be admitted to hospital and cannot be compulsorily detained because of the absence of a mental disorder, as defined by the existing legislation. Should such a person be allowed to leave and subsequently complete suicide, the clinician may face criticism from her/his peers and the courts, because of a perceived failure of duty of care. Another difficult situation is when a person presents at an emergency medicine department seeking admission and making claims of suicidal thoughts, but lacks evidence of mental disorder and appears to be seeking admission for social reasons. Should admission be refused and suicide ensue, the clinician is again exposed to criticism.
Defensive medical practice is to admit everyone who presents to hospital requesting psychiatric admission, but this works to the detriment of people with serious mental disorder.
Finally, the law of the land does not automatically dictate medical/nursing ethics. While medical and nursing personnel must, of course, obey the law, they must also operate within the confines of professional ethics. These ethical considerations include an embargo on providing intrusive interventions to people who lack an illness/disease/disorder. Thus, medical/nursing personnel frequently find themselves minced in a clash between the law of the land and their professional ethics.
When determining, in the aftermath of a suicide, the appropriateness or otherwise of the activities of mental health workers, these factors need to be considered: 1) suicide can occur in the absence of mental disorder; 2) diagnosis is hampered by imprecise definitions and clinician opinions frequently differ (the more so when one has the wisdom of hindsight and another not); 3) the invention of the category of ‘mental health problem’ implies that health departments have a responsibility for persons with no mental disorder who are having a reaction to the stresses of life (something for which they are not trained, and in general do not believe to be their core business); 4) wherever possible, responsibility should rest with the individual (individual autonomy is preferred over the paternal response); 5) there may be conflict between the dictates of state law and professional ethics; and 6) health as defined by the WHO is not simply the responsibility of health departments, but involves a contribution to public life of all public agencies and community members.
