Assessing or treating a suicidal patient represents probably the most challenging task clinicians have to face in their professional life, both intellectually and emotionally. Not many clinical interactions generate emotional responses quite as intense as dealing with an individual who has recently survived a serious suicide attempt or is contemplating a fatal act in the near future. There are multiple reasons behind these responses. Firstly, as clinicians we have been trained to objectively and systematically assess symptoms and signs of illness in order to reach a sound diagnosis. On the basis of this diagnosis—whether provisional or confirmed—we have the responsibility of formulating and evaluating an effective treatment plan. In the case of suicidality, what we are often faced with is not a diagnosable illness but a behaviour to which many different psychiatric illnesses of various severities may contribute. In some cases, no detectable psychiatric illness is present. Secondly, as clinicians we have been invested with the responsibility of ensuring the well being of our patients and, as much as possible, to avert complications arising from illnesses. In recent times, clinicians have experienced an increasing level of scrutiny into the validity of the diagnoses attributed to patients and the effectiveness of the treatment strategies implemented. The threat of litigation has greatly affected the way clinicians approach clinical interactions with their patients. As a result, being able to manage the countless issues surrounding the assessment and management of individuals presenting with suicide risk represents one of the fundamental skills clinicians have to develop and preserve throughout their professional life. In this article, we will attempt to describe and clarify the most salient aspects of dealing with suicidality in every-day clinical practice.

The first obstacle faced in approaching the suicidal patient is certainly how to appropriately define the behaviour we are about to assess. In recent years, there has been considerable debate surrounding the validity of current definitions, in particular for what relates to our ability to differentiate suicidal behaviour from self-harm behaviour without suicide intent [1,2]. In this chapter the term “suicidal” will be used to describe any behaviour or ideation involving harm to self irrespective of the level of intent to die. The reason for approaching the subject from this stance is two-fold. On one hand, as stressed by [3], the motives behind self-injurious behaviours and suicidal ideation are sometimes so difficult to establish that clinicians will never be in the position to exclude suicide intent unequivocally [4]. The other consideration to be made is that despite the many differences observed between individuals with self-harm and suicidal behaviour, the absence of reported suicidal intent is not sufficient to exclude the risk of completed suicide. On the contrary, the presence of self-injurious behaviour significantly increases the risk of suicide later in life. In a recent cohort study of individuals presenting to an emergency department following self-harm, the authors found that in the 4-years following presentation the rate of completed suicide was 30 times greater than the one expected in the general population [5]. In addition, mental health clinicians should be aware of the frequent fluctuations in suicidality individuals may manifest, these ranging from death wishes to completed suicide, with intervals free from any suicide ideation [6,7].

Before delving into the clinical aspects of suicide, it is important to make it clear that suicidal behaviour cannot be fully understood if interpreted only as a consequence of psychiatric illness. In fact, suicidal behaviour is a multifactorial event, extending beyond the psychiatric or psychopathological realm of human experience.

The global mortality rate from suicide in 2000 was 16/100,000, with approximately one million individuals taking their life that year. For countries that report mortality data to the World Health Organization (WHO), the rate of suicide has increased by 60% over the last 45 years [8], with suicide being represented in the top ten leading causes of death for all ages, but in the top three for the age group of 15 - 34 years [8]. Sadly, for every completed suicide at least six other people related to the deceased remain deeply affected [9].

“Attempted suicides” (non-fatal suicidal behaviours) are estimated to be up to 20 times as frequent as suicide [6]. Data on the incidence of suicide ideation is unfortunately not as reliable as we would hope, but we have some information regarding the number of people contemplating suicide during their lives. In a recent international multi-centre community survey study, up to 25% of the interviewees reported having contemplated suicide during their life, with up to 15% going as far as planning a suicide act. It is also worth considering that in the same study, between 0.4% and 4.2% of the people surveyed reported at least one suicide attempt during their life [10].

Without a doubt, the most important clinical tool psychiatrists require in approaching a suicidal patient is the ability to conduct an efficient and effective risk assessment. In this procedure, the various aspects of patients’ demographic data and personal and family history are translated into risk or protective factors on the basis of what available research data reveal. The ultimate objective of the interview is to transform the information gathered into a statement that includes the overall level of risk for that individual. In doing that, we have to remember that the amount and reliability of the information we obtain during a risk assessment will be closely related to the strength of the therapeutic relationship with the patient we are interviewing: the longer the relationship (and the therapeutic alliance), the higher the probability of a positive impact on the outcomes.

Before going any further, it is important to understand that a risk assessment is in no way a tool we can utilize to predict suicide. Unfortunately, even the most comprehensive risk assessment performed will only be useful to gauge the level of risk for that particular individual at that particular time. In fact, while some risk and protective factors do not change over time, many others do, with a significant influence on the choices clinicians have to make in managing patients. This means that a patient’s risk assessment has to be regularly revised and adjourned to reflect the developments in the individual’s clinical presentation, life circumstances and treatment, if implemented [11]. In recent years, a number of professional bodies have published and regularly updated practice guidelines for the assessment and management of individuals at risk of suicide, such as those produced by the National Institute for Clinical Excellence [12], the American Psychiatric Association [13], and the American Academy of Child and Adolescent Psychiatry [14].

In assessing risk, we soon realize that some of the variables we observe during the interview are beyond our control. For example, there are a number of demographic variables that have a strong correlation with suicide rates. One of the most significant is gender. In fact, even though women are more likely to be admitted to a hospital following a suicide attempt, men complete suicide in considerably greater numbers, with around three to four men completing suicide for every woman [15]. These differences appear to have been consistent over time and apart for very few exceptions, consistent across cultures [15]. A number of variables have been hypothesized as linked to this discrepancy (the “gender paradox”): men appear to choose more lethal methods; they are more likely to manifest impulsive behavior, and to abuse substances. Moreover, men are much less prone to seeking help if in need [7,16].

Among demographic data, we also find age as a significant variable needing consideration when assessing risk. In fact, suicide rates steadily increase with age, to reach the level of 50/100,000 in the male elderly group as represented in Figure 1, which refers to the average rate of suicide in the world. 

As also noted in the figure, whereas suicide rate for women shows a gradual increase with age, men’ suicide rate first peaks in middle adulthood, showing a marginal decline until the dramatic increase for the 75+ group [17]. Marital status is another important factor to be carefully examined even if, because of barriers in collecting reliable data, it appears to have been researched less than the risk factors discussed so far. Despite this, there is sufficient evidence to suggest a link between relationship status and suicidal behaviour. In general, research supports a positive association between being single and suicide rates. The association appears stronger if we consider divorce and—especially—separation [18,19]. A study found separated men to be about six times more likely to die from suicide than their married counterpart [20]. The impact of being single or separated on suicide rates is for men far greater than for women [18]. It has been suggested that in women the effect of being single or separated is mitigated by the protective role of parenthood,

which has been found to be stronger in female subjects [21]. Also, women would generally benefit from far greater social networks than men, and this would offer both support for emotional ventilation and opportunity for new relationships [19].

Beyond demographic characteristics, suicide risk will also be affected by the genetic make-up. For example, it has been observed that the risk of suicide in individuals with family history of suicide is twice bigger than in the general population, independently from the presence of severe mental illness [22]. Secondly, the improvement of genetic modeling has allowed confirming that adopted individuals who complete suicide are more likely to have a family history of suicide among their biological relatives. Again, the impact of genetic predisposition was independent from psychiatric disorders commonly associated with suicide [23].

It is beyond the scope of this article to discuss all possible risk factors linked to an increased likelihood of suicide in the individuals we assess and treat (Table 1). Our objective is primarily to alert clinicians about the fact that suicide risk assessment requires flexibility and constant adaptability in our perspectives during the interview. Many different and multifaceted aspects of human experience have to be taken into account to carry out a thorough suicide risk assessment; if we do not maintain a “wide open angle” in our examination, we run the risk of becoming excessively focused on a partial view of the individual’s own experience, thus losing our ability to perform a valid evaluation.

Having considered some of the demographic and biological variables impacting on the level of risk, we now come to examine probably the most significant group of unmodifiable risk factors clinicians cannot afford to ignore. A history of attempted suicide, suicide ideation or even having been hospitalized for a major mental illness

significantly exposes individuals to an increased risk of suicidal behaviour. An overwhelming amount of research data confirms the magnitude of the impact those factors have on our ability to predict future behaviour. For example, around 5% of individuals who attempted suicide were found to complete suicide within 10 years [25]. Not only previous history of suicide attempts should raise the level of clinical alert, but also individuals’ past history of suicide ideation. [26], in a population of outpatients, measured the level of suicide ideation at the time of presentation and the severity of the suicide ideation when at its worst in the life span. The authors found that past suicide ideation was a far better predictor of future suicide than the level of suicide ideation at the time of assessment, with odds ratios of 13.85 vs 5.42. More recently, a great deal of attention has also been focused on the relationship between hospitalization and suicide risk. Dramatic changes have occurred over the last 30 years in the way mentally ill individuals are treated, with a significant shift in the emphasis placed on the utilization of hospital/inpatient treatment [27]. We need to be mindful of the fact that several studies have established a clear relationship between suicide and certain aspects of hospital treatment. Firstly, suicide is highest the first few days following hospital admission and the first few days following discharge. Secondly, suicide risk increases along with number of admissions to an inpatient facility [28,29]. There is much speculation over the reasons for such relationships, which are extremely relevant to ‘front line’ psychiatrists, often the only clinicians involved in the decision over whether to admit or discharge an individual at risk of suicide. We are used to think that hospital treatment represents the best way to avert suicide when individuals are considered to be at high risk. Unfortunately, in many occasions the hospital environment is inadequate to appropriately support the suicidal patient, Moreover, the stigma attached to psychiatric hospitalisation often hinders the process of recovery. The result is, in some cases, that the suicidal person perceives the prospect of a hospital admission as the “last straw”, paradoxically facilitating suicide completion. Actually hospitalisation should always be seen as one of the components of a sound and well thought out management plan, and not as an easy way to prevent suicide or speed up the healing process [30].

Modifiable risk factors are those factors we as clinicians can have some impact on. The same way as for the unmodifiable risk factors, we will only concentrate on those we cannot afford to ignore in any circumstance.

Needless to say, any form of suicide ideation is regarded as clinically important, and as such requiring the utmost diligence in assessing the nature and the extent of the suicidal thoughts. As mentioned above, individuals who present with suicide ideation are more likely to complete suicide regardless of the nature and severity of the mental illness they suffer from [26]. Very often individuals perceive those thoughts as distressing, but because of the stigma attached to suicidality, they may not be as forthcoming as we would hope. That is why a careful and tactful exploration of patients’ suicide ideation requires considerable skills and experience, in particular with specific populations such as the elderly and the adolescents. It is important to stress that exploring suicide ideation does not to increase patients’ risk of suicide. In fact, frequently patients find comfort in being able to discuss their struggle and ventilate their self-destructive thoughts. In the evaluation of patients’ level of ideation, it is essential to investigate the presence of any intent to act on those thoughts and even more any plans enabling the patients to perform the act. Moreover, it is the clinician’s responsibility to carefully explore the availability of lethal means to carry out the suicidal act and ensure steps are taken to limit access to them. It cannot be stressed enough that the more detailed the suicide plan, the higher the risk of suicide. However, it is also important to consider that, in many cases, individuals who have a high level of intent may be reluctant to disclose any information that could lead to their suicide plans being hindered.

Of all risk factors for suicide that have been considered so far, researchers have identified mental illness as the most significant. Over 90% of individuals completing suicideare affected by at least one mental illness, with depression as the most commonly reported (at least in 30% of cases). Substance abuse disorders are also frequent (17.6%), as well as schizophrenia (14.1%), and personality disorders (13.0%) [31]. Lifetime risk of suicide in psychiatric illnesses such as depression, schizophrenia and alcohol use disorder is reported to be around 5% - 15% [31]. It is also important to consider that even though a significant number of individuals who die from suicide have never had any medical contact, between 40% and 60% have seen a clinician in the month preceding suicide [32]. Especially in the western world, the relationship between suicide risk and mental illness is so strong that no prevention strategy would succeed without effective treatment of the disorder individuals present with [33,34].

A risk assessment cannot be considered complete until both risk and protective factors are carefully evaluated. It has been well established how some factors such as responsibility to family and children, religiosity and social support can have a positive impact on suicide rates [35]. A self-report questionnaire, the Reasons for Living Inventory, has been found useful in clinical practice to document the presence of possible preventive factors [36]. Among the protective factors explored by researchers over the years, therapeutic alliance is probably one of the most relevant to us as clinicians [37]. Regardless the nature of the treatment offered, the interview with a suicidal individual requires the psychiatrist to make every possible effort to build trust and a sense of co-operation. It is only when sufficient mutual respect is established that the patient’s experience can be fully explored and a reliable assessment carried out. Empathy is still one of the most powerful tools clinicians can utilize in caring for suicidal individuals. However, it may be difficult to maintain a sufficient level of empathy when our therapeutic efforts are contrasted by the need to ensure the safety of our patients [38].

To this date, there is no recognized tool able to reliably predict suicide. The only certainty is that all clinicians are expected to perform a competent risk assessment, no matter what the circumstances leading to the assessment are [39]. To facilitate a productive interviewing, we need to provide a safe and comfortable environment. In fact, where the interview is performed will have a significant impact on the outcome. A suitable room should be selected to ensure that confidential and emotionally sensitive information could be comfortably disclosed to the clinician without the risk of distraction or intimidation. As much as possible, the clinician should put in place measures to be able to dedicate to the patient the time required for the assessment without interruptions (and, helas, sometimes this time is well beyond the hour!). In preparing for a risk assessment, it is important to ensure that the mental and physical state of the patient are compatible with the interview that is about to be undertaken.

Besides information deriving from the interview with the suicidal person, itself, a sound risk assessment could benefit from the availability of a “collateral” history. Frequently, it is on the basis of information disclosed by partner or spouse, relatives or friends that individuals first come to the attention of clinicians. Despite this, just as often we fail to seek or acknowledge the invaluable information significant others can provide. In fact, on many occasions the collateral history we obtain will be the only reason for deeming the patient at high risk and initiating treatment. Common examples of this case scenario are patients who purposefully conceal their suicide intent and plan, or those who are affected by mental illness so severe that they are unable to divulge their thoughts and feelings in an understandable way.

There is yet another important task clinicians have to master in modern day practice, which is the one of thoroughly documenting the information gathered. It goes without saying how essential it can be for a psychiatrist or a mental health service to have sufficient information in the clinical file in the case of litigation or scrutiny [40]. Moreover, in current times it is most unusual for a patient presenting to a mental health service to receive psychiatric care from a single clinician. Usually, several doctors and mental health professionals become involved —in different ways and at different stages—in the management process. As a result of this, each clinicians has the added responsibility to communicate as effectively as possible the information collated and what has been done up to that point to avoid unnecessary delays or omissions.

Depressive illnesses affect around 5% of the population. The risk of completed suicide has been unequivocally linked to depression in its various forms [31]. An estimate of the suicide risk in a population presenting with mood disorders compared to the general population was 20-fold for major depression, 15-fold for bipolar disorder and 12-fold for dysthymia [41]. More recent studies have confirmed the role of depression as a precursor of suicide, especially when protracted and refractory to treatment [42]. In the assessment of a depressed individual, hopelessness requires special consideration. Its significance is not only related to its diagnostic value but also as the expression of emotional turmoil individuals experience, regardless of the specific illness being treated. In fact, a significant number of studies have confirmed a close relationship between the experience of hopelessness and suicide risk independently of the severity of the illness itself [43]. So much so, that the Beck Hopelessness Scale is still one of the most commonly used and valued psychometric tools utilized in clinical practice [44]. As mentioned above, suicide risk is significantly high in bipolar disorder, where episodes of depression alternate with episodes of mania of varying frequency and severity. Even though suicidal behaviour can occur during a period of mania, the vast majority of completed suicides in individuals with bipolar disorder have been found to occur during a depressive episode [45].

Schizophrenia is one of the most severe and disruptive mental illnesses, affecting around 0.5% - 1% of the population. It is characterised by impairment in perception, cognition, and social and occupational functioning. Patients typically manifest a varying combination of positive and negative symptoms requiring ongoing antipsychotic treatment. The lifetime suicide risk for individuals with schizophrenia has historically been quoted at around 10% of affected individuals. [46] have recently challenged that notion and estimated that risk at about 5%, also finding most of the suicides to occur in the period following the illness onset. In the case of schizophrenia, it is well recognised that the older the patient, the smaller the risk of suicide. The rate of suicide in schizophrenia is influenced by some factors that are illness specific. It has been found that factors such as motor restlessness fear of mental disintegration and poor adherence to treatment impact on the rate of suicide just as much as more general factors like depressive illness or previous suicide attempts [47]. It is also important to stress again how the risk of suicide in schizophrenia peaks in the first few days following admission to hospital and following discharge. Even though we institute treatment as effectively as we can, with the best of intentions, the process of hospitalisation can be extremely traumatic. In the same way, the period after discharge can be particularly difficult, especially in terms of adjusting to life with a newly diagnosed illness that carries little prospects of satisfactory recovery and strong stigmatization from community members. Monitoring of mental state and compliance enhancing strategies are particularly important in this population.

Substance abuse (alcohol abuse in particular) represents a social phenomenon with significant repercussions on people’s physical and mental health. In regard to suicide risk, alcohol abuse is considered second only to depresssion as a precursor to completed suicide [48]. The lifetime risk of suicide in alcoholics varies considerably depending on the reporting habits and the overall rate of suicideof the country involved, but it is estimated to be around 3% [49]. Most attempted and completed suicides in alcoholics appear to be related to a depressive episode, stressful live event, loss of support or disruption of interpersonal relationships [50]. The impact substances have on patients’ level of rationality and impulsivity makes the assessment of suicide risk all but secondary in the management of substance abuse.

Personality disorders are defined as enduring, inflexible and pervasive patterns of behaviour and inner experience that seriously deviate from cultural norms and cause significant distress or impairment in various areas of individuals’ functioning [13]. Of the various personality disorders known in clinical practice, the most frequently encountered in most settings is certainly Borderline Personality Disorder, which features suicidality and self-injurious behaviour as essential criteria of illness. Individuals affected by one or more personality disorders attempt and complete suicide just as much as individuals diagnosed with a depressive illness [51]. In fact, a review of psychological autopsy studies revealed that around 57% of individuals who had completed suicide had a diagnosis of personality disorder [52]. Moreover, personality disorders have been found to represent the second most common diagnosis after adjustment disorder in a population of suicide attempters [53]. The difficulty in fully appreciating the impact personality disorders have on suicidality is the fact that they very often coexist with other mental illnesses [54]. Moreover, the presence of one or more personality disorders represents a significant challenge for clinicians and mental health services, both because of the disruptive behaviours they often manifest and because of the permanent nature of the disorder.

Anxiety disorders represent the most common group of mental illnesses in the community, with almost 30% of the population affected by at least one of these disorders at some point in their life [55]. Even though trait anxiety is often reported as one of the factors increasing suicide risk, anxiety disorders have never been seen as independent risk factors for completed suicide. However, a recent review found that some anxiety disorders may increase suicide risk, especially when associated with other mental illnesses such as depression, bipolar disorder, or schizophrenia [56]. In view of this, it is essential to make every possible effort to treat any anxiety disorder individuals present with, also considering that their response to psychiatric interventions is often much better than for other mental illnesses.

Completed suicide represents the third leading causes of mortality in the adolescent population [8] Just under 15% will self harm and up to 45% report suicidal ideation during their life [47]. More than the adult population, adolescents are vulnerable to the effects of relationship breakdown, and physical, sexual and emotional abuse [57,58]. Most studies highlight how adolescents who attempt suicide show reduced distress tolerance skills and poor mood regulation [14]. Moreover, adolescents manifest higher levels of impulsivity and aggression compared to their adult counterpart with a significant impact on the likelyhood of self-destructive behaviours [59]. In an attempt to facilitate the assessment and treatment of adolescents presenting with suicidal behaviour [14] have developed practice guidelines to support clinicians in their effort to reduce the likelihood of completed suicide in this very vulnerable population.

The elderly population is probably the one presenting with the most challenges for mental health clinicians [60]. As already highlighted, men over 75 years of age carry the highest suicide rate of all, and have done that consistently for a long time. At the same time, the elderly population will be the most likely to have been seen by a medical practitioner in the month preceding their completed suicide. A number of factors make detection and management of suicide risk extremely difficult [61]. The most significant risk factors for suicidal behaviour identified in the elderly population are in fact factors such as physical illness, social isolation, relationship problems, and bereavement or loss [62]. As one can appreciate, it would be extremely difficult to identify at risk individuals on the basis of almost universal problems present in this population, like those listed above [63].

If the assessment of suicide risk can be perceived at times as challenging and overwhelming, the treatment of individuals at risk of suicide represents for clinicians a real test of psychological strength. In fact, managing suicidal individuals is often emotionally taxing to the point of deterring some clinicians from accepting referrals involving suicide risk. Unlike other medical disciplines, death is not generally acknowledged as a possible complication of illness in mental health. To cope with the increased sense of pressure, mental health clinicians often become vulnerable to counter-transference reactions that instead of facilitating treatment progress create more angst and anxiety. Rescue fantasies and attempts to reparent patients are only examples of the many forms counter-transference can take [64]. Effective anxiety management and consistent supervision are imperative for any clinician intending to venture into this treacherous field of medicine.

Dealing with suicidal patients represents a challenging clinical task, which requires competence and dedication. Apart from knowledge of protocols and therapeutic tools, clinicians are particularly requested to care. Being perceived by the suicidal patient as an insufficiently caring professional might constitute the last ring of a long chain of disappointments.