Factors in the Assessment of Suicidality in Youth

Abstract

Suicide remains a leading cause of death among youth, and suicide ideation and behavior are relatively common in both normal and clinical populations. Clinicians working with young people are often required to assess for the presence of suicidal ideation, suicidal behavior, and other risk factors, and to determine the level of risk. This paper provides the clinician with a summary of risk factors for youth suicide, as well as providing standardized terminology to enhance the clinician's assessment of suicidal ideation and behavior.

Introduction

Youth suicidal behavior and ideation (suicidality) remain a major public health concern because of their life-threatening nature and widespread prevalence. It is an issue that exists across psychiatric diagnoses and in varied socio-cultural populations. In fact, suicide is the third leading cause of death among youth (10-24 year olds), responsible for an estimated 4,600 deaths during 2004 in the US alone.¹ The importance of the problem of suicidality is highlighted by its prevalence in the general population. A 2005 national survey by the US Centers for Disease Control and Prevention (CDC) found that approximately 17% of adolescents had seriously considered suicide in the past year, 13% had made a suicide plan, and 8% had attempted suicide at least once.² Predictably, rates are even higher in depressed youth with approximately 60% reporting having thoughts of suicide and 30% actually attempting.³

Assessment of Youth Suicidality

Accurate assessment of suicidal ideation and behavior is a critical and necessary component of a comprehensive clinical evaluation of children and adolescents. Knowledge of the risk factors for suicide is a key prerequisite for assessment of risk. Risk factors have been identified by studies of clinical and normal populations as well as case control and psychological autopsy studies, and have been shown to vary with gender^4,5 and age.⁵

Risk Factors for Youth Suicidality

Psychiatric Diagnosis

Up to 90% of young people who complete suicide have at least one psychiatric diagnosis as determined by psychological autopsy, and up to 70% experience two or more diagnoses.⁴ Depression is the most common diagnosis in adolescents who complete suicide⁴ and is highly prevalent in those with suicidal ideation and attempts.⁶ Other major suicide risk factors that should be assessed and potentially targeted for treatment include anxiety disorders, substance abuse, and conduct and antisocial disorders, the latter two diagnoses being significantly more prevalent in male suicide completers.^4,5 Brent and colleagues⁵ found that the suicide rate was greater in older adolescents (16 years and older) as they experienced higher rates of psychopathology, in particular substance abuse and greater motivation for suicide. While research evidence on the association between psychosis and suicidality is limited,⁷ risk of suicide for individuals with schizophrenia has been shown to be high during the first episode, ⁸ which often occurs in late adolescence, and during the five years following initial diagnosis.⁹

Past Suicidal Behavior

Any suicide attempt, above and beyond being depressed, is a major risk factor for completed suicide^4,5 and a further attempt.^10,11 Past suicidal ideation has also been found to increase risk of future suicidal ideation, suicide attempt, and depression.¹¹ Adolescents who are both depressed and have attempted suicide are at extremely high risk for both recurrent suicidal behavior and for completed suicide.^7,11,12,13

Family Factors

Adolescents who complete suicide are more likely to come from a family with a history of suicide.¹⁴ They are also more likely to live in non-intact families and home environments characterized by high levels of conflict, poor attachment, and problematic communication.^11,14 Assessment of parental depression and substance abuse is also indicated, as they are associated with adolescent suicide.¹⁴ Familial transmission of suicide almost invariably occurs with familial transmission of mood disorder, suggesting that clinicians also be aware of family history of mental illness, mood disorder in particular.¹⁵

Additional Risk Factors

Life stressors have been found to be significant risk factors for completed and attempted suicide in adolescents, particularly problems with authorities, academic difficulties, and relationship troubles (e.g. breaking up with a girlfriend or boyfriend).^10,16,14 Other risk factors for suicidal behavior to keep in mind include sexual abuse,^11,17 particularly with comorbid psychopathology,¹⁸ physical abuse,^19,20 low socio-economic status, impulsivity,¹¹ hopelessness,⁴ aggression,²¹ and poor social problem-solving skills.²² Gay, lesbian, and bisexual youth have been demonstrated to be at greater risk of suicidal ideation, ²³ and suicide attempt ^23,24 compared with heterosexual youth. Risk appears to be stronger in males than females ^24,25.

Defining Suicidality

The ability of clinicians to accurately assess suicidality has been compromised by a lack of well-defined terminology and understanding as to what constitutes suicidal behavior.²⁶ Consequently, behaviors that are not suicidal are labeled as such, while suicidal behaviors may be missed, leading to misinterpretation in both clinical and research settings. Non-suicidal self-harm behaviors, such as “self-mutilation”, done purely for reasons other than ending one's life, and suicidal acts are frequently mistaken for one another.

The use of a standardized method of classifying the spectrum of suicidal ideation and behavior is recommended.²⁷ Adherence to a standardized system enables the use of consistent terminology which may help track change in a patient's suicidal behavior over time. Its application may also help to avoid the use of inaccurate synonyms for suicidality and promote more precise communication between clinicians involved in a patient's care. The following is a description of terminology defining suicidal ideation and behavior. Behavioral definitions were adapted from the Columbia Suicide History Form.²⁸ This approach is hierarchical, ranging from suicidal ideation to attempted suicide. Evidence supports a hierarchical model of suicidality in youth.²⁹ The terminology and related factors to consider are summarized in Table 1.

Table 1.

Key definitions and important elements in addressing suicidality in youth.^*

Important Clinical Events To Assess	Definitions	Tips and Details to Ask About
Suicide Attempt	A potentially self-injurious act committed with at least some wish or intent to die as a result of the act. Only Some Intent is Needed: The behavior was, at least in part, thought of as a method to kill oneself. Intent does not have to be 100% suicidal. If there is any intent or desire to die associated with the act, then it can be considered an actual suicide attempt. For example, if any small part of them was trying to hurt themselves because they wanted to die, even if 99% of them wanted to make their parent angry or feel sorry for them, the behavior is classified as a suicide attempt. No Injury Needed: There does not have to be any injury or harm, just the potential for injury or harm. For example, if a person pulls trigger while gun is in the mouth but the gun jams and no injury results, this is considered a suicide attempt; or in the case of overdose, once any pills are ingested, it is considered an attempt. Intent Can Be Inferred: Even if an individual denies intent/wish to die, it may be inferred clinically from the behavior or circumstances. For example, if someone denies intent to die, but they thought that what they did could be lethal, intent may be inferred.	Although injury is not needed to classify an act as suicidal, both actual and potential medical injury/lethality should be assessed in detail as they are important aspects of clinical severity; Get as many details about method as possible (how many pills, what kind, what was used to cut); Impulsive or planned (How long had they been thinking about it?); Motives: Why they tried to hurt self (to die, end pain, to get attention or revenge, make someone angry). Children and adolescents often have mixed motives and it is helpful to discuss in detail why they did what they did; Attempt to find out how they felt afterwards. For example, relief may suggest a decreased risk of reattempt while disappointment might suggest greater risk of reattempt.
Interrupted Attempt	When the person is interrupted (by an outside circumstance) from starting the potentially self-injurious act. Examples include: Overdose: Person has pills in hand but is stopped from ingesting them. Shooting: Person has a gun pointed toward him-/herself; gun is taken away or the person is somehow prevented from pulling trigger by someone else. Jumping: Person is poised to jump, but is grabbed and taken down from the ledge. Hanging: Person has a noose around their neck but has not yet started to hang, and is stopped by someone.	Probe the individual's reaction to being stopped from attempting. (“What was your reaction to being stopped from attempting to kill yourself?”; “How did you feel?”) Enquire about any plans to reattempt. (“Have you been thinking about trying again?”; “Have you taken any steps toward another attempt?”)
Aborted Attempt	When person begins to take steps toward making a suicide attempt, but stops self before engaging in any self-destructive behavior. Similar to interrupted attempts, except that the individual stops him-/herself, instead of being stopped by someone or something else.	If they stopped short of making an attempt, inquire why they did so (got scared, second thoughts); Enquire about any plans to reattempt. (“Have you been thinking about trying again?”; “Have you taken any steps toward another attempt?”)
Other Suicidal Behaviors : Preparatory Acts	Preparatory acts are actions taken towards a suicide attempt or death. Examples: Collecting or buying pills, purchasing a gun, giving things away, writing a will, writing a suicide note.	If an individual has suicidal thoughts that include a plan, enquire as to whether they have made any preparations for a suicide attempt or their own death. (“Have your plans for suicide included any actions?”; “Have you prepared for a suicide attempt?”; “Have you written a suicide note?”; “Have you gathered any things to be used in a suicide attempt?”)
Non-Suicidal Self-Injurious Behavior	To Be Distinguished from Suicidal Behavior Self- injurious behavior with no intent to die. Behavior is intended purely and 100% for other reasons, either to relieve distress (“self-mutilation” e.g. superficial cuts or scratches, hitting/banging, or burns); or to effect change in others or the environment (“instrumental” e.g. to get attention, sympathy, get out of the house or school activity). Hence, it is distinct from suicidal behavior. Non-suicidal self-injurious behavior is commonly thought of as “self-mutilation” and “instrumental behavior”	“Did you want to die, or was there another reason for harming yourself?”
Suicidal Ideation	There are two types of suicidal ideation: Passive: thoughts about a wish to be dead Active: thoughts of wanting to end one's life, commit suicide	Suicidal ideation can be probed through the use of questions such as “Have you had any thoughts about wanting to be dead?” or “Have you had thoughts about ending your life?”. Other ideation features to ask about include whether there has been a plan for suicide, and if so, has there been any intention to act upon the plan. Aspects of suicidal ideation to inquire about include: Frequency (How often do thoughts occur?), Duration (How long do thoughts last?), Deterrents (What stops the individual from wanting to be dead? Such as, family, religion, friendship, pet, event in the future.). If active thoughts are present, inquire about: Methods (ways to kill oneself), Plan for specific suicide attempt, Intent to act on thoughts (How likely is the individual to carry out suicide attempt?)

^*Behavioral definitions adapted from The Columbia Suicide History Form developed by John Mann and colleagues.²⁵ Suicidality as defined in this table can be assessed in a brief, low burden way using The Columbia – Suicide Severity Rating Scale (available from posnerk@childpsych.columbia.edu).

Suicidal Ideation

Two broad types of suicidal ideation have been described: passive suicidal thoughts characterized by the wish or desire to be dead, and active suicidal thoughts characterized by specific thoughts about the act of killing oneself or committing suicide. Both types of ideation have been found to be predictive of completed suicide³⁰ suggesting the importance of noting passive as well as active thoughts. The severity of suicidal thoughts can be briefly measured by their frequency and duration, the existence of deterrents to acting upon thoughts, and reasons for the thoughts. If thoughts of suicide are evident asking about the existence of suicide plans and any intent to act upon them is indicated.

Suicidal Behavior

Beyond thoughts about suicide are a range of suicidal behaviors, including suicide attempts, interrupted attempts, aborted attempts, and other preparatory suicidal acts. Preparatory acts include any behavior towards a suicide attempt (e.g. collecting materials, visiting a potential site for a suicide attempt) or death (e.g., giving away possessions, writing a suicide note). A suicide attempt is defined as not only self-harming behavior but also the intent behind the behavior, that is, the reason for engaging in a potentially self-injurious act (e.g., Why did they engage in the behavior? To kill themselves or for some other reason?). Hence, assessing suicidal intent becomes an important component of determining suicidal risk. Common misunderstandings about a suicide attempt are that physical injury must occur and suicidal intent has to be 100% of the motivation. Children and adolescents often have mixed motives, that is, suicidal and non-suicidal motivations, for their self-injurious behaviors (e.g., attempting suicide as a way of coping with intolerable pain and a method of expressing anger with another's behavior). Any amount of intent to die, however, is sufficient to label an act as suicidal (e.g., “Part of me wanted to die but a big part of me wanted to make my Mom angry”). A behavior should be labeled as non-suicidal self-injury, then, if the individual committing the act is doing it purely for reasons other than to end their life (no intent to die) (e.g., “self-mutilation” when an individual tries to hurt themselves to relieve pain or to stimulate feelings, such as superficial cuts to the upper thigh made to feel “real” or feel “something”). Suicidal behavior and non-suicidal self-injury are therefore distinguished by the presence or absence of suicidal intent. In some cases suicidal intent may be concealed or denied (e.g., a youth who jumps from a high story) whereby the only reasonable intention that can be inferred is suicide), however the clinician has the scope to infer suicidal intent, given details of the attempt that suggest that suicide was the only plausible explanation for the behavior. Suicidal intent can also be inferred if the young person denies intent yet believes that their self-harming actions could have caused their own death ²⁵.

Despite their level of intent, children and adolescents sometimes do not have an accurate comprehension of the lethality of the means employed for self-harm. For example, a young adolescent may believe a small overdose will lead to certain death. Clinicians should be careful not to dismiss any attempt as being trivial or non-serious. At times, suicidal occurrences are inappropriately dismissed because of seemingly low levels of lethality (e.g., taking three pain relievers). Evidence from an adult study suggests that the relationship between suicidal intent and the severity of self-injury is moderated by the level of accuracy in the person's expectations about the likelihood of dying as a result of his or her behavior.²⁸ Higher levels of suicidal intent are therefore associated with increased lethality for individuals who have more accurate conceptions of their likelihood of dying than those individuals who have inaccurate expectations. This finding sheds light on the otherwise counter-intuitive claim that there is no direct association between intent and lethality.^31,32

In some cases a youth may take steps towards a suicide attempt but is interrupted by someone or something before the potential for harm occurs. An example of such an “interrupted attempt” is a case in which a teenager has a gun in his hand and the gun is grabbed and taken away by his mother. This case can be contrasted with an “aborted attempt”, in which a youth stops him- or herself from attempting. For example, a teen becomes overwhelmed while preparing for a suicide attempt using a handful of stockpiled pills and flushes them down the toilet.

Assessment Considerations

Research indicates that information garnered from a young person regarding suicidal intent is likely to be reliable.³³ Studies on school-based screening strategies for suicidality also show reliable and valid self-reported indications of suicidal behavior and ideation. Clinicians should typically rely on the young person as the primary informant given consistent findings that parents are often unaware of their child's suicidality.^34,35 Furthermore, studies that inquire about suicidal ideation and behavior in normal populations appear to be innocuous, as results provide no indication of iatrogenic effects of suicide screening. Students, even those who are high-risk (reporting depressive symptoms, substance-use problems, or previous suicide attempts), do not generally experience increased distress or suicidal ideation as a result of being questioned about suicidal feelings.³⁶ These findings provide firm evidence that challenges the myth that asking about suicidality promotes suicidality. Young people should be asked about their access to lethal means as this is an indicator of risk. Parent or guardian's involvement during assessment is important to provide information about suicidality (e.g., verbal threats to self harm, self inflicted wounds, morbid writing or drawings) and psychiatric symptoms that the young person may be experiencing but not able to detect, such as psychomotor agitation, or be unwilling to disclose, such as antisocial behaviors. Parents may be directed to restrict access to lethal means, such as knives or firearms, and have a key role in monitoring for ongoing suicidality and maintaining close contact with health professionals while risk is elevated.

Helpful Assessment Tools

Although there are many measures of suicidality they are typically either high burden, do not address the full spectrum of suicidality, or are not directed at non-research settings, clinical practice, or primary care. Comprehensive reviews of assessment measures can be found elsewhere ^37,38. Two selected measures that may be helpful to clinicians and practitioners are described, A low-burden measure of the spectrum of suicidal ideation and behavior, the Columbia – Suicide Severity Rating Scale (C-SSRS) was developed in the NIMH Treatment of Adolescent Suicide Attempters Study to assess severity and track suicidal events through any treatment (Inquiries: posnerk@childpsych.columbia.edu). It is also the prospective counterpart to the system developed by Columbia investigators for the Food & Drug Administration in their analysis of the association between suicidality and antidepressant treatment in pediatric populations.³⁹ It is a clinical interview providing a summary of both ideation and behavior that can be administered during any evaluation or risk assessment to identify the level and type of suicidality present. It can also be used during treatment to monitor for clinical worsening.

Self-report questionnaires can be useful for adolescents who refuse to talk or have difficulty expressing their thoughts and feelings verbally. Assessing factors associated with risk for suicidal behavior includes not only evaluation of suicide ideation but also related constructs. One measure that has been developed to assess these constructs and to be sensitive to changes in level of suicidality is the Suicide Assessment Scale.⁴⁰ This measure includes constructs that have been found to be associated with imminent suicide risk, such as poor frustration tolerance, lack of resourcefulness, sadness, hypersensitivity, and perceived loss of control. The scale has been shown to differentiate suicide attempters from non attempters, to be predictive of future suicide and to be sensitive to change in suicidal state.^41,42

Clinical Considerations

Assessing suicidality in youth is a critical clinical task. In conducting an assessment, the clinician is recommended to complete a thorough diagnostic assessment given the association between psychiatric diagnosis and suicidality. It is recommended that the clinician assess identified risk factors for suicidality described above, as well as take a detailed account of the patient's history of suicidality given the association between past and future suicidal behavior. Clinicians need to take a comprehensive family history of mental illness, psychiatric treatment, and attempted and completed suicide. Discussing the limits of practitioner-young person confidentiality is a necessary component of an assessment. If at all possible, young people should be given advanced warning of a breech in confidentiality and the reasons for it (e.g., ensure the safety of the patient). Adoption of standardized definitions of suicidality is advised to increase the accuracy of suicidal terminology, which in turn will benefit the young person and caregivers. Suicide risk assessment may be informed by administering measures such as the C-SSRS and Suicide Assessment Scale, which enable the estimation of change over time. Given the recent controversy around suicidality during antidepressant treatment (e.g., Hammad and colleagues³⁴), regular and systematic monitoring of suicidality, commencing with a pre-treatment baseline, is recommended instead of relying on spontaneous reports from the patient. This may assist in tracking treatment progress and in determining whether any suicidality experienced during treatment is treatment- or disease-specific thus enabling a better understanding of suicidal state and risk.