Conducting an accurate suicide risk assessment (SRA) is no simple task as there are a number of factors that influence an individual's level of suicidality and his/her willingness to share this information. Therefore, it is imperative that practitioners adopt a systematic approach to conducting and documenting the foreseeability that a patient will commit suicide.
PubMed was used to search for articles published in MEDLINE journals using the following keywords: suicide, risk assessment, measure, scale. Randomized trials and pilot, proof-of-concept publications investigating the use of specific SRA measurements were included in the review. The scales are described based on the prevailing opinions in psychiatry from the American Psychiatric Association's Textbook of Suicide Assessment and Management (2nd ed., 2012).
Although various SRA scales exist, experts in the field have repeatedly concluded that there is not any one scale that can predict who will commit suicide to any useful degree. However, when used along with the clinical interview, standardized suicide risk factor components of clinical and research scales remain crucial to gaining information often omitted by patients regarding thoughts and preparation for suicide. A summary of the most widely cited scales and approaches used in SRA is provided.
SRA remains a challenge largely due to the fact that suicidal behavior is multifactorial. As a result, risk formulation is a process that should involve both standardized measures and detailed clinical interviews repeated over time.
For mental health professionals, one of the most important responsibilities is to accurately and efficiently identify those patients who are at the highest risk for suicide. Conducting an accurate suicide risk assessment (SRA) is no simple task as there are a number of factors that influence an individual's level of suicidality and his/her willingness to share this information. Therefore, it is imperative that practitioners adopt a systematic approach to conducting and documenting the foreseeability that a patient will commit suicide.1 Fortunately, research and clinical practice have yielded a number of useful tools and approaches that enable assessments of patients with greater confidence.
It has been estimated that 20% of people who commit suicide have been seen by a mental health professional within 30 days of killing themselves.2 Studies of patients who committed suicide have shown that the majority of these individuals denied suicidal ideation (SI) in their last recorded communication with a health care provider.3 Furthermore, patients who died by suicide were more likely to have been categorized as low risk during the documented assessment before their deaths.4 Moreover, research has consistently shown that individuals determined to kill themselves will not readily admit this to their mental health provider.5 Given this information, conducting an accurate SRA can be quite daunting and difficult. Thus, relying only on one's “clinical intuition” when determining risk for suicide is dangerous and irresponsible.5 The individual, clinical, situational, demographic, and interpersonal circumstances present for each patient must be considered.1 The importance of this assessment cannot be understated, and without a systematic assessment method, important factors can be overlooked.1
The overarching goal of this review is to provide the busy clinician with an overview of the scales and approaches suggested for use as adjuncts to clinical interviews when trying to determine an individual's risk for suicide. It is not intended as an analysis or review of the validity of the scales described.
PubMed was used to search for articles published in MEDLINE journals using the following keywords: suicide, risk assessment, measure, scale. Randomized trials and pilot, proof-of-concept publications investigating the use of specific SRA measurements were included in the review. References from these original articles were used to further guide a more in-depth literature search for additional publications and review articles about SRA. Only articles published in English were included. There were no date restrictions placed on the literature search for publications. The first five chapters of the American Psychiatric Association's Textbook of Suicide Assessment and Management by Simon and Hales (2nd ed., 2012) were also used to identify methods on how to incorporate evidence-based SRA scales into everyday practice. The scales will be described based on the prevailing opinions in psychiatry from the Textbook of Suicide Assessment and Management.
In 1963, the Los Angeles Suicide Prevention Center began constructing SRA scales to help assess the risk level of callers.6 Although various SRA scales exist, most lead to a large number of false positive and false negative results due to the relative infrequency of completed suicide.1,7 Because suicide is a statistically rare event, there exists a relative lack of opportunities to cross-validate scales and replicate their findings.1,7,8 As such, experts in the field have repeatedly concluded that there is not any one item on a scale or combination of scale items that can predict who will commit suicide to any useful degree.1,4,9 Unfortunately, current risk assessment practices remain highly variable across providers, facilities, and practice settings.1,5
Although not routinely used in clinical care, standardized suicide risk factor components of clinical and research scales are crucial to clinical assessment; however, these scales cannot be used alone or as a substitute for a full clinical assessment for the reasons previously mentioned.10 The following is a summary of standardized scales and approaches to use when determining a patient's risk for suicide. The measures and methodologies described can be used to gain additional insight and gather information that may be omitted by the patient regarding actual suicidal thoughts and intentions.
The Beck Depression Inventory-Revised (BDI-II) was constructed to measure the severity of self-reported depression according to Diagnostic and Statistical Manual of Mental Disorders, 4th Edition criteria for depressive disorders.11 It consists of 21 items. Each item is rated on a 4-point Likert scale with 0 indicating no reported symptoms and 3 indicating extreme symptoms. Total scores range from 0 to 63 with the following breakdown: 0 to 13 = minimal depression, 14 to 19 = mild depression, 20 to 28 = moderate depression, and 29 to 63 = severe depression. The BDI-II is generally completed in approximately 5 minutes and can be scored by any trained mental health professional (TMHP).1 This inventory is designed to assess feelings and behaviors over the previous 2 weeks and can be used to track depressive symptom severity over the course of treatment.1 The BDI-II has been validated for use in adolescents and adults, in an outpatient or inpatient setting, and has become one of the most widely used instruments for depression assessment.1 Item 9 on the BDI-II directly assesses SI. Item 2, the pessimism item, has been shown to be a more powerful indicator of suicidal behavior than SI; however, both of these can be important indicators in suicide risk.1
The Beck Hopelessness Scale (BHS) was developed based on the known association between pessimism, hopelessness, and suicide. The BHS is a 20-item, true-false, self-report inventory and scale that takes 5 minutes to complete.1 Total scores range from 0 to 20. The BHS measures three major aspects of hopelessness: feelings about the future, loss of motivation, and expectations. It can be scored by any TMHP and has been studied for use in children, adolescents, and adults. A score of 10 or greater correctly identified 91% of suicides within a 10-year follow-up in an inpatient psychiatric setting of patients with reported SI.12 In addition, a score of 9 or more accurately identified outpatients who eventually committed suicide.13 Beck's studies of this particular scale suggest that hopelessness is an even stronger predictor of suicidal intention than the severity of the patient's depression.13 Scores obtained on the BHS have been more strongly related to suicidal behavior as compared to BDI-II scores, but it is recommended that both be used in combination when assessing suicide risk.1
The Beck Scale for Suicide Ideation (SSI) measures active and passive suicidal desire as well as suicidal preparation.14 It is a 19-item scale that aids in measuring the intensity of SI, taking 5 minutes to complete. Each item is graded on a 3-point Likert scale (0 for low suicidal intensity to 2 for high suicidal intensity). Total scores range from 0 to 38. Additionally, there is a self-report version available.15 The Scale for Suicide Ideation reliably measures the severity of SI in both inpatient and outpatient settings and has been studied in adolescents and adults.14,16 This scale should be used within the construct of information gathering about an individual's suicide risk through the clinical interview. As such, clinicians should avoid using the obtained cutoff score as an indicator of a patient's risk of suicide. Instead, positive responses should be used as a prompt for further detailed inquiry.1
Motto's Risk Estimator for Suicide was developed empirically after 15 variables were identified as significant predictors for suicide in nearly 3000 individuals admitted to an inpatient psychiatric setting for depression or SI.7 Examples of the 15 predictors include emotional disorder in the family, threatened financial loss, and ideas of persecution or reference. Points are assigned to each of the 15 risk factors, and higher scores correspond with higher risk for suicide. This scale can be administered and scored by any TMHP in about 5 minutes.1 It provides a risk rating for suicidal behavior within the subsequent 2 years and has only been studied for use in adults. Although its predictive validity has not been well established, it can provide valuable supplemental information to aid clinical judgment and help provide a quantitative expression of suicide risk.1,7,8
Linehan's Reasons for Living Scale takes a positive approach to SRA by focusing on an individual's reasons for not committing suicide (ie, protective factors).17 The original Reasons for Living Scale is a 48-item, self-report measure for use in adults; however, various versions exist in the public domain, including a 72-item version and the Brief Reasons for Living Inventory for Adolescents.18 This scale asks patients to rate how important each reason for living would be if suicide were being contemplated on a 6-point Likert scale with 1 = not at all important to 6 = extremely important. The Reasons for Living Scale takes a patient approximately 10 minutes to complete and can be scored by any TMHP.1 It can reliably help identify individuals with SI compared to those without and is recommended for use to help monitor chronic suicidality in high-risk patients over time.17 Use of this scale may help patients identify personal strengths and reasons for living that may be otherwise unthinkable without prompting.1
The Columbia-Suicide Severity Rating Scale was developed by a team of investigators from Columbia University, the University of Pennsylvania, and the University of Pittsburgh to quantify the severity of SI and behavior.19 Because research has suggested that the worst-point ideation is a stronger predictor of subsequent suicide attempts than current ideation or the intent and lethality of previous attempts, the Columbia-Suicide Severity Rating Scale examines an individual's lifetime to pinpoint when the worst-point ideation occurred.19 It focuses on 4 major constructs, which are rated by the patient: severity of SI, intensity of SI, suicide behavior, and lethality. Any TMHP can administer it in about 10 minutes.1 This assessment also focuses on preparation behaviors for suicide acts, which have been closely associated with eventual suicide attempts.1 The Columbia-Suicide Severity Rating Scale has been studied and validated in adolescents and adults and is useful in detecting changes in suicidal thinking and behavior over time.19,20
The Chronological Assessment of Suicide Events, or the CASE Approach, is an interviewing strategy rather than a discrete scale for assessing an individual's risk for suicide. It recommends using 4 chronological time frames as a focus for individuals' suicidal thoughts and behaviors in order to minimize the omission of important information.21 The clinician should ask the patient about the following time frames in this order: present suicide events (past 48 hours), recent suicide events (previous 2 months), past suicide events (back in time prior to 2 months ago), and immediate suicide events (during the interview itself).22 Also at its foundation is the Equation of Suicidal Intent, which consists of the following: Real Suicidal Intent = Stated Intent + Reflected Intent + Withheld Intent.21,22 It aims to maximize the validity of an individual's stated and reflected intent on committing suicide while, at the same time, minimizing the amount of information withheld from the clinician.22 Reflected intent is used to help unlock a person's real intent and is defined as the quality and quantity of the suicidal thoughts, desires, plans, and extent of action taken to complete the plans.21,22 The CASE Approach employs several methods to elicit valid responses from patients about suicide. Table 1 includes examples of how to ask patients about suicide in a way that normalizes and attenuates the shame associated with taboo subjects such as suicide.22 Other methods of eliciting truthful ideations about current and past thoughts and acts of self-harm are shown in Table 2. Behavioral Incident is one method used to draw out specific details about past and current acts of self-harm.22 Gentle Assumption is another manner of inquiry used to normalize risky or self-destructive behavior.22 A clinician can use Denial of the Specific types of questions to ask about explicit self-harm ideations, and Symptom Amplification is a method used to avert an individual's tendency to minimize reports of disturbing behavior.22 The CASE Approach can be used by TMHPs, can be completed in 5 to 10 minutes, and is not recommended for interviewing children.1 With routine use, clinicians can become adept at utilizing the CASE Approach, which can be flexibly adapted to the psychopathology and personality traits of individual patients.1
The Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is a 567 true/false–item inventory that consists of 10 basic clinical scales and 14 content scales. Interpretation is based upon a code type, consisting of the two highest scores on the clinical scales. Research has shown that individuals preoccupied with suicide and death may yield a profile with the highest elevations on the Depression Scale (Scale 2) and the Psychasthenia Scale (Scale 7).1 Although a 2-7/7-2 code type is the most correlated with suicidality, there are several other code types that have been associated with a high-risk for suicide.1 Furthermore, the MMPI-2 contains content component scales that have also been associated with suicidal thinking. The Suicidal Ideation (DEP4) and the Suicide/Death Ideation (SUI) subscales have been particularly useful in identifying individuals considering suicide.23 Finally, inspection of the validity scales found on the MMPI-2 has been cited as useful in identifying and categorizing individuals who present as high risk for suicide.1 In particular, individuals with elevated L scores (Lie Scale) should alert clinicians to the tendency of an individual to minimize or outright deny any symptoms, including SI, indicating a need to be more thorough and careful when conducting a risk assessment.1 The MMPI-2 must be administered by a clinical psychologist and generally takes 90 minutes to administer and score.1 The results provide important subjective information from the patient's perspective that is often not gathered during the course of a clinical interview.1
The Rorschach Inkblot Test was, at one time, the most commonly used method for estimating suicide risk.1 The Rorschach contains 10 stimulus cards that are presented to patients with the directive to share what the inkblot might represent. The idea behind the use of this test is that patients will project their inner world onto the ambiguous shapes seen on the cards. A patients' interpretations of the innocuous inkblots can provide a glimpse into the patient's way of thinking and interpreting the world. Examples of responses to Rorschach cards that might be indicative of suicidal thinking include the following: “It looks like staring into a well … so black and deep … disgusting … can't get out of there.” Such statements can be interpreted as painful self-reflections and may be indicative of shame, despair, or an inability to find alternatives to current circumstances, all of which can be considered characteristics of patients at higher risk for suicide.1 The Suicide Constellation is now included among the Rorschach special indices and consists of 12 variables that highlight certain features common in Rorschach protocols of individuals (N = 101) who completed suicide within 60 days of testing.1 Patients who endorse 6 or more Suicide Constellation variables should prompt the clinician to rescore the protocol because there may be a possibility of self-destructive preoccupation present, and scores of 7 or more Suicide Constellation variables have been utilized to predict near-lethal suicide attempts.24 This test can only be administered by a clinical psychologist, can take an hour or more to administer, and is only recommended for use in individuals 15 and older.1 The Rorschach is considered to be helpful in that it is a relatively nonthreatening task without any right or wrong answers. Although its use has declined in recent years, it remains a potent tool in evaluating SI.1
The Firestone Assessment for Self-Destructive Thoughts is based on the observed association between the strong influence on negative thought processes and suicidal and self-destructive behavior.1 It is an 84-item, self-report questionnaire designed for use as a screening instrument.25 Patients rate each item on a 5-point Likert scale (0 = never to 4 = always) reporting the frequency of their negative thoughts.25 Higher scores reflect higher levels of self-destructive thinking. The Firestone Assessment for Self-Destructive Thoughts needs to be administered by a clinical psychologist and takes about 20 minutes to administer and score.1 It is designed to help clinicians determine the level at which the patient is experiencing the highest intensity of self-destructive thoughts. This determination can provide guidance to focus interventions toward the area(s) in which the most psychological pain is being experienced.1 Predictive validity for suicide and Firestone Assessment for Self-Destructive Thoughts scores do not exist at this time; however, its potential for identifying specific maladaptive cognitions related to SI can enhance the clinical interview when trying to determine one's overall risk for suicide, especially when repeated over time.1
Table 3 includes brief summaries of additional tools and resources that may be used to help with SRA.
Given the dynamic nature of suicidality and suicidal behavior, clinicians will always be faced with inherent difficulty when attempting to accurately predict suicide in their patients.10 This is further complicated by the time constraints placed on clinicians in busy practices. As a result, there is a significant need for instruments that are both thorough and efficient when assessing suicide risk. Research has consistently shown that a clinician can greatly improve the accuracy of any risk assessment by utilizing a multimethod approach consisting of a careful semistructured clinical interview complemented by the use of standardized measures.1
An individual's risk of suicide is a phenomenon that one cannot identify by simply using a static set of clinical features.8 Clinicians must consider the accumulation of suicide risk and protective factors over time to more accurately determine an individual's imminent suicide risk.45 In fact, the importance of conducting SRAs throughout the patient's illness and recovery cannot be overemphasized.46 Repeated assessments over time, factoring in a broad array of clinical features, are suggested to develop adequate and replicable prediction systems for suicide.8 A thorough review of records of reported SI and action is not only essential to determine the credibility of a patient's self-report, it is also needed in order to identify patterns between the current report and previous reports of behaviors, the duration and intensity of thoughts, and prior plans.21,22 Consideration of important dynamic patient factors coupled with the information gained from a standardized assessment may provide a more accurate estimate of one's suicide risk.8
When conducting an appropriate SRA, a clinician is expected to make an educated judgment of near-term risk for suicidal behavior and document this according to clinical observations, patient responses, and the clinician's understanding of what heightens suicidal risk.46 Practitioners will not be held liable for incorrect judgment; however, the general consensus is that any judgments made must be reasonable, based upon observed data, and indicate that data was reasonably sought.47 As such, simply documenting “Patient denies SI/HI” is inadequate, and the phrase “patient contracts for safety” has no evidence in decreasing suicidal behavior.47,48 Therefore, obtaining a “no harm” contract from a patient is not the same as and should not be considered to be adequate in assessing and charting risk for suicide.48 Instead, clinicians should clearly document all relevant risk and protective factors from the first contact with the patient and then review and document for any changes in these factors at subsequent visits.1
Clinicians should be aware of the unintended consequences of high-risk-suicide categorization, such as unwarranted detention for some patients as well as the stigma that arises after patients are labeled as dangerous to themselves.49 If positive testing results are communicated frankly with patients and with a genuine concern for their well-being, this should not damage the therapeutic relationship. The desired outcome of the clinical encounter to assess suicide is to engage in a therapeutic alliance that will acknowledge and assess risk factors, protective factors, precipitating events, and warning signs and address them by developing and implementing a safety plan while continuing to monitor and manage suicide risk.
“There are no shortcuts or quick fixes in conducting a competent SRA.”10(p291) An evidence-based approach to SRA remains a challenge largely due to the fact that suicidal behavior is multifactorial. No one characteristic accurately identifies a person's risk, and at this time, there is no one tool or scale that is sensitive or specific enough to predict suicide.1 As a result, risk formulation is a process that should involve both standardized measures and detailed clinical interviews repeated over time.5
Clinicians must take a proactive approach to SRA. Mental health practitioners should not wait for patients to express suicidal thoughts to begin assessing an individual's risk for suicide, nor should one wait until after a suicide attempt or for overtly suicidal behavior. Instead, providers need to adjust historical views of the role for SRA and see it as something more than a simple inquiry with the documentation of “denies SI” in a mental status exam. An adequate and appropriate risk assessment should be a dynamic, ongoing process and needs to become part of documentation at intake and at follow-up encounters. This can be accomplished by combining information obtained from standardized measures, such as the BDI-II with BHS, Columbia-Suicide Severity Rating Scale, or Reasons for Living Scale, with the material gained from the clinical interview. It is further recommended that if a patient is reticent to share information about suicidality or is evasive in any way throughout the course of therapeutic intervention, a more in-depth means of assessment should be employed to assess the exact nature of SI (eg, Rorschach or MMPI-2).1 Furthermore, dynamic risk and protective factors should be included in each SRA. The best approach to SRA is to utilize this cumulative information to help identify patterns and important changes in patients that raise their risk for suicide and prompt action, such as suicide safety planning, when necessary. Overall, it is recommended that the use of an in-depth risk assessment upon intake and longitudinally throughout the course of treatment can ensure that risk assessment is a practical, efficient, and routine part of every clinician's practice.