Continuing Education Activity

This activity describes the research evidence and current practice recommendations associated with the evaluation and treatment of the heterogeneous group of individuals who endorse suicidal ideation. It highlights the role of the interprofessional team in assessing, managing, and improving care for people with suicidal ideation.

Objectives:

• Summarize the current theories associated with suicidal ideation.
• Describe the risk screening and clinical evaluation considerations when caring for an individual with suicidal ideation.
• List the evidence-based treatment recommendations for individuals endorsing suicidal ideation.
• Discuss interprofessional team strategies for improving care coordination and communication to meet the diverse needs of suicide ideators and improve outcomes.

Introduction

Suicidal ideations (SI), often called suicidal thoughts or ideas, is a broad term used to describe a range of contemplations, wishes, and preoccupations with death and suicide. There is no universally accepted consistent definition of SI, which leads to ongoing challenges for clinicians, researchers, and educators.[1] [2] For example, in research studies, SI is frequently given different operational definitions. [3]This interferes with the ability to compare findings across studies and is frequently mentioned as a limitation in meta-analyses associated with suicidality. [4] Some SI definitions include suicide planning deliberations, while others consider planning to be a discrete stage. 

Beyond the lack of clear nomenclature, there are other concerns. A systematic review of the numerous interprofessional clinical guidelines for suicide yielded no consensus on a clinical gold standard for assessing and managing SI or people at risk of suicide. [1] Although scales to measure depression, SI and risk for suicide exist, none produce a score that is sufficiently reliable or clinically useful in predicting the very small subgroup of suicide ideators whose death by suicide is imminent. [5](The American Psychiatric Association Practice Guidelines for Psychiatric Evaluation of Adults, 3rd ed. 2016, p. 19).

It is evident that suicidal ideations present in a "waxing and waning manner" [6], so the magnitude and characteristics of SI fluctuate dramatically.[7] It is critically important for healthcare professionals to recognize that SI is a heterogeneous phenomenon. It varies in intensity, duration, and character. As there is no "typical" suicide victim, there are no "typical" suicidal thoughts and ideations. Unfortunately, healthcare records often document SI in a binary yes/no fashion, although it encompasses everything from fleeting wishes of falling asleep and never awakening to intensely disturbing preoccupations with self-annihilation fueled by delusions. Therefore, thoroughly assessing and monitoring the pattern, intensity, nature, and impact of SI on the individual and documenting this accordingly is important for all healthcare professionals. It is also important to reassess SI frequently due to its fluctuating pattern.

The magnitude of SI fluctuations was studied using an ecological momentary assessment method. Individuals who attempted suicide in the past year plus a sample of suicidal in-patients recorded the intensity of their suicidal thoughts from hour to hour for four weeks. Analysis of these data showed dramatic fluctuations in the intensity of SI by all participants. All participants had SI, which varied in its intensity, either upwards or downwards, by one standard deviation on most days. Many had one standard deviation fluctuations several hours apart within the same day.[7] This knowledge is important for all healthcare professionals to consider and highlights the need to monitor fluctuations and not dismiss the possibility of sudden increases in suicidal urges, even when the current level is mild, and the individual currently has control over them. Additionally, SI is considered a better predictor of lifetime risk for suicide than imminent risk, so assessments should include describing the characteristics and impact of prior SI as well as current.[8]

The Center for Behavioral Health Statistics Quality publishes the results of the American nationwide household survey, the National Survey of Drug Use and Health (NSDUH). Piscopo's 2017 publication summarized the results from the 2009-2014 surveys, which show that 6% of 18-25-year-olds respond affirmatively to the survey question, "At any time in the past 12 months, did you seriously think about trying to kill yourself?" In contrast, the lowest rate of SI was 1.6% in those aged 65 years and above. There is no clear association between endorsing SI and attempting suicide. For every 31 Americans with SI, only one individual will attempt suicide. The rates of suicide deaths also vary by gender, age, race, and other demographic variables. Further evidence of the weak association between reported SI and fatal suicides is apparent when comparing the NSDUH results to CDC mortality records. Despite the low prevalence of SI in white males over age 75 years, they have the highest rate of fatality by suicide (approx. 40 per 100,000). Meanwhile, females over 75 years have much lower rates (4 per 100,000). The suicide ideators in the 18-25-year-old group had significantly fewer suicide deaths (approx. 17.5 per 100,000 for males and 4 per 100,000 for females). 

Most people have control over SI and do not attempt suicide, even when reporting SI. Psychiatric Association Practice Guidelines for Psychiatric Evaluation of Adults (2016, p. 19) points out that SI is a symptom of another primary psychiatric diagnosis and suggests that 90% of people who end their lives by suicide meet the diagnostic criteria for one or more psychiatric diagnoses. However, data clearly show that numerous medical illnesses are associated with increased odds of suicide, and that suicidal death extends through all demographic groups and includes virtually all psychiatric and medical diagnoses. The CDC's mortality records for 2017 reveal over 50% of deaths by suicide were by people with no known psychiatric illness. Some criticisms have been made that suicidality should be regarded as a distinct psychiatric diagnosis, with its symptoms and unpinning pathological processes.[2][1] Although this is beyond the scope of this paper, this contention is becoming more widespread.[6]  After reviewing all of the existing clinical guidelines, Bernert et al. concluded there is an urgent need for "easily-accessible best practice guidelines, adaptable to diverse fields of medicine and clinical specialties, that may be the first point of contact for risk detection, intervention, and prevention."[1]

Although this paper aims to focus primarily on suicidal ideation, it is important to provide context. Therefore, while the intention is not to broaden the focus to suicide, it is impossible to address the significance of SI without also discussing suicidal behaviors and outcomes to some extent. It is estimated by the Center for Disease Control and Prevention (CDC) that in 2017 there were approximately 10 million people in the USA who experienced suicidal thoughts. Fortunately, the majority of ideators in the USA and globally will never attempt suicide, and fewer will use lethal means that result in death. Of the 10 million Americans with SI, it is estimated there were 1.4 million suicide attempts in 2017, but healthcare was only sought by approximately one-third of those who attempted. The degree of suicidal intent and the lethality of means used during attempts vary tremendously.  One-half of the 47,000 suicides that occurred in America during 2017 were caused by firearms. (CDC).

Globally, the World Health Organization (WHO) collects mortality data, including the prevalence and means of suicides, for all member nations. Beginning in 2013, after declaring that the rising suicide rates constituted a "global public health crisis," they advocated for evidence-based strategies to prevent suicides globally. In developing nations, where the ingestion of pesticides was the leading cause of fatal attempts, suicide prevention efforts promoted using less toxic pesticides. Evidence exists that reductions in suicides can be achieved by reducing access to lethal means, but this requires a comprehensive systemic approach that includes collaboration between policy-makers, healthcare professionals, and interventions to reduce modifiable risk factors.[1]

Primary Care Professionals Concerning SI

A recent meta-analysis of 44 studies of healthcare services used by suicide victims showed a chief reliance on primary care professionals in all countries. Only 31% of suicide decedents received inpatient or outpatient mental health services in the year before their death; 57% of the decedents had contact with mental health services at some point during their lifetime.[9] Primary care professionals are more apt to have an established relationship with patients and have a more complete understanding of their health history. Furthermore, due to having a pre-established relationship, when patients experience worsening SI or stressors that may precipitate suicidal behaviors, they are more likely to seek help from primary care professionals. Studies show that 80-90% of suicide decedents increased their contact with primary care professionals in the year and months before their suicide; 44% of those who died by suicide had contact with primary care in their last month of life.[9][10] Although the impetus for increased contacts is undoubtedly variable, it does indicate opportunities exist for healthcare professionals to identify any new risk factors for suicide and offer treatment options to address modifiable factors. 

During their final visit with primary care professionals, 90% of successful suicides in a UK study disclosed their SI to primary care providers.[11] These same providers were interviewed about their experience losing a patient in their practice to suicide (n=39). In each case, the primary care physician had referred their patient for psychiatric services, so the physician's communication with both their patient and the psychiatric service professionals (physicians and community psychiatric nurses) could be examined in the study. Twenty percent of the physicians who heard their patients endorsing suicidal thoughts stated they were concerned about their patient's safety during their final appointment.[11]

The researchers noted that many of the primary care providers were uncertain about how to interpret the meaning of SI when expressed by their patients. This was particularly evident when the patient had a history of voicing SI. Examples of statements from the primary care physicians include;

• "Although we put her down as a moderate suicide risk, none of us thought she’d ever do it because she talked about it so much." (p. 263)
• "He’d done this on numerous occasions. Taken overdoses, not as a suicidal attempt but in an attention-seeking, in a [state of] mental distress, help me, [a] cry for help ... there was never a disorder" [11] 

The limited understanding of how to assess and treat people with SI was apparent in some of the interviews in this UK study, although there was substantial variability between physicians. However, similar gaps in knowledge are also evident in other studies involving primary care professionals.  An observational study of primary care providers in the Netherlands showed SI was assessed in only 44% of clinically depressed patients and 66% of new-onset depression patients.[12] The frequent lack of suicide risk prevention policies in primary care practices is also apparent, and even when they do exist, there may be uncertainty regarding what they entail.[13] Julie Goldstein Grumet, director of US Health and Behavioral Health Initiatives for the Suicide Prevention Resource Center and the director of the Zero Suicide Institute, published preliminary research data in the Jan. 2019 Journal of Health Care Compliance. Less than one-third of healthcare providers (n=15,000) who completed an optional self-test for the Zero Suicide initiative in the USA reported feeling knowledgeable about suicide risk warning signs. Similarly, only 1 in 3 knew their organization's procedures for patients at increased risk or felt confident in their ability to respond to a suicidal patient. Although standardized tools, instruments, and rubrics do not provide a clear indication of imminent suicidal risk, the American Psychiatric Association Clinical Guidelines for Evaluating Suicidality suggests these tools can be useful as prompts when interviewing to ensure thoroughness in the questions asked during the risk assessment. Grumet et al. reported that only 35.5% of the providers stated they use any of these available tools when screening or assessing. Furthermore, only one-third of the healthcare professionals responsible for delivering treatment (n = 4,101) indicated they "strongly agreed" when self-rating their level of confidence or comfort in treating patients with elevated suicide risk. 

Although additional training is recommended by many to improve the competence of healthcare professionals, the outcomes from an interprofessional course which included how to assess SI produced modest results. Students were taught the importance of SI assessments during an online module and then required to practice completing suicide risk assessments on standardized patients. However, later in the semester, only 55% of the students (65.5% of medical; 54.5% of nursing; 46.4% of pharmacy) completed an SI assessment on the depressed standardized patient during their final objective clinical standardized exam (OSCE).[14] This emphasizes the importance of reinforcing learning in academic settings and monitoring competence and compliance in healthcare organizations.

There were additional findings in the study of UK primary care physicians that are pertinent to consider. Multiple barriers with psychiatric services were described by the primary care physicians. The primary care providers commented that they often felt frustrated when they were left to manage suicidal patients alone or when the system created obstacles and referral mazes which made it difficult for them to advocate for what they believed was in their patient's best interests:[11] Examples of statements from the interviews include;

• "Because the patient did not attend his last psychiatric appointment, the psychiatrist discharged him and sent a letter to us stating this. I disagreed [because the patient needed the psychiatrist's expertise]" p. 264.
• "Sometimes we feel like we have to manipulate the system [to expedite referrals] just to get a patient assessed ... we, therefore, refer to hospital [emergency department] as patients will get seen and assessed on the day and they do follow-up as their referrals go to the CMHT (Community Mental Health)" p. 265.[11] 

Almost all of the physicians (90%) said they knew their deceased patient "well," but this knowledge may have been under-utilized during the subsequent psychiatric treatment planning. One of the physicians was frustrated about being excluded from providing input into the psychiatric treatment plan - a plan he disagreed with and which he suggested may have contributed to the suicide.[11]

The issues and conclusions identified in this UK study are consistent with other studies. A review of literature focusing on improving primary practice professionals ability to detect and treat SI and prevent suicide produced four major recommendations:

i) educating practitioners on risks for suicidal thoughts and behaviors

ii) providing patient screening to identify suicide risk and/or mood disturbances

iii) using evidence-based interventions, including collaborative, multi-disciplinary teams, to manage depression

iv) assessing for the presence of suicide risk factors and managing suicide risk when symptoms arise.[15]

These recommendations are similar to the reaction of the American accreditation agency, the Joint Commission (JC), to address frequent suicides following contact with ED, PCP, and MH services. The Joint Commission’s Sentinel Event database received 1,089 reports of suicides from 2010 to 2014 among patients receiving care in an accredited hospital or within 72 hours of hospital discharge or release from an emergency department. Shortcomings in the patient's psychiatric assessment were the most frequently cited root cause. As of July 1, 2019, the JC requires the use of validated screening tools to assess any patient whose primary reason for seeking treatment or evaluation is for a behavioral health problem. However, universal screening was not mandated, although many organizations may elect to do this. The JC, working with other suicide reduction organizations, has numerous resources on its website to assist members to meet this accreditation requirement. Their stated rationale for new requirement included rising suicide rates, increased empirical knowledge and available risk assessment tools, and the non-compliance of over 21% of accredited behavioral health organizations and 5% of hospitals to meet JC's National Patient Safety Goal (NPSG) 15.01.01 "Element of Performance 1 – Conduct a risk assessment that identifies specific patient characteristics and environmental features that may increase or decrease the risk for suicide" (Joint Commission, NPSG 15.01.01)[16]

The JC advised that all accredited organizations should do the following:

1) Review each patient’s personal and family medical history for suicide risk factors.

2) Screen all patients for suicide ideation using a brief, standardized, evidence-based screening tool.

3) Review screening questionnaires before the patient leaves the appointment or is discharged.

4) Take action based on the assessment results to inform the level of interventions needed. (The Joint Commission, 2016, p. 3)[17]

Active and Passive Suicidal Ideation

"Active" suicidal ideation denotes experiencing current, specific, suicidal thoughts. Active SI is present when there is a conscious desire to inflict self-harming behaviors, and the individual has any level of desire, above zero, for death to occur as a consequence. The probable lethality of their actions, based on the means used for the suicide attempt, is not the focus. Rather, the individual's expectation that their attempt could produce a fatal outcome is the key consideration.[18] 

Example of an Active SI assessment item

Miller et al. (1991) Modified Suicidal Ideation Scale

• "Over the past day or two, when you have thought about suicide, did you want to kill yourself? How often? A little? Quite often? A lot? Do you want to kill yourself now?" 

"Passive" SI refers to a general wish to die but when there is no plan of inflicting lethal self-harm to kill oneself. Passive SI includes indifference to an accidental demise which would occur if steps are not taken to maintain one's own life. Passive SI receives less attention from clinicians and researchers than active SI. Although most research studies do not distinguish between active and passive SI, few studies focus on passive ideations. One author pointed out the underlying assumption of healthcare professionals is that the desire for death is not typically thought of as a harbinger of more severe suicidal outcomes.[19] 

Examples of Passive SI assessment items

Beck et al. (1979) Scale for Suicidal Ideation (SSI) was the first to measure "passive suicidal desire"

• 0 = Would take measures to save [one's own] life
• 1 = Would leave life/death to chance
• 2 = Would avoid steps necessary to save or maintain life[20]

European Depression Scale item,

• "In the past month, have you ever wished you were dead?"[21] 

Miller et al. (1991) Modified Suicidal Ideation Scale 

• "Would you deliberately ignore taking care of your health? Do you feel like trying to die by eating too much (too little), drinking too much (too little), or by not taking needed medications?" 

Passive SI, Death Wishes, and the Older Adult Population

Assessing SI is an essential component of suicide risk assessment for individuals extending beyond those with known psychiatric conditions, especially in the older adult populations around the world. Individuals who endorse SI have a higher lifetime risk of future suicide than individuals who have never experienced any SI, although the prediction value is only weak. The value of SI in predicting imminent suicide risk has not been shown but does factor into the overall assessment of protective versus risk factors in short-term, imminent suicide risk. 

A common misconception is that passive SI has less clinical importance. Large population-based studies (n>85,000) that compared the odds ratio to predict suicide attempts based on reported passive SI or reported active SI show there is no significant difference.  Including questions to assess both active and passive SI was recommended as the best clinical practice to predict risk.[19]  

Compared to younger populations, older adults are more apt to endorse passive SI and less inclined to express active SI or seek mental health care. Also, the majority die on their initial attempt.[22] White males over 85 years have the highest rates, largely due to the use of lethal means like firearms. A systematic review revealed that older adults who died by suicide have very different personality profiles than younger suicide victims. Overall, older suicide victims had less evidence of maladaptive personalities, and the majority did not meet the threshold for psychiatric diagnosis. The only significant association was with a relatively small number of older suicide victims who had obsessive-compulsive and avoidant personality disorders. The researchers suggested these personality traits may have made later-life changes and transitions more difficult. They also noted that older suicide victims were more heterogeneous in both their risk factors and experiences compared to early-life suicide victims.[23] 

Overall, there is a paucity of research addressing the nature of SI in older populations, although passive SI is understood to be more associated with older adults. However, when searching the literature using analogous terms like "death ideation," "death wish," "self-chosen death," and "wish to hasten death" (WTHD), it becomes more clear that this terminology has been ascribed to older adults' ideations. Healthcare professionals should bear in mind the social constructs and norms that influence the way suicidality is addressed and indirectly minimized by the use of these terms to describe SI in this age group. Without drifting too far into this literature, several examples of recent studies may help illuminate this relatively well-researched area of study.

• Death wishes:  A death wish was expressed in 9.5% of a large sample (n= >35,000) of New Zealanders aged 65 and older who were being evaluated for home care services. Depression, poor self-reported health, and loneliness were each independent, predictive variables of death wish.[24] 
• Self-chosen death:  Interviews with Dutch older adults (n=25) who were 70 years or older (mean age of 82), who wanted to die because they considered their lives complete and no longer worth living. All of these ideators had age-related debilitation, but none had a terminal disease. They considered their death wish to be reasonable and wanted to have the same ability as those with terminal illnesses to chose death based on the Dutch euthanasia laws.[25]
• Wish to hasten death: A systematic review of 16 studies examining WTHD in patients with advanced illnesses showed that feeling like a burden contributed and may have triggered the WTHD.[26]

A study of incarcerated prisoners aged 50 years and older (n=124) found past alcohol dependence and self-rating one's health as poor/fair were equally associated with both passive (10%) and active (11%) SI. Compared to inmates who denied any SI, both groups of ideators had significantly higher incidences of previous suicide attempts and/or major depressive episodes.[27]

A European study of retired middle-aged and older adults (n >35,000) examined the association between health status and passive suicidal ideations. They found increased odds of passive suicidal ideation when the participant had been diagnosed with a heart attack, diabetes/high blood sugar, chronic lung disease, arthritis, ulcer, and hip/femoral fractures.[28] Other studies show elevated odds ratios for suicide with hepatic disease [29], CVA [30] physical disability [31]). These findings suggest a need for all healthcare professionals to be aware that the lack of adherence to the medication or dietary regime may warrant further exploration concerning possible SI. 

Research findings from a community sample of older adults (n=1,226) who participated in the PROSPECT study (Prevention of Suicide in Primary Care Elderly: Collaborative Trial) showed that SI was present in 29% of participants with major depression; 11% with minor depression; and 7% without depression. The findings from this study underscore the need to assess SI in older patients, including those who do not have signs and symptoms of depression.   [32] A systematic review of self-harm in older adults concluded that more research needs to be done better to understand this population's unique characteristics and needs. Suicide attempts in this age group are usually fatal in their first attempt due to the lethality of their means, existing fragility, and lack of desire or opportunity for rescue. Because older adults typically have frequent contact with PCPs, opportunities to detect SI and provide appropriate interventions.[33]

Suicide Risk Assessment and Suicide Risk Formulation

The suicide risk assessment (SRA) focuses on identifying the risk factors and protective factors for any given individual. This is followed by the suicide risk formulation (SRF), which assigns a level of imminent suicide risk. The subsequent triage and treatment plans are based on the SRF. One of the concerns discussed in the literature is the emphasis on the patient's communication of suicide ideation.[34] The American Psychiatric Association (2016) Practice Guidelines for the Psychiatric Evaluation of Adults states, "When the clinician is communicating with the patient, it is important to remember that simply asking about suicidal ideas or other elements of the assessment will not ensure that accurate or complete information is received." (p. 21). 

Not all ideators are apt to share their SI. Over a dozen research studies have shown that 75% of patients who die by suicide denied SI the final time they were asked by a healthcare professional. Typically, their death by suicide occurred within the month of their last visit [8]. Berman completed chart reviews of 157 patients throughout the USA who died by suicide within 30 days of being evaluated by a healthcare professional. All victims were either receiving in-patient or out-patient mental health care; or were evaluated in an emergency department or by their primary care professional. The Joint Commission requires healthcare professionals in these practice settings to assess SI for anybody at risk of suicide. However, despite being asked, the majority (66%) denied SI. Within two days, 50% of these individuals who had denied SI ended their lives by suicide.[8] Berman noted that the denial of SI provides a basis for patient discharge if the individual was admitted due to SI. While this may be an incentive for a patient to deny SI, particularly if they want to be discharged, caution should be exercised. Berman states that too frequently, clinicians assume that SI must exist for suicide to occur when SI is only a risk factor for suicide. Additionally, SI is a weak predictor of increased lifetime risk, it does not predict imminent risk -- but, then again, nothing does. 

Ribet et al. examined the root causes that may have contributed to 141 veteran suicides within a week of their hospital discharge. Flaws in communication were frequently cited. It was also noted that almost half of the suicides occurred following an unplanned discharge.[35]  The Joint Commission released multiple sentinel event warnings over the past decade based on reports of patient deaths in hospitals or shortly after discharge from mental health units or release from emergency departments.  They stated, "there is no typical suicide victim” and cautioned against assuming only certain individuals are at risk based upon their diagnosis or treatment setting.[17] 

Emergency Department Suicidal Ideation Assessments

Beginning July 1, 2019, healthcare professionals are required by the Joint Commission's NPSG 15.01.01 to use a validated tool to assess suicidal risk for all patients whose primary reasons for seeking healthcare is the treatment or evaluation of a behavioral health condition.[17] The ability to accurately triage patients is contingent on the reliability of the instruments and also the clinician's clinical judgments. Much remains to be learned about the risk factors for imminent, short-term, and long-term suicidal behavior and the best way to identify risk.  A recent prospective study in Canada compared the risk of suicide attempt within 6 months for individuals presented in ED with SI but whose presentations differed, as evident in their responses to screening questions (n= 5,655). During ED triage screening, some individuals primarily endorsed SI characterized by an "ambivalence about living" while others expressed active SI. Within 6 months, 3% of the initially screened sample presented again in the ED with a suicide attempt. Individuals who initially identified with "ambivalence about living" had more than double the risk of suicide attempts (odds ratio [OR] = 2.57, 95% CI = 1.64-4.02, P < 0.001). Those with active suicidal ideation had more than triple the risk of an attempt within 6 months compared to non-SI individuals (OR = 3.75, 95% CI = 2.61-5.34, P < 0.001) Both active suicidal ideation and ambivalence about living are concerning presentations associated with risk of attempt within 6 months. Clinicians should be mindful that differentiating between active suicidal ideation and ambivalence about living are presentations that warrant follow-up due to the increased 6-month risk of attempts.[36][37] 

Data show that 10% of people who ended their lives by suicide visited an emergency department within two months of inflicting fatal self-harm.[38] People who have psychiatric histories, substance use disorders, or depression were most apt to be assessed for SI, but this negates the significance of many other known socioeconomic factors. Chart reviews of suicide victims who ended their lives within hours to days of being assessed by a healthcare professional showed the pitfalls of relying too much on patients' admissions of SI.  Berman's review of the victims' charts showed strikingly similar profiles between the patients who admitted to SI or denied the presence of SI.  There were no significant differences in their diagnoses, current presentations, or current circumstances. Almost all suicide decedents' charts had documentation showing current anxiety/agitation and sleep problems, current interpersonal problems or job/financial strain, current comorbid diagnoses, current social isolation/withdrawal, plus a history of SI/prior attempts plus many had a family history of a mental disorder. Healthcare professionals should bear in mind that relying on verbalized or reported SI as a gateway to assessing suicide risk may be inadequate, especially when SI is denied. [8] 

A 2013 study indicated that 0.6% of emergency department (ED) visits were due to suicidal thoughts, but when screening for SI was done, incidental, occult suicidal ideation was found in over 11% of patients who arrived due to medical complaints. Although SI was identified in medical patients and communicated, no follow-up regarding their SI occurred while on the medical unit.[39] The Joint Commission does not require universal screening of all patients, but some suicide risk reduction programs, professional organizations, and healthcare systems are advocating and implementing policies for universal screening in ED. A recent study of ED nurses and physicians in ED showed that most felt confident completing SI screenings, but only 7% of physicians (residents and attendings) and 37% of nurses reported they did so all of the time or most of the time.[40] Focus group interviews with ED nurses produced skepticism that the risk for suicide can be more reliably assessed with a brief screening tool question, such as 'Do you have thoughts or plans to harm yourself?' compared to a clinician's judgment. The participants described their efforts to improve suicide screening during ED triage required an ongoing iterative process of assessing for SI, which included probing, eliciting, evaluating, and reacting to identify occult SI.[41]

Children ages 10-12 years who presented at the ED were screened with the Ask Suicide-Screening Questions (ASQ) and Suicidal Ideation Questionnaire. Positive screen results were present in 54% of patients whose chief complaint was psychiatric, but screening also showed positive results for 7% who presented to the ED with chief medical complaints. The overall rate of SI for these pre-teens was 29%, with 17% reporting engaging in prior suicidal behaviors. Although this sample was small, these findings highlight the potential value of screening children as young as 10 years for SI, including those who present with medical concerns.[42] These findings suggest that a substantial number of individuals who present in ED may have occult SI. The interprofessional ED team should bear in mind that some individuals may express suicidal thoughts differently or deny SI when presented in the form of a screening tool and feel more comfortable disclosing it when approached privately in a supportive, direct manner.  The literature suggests that all healthcare providers in ED will benefit from additional training to increase their knowledge, skills, and confidence.[43]

A large meta-analysis (71 studies N= 4,669,303 individuals) included inpatient and non-inpatient adult populations to examine whether expressing SI was associated with subsequent suicides. Only limited sensitivity of SI for suicide was found (41% at 95% Confidence Interval (CI) 35–48), which means approximately 60% of suicide victims did not report experiencing SI. These authors examined whether there was a difference between using a structured instrument to assess SI versus relying on the healthcare professional's clinical judgment. Using structured instruments to assess SI was associated with a non-significantly lower pooled odds ratio (2.38, 95% CI 1.14–4.99) than when SI was clinically defined (OR = 3.72, 95% CI 2.96–4.67), but a great deal of heterogeneity in the studies. An important finding was that having a suicide plan, which was reported in only four studies, did significantly increased eight-fold (OR = 8.51, 95% CI 5.51–13.06). Two studies used an expressed wish to die as their operational definition of SI (OR = 3.01, 95% CI 1.49–6.06). 65 studies did not specify the individual's level of intent or planning when describing SI. There was a moderately strong but highly heterogeneous association between suicidal ideation and later suicide (n = 71, OR = 3.41, 95% CI 2.59-4.49, 95% prediction interval 0.42-28.1, I2 = 89.4, Q-value = 661, d.f.(Q) = 70, P ≤0.001).[44]

Suicidal Ideation and Psychosis

Chapman's (2015) meta-analysis examined whether expressing SI was related to subsequent suicides for two different groups of individuals -- adults diagnosed with mood disorders (11 studies reporting on 860 suicides) and adults diagnosed with schizophrenia spectrum psychotic disorders (14 studies reporting on 567 suicides).  Results showed that people with schizophrenia spectrum psychosis who expressed SI had over a six-fold increase of suicide [14 studies; Odds ratio (OR) 6.49, 95% confidence interval (CI) 3.82-11.02]. Meanwhile, the association between expressing SI and suicide among patients with mood disorders was not significant (11 studies; OR 1.49, 95% CI 0.92-2.42).[45] 

Another meta-analysis used data from 50 longitudinal studies that followed individuals who had experienced psychotic symptoms. These researchers attempted to differentiate between the impact of positive and negative symptoms of psychosis and their association with SI. Findings showed that positive symptoms in psychosis were weakly associated with SI (50 studies; OR = 1.70, 95% CI 1.39-2.08). On the other hand, negative symptoms in psychosis failed to show significance with SI and were found to be protective factors against death by suicide.[46] 

A meta-analysis and systematic review of longitudinal studies of individuals in the general population (n > 84.000 representing 12 samples from 23 countries) showed that people who reported having at least one lifetime psychotic experience had double the odds of experiencing SI in the future (5 articles; n = 56,191; OR 2.39, 95% CI,1.62-3.51); triple the odds of a future suicide attempt (8 articles; n = 66,967; OR = 3.15 95% CI, 2.23-4.45), and four times the odds of future suicide death (1 article; n = 15,049; OR= 4.39 95% CI, 1.63-11.78]. The authors concluded these elevated risks exceeded what could be explained by co-occurring psychopathology. suggesting healthcare professionals should be alert to the risk of SI in anybody with a history of psychotic experience.[47] These findings emphasize the need for healthcare professionals to recognize the importance of psychosis as a risk factor for SI and suicidal behaviors. 

Etiology

For over 50 years, researchers have been investigating possible risk factors and mechanisms that lead to suicidal ideations (SI) and behaviors. It turns out there are thousands of risk factors associated with suicide ideation, but not a single one of these SI factors or grouping of SI factors is a reliable, clinically useful predictor of who will end their life by suicide.  The current theories suggest that suicidal thoughts and behaviors occur in response to complex interactions between psychological, biological, environmental, and cultural factors. Brief summaries of psychological and biological theories are provided in this section. Readers are encouraged to review the primary sources for a more comprehensive understanding. 

Diathesis-Stress Models of Suicide

This model proposes that stress alone is an insufficient explanation for suicidal thoughts and behaviors because, given the same stressful life events, only a small percentage exhibit SI and suicidal behaviors. The diathesis models suggest that variations and interactions between biological and/or psychological risk factors predispose some individuals to a vulnerability, referred to as a diathesis. When predisposed individuals encounter particular life stressors, the resulting confluence of distal and current risk factors can trigger suicidality. [48] 

A systematic review of 40 years of research studies showed a relationship between negative life events and suicidal ideations. However, the relationship between negative life events and SI was most evident in people with severe ideations. The authors concluded that better research designs are needed because most studies in their review were limited by poor design and ambiguous terminology. No evidence was found that supported the notion that positive life events reduce SI. [49] 

More than 3000 constructs/factors have been proposed and tested as possible risk factors for suicidal ideations and behaviors over the past 50 years. Franklin et al. explained the only way to accurately determine whether a construct/factor is a key risk factor is by using a longitudinal research design to examine the outcomes of the factor's effect years later. They completed a meta-analysis of every longitudinal study completed in 50 years. The average follow-up was 4 years, although some were as long as 10 years duration. When reviewing the hundreds of studies, these researchers noted that a preponderance of studies (almost 80%) focused on five broad categories of risk factors for SI and suicidal behaviors. The five categories and examples of factors studied within these categories are summarized below:[50]

• internal psychopathology (e.g., anxiety disorders; mood disorders; hopelessness; emotion dysregulation; sleep disturbances)
• demographic factors (e.g., age; education; employment; ethnicity; gender; marital status; religion; socioeconomic status)
• prior suicidal thoughts and behaviors (e.g., prior deliberate self-harm, nonsuicidal self-injury, suicide attempt, suicide ideation)
• external psychopathology (e.g., aggressive behaviors; impulsivity; incarceration history; antisocial behaviors; substance abuse)
• social factors (e.g., abuse history; family problems; isolation; peer problems; stressful life events) [50]

Coding the constructs/factors that were tested in the 365 studies revealed (n=495) "protective factor" cases and (n=3428) "risk factor" cases for meta-analysis using random effect models. The results showed that no risk factor alone or in the combinations that have been studied has predictive abilities that are much better than chance. The only two factors that emerged were hopelessness and previous SI and suicidal behaviors, but these were weak predictors of future suicidality. The researchers concluded there is a need to improve the quality of research being conducted and to consider new approaches such as using machine learning algorithms to identify combinations of risk factors that are predictive of suicidality. A half-century of research has not provided empirical knowledge that is needed to identify clinically important risk factors for SI and suicidal behavior.  [50]

The Ideation-to-Action Framework of Psychological Theories

The ideation-to-action framework is the basis for several theories that share the same basic underpinnings.  These theories contend there are separate, distinct stages that mark the development of SI and its progression from SI to lethal suicide attempts. Each stage has its own risk factors, processes, and explanations. Importantly, all of these theories share the belief that suicidal desires do not automatically lead to suicidal actions. This is promising because it offers hope that ideators can be helped before they acquire the capacity to inflict life-ending self-harm. [51]

The Interpersonal-Psychological Theory of Suicide (IPTS) developed by Joiner in 2005 predicts that the combination of feeling like a burden to others ("perceived burdensomeness") and being socially alienated ("thwarted belonging) will produce death desires or SI. [52]  However, individuals will not act on these desires unless they overcome the normal and protective self-preservation mechanism -- the fear of death.  This is referred to as acquiring ("the capability") to die. This capability develops with repeated exposure to painful, provocative events and/or fearful experiences, which creates habituation. Examples of exposures that can produce habituation to the fear and pain involved in self-inflicted harm include childhood maltreatment, combat exposure,  self-starvation seen in anorexia, non-lethal suicide attempts, [53] physical pain, plus many more. The interactions between low family support (thwarted belonging) and perceiving that one doesn't matter (perceived burdensomeness) predicted current active SI beyond depression indices in Joiner's original test of the theory. This theory was the first to explain the mechanism of developing SI and explain why most ideators did not attempt suicide because they did not acquire the capability. [52] The theory proposes that active SI occurs when there is the interplay between perceived burdensomeness and thwarted belonging; passive SI occurs when only one is present. [51]  A systematic review of 58 articles that tested IPTS constructs and mechanisms showed that the relationship between perceived burdensomeness and its influence on SI was most frequently tested in studies and best supported. Interaction effects between thwarted belonging and SI were modestly supported. The capability for suicide showed weak associations [54], but some may be attributed to the inconsistent operational definitions used in the studies. [53]  A meta-analysis of the theory was completed using hypothesis-driven random effects models from 122 distinct samples. The researchers concluded their findings supported IPTS. The interaction between the constructs of thwarted belongingness and perceived burdensomeness was significantly and robustly associated with suicidal ideation. Additionally, the interaction between thwarted belongingness, perceived burdensomeness, and the capability for suicide was significantly related to an increase in the number of previous suicide attempts. However, the effect sizes for these interactions were modest. Inconsistent definitions of the capability for suicide may have contributed. [53] Alternative configurations of theory variables were also shown to be useful for predicting suicide risk as theory-consistent pathways. The researchers recommend IPTS as a framework for understanding suicidality but offered suggestions and limitations. [55]

Three Step Theory (3ST) by Klonsky and May, published in 2015, proposes that pain and hopelessness are the primary factors that produce SI. If the cause of pain (psychological, interpersonal, and/or physical acute or chronic pain) resolves, or when there is hope that the pain will diminish with time or effort, SI will dissipate because the individual's focus will be on a better future rather than suicide. [53] The second step in 3ST occurs if the pain does not resolve but escalates instead, the individual's sense of connectedness becomes overwhelmed. The connectedness can be to loved ones, valued roles, or anything that provides the individual with a sense of meaning and purpose. If the balance shifts, so pain is greater than connectedness, SI increases in intensity, moving from passive ideation to active SI. The third step in 3ST occurs when the capacity to attempt suicide occurs. Dispositional characteristics that contribute to capacity include a genetically high pain threshold or low fear of death. Acquired contributors include the habituation of painful or provocative events. Practical contributors include having knowledge and access to lethal means, such as soldiers and gun owners; healthcare professionals having access to and knowledge of medications; [53] use of social media to learn novel approaches [56]. Research studies using the 3ST show the interaction between pain and hopelessness was able to predict SI better than perceived burdensomeness robustly and thwarted belonging in IPTS; the degree of connectedness successfully predicted SI even in the presence of pain and hopelessness; and finally, separating suicide capacity into dispositional, acquired, and practical contributors were shown to predict suicide attempt histories over and above SI. The 3ST was used with a sample of Chinese university students (n = 1,097). Each participant completed a battery of questionnaires indexing psychological pain, hopelessness, connectedness, suicide capacity, suicide ideation, and suicide attempt. The prevalence of suicide ideation was 21%, and the model successfully predicted relationships amongst the constructs and risk of SI and attempts. [57] Klonsky reported the findings of another study which provided participants with a list of 42 factors (including sleep problems, agitation, giving away possessions, family conflict, disengagement from social activities, anger, guilt, and shame). The two most frequently reported factors that preceded suicide attempts and deaths were the presence of pain (emotional misery or pain’) and feelings of hopelessness about the future.  [53] These findings suggest healthcare professionals should ask patients about their hopefulness, particularly when experiencing emotional or physical pain, to determine how these constructs are balanced. Furthermore, understanding the factors that contribute to the capacity for suicide may be warranted.

The Integrated Motivational-Volition Theory (IMV) model by O'Connor in 2011 differs from the IPTS in several ways. First, the pathways to SI are defeat and entrapment instead of belongingness and burdensomeness. A meta-analysis of 40 studies (n > 40,000) found similar-sized, strong relationships for defeat and entrapment in patients with depression, anxiety problems, PTSD, and suicidality. [58] Additionally, findings show robust associations between SI and experiencing defeat and entrapment. [59] The IMV theory expands the theoretical framework for the capability of suicide by specifying additional factors such as impulsivity that may contribute to the progression from SI to suicidal behaviors, although there are insufficient data to support this at present. Similarly, although there are preliminary findings that support social contagion and future flashes/mental imagery of suicide, further studies are needed to verify this pathway. [53] [60] 

Implications for Healthcare Professionals

Given the moderate to strong associations between the risk factors in these theories and the presence of SI, healthcare professionals may consider including SI assessment interview questions to elicit a patient's physical or/and emotional misery or pain; feelings of defeat and entrapment; their level of hopefulness and sense of connectedness. Additionally, reducing the capability for suicide by restricting access to lethal means (e.g., firearms) is recommended in suicide risk reduction guidelines, although more research is needed to determine the best way to limit access. Incorporating strategies to address cognitive distortions and build hopefulness and connectedness may protect overwhelming perceptions of defeat and entrapment.  

Biological Factors 

• Researchers have investigated a variety of biological factors hoping to find clinically useful biomarkers to detect severe SI before individuals make lethal attempts.  
• Substantial knowledge has been gained in the past decade, but additional research is needed to understand the mechanisms and identify relevant biomarkers.

HPA Axis Dysfunction

Dysregulated hypothalamic-pituitary-adrenal (HPA) axis function was initially proposed decades ago within the stress-diathesis model. More recently, it has been discussed in the context of epigenetic factors, which may link childhood trauma and adverse events to subsequent conditions thought to be associated with HPA dysregulation. [61] To briefly review, environmental stressors trigger the central release of serotonin, dopamine, and norepinephrine; stress also leads to the production of glucocorticoids via the HPA axis. This process begins when the hypothalamus produces corticotropin-releasing hormone (CHR), which binds to anterior pituitary receptors. This prompts adrenocorticotropic hormone (ACTH) to be secreted by the pituitary gland into the circulatory system. The adrenal gland responds to ACTH by producing and releasing glucocorticoids (GC). Circulating GC binds to the GC receptors in the hippocampus and anterior pituitary gland and inhibits the release of CRH. This negative feedback loop is essential for proper homeostasis. The FKBP5 gene regulates and influences the sensitivity of the GC receptors. Polymorphisms of the FKBP5 gene produce differences in GC receptor sensitivity. [62] Dysregulation of the HPA axis occurs when the sensitivity of the GC receptors in the hippocampus and anterior pituitary is reduced. This leads to hypersecretion of CHR, which causes an overproduction of ACTH, consequently increasing the production of GCs.[62] Research studies have had mixed findings when examining the relationship between SI and HPA axis dysregulation. Some investigators have queried whether this could be explained by the heterogeneity of suicide ideators and the possibility that different subgroups of the SI population have different cortisol responses. A recent study showed that individuals with only a short duration of SI had greater increases in cortisol levels than individuals with prolonged/continuous SI. However, baseline cortisol levels and total output were not related to either the duration of SI or the severity of SI (as measured by SSI scores). [63] Another study (n=160) found a family history of suicide or a suicide attempt was associated with lower cortisol levels in response to elevated stress. Individuals who reported SI had lower cortisol levels compared to those without SI.[64] 

Neuroinflammation and Immune System Dysfunction

Microglia are the primary immune system cells in the central nervous system, and they function like peripheral macrophages. When injury or pathogens are detected, the microglia become activated. They rapidly change their morphology, quickly proliferate and begin to migrate to the location of the injury/pathogens to phagocytose and destroy pathogens and remove damaged cells. Like the peripheral immune system, they release a multitude of pro-inflammatory and anti-inflammatory mediators (cytokines and chemokines). Post-mortem examinations of suicide victims' brains show an increased density of microglia in the anterior cingulate cortex, the dorsolateral prefrontal cortex, and the mediodorsal thalamus regions. [65] In living subjects, PET scans show increased microglial activation in the presence of SI. [66]  Microglial activation affects the regulation of the tryptophan-kynurenine pathway and increases the ratio of neurotoxic metabolites relative to neuroprotective metabolites. [65]

Kynurenine pathway (KP) dysregulation has received considerable attention recently. The amino acid tryptophan is primarily catabolized in the kynurenine pathway. The enzymes that regulate kynurenine metabolism in the CNS are upregulated in response to inflammation, which incidentally lowers the availability for serotonin.  The tryptophan is converted to l-kynurenine, which is then enzymatically processed into neuroactive metabolites, including kynurenic acid in the astrocyte cells and quinolinic acid in the microglial cells.  The kynurenic acid is an N-Methyl-D-aspartate (NMDA) receptor antagonist with neuroprotective and anti-convulsive properties.  Quinolinic acid, on the other hand, is an NMDA agonist which has been described as neurotoxic. During neuroinflammation, increased production of quinolinic acid creates an imbalance with the neuroprotective metabolites. This is thought to create downstream effects in the glutamate systems. [67] Glutamate neurotransmission is essential for normal cognitive and emotional processing.  Alterations in glutamate/glutamine and GABA levels have been found in the prefrontal cortex of unmedicated depressed patients. [68] Additionally, alterations in the regulation of a variety of glutamate receptors have led to suggestions that dysregulation in the entire glutamate signaling system within the brain may contribute to a lack of cognitive flexibility, increased impulsivity, poor working memory, depressed mood, and suicidality. [68] Targeting the KP enzymes may be effective to treat treatment-resistant depression and suicidality. [67]

Translocator protein (TSPO) is upregulated in activated microglial cells and is an indication of neuroinflammation. Positron emission tomography (PET) scans were used to compare the level of TSPO in the brains of depressed patients and healthy controls. The scans revealed a significant increase in the availability of TSPO in depressed patients vs. healthy controls (p = .005).  The elevation of TSPO (large effect size) was found in the anterior cingulate cortex.  TSPO was not elevated in the patients who did not experience suicidal ideations but was shown to be significantly increased in those with SI, most robustly in the regions of the anterior cingulate cortex (p = 0.008) and the insula (p = 0.023). [66] 

Brain-derived neurotrophic factor (BDNF) is involved in neurogenesis and synaptic plasticity.  Some studies have shown that lower concentrations of BDNF in the CNS and peripherally may be useful biomarkers for suicidality. A meta-analysis concluded there was insufficient data and recommended additional research. [69] More recently, a study of US Army and National Guard troops who served in Afganistan or Iraq (n=3,889) found that current SI was associated with lower BDNF (OR = 1.5, 95% CI = 1.5-1.8, p=0.0002). [70]

Genetic Factors

An emerging area of research focuses on genetic factors. Variants of the FKBP5 gene have been widely studied and implicated in depression and SI and behaviors. A recent systematic review and meta-analysis focused on studies that examined the role of the FKBP5 gene variants rs1360780, rs3800373 and rs4713916. Results showed that the variants rs3800373 and rs4713916 were associated with a significant increase in the risk for depressive disorders when modeling with heterozygous and dominant configurations. A stratified analysis based on ethnicity showed the same variants were significant for increased risk of depression in Caucasians. However, when the FKBP5 gene variants were analyzed concerning suicidal behavior, the rs1360780 was a significant risk for suicidal behavior risk within the overall population, and rs3800373 was significant in the subgroup who were known to have completed suicide. Although the evidence shows FKBP5 gene polymorphisms are associated with the risk of depressive disorders and suicidality, the authors recommended additional studies with larger sample sizes to confirm their results. [71]

Epidemiology

Globally

There are approximately 800,000 deaths from suicide annually, which, as the World Health Organization points out, exceeds the number of people who die each year from homicide and wars throughout the world. Suicide was the second-leading cause of death for 15-29-year-olds globally in 2016. In 2013, the World Health Organization declared suicide a global public health crisis due to the rising rates and urged all member nations to implement evidence-based prevention strategies. Approximately 79% of the world's suicides occur in low-income and middle-income countries where inadequate access to services, poverty, and mental health stigma are cited as barriers to seeking care. In developing countries, most suicides are completed by people living in poverty, most often young single women who lack education and struggle to support themselves.  Older adults are among the high suicide rates throughout every region of the world.

Due to their large populations, India and China contribute to over half of the total global suicides. However, when comparing 2016 suicide rates per 100,000 of the population, the WHO reported India (18.5/100,000) and China (7.9/100,000). The highest rates are in Eastern European Baltic countries (Russian Federation 48/100,000; Kazakhstan 40/100,000 Ukraine 34/100,000) and South Korea (30/100,000). Throughout the world, countries with predominantly Muslim populations have the lowest reported rates (e.g., Pakistan 3/100,000; Saudi Arabia 4.6/100,000). Latin American countries (Honduras 5.3/100,000; Mexico 8.2/100,000) also have low rates compared to European nations (France 17.9/100,000, Germany 13.6/100,000 and the UK 11.9/100,000) or North Americans (Canada 15/1000,000 and USA 21/100,000). Some literature suggests cultural differences and religious beliefs about suicide influence suicide rates. In middle-income and high-income nations, male suicide rates are much higher than female rates, with males surpassing females by 4:1 or higher.

A large international comparison of the risk factors for SI and suicide attempts was obtained from data collected in the WHO World Mental Health Surveys, which involved structured face-to-face interviews with participants (n >108,000) from 21 countries. [72] Diagnosed mental health disorders were associated with SI and suicide attempts in all countries. However, there was a key difference. Mood disorders were the strongest predictors for SI and non-fatal suicide attempts in developed nations. Meanwhile, substance use, impulsivity, and post-traumatic stress disorder were the strongest predictors of SI and non-fatal suicide attempts in developing countries. The importance of SI was evident in the study's finding that mental health disorders were primarily associated with predicting the onset of suicidal thoughts but not predicting the progression from thoughts to attempts. [72]

SI and suicide methods and rates vary by country, although there are some patterns that cross all nations.  Economic recessions and job losses [73][74]correspond to the increased prevalence of SI and suicides.[75] The WHO collects global data for mortality, including the causes of suicide deaths. This enables the collection and identification of regional and global trends, so appropriate surveillance and prevention measures can be implemented. Ingestion of pesticides was the leading means of suicide in low and middle-income nations. Efforts have been undertaken in many countries to reduce access to toxic pesticides and highly-lethal means, such as firearms. [76] This requires a collaborative effort between healthcare professionals, community health services, policy-makers, advocacy groups, industries, and public media. 

There is an important role for epidemiologists to monitor risk factors for suicide and emerging trends, including new and unusual suicide methods. For example, within a decade, there was a 65-fold increase in suicides by young people in Taiwan who died in charcoal-burning suicides. This involved burning charcoal fires indoors in small-contained spaces. The practice spread regionally and was adopted by different demographic groups, becoming a preferred means for young females in some locals.  Several years later, charcoal burning deaths appeared in England, and epidemiologists traced it to the dissemination of this highly lethal means via social media. [77] The UN and member nations are now collaborating on strategies to reduce global suicide rates and advocating for better monitoring of social media platforms and responsible media coverage of suicides to reduce sensationalizing. 

United States

The CDC records show over 47,000 suicides occurred in 2017. From 1999-2016, American suicide rates increased by nearly 30%. Despite the assumption that individuals who take their own lives have psychiatric histories, the data show that in 2015, over 50% of suicide victims had no known mental health diagnosis. This means other factors need to be more fully appreciated as key risk factors for lethal self-directed violence. Research studies point to the significance of losses or difficulties with relationships, financial troubles, employment problems, and pending legal issues as key contributors. Rural areas of the country, such as the western states, have higher suicide rates compared to more urban areas in the east. The prevalence of firearms, remote distance from support systems/mental health services, and financial stressors may contribute.

Firearms account for 50% of suicides in America, followed by suffocation (28%), drug poisoning (11%), and non-drug poisoning (3%), according to the CDC WISGARS database. The crude suicide rate is highest (20.1/100,000) for individuals aged 45-59 years and those aged 85 years and above. Males have higher rates of suicide than females, largely due to the differences in the lethality of means used during suicide attempts. Older white males in rural American states usually die on their first suicide attempt due to their use of firearms. Females in the USA are more likely to attempt by overdose and be rescued.

However, it is the increased prevalence of suicide in youth and younger adults that garners attention.  CDC data (2008-2017) shows an increase in the crude suicide rate for ages 18-35 years as 12.7/100,000 rising to 17.0/100,000 and affecting both males and females. Beyond the dramatic increase in completed suicides, there is a corresponding increase in suicide attempts. It is estimated by the American Foundation for Suicide Prevention (AFSP) that in 2017 there were 1,400,000 suicide attempts and almost 10,000,000 people with suicidal thoughts in the USA. 

The American Psychiatric Association Assessment and Management of Risk for Suicide Working Group 2013, p. 22 cites the risk of suicide increases with:

• Psychiatric diagnoses:  especially mood disorders, psychotic disorders, anxiety disorders, posttraumatic stress disorder, substance use disorders, and disorders associated with impulsivity
• Medical conditions: particularly those that are chronic, debilitating, disfiguring, or painful

History and Physical

Screening for Suicidal Ideation

A variety of SI screening and suicide risk assessment scales have been validated and meet the Joint Commission's requirement for primary care, ED, and behavioral health professionals to assess individuals with behavioral health issues. However, an over-reliance on any scale should be avoided.  Literature shows the intensity of SI fluctuates, but more importantly, no scale has been shown to predict imminent suicide risk accurately.  Depending on what scales are used, the same individual often scores positive for SI on one scale and negative on others. [3] This emphasizes the need to understand the limitations of any screening or assessment tool, as well as the importance of clinical reasoning and good interviewing skills.  Tools should be consistent with the patient's age, the setting, and organizational policies. For example, ED screening tools should aid in triage and consist of 5 or fewer items. [78] Although universal screening is not mandated by the Joint Commission, research shows there is a high prevalence of occult SI which may be missed in ED if not screened.  The Emergency Nurses Association (ENA) Clinical Guidelines (2017) recommends that ED nurses provide universal SI screenings to all patients. The validity and reliability of screening tools rest on clinicians using the instruments as they were designed, so it is important to never deviate from the tool's wording.  [78] 

Healthcare professionals must be mindful to guard against assumptions, stereotypes, and behaviors that suggest impatience, negativity, or minimize the patient's concerns. [79] A study compared the suicide risk rating scores that patients assigned to themselves and the scores that the clinician assigned. The clinicians' scores were significantly lower, which suggests they may underestimate the seriousness of an individual's suicidality. [80] Standardized risk assessment tools may be inadequate measures of suicidality in different cultural groups or for people with cognitive decline and dementia. [81]

Initial Assessment of Suicide Risk in ED:  2017 Emergency Nurses Association Clinical Guidelines reported a moderate amount of evidence (Level B) that supports the validity, feasibility, and reliability of these tools for use in ED) [78] 

• Ask Suicide-Screening Questions (ASQ) is a 4-question screening tool for pediatric/young adults who present to ED with medical complaints. It is recommended to administer without the parent/guardian being present. 
• Manchester Self-Harm Rule (MSHR) uses 4 questions to identify the ED patient's risk of suicide or repeating self-harm based on their history.
• Risk of Suicide Questionnaire (RSQ) is a 4-question screening tool suitable for 8 years through adult; it takes 90 seconds to complete.

Discharge/Disposition Assessment from ED: 2017 ENA Clinical Guidelines report a moderate amount of evidence (Level B) for these instrument in the ED setting: [78] 

• Behavioral Health Screening Emergency Department (BHS-ED) is a shorter 10-minute web-based version of the comprehensive BHS tool. The BHS-ED focuses on depression, suicidal ideation, posttraumatic stress, risk behaviors, and stress. 
• Columbia Suicide Severity Rating Scale (C-SSRS) Assesses for risk behavior, SI and guides the assessment of the lethality of the suicide risk. Joint Commission states that it can be used in all settings, with all ages, however using "the C-SSRS Full Version without the risk assessment is not sufficient to qualify as an evidence-based suicide risk assessment process. Assessment of the risk and protective factors, in a structured or unstructured way, is required in addition to the suicide inquiry" is stated on the Joint Commission Suicide Prevention Resources website.
• Geriatric Depression Scale (GDS) uses a 15-question depression scale to evaluate and depressive symptoms and SI. Some evidence this relationship between depression and suicidality was stronger in the old-old than in the young-old.
• The ReACT Self-Harm uses data collected based on 4 elements of concern to assist with disposition decisions related to follow-up. 

Additional screening and assessment tools:  The Joint Commission included several of these in its list of validated instruments. 

• Beck Hopelessness Scale (BHS) - A 20-item questionnaire measuring pessimism/hopelessness, a construct related to SI. [82]
• Beck Scale for Suicide Ideation (BSI) - Joint Commission approved BSI, a 21-item self-report instrument for detecting and measuring the current intensity of the patients’ specific attitudes, behaviors, and plans to commit suicide during the past week. The BSI was developed as a self-report version of the interviewer-administered Scale for Suicide Ideation. The first 5 questions can be used as a screening tool.  [83]
• Scale for Suicide Ideation-Worst (SSI-W) Joint Commission approved SSI-W for in-patient and out-patient areas use: This 19-item rating scale developed by Beck is interviewer-administered and measures the intensity of patients’ specific attitudes, behaviors, and plans to commit suicide at the period when they were the most suicidal. The instrument is validated and reliable and takes 10 minutes to complete. [84]
• Death/Suicide Implicit Association Test (IAT)  [85] (IAT) 
• Geriatric Suicide Ideation Scale (GSIS) [86]
• Nurses Global Assessment of Suicide Risk (NGASR) identified as a reliable tool, particularly useful for novices, to aid in overall clinical assessment [87]
• Patient Health Questionnaire (PHQ-2 and PHQ-9) [88]
• Scale for Suicidal Ideation (SSI), a classic 19-item questionnaire for clinicians or self-ratings (paper or computer-based), validated translations to numerous languages [89]

Lacks sufficient validity for ED use: Research evidence suggests these scales currently lack sufficient validity to use for screening or assessing the risk of suicide. 

• SAD Person Scale (SPS) Systematic review concluded this tool could not currently predict suicide. [90]  
• Modified SAD Persons Scale (MSPS) Not useful for predicting suicide in ED patients [91] Do not use MSPS (or other risk classification scales) alone to determine treatment or predict future risk due to poor sensitivity [92]

Clinical Interview Considerations 

In addition to the patient's current expression of SI, it is important to determine if there is a personal history of suicide attempts, interrupted or aborted attempts. Additionally, even if no prior SI or plans/attempts have occurred, any reckless, self-endangering behaviors or self-inflicted harm, such as cutting, should be queried. The lifetime risk of suicide increases with prior self-harm, but the reason is unknown.  Some theories suggest that prior self-harm habituates the individual to tolerating pain and normalizes self-harming behaviors, thereby increasing one's capability for suicide. [93] Beck (1998) suggested that eliciting information about an individual's worst-ever episode of SI and behaviors is the best lifetime predictor of future suicide. The more lethal and detailed the plan, as rated with the Scale for Suicide Ideation (SSI-W)[84], the higher the lifetime risk. [93]  Franklin et al. highlighted the difficulty in predicting future behavior given the lack of empirical evidence to reliably connect past and current thoughts and behaviors to future actions. Healthcare professionals should bear in mind that a prior suicide attempt is a predictor for subsequent lethal attempts, but it is only a weak predictor. Unfortunately, there are no strong predictors to rely upon. [50] A different approach, which needs further study, suggests that a patients' projections of themselves in the future may provide useful assessment data. Ideators with PTSD, BPD, and to a lesser extent MDD reported experiencing mental imagery (or a flash-forward) that depicted visual images of killing themselves. [94] An Asian study found that visual imagery of suicide was only present in people with severe SI and that the mental imagery subsided when the SI subsided. [95] 

Interviewing to Assess Suicidal Ideation  

Literature shows nurses, particularly those who do not work in mental health, frequently have negative views toward people who self-harm, but education and reflective learning activities produce more favorable attitudes. [96] Physicians' beliefs about suicide may also include misinformation, such as people who talk about suicide do not kill themselves.  [97] Therefore, healthcare professionals should be mindful of their assumptions about suicidal individuals because this can impact the quality of assessment and treatment. 

Examples of SI assessment questions from literature, including VA/DoD Clinical Practice Guideline for Assessment and Management of Patients at Risk for Suicide, Registered Nurses Association of Ontario Best Practice Guidelines for Assessment and Care of Adults at Risk of Suicidal Ideation and Behaviour (2009), Clinical Guidelines from the American Psychiatric Association Clinical Guidelines for Assessment of Suicidal Adults (2016): 

Characteristics of Current SI - onset, frequency, duration, intensity, triggers, associated factors, ability to control, attribution, passive or active SI thoughts

a) Suicidal Ideation – Ask questions to elicit thoughts on living and dying. Distinguishing between passive and active SI is typically done to identify if there is an imminent short-term risk, although literature shows attempts do occur without prior expressing of SI, and long-term risk is equal with passive and active SI. The nature of SI can fluctuate rapidly, so assessing worst-ever and more recent fluctuations is advised. Age of ideator may influence the character of ideations.

• Ambivalent thoughts between continuing to live vs. wanting to die are common.
• Often death is not the goal but the only option the ideator can envision as a solution to present or upcoming crisis. (e.g., financial issues, job loss, legal problems, relationship issues) Executive functioning issues may limit the ability to problem-solve or identify other options
• What is the character of ideations? (passive wishes to never awaken and/or actively kill oneself)
• How frequent are ideations? Recently increased? (several times/week, daily, hourly) 
• How long do the thoughts/urges last? (minutes, hours, constant)
• What are triggers that are associated with the onset, duration, and intensity of these thoughts? (e.g., when alone, at night, after arguments, while drinking?)
• How much control do they have over the thoughts? (e.g., controllable except when drinking?)
• What does death represent? (e.g., punishment, reunion, escape)

b) Plan –  Asking about a plan is crucial because the presence of a plan is a key variable that is associated with suicide risk. Detailed plans are associated with more serious attempts

• Inquire if the person has thought of possible plans, considered options, or chosen a method? (e.g., overdose, firearm, suffocation, jumping, MVA, etc.)
• If a present plan is denied, inquire about any previous plans (Lifetime risk is associated with SI/plan at worst-ever, low-point)
• Has the person taken steps to put a plan in place? (e.g. identified a location(s), considered logistics, etc.)
• How detailed and specific is the plan? (e.g., precise or vague locations, etc.)
• Have they done anything to prepare? (e.g., hoarding medications, purchased rope, etc.)
• Have they rehearsed the plan mentally or role-played any aspect(s) of the plan? (e.g., putting a rope around the neck, holding/positioning firearm, etc.)
• Is there a specific time/date or event that was chosen for the attempt? (e.g., an anniversary, a red line)
• Motivation - (e.g., suicide pact, bereavement; attention, to escape humiliation/shame; delusional thoughts or command hallucinations; revenge, etc.)

c) Access to Means – Assess to lethal means increases the risk of a lethal attempt.

• Does the person have access to the chosen method? (e.g., access to a firearm, medications, etc.)
• Does the individual have career-based knowledge or familiarity? (e.g., military, police, anesthesiologist, etc.)
• Is the chosen plan plausible? How easy would it be for the patient to access such means or put plans in place?

d) Intent – To what extent does the patient intend to die? Note: Anything above zero intent to die is considered to affirm suicidal intent.

• What is the primary reason for self-harming thoughts? (e.g., attention, punish self or other(s), insurance benefit, etc.)
• What is the intensity of the person's desire to die? (level of ambivalence with life/death?)
• Has the person made preparations for death? (e.g., put affairs in order/will/note, prepared the location, gave away gifts/possession(s), etc.)
• Does the plan include steps to minimize being discovered? (e.g., time of day and/or location where unlikely to be rescued is associated with higher lethality)
• Has the person shared intent with anybody? (e.g., sought medical care, confided to anybody about their hopelessness, SI)

e) Lethality – The healthcare professional's objective appraisal of the lethality of the plan or attempt is not as important as whether the person believes it could be lethal.

• Did the person believe that the chosen method and plan would be lethal?
• Would the chosen method and plan allow for discovery and lifesaving intervention?
• Does the plan include the use of alcohol or drugs? 

f) Protective factors – Are there any people or circumstances that allow the patient to want to go on living?

• Are there any barriers that prevent the person from taking their life?
• Would the person want to continue living if certain issues/factors could be resolved? 
• Social factors (sense of responsibility to others; cultural and religious beliefs?)
• Quality and strength of therapeutic alliance with a counselor, healthcare professionals, etc. 

g) Previous attempts or aborted/interrupted attempts? (Prior attempts are associated with increased risk future attempts, literature shows a significant but weak correlation)

• Ask about past or aborted suicide attempts? (When? What were the precipitants? What was the method? What was the medical severity? Were alcohol or drugs involved?
• Has the person had prior intentional self-injury without suicide intent? Ask when. What was the precipitant? What was the medical severity? Where drugs or alcohol involved?

Present/Past Risk factors that Should be Assessed

Anxiety symptoms and panic attacks; Impulsivity; Psychotic thoughts; Anger/aggression; Mood - depression, hopelessness; Substance use (alcohol or other); Trauma; Prior history of a psychiatric hospitalization; a history of suicidal behaviors in biological relatives

Health and psychosocial stressors - Painful, chronic, debilitating, disfiguring or terminal medical illness; insomnia; head injury; neurological disease; financial, legal, occupational/school or relationship problems

Evaluation

As discussed various screening tools may assist the practitioner in identifying at-risk patients. The sensitivity and specificity of screening tools are generally highly variable and are partially dependent on the experience of the examiner.

Treatment / Management

Although SI with serious intent to die is a clinical emergency, to date, there is no evidence-based treatment to manage it. [98]  A systematic review of interprofessional clinical practice guidelines (n=10) for suicide prevention, assessment, and management of suicidal thoughts and behaviors showed tremendous variations from one guideline to the next. All formal guidelines included detailed recommendations for using evidence-based risk factors and protective factors, but only 80% provided recommendations for assessing suicidal intent, and when this was mentioned, it was only briefly addressed. A gap in all of the guidelines except the DOD/VA guideline was the lack of inclusion of patient-driven safety plans to help identify supports, resources, and coping strategies. Another gap in most guidelines was the omission of outpatient intervention safety strategies such as restricting access to means. [1] These gaps will be addressed in this section.

No-harm Contracts vs. Crisis Safety Plans  

• No-harm contracts, sometimes called no-suicide contracts, were advocated as best practices and taught to healthcare professionals for many decades. The value of these began to be questioned over a decade ago, and some findings suggested they may have detrimental effects.
• Current evidence supports the use of Safety Plans, which are made in collaboration with the client and personalized to help them identify triggers and use internal and external coping strategies.

No-harm Contracts vs. Safety Planning: A recent RCT compared the use of no-harm contracts with standard or enhanced crisis safety plans. Active-duty soldiers (n=97) who presented with SI and were being treated for their behavioral health concerns were randomly assigned to one of three groups. At the 6-month follow-up, approximately 5% of the participants who had received crisis safety plans and 19% who received no-harm contracts attempted suicide (log-rank χ(1)=4.85, p=0.028; hazard ratio=0.24, 95% CI=0.06-0.96). This suggested a 76% reduction in suicide attempts in the groups with either the standard or enhanced safety plans (both forms of safety plans were equally effective). Additionally, the safety plans were associated with a significantly faster decline in suicide ideation (F(3,195)=18.64, p<0.001) and fewer inpatient hospitalization days (F(1,82)=7.41, p<0.001). [99] The evidence from this study suggests safety planning to develop personalized strategies to deal with crises that provoke SI is more beneficial than no-harm contracts. 

Comparing Quality of the Safety Plan to Suicide Attempts: The quality of the safety plan has been shown to affect patient outcomes. DOD/VA guidelines for the management of SI include developing a client-focused safety plan. Researchers examined the documented safety plans for Veterans and rated the quality of the plan based on whether the plan was fully completed and the degree to which it showed personalization. The plans were compared this longitudinally to subsequent suicide-related outcomes among the Veterans for whom the safety plans were developed. They found that poor quality plans, which were not personally relevant or were incomplete, predicted an increased likelihood of suicidal behavior. Higher quality safety plans, particularly plans with higher scores for part 3, which includes places and people that serve as distractions, had a lower probability of suicidal behaviors. The authors concluded that many of the plans were of poor quality and suggested that more training about how to complete safety plans should be done with healthcare professionals. They also highlighted the importance of personalizing the plans and ensuring completeness in the area that focuses on creating a list of people and places that may serve as distractions. [100]

Creating a Safety Plan: Best Practice

• The safety plan is created after the risk assessment is completed and includes information gathered by the healthcare professional during the interview.
• Collaboration between the client and the healthcare professional is essential to ensure the plan is specific and personalized for each individual. [100].

An example of the Safety Planning used by VA and published in Green et al., (2018) [100]. Some examples are from Green et al., others have been added. 

1. Warning Signs  (e.g., isolating myself, drinking alcohol, thinking I'm unlovable, feeling like a burden)

2. Internal Coping (e.g., playing guitar, meditating, playing with my dog, listening to music)

3. Social Contacts/Locations that serve as Distractions (e.g., John Smith [phone #], Jane Doe [phone #], go to Big Lake, walk downtown)

4. Family or Friends that may offer Help (e.g., Mother [phone #], Father [phone #], Friend Sam [phone #])

5. Professionals or Agencies (e.g. Therapist [phone #], nearest VA Urgent care or ED [address & phone #], National Suicide Help Line 1-800-273-TALK)

6. Reducing the Potential for Lethal Means (e.g., Brother will keep handgun until therapist and Veteran agree Veteran is safe; Provider will work with VA pharmacy for meds to be dispensed in blister packaging; Wife will help manage Veterans meds keeping them stored and locked at all times; Carbon monoxide detector that sends smartphone alert will be purchased)

Technology to Reduce Suicidality

• With suicide rates higher in rural areas, the WHO recommended developing technologies that may be beneficial to support individuals experiencing SI and who are at risk of suicide.
• Further research is needed to test the effectiveness of such strategies. Several examples of technology-driven management tools are provided.

Safety Plans Added to the Patient Portal of Electronic Medical Record - An outpatient behavioral health service integrated the collaboratively developed safety plans of patients into the patients' portal area within the system's overarching electronic medical record (EMR). This enabled both patients and all providers to access the plan 24/7 and gave the patients the ability to securely message healthcare providers as needed through the portal. The ability to modify the plan as a living document was another benefit. Limitations were also addressed, such as the need to log in to access the plan and the questionable feasibility of doing this while in crisis. [101]

Internet-based Self-guided Safety Planning - Individuals (n=150) were recruited from a depression/SI screening website to complete a self-guided Internet-based safety plan and then evaluate its usefulness. Findings showed moderate support for online safety planning, but the majority believed the tool might be more helpful for others than for themselves (p < .001). There were many limitations to this study, and the quality of the safety plans was not assessed. Currently, there is insufficient evidence to recommend this approach. [102] 

Text-based helpline for children and youth - A recent intervention in Denmark uses text-based communication rather than phone calls to assist children in distress. Approximately 13.5% of the text messages involved SI, and these were typically sent by older girls. Within the group of ideators, 26% had severe suicidality. The helpline staff assisted the youth in developing action plans, and over half of the sessions ended with an action plan in place. More of the ideators than non-ideators recontacted the helpline to follow up. Almost 36% of the children who had SI felt better immediately after their initial text session; 24% reported feeling better when they followed up two weeks later, but 37% indicated they felt worse. The interventions that were most associated with positive impact were concluded to be discussing emotions, expressing empathy, and encouraging the children to talk to somebody. When boundaries were set, the impact was more negative. [103] This study suggests that young people may be more comfortable using text messaging than talking on a telephone, especially when experiencing SI. It also indicates that some level of safety planning can be achieved through communication approaches that appeal to this generation. This is consistent with a 2015 Cochrane review of interventions for self-harm in children, which stated, "It is increasingly apparent that the development of new interventions should be done in collaboration with patients to ensure that these are likely to meet their needs. [The] use of an agreed set of outcome measures would assist evaluation and both comparison and meta-analysis of trials" [due to the very low-quality evidence and heterogeneity of studies at the time of the review]. [104]

Limiting Access to Lethal Means

• Reducing access to lethal means is best practice and consistent with several theories of suicidality.
• Documentation suggests few SI patients/families are being counseled to remove or restrict access to lethal means.
• Clinical guidance is needed to identify what to include in lethal means counseling and to train more healthcare professionals to provide it. 

Documentation in EMR of a Discussion to Reduce Access to Lethal Means - A recent study reviewed ED charts (N=800) of patients who screened positive for SI and suicide risk. There was no documentation that any healthcare professional assessed access to lethal means before discharging to home the majority of people who screened positive for SI (n=545, 85%). This study revealed that only 18% (n=145) had any documentation of an assessment of lethal means.  However, for the small group of ideators who were asked assessed for lethal means, only 8% (n=11) had documentation that a healthcare professional discussed a safety/action plan to reduce access to lethal means. This most frequently entailed changing home storage or moving objects out of the home. [105] These findings show the need to document lethal means assessments for all individuals who screen positive for SI, plus the need to have discussions about the removal of lethal means. This may require adding prompts for these assessments to the electronic documentation system.  

Reducing Access to Firearms - A systematic review of studies (n=70) shows that reducing access to firearms corresponds to reduced firearm suicides. Reduced access through legislation has been effective in other countries, but the 2nd amendment in the United States curtails the feasibility of this. Most firearm suicides are from guns purchased years earlier, so restricting purchases may be ineffective as well. The probability of a firearm suicide increases for everybody in the household when a gun is in the home, so education about firearm safety should be provided. This includes keeping the weapon stored unloaded and in a locked location, preferably out of the home if any family member has SI. Ammunition should never be stored with the firearm. The efficacy of smart gun technologies that use fingerprints for locking has not been systematically evaluated yet, although few gun-owners have purchased this technology, so this is a limitation. Firearm suicide rates for males are higher than females throughout the world, and approximately ten times higher in the United States, with older white males having the highest rate. This group is less likely to seek psychiatric care, so discussions about removal or safeguards to reduce lethal means should be initiated by primary care professionals, ED staff, and clergy have been suggested. More high-quality research is needed to identify the most effective measures and the best practices for firearms safety counseling.  [106] [107]

Firearm Safety Counseling and Legal Implications - The American Medical Association (AMA) authored "A Physician's Role in Firearm Safety." This March 2017 document summarized the AMA's position and addressed legal concerns related to counseling individuals about gun safety. The AMA indicated that physicians and other health professionals should be trained to assess and respond to individuals at risk of suicide. Although Florida law does not permit providers to ask about gun ownership, there are no restrictions in any state to providing “lethal means counseling” or asking about firearms in Florida if imminent safety concerns exist. Discussing firearms safety measures is recommended universally as part of pediatric safeguarding risks in the home and should include proper storage and use of devices to prevent accidental firing. When an individual is at risk of suicide, working with them, their family, and support systems are recommended to limit access until the suicide risk subsides. 

Healthcare Professionals Perceptions about Providing Firearm Safety Counseling - A majority of healthcare providers endorse screening for firearms and providing safe storage counseling. However, they had neutral positions on whether to provide patients with firearm locks, and only 15% provided caregivers free safety locks, which were available for distribution. [108] The rate of American firearms suicides is the highest in 8 western states. Approximately half of the hospitals in these states (n=190) participated in a study regarding their discharge counseling practices from ED for suicidal patients. Approximately one-third of the hospitals stated suicidal patients are always either admitted or transferred. Of the EDs that do discharge people with SI, most of the EDs addressed the presence of firearms in the house (80%); access to alcohol in the home (68%), safe storage of medications (65%), and all three of these topics were covered in (52%) of EDs. Written lethal means counseling protocols were present in less than half of the EDs, but the presence of written protocols corresponded to more thorough safety counseling. [109] 

Training Healthcare Professionals - formally training staff on how to counsel patients and families to reduce assess to lethal means has been shown to be beneficial. The message is more likely to be received well by the recipient(s), but without training, nurses report be reluctance because they do not want to appear confrontational. [110] A training program, Counseling for Access to Lethal Means (CALM), was provided to the case managers at an Area Agency on Aging. At the 3 month follow-up, 38% of the case managers had provided counseling to their aged clients about firearms access. [111]  Nurse managers in ED or their Chief Nursing Officer (n=190) from 8 Western US states participated in phone interviews regarding safety counseling. Most (74%) thought their ED providers did excellent safety/lethal means counseling; 77% believed that suicidal patients accepted safety counseling. However, fewer (64%) nurse managers and CNOs reported that safety counseling was supported by hospital administration. Only 69% thought firearm safety counseling was an effective strategy to prevent suicide and the majority (60%) indicated they had doubts about whether suicide is preventable. This emphasizes the need to provide further education about suicide, including evidence-based strategies and addressing the skepticism about the inevitability of suicide. [112]

Environmental Safety Considerations

Home/residential setting:  Treatment of suicidal ideation focuses on keeping the individual safe while treating the underlying causes. The treatment setting will vary based on the severity of the ideations and suicide risk. [78] Knowing the nature of SI triggers for the individual and their intended plan is important to know how to safeguard their environment, especially when the individual is discharged home. Completing a safety plan before discharge from ED is helpful to enable those identified to be contacted and to help make the home environment safe. This may include moving firearms out of the house, leaving them stored with a friend or at a gun club, locking medications or knives, and so on. If certain situations are triggers, such as driving alone, this should be avoided. 

Medical unit:  Often, patients who have attempted suicide require care on medical units until they are stable for transfer to in-patient psychiatry or discharge to another location.  Approximately 25-30% of hospital suicides occur someplace other than in an in-patient psychiatric unit. Bathrooms are the most common locations for suicide in hospitals, especially because of the privacy and abundance of ligature tie points hanging. Staff in ED and on medical units should be aware of this and monitor the patient's environment for equipment and supplies that could provide lethal means. Whenever possible, patients who have attempted suicide, or have a history of SI or attempts, should be in a room where they can be observed from the nurses' station. Universal SI screening is not typically done when admitted to a medical unit for a non-behavioral health concern.  Healthcare professionals should bear in mind that SI can be triggered by painful and/or chronic conditions, anxiety, insomnia, or adverse reactions to medications. Similarly, complications during treatment, such as ICU delirium, can produce fear and provoke aggression or self-harm. Assessing and documenting each patient's mental status each shift is important for staff on all care units. In behavioral health settings, environmental management includes maintaining close observation (typically every 15 min) or constant observation of suicidal individuals. Checking and storing all possessions is important during the admission process, but it is equally important to ensure that no potential means are brought into the patient during a visit.

The majority of suicides on behavioral health units occur in bathrooms and involve asphyxiation using door handles, hinges, or other areas fixation points. Approximately a decade ago, the VA urged staff to conduct thorough environmental risk assessments to detect and correct safety concerns that contributed to suicide risks. The report generated hundreds of ligature tie points and physical materials that could cause harm. The VA system initiation extensive architectural and structural changes to their units, resulting in a suicide rate that dropped by over 80%. [113]  Although the Joint Commission urged similar environmental risk assessments and corrective actions, it was not mandated until 2019. Now all in-patient psychiatric units must remove ligature fixation points, such as door handles/hinges or exposed piping behind sinks, to comply with the Joint Commission policy. 

Pharmacologic Therapy

As mentioned above, most patients with SI have psychiatric diagnoses. Proper treatment of depression, bipolar disorder, and borderline personality disorder is essential. This can include selective serotonin uptake inhibitors, serotonin-norepinephrine reuptake inhibitors, tricyclic antidepressants, lithium, antipsychotics, especially olanzapine can be of benfit.

Electroconvulsive Therapy

Electroconvulsive therapy (ECT) has been used for decades to provide rapid treatment of unipolar or bipolar depression with severe SI, which has not responded to other treatments or when a delay is too risky. However, the mechanisms of ECT that are associated with improvements in depression and reduction of SI are poorly understood. A recent study supports the hypothesis that at least some benefit may be attributable to changes in the immune-moderated tryptophan-kynurenine pathway (KP). As discussed in the pathophysiology section of this paper, the KP metabolites are characterized as neuroprotective or neurotoxic. Analysis of serum blood levels of the neuroprotective metabolites showed increases in both the quantity and ratio of neuroprotective metabolites throughout ECT treatments. The authors concluded their study provides the first evidence that some of the positive effects of ECT on depression are related to changes in the supply and balance between the neuroprotective and neurotoxic metabolites of the KP. [114]

Ketamine for Suicidal Ideation

For half a century, the pharmaceutic treatment of major depressive disorders (MDD) has been driven by the monoamine hypothesis of depression. This theory proposes that a deficiency of synaptic neurotransmitters (e.g., serotonin, dopamine, norepinephrine) produces depressive symptoms, so replenishing the synaptic deficiency would alleviate the symptoms. However, this model does not explain the latency period before benefit is achieved from antidepressant medications nor the reason that approximately one-third of MDD patients do not have symptom reduction with meds. Advances in neuroimaging have enabled investigators to identify structural and functional changes associated with MDD and SI. More recent hypotheses offer explanations that help to explain the limitations of current antidepressants and provide a more comprehensive explanation of possible underlying mechanisms. The difference in mechanisms between acute and chronic stress responses is of central importance, as is the role of glutamatergic pathways. Glutamate is the primary excitatory neurotransmitter in the brain and essential in the regulation of emotion, cognition, and behavior. Exposure to acute stress triggers an increase in glutamate within the prefrontal cortex.  This leads to a brief elevation in the extracellular glutamate and sustained elevations in synaptic strength and activation of the NMDA receptors and α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptors. The AMPA receptor activation increases the release of brain-derived neurotrophic factors (BDNF). However, in chronic stress responses, the opposite occurs. Neuronal hypertrophy in the nucleus accumbens and amygdala is associated with increased alarm responses while executive functioning becomes compromised. Prefrontal glutamate levels are decreased, synaptic strength is reduced, and NMDA and AMPA receptors are less activated, and extracellular glutamate remains elevated. This interferes with normal glutamate neurotransmission by reducing intracellular signaling in the prefrontal cortex and hippocampus. The decreased levels of neuroplastic and neurotrophic factors contribute to the morphological neuronal changes evident in MDD and suicide victims. The neuronal dendrites retract, branching in the prefrontal cortex diminishes, the spine density of neurons in the hippocampus and prefrontal cortex is reduced, resulting in decreased synaptic strength and neurotransmission. [115] [116] 

Ketamine was introduced in 1962 as a medication that provided dose-dependent analgesia, sedation, and anesthesia without suppressing respiratory function or producing hypotension. Despite these advantages, a serious limitation of ketamine was also evident - it produces psychoactive experiences, including depersonalization and hallucinations. Ketamine is an AMPA and NMDA receptor antagonist. Therefore it increases BDNF and promotes a rapid burst of glutamate after administration. This leads to an increase in synaptic strength and activation in the prefrontal cortex, promotes glutamate neurotransmission, and improves prefrontal synaptic connectivity. In 2010, a double-blind RCT of in-patients (n=33) with treatment-resistant MDD  showed rapid and robust treatment effects, with significant improvements evident within hours of administration of IV ketamine (0.5mg/Kg). All of the participants had extensive histories, with almost one-third having attempted suicide and 61% having endorsed SI. The participants had been ill for an average of 26 years, so the significant improvements in hopelessness, anxiety, and depression were clinically important. [117] Subsequent trials of sub-anesthesia IV doses of ketamine (0.5mg/Kg) consistently showed rapid and robust antidepressant effects on patients who had failed to respond to antidepressants, but there were also side effects, including dissociation. A 2018 systematic review and meta-analysis of the response to ketamine (n= 5 studies with 99 subjects) showed a single dose of ketamine, whether administered as an IV bolus or infusion, produced significant reductions in SI within 4 hours. Meta-analyses based on random-effects modeling showed a large effect (SMD=-0.92; 95%CI: -1.40 to -0.44; p<0.001). [98] 

Whether SI reductions were independent outcomes or secondary to improvements in mood was investigated in a recent systematic review and meta-analysis of individual participants' data (n=167) from 10 RCTs. All individuals had active or passive SI at baseline (measured on a validated scale) and were treated with a single dose of IV ketamine. Twenty-four hours after administration, there was a significant decrease in the severity of most participants' SI on both the clinician-administered (p<0.001) and self-reported outcome measures (p<0.001). Effect sizes were found to be moderate-to-large (Cohen’s d=0.51–0.85) at all time points. Sensitivity analysis showed ketamine was associated with significant benefits on the SI items on all scales (all p<0.001) but not on the Beck Depression Inventory (p=0.080). After controlling for the severity of the participants' depressive symptoms, ketamine’s beneficial effect on SI remained significant. [118] This suggests that ketamine may be beneficial in the treatment of SI independent of depression; however, further research is needed. 

Ketamine consists of two mirror-image molecules referred to as enantiomers. In March 2019, one of the mirror image molecules, the s-enantiomer of ketamine, called esketamine (trade name Spravato), received fast-track and breakthrough treatment designation by the FDA who approved its use as a nasal spray for treatment-resistant MDD in adults. Side effects during esketamine trials included sedation, dissociation, hypoesthesia, vertigo, anxiety, nausea, drunk-like feelings, and hypertension. Ketamine has a high risk for abuse as a recreational drug, so esketamine is only available to certified providers within a monitored distribution system. The potential for adverse reactions to esketamine, including the possibility of a paradoxical increase in SI or CV-related complications, prompted the FDA to require a risk evaluation and mitigation strategy (REMS). Patients must continue taking another anti-depressant while using esketamine, and patients will be required to administer their dose at a certified medical office where their response will be observed for two hours after inhalation. Both the patient and provider must sign the patient enrollment form that outlines post-administration safety precautions and an acknowledgment regarding esketamine intended use and possible risks. [117] [115] The long-term effects of esketamine are unknown, so the outcomes of ongoing treatment with esketamine will continue undoubtedly keenly observed. For example, it is unknown whether the decrease in SI will translate to a reduction in attempts and suicides. 

Clozapine for Suicidal Ideation in Schizophrenia

Clozapine, the first atypical antipsychotic, was approved by the FDA in 1972. Although it has life-threatening adverse reactions, including agranulocytosis, cardiac toxicity, and CNS alterations, anecdotal evidence suggested it reduced suicides in people treated for schizophrenia. This prompted a series of RCTs throughout the 1990s. Researchers concluded there were significant reductions in suicide rates when treated with clozapine compared to olanzapine or haloperidol. Based on these findings, in 2003, the FDA approved the use of clozapine as the first and only medication specifically labeled to treat suicidality (in schizophrenia). A meta-analysis based on six studies showed significant reductions in suicides (p<0.0001) and approximately a three-fold reduction in suicide-related behaviors. [119] However, the mechanism that produced the beneficial outcomes was not understood. Some suggested that the closer monitoring of patients due to the life-threatening complications had a positive effect on patients by increasing their contact with healthcare professionals. The evidence that clozapine reduces suicide is no longer considered to be as strong as first believed. [120][121] Nonetheless, clozapine is notable because it remains the only medication that the FDA has labeled for suicide prevention, albeit only in schizophrenia. 

Other Interventions

Few studies of SI treatments are longitudinal, and many have a high risk for bias. The majority of systematic reviews and meta-analyses of treatment interventions acknowledge this limitation. The following findings are from recent systematic reviews and meta-analyses of SI treatments.

Online and mobile telephone apps for the management of SI and self-harm:  Authors of a systematic review and a meta-analysis of 14 studies, all randomized controlled trials (RCTs), quasi-experimental and pre/post observational studies (n>3300 individuals), noted all study designs that had one or more aspects that introduced the risk for performance and detection bias. They concluded that digital apps for SI and self-harm might be more effective than waitlist but also stated it is "unclear whether these reductions would be clinically meaningful at present." [122]

Dialectical behavioral therapy (DBT) for SI and suicidal behaviors: Authors of a meta-analysis concluded that dialectical therapy reduced suicidal behaviors but did not significantly reduce SI (d = -.229, 95% CI = -.473 to .016). They suggested this could be because DBT is aimed at focusing on behavior changes more so than thoughts.  [123]

School-based prevention programs for youth: An expedited knowledge synthesis of studies (n=7) which were RCTs or controlled cohort studies (CCSs), was done to facilitate evidence-informed decision-making. No changes in suicide death rates occurred, but improvements in there were reductions in SI and attempts reported.  [124] 

Interventions for older adults: A systematic review of research studies (n=21) included a variety of programs, mostly aimed at depression, including telephone counseling, community-based education programs, gatekeeper training, and group activities. The researchers concluded the quality of the studies was low. Several RCTs were of higher quality and suggest multi-faceted primary-care-based depression screening and management with pharmacotherapy and psychotherapy may be effective at reducing SI, but more research is needed to verify this. [125]

Differential Diagnosis

• Chronic fatigue syndrome

• Dissociative disorders

• Hypochondriasis

• Hypoglycemia

• Hypopituitarism

• Schizoaffective Disorder

• Schizophrenia

• Somatic symptom disorder

Pertinent Studies and Ongoing Trials

The 2019 FDA approval of esketamine for adults with treatment-resistant MDD means some people with coexisting SI may qualify for this treatment. Meanwhile, investigators are turning their focus towards determining whether ketamine will provide a measurable and clinically important reduction in SI. Alternative doses and routes of administration are also being examined. Most completed studies included a ketamine IV dose of 0.5 mg/kg administered over 40 minutes. Samples were drawn from individuals with treatment-resistant MDD who had varying SI severity. Fewer studies have examined the efficacy of ketamine administered by oral, IM, or intranasal inhalation. [126] Currently, researchers are recruiting participants with suicidal thoughts for RCTs that use intranasal and oral administration of ketamine. 

Consultations

Psychiatric consultation is often indicated to complete the suicide risk assessment and establish a treatment regime. Psychological testing may be warranted. Social work, occupational therapy, and other disciplines may be included in planning care to address modifiable risk factors.

Deterrence and Patient Education

Counseling patients and families to restrict access to lethal means (locking medications, removing or locking firearms, etc.) during episodes of SI is encouraged in the literature. However, chart reviews show this is frequently not done. A training intervention for healthcare professionals in ED improved the frequency they initiated counseling against lethal means (CALM). [127] When ED nurses did not have a designated person available to provide counseling, patients may not receive counseling because many nurses were unsure how to do so without appearing confrontational. [110] Further research and guidelines to identify the best way to provide patient education and counseling are needed.

School-based educational programs may reduce SI, but a recent systematic review determined the quality of evidence is low. [128]  There is insufficient evidence to support gatekeeper training programs at present. [128] A project in rural India trained lay-persons to work within their communities to improve the detection and treatment of mental health issues. Additionally, community education successfully reduced the stigma of mental illness in these rural communities. [129] Similar approaches may be beneficial to target high-risk groups. For example, a systematic review showed people who are bereaved by the death of a loved one by suicide have very high rates of SI (15-49%).[130] Education to reduce the stigma and fear of appearing weak or unfit for duty may improve the reporting of SI by groups known to have high rates of suicidality but who generally avoid disclosing it. (e.g. first responders, [131] military personnel [132] [133]and medical students/providers). [134] [135] 

Helping the ideator to formulate a safety plan that includes appropriate supports, resources, distractions, and safeguards to prevent assess to lethal means are best practice interventions. 

Enhancing Healthcare Team Outcomes

For over 50 years, suicidality research was dominated by efforts to identify single and groups of risk factors that would reliably predict an individual's risk for long-term (lifetime) or near-term (imminent) suicide. As it turned out, thousands of risk factors were very weakly associated with having suicidal ideation (SI), but no single or combination of risk factors are clinically useful to predict a particular individual's suicide risk. There are data that suggest various factors may weakly increase the lifetime risk of suicide within populations, but there is no way to reliably predict an individual's lifetime or imminent risk for suicide. 

One of the reasons for the inconclusive findings and low-quality evidence from decades of studies is because there is no typical SI presentation. The heterogeneity of ideators has implications for both researchers and clinical practitioners. Each individual presents differently, with their own unique experiences, risk factors, and protective factors. Therefore, a person-centered approach is crucial to achieving positive outcomes. 

More recently, researchers turned their focus to the social and healthcare-related factors that may contribute to rising rates of SI and suicide. Literature shows that social stigma, media coverage, and traditional healthcare practices all contribute to poor outcomes. Studies of people who sought treatment for SI and suicidal behaviors suggest how healthcare professionals respond and treat patients when they disclose their suicidal ideas is vitally important. When the ideator feels negatively judged or if there is a delay in follow-up care, their outcomes worsen. There is also abundant evidence that individuals who die by suicide were seen by healthcare professionals in the days, weeks, and months preceding their death. This means that opportunities exist in most situations for healthcare professionals to impact outcomes by assisting the ideator during a crisis.

Many healthcare professionals, regardless of their discipline, are doubtful they have sufficient knowledge and skills to intervene. More training is recommended by professional associations, researchers, and the Joint Commission to increase evidence-based knowledge, the proper use of objective instruments, and to increase healthcare professionals' self-confidence in assessing and treating ideators. Counseling to limit access to lethal means is recommended, but at present, there is a lack of research evidence to guide training programs. Further research to identify the best strategies and which professionals are best able to provide this counseling is needed. [127] The literature also suggests the need to address negative attitudes and myths such as patients who talk about SI are attention-seeking or that suicide cannot be prevented. [110]

Because suicidality is believed to arise in response to a complex interplay of an individual's unique bio-psycho-social-cultural-spiritual risk factors, it follows that outcomes can be improved through a holistic approach that incorporates the expertise of each discipline. Traditionally, this has not occurred. For example, there are dozens of clinical guidelines that have been published by each healthcare profession, but the recommendations differ. This is largely due to the low quality of evidence that supports recommendations and the lack of gold standards for prevention, assessment, and treatment. However, it has become clearer in recent years that quality and safety are improved when an interprofessional approach is used. Therefore, using an interprofessional approach that includes comprehensive strategies to address social and healthcare practices as well as individual patient factors will have the most positive impact on patient outcomes.

Universal screening for SI by ED nurses has been recommended by the Emergency Nurses Association to detect individuals who have occult SI but whose presenting problem is medical and not behavioral. [78]  There is evidence that youth and elderly people often present with occult ideation. This screening recommendation is based on expert opinion. 

Zero Suicide http://zerosuicide.sprc.org/ is an example of a programmatic approach that uses a multi-level systems approach to prevent patient suicide in health and behavioral health systems. They emphasize the need for an organizational commitment to the aspirational goal of eliminating suicides by increasing the competence of healthcare professionals and using continuous safety and quality improvement strategies. In order for this to be successful, it is important that all healthcare professionals feel supported and that changes in the policies and processes encourage better teamwork, communication, and collaboration. The zero suicide website summarizes the approach as follows: [136]

1. Lead system-wide culture change committed to reducing suicides
2. Train a competent, confident, and caring workforce
3. Identify individuals with suicide risk via comprehensive screening and assessment
4. Engage all individuals at-risk of suicide using a suicide care management plan
5. Treat suicidal thoughts and behaviors using evidence-based treatments
6. Transition individuals through care with warm hand-offs and supportive contacts
7. Improve policies and procedures through continuous quality improvement

The Therapeutic Risk Management of Suicidal Patient (TRMSP) model, a legally and clinically sound framework to aid psychiatrists in suicide risk assessment and treatment management, was adapted for use with interprofessional teams. [137] This model incorporates three-tiers to assess suicide risk, document the risk, and then develop a patient-centered treatment plan for individuals who are known to endorse SI. The authors suggest this model could provide interprofessional teams with a much-needed interprofessional, evidence-based model to manage suicidality. 

Advocates of the TRMSP interprofessional model suggest that incorporating it into practice would increase the evidence-based knowledge and skills of nurses, social workers, and physicians. It would increase the number of providers who are knowledgeable of suicide risk assessment, thereby improving access to care and reducing existing gaps. Most ideators are not under the care of a psychiatrist, so limiting requisite knowledge and skills to psychiatrists limits the quality and timeliness of care that ideators receive. [137] 

Reducing communication gaps during transitions in care settings and between disciplines is recommended as a strategy to improve outcomes. Adapting the TRMSP model for interprofessional use improves the quality of communication by providing opportunities for better information exchange and more consistent documentation (including both quantitative and qualitative information in the electronic medical record). Ultimately, the continuity of care can be enhanced if all disciplines understand and share a common model that permits better collaboration and teamwork. The creation and documentation of acute and chronic suicide risk stratification enable all disciplines to better understand the patient's status over the trajectory of care. 

Interprofessional TRMSP Model of Care [137] 

1. Risk Assessment Based on Objective Measures

The use of standardized, valid, and reliable SI and suicide risk assessment measures can be used by a wide array of healthcare providers as long as they are taught how to use the tool and what the strengths and limitations are for each instrument. Regardless of who completes the risk assessment, a detailed interview that gathers and documents the patient's explanation of their situation is essential. Thus, interviewing skills and gathering relevant information from the patient, family/friends, and medical histories is imperative to establishing the context because scores alone, from any standardized measure, are inadequate. These authors suggested that the C-SSRS is a richer tool than some of the others, but even with the C-SSRS, clinical judgments need to be based on more than the standardized risk assessment results. The authors stated, "From the medicolegal perspective, these instruments can and should populate the medical record, but must be balanced with individualized narratives of the patient’s situation." [137] 

2. Acute and Chronic Suicide Risk Stratification

The assessment and documentation of the level of both acute suicidal risk as well as long-term risk is important. This provides all providers with a more thorough, nuanced, and accurate depiction of the patient's suicide risk. By documenting a stratified suicide risk assessment, it becomes easier for all healthcare professionals to understand the patient's fluctuations in risk. For example, an individual with a low acute suicide risk designation but who has a high chronic suicide risk may require additional safety precautions if distressing life circumstances arise. The stratified assessment documentation enables other healthcare professionals to better predict how acute risk may suddenly increase given certain triggers. [137] 

3. Development of an Individualized Safety Plan with Collaboration of the Patient

Safety plans are considered best practices and have replaced suicide contracts (which are no longer recommended). [99] [100] The suicide plan empowers the individual to recognize their warning signs and triggers so they can manage a crisis in a constructive manner. The safety plan helps the individual to reduce the risk of acting on suicidal thoughts by taking measures to reduce assess to lethal means and promoting their coping and resource utilization during stressful situations. The safety plan belongs to the patient but should be made visible to all providers within a network of care so it can be reinforced and implemented consistently. Similarly, updates to the safety plan can be done in a collaborative manner between the patient and any of the interprofessional team members. This helps to ensure that the approach and messages are consistent and patient-centered to meet the needs of each ideator. [137] 

There is low-quality research evidence from cohort studies and expert consensus to support these guideline recommendations.