Introduction
The suicide rate in the United States was 14.2 per 100 000 people in 2019.Reference Xu, Murphy and Kochanek 1 In the past 20 years, there has been a 33% increase in the rate of suicide deaths in the United States, with the rate increasing more rapidly in rural counties than in large metropolitan areas.Reference Hedegaard, Curtin and Warner 2 According to 2019 data from the National Survey on Drug Use and Health, approximately 12 million adults in the United States had seriously considered killing themselves in the past year. A cross-national study conducted by the WHO determined that 60% of the transitions from thoughts to plans, and from plans to attempts, occurwithin a year after the first onset of ideation.Reference Schreiber and Culpepper 3 U.S. males are 3.7 times more likely than females to die by suicide, but females are more likely to make suicide attempts than males. According to the Centers for Disease Control and Prevention, by May 2020, the number of emergency visits for suspected suicide attempts among adolescents (ages 12-17) had increased by 31% from the previous year. There was a nearly 50% increase in emergency visits from adolescent girls aged 12 to 17 with suspected suicide attempts between February and March 2021 compared to the same period in 2019.Reference Yard 4
In the current realm of clinical practice, the modifiable and nonmodifiable risk factors are incorporated into risk assessments, as well as suicide screening tools mandated by the Joint Commission. Since suicide is a rare event, and unable to be accurately predicted, many successful suicides are unfortunately unpreventable. The predictive ability of the existing suicide risk tools is only slightly better than chance, based on a recent meta-analysis of the last 50 years of research.Reference Franklin, Ribeiro and Fox 5 A systematic review also has raised numerous concerns about risk assessment tools. First, the known risk factors are so common that their utility has been questioned. Second, overreliance on risk assessment tools may lead to lowered engagement between the patient and the provider, which may at times paradoxically increase suicide rates.Reference Chan, Bhatti and Meader 6 Another meta-analysis found that 95% of high-risk patients will not die by suicide and that 50% of patient suicides came from the lower risk categories. A highly provocative study reported there was no improvement in the accuracy of suicide risk assessment over the past 40 years.Reference Large, Kaneson and Myles 7 The U.K. national guidance and advice body, The National Institute for Health and Care Excellence, recommended that assessment tools and scales designed to give a crude indication of the level of risk (eg, high or low) of suicide should not be used.Reference Murray 8 On the contrary, the American Psychiatric Association further clarified that the process of identifying risk and protective factors in an assessment does not have predictive value, but that it may be used to mitigate the modifiable risks and plan interventions as part of a more informed decision-making. 9
Furthermore, although the evaluation of suicide risk is a core competency for psychiatrists, many fail to perform or document a systematic suicide risk assessment (SSRA) during patient encounters. Late Distinguished ex-President of the American Academy of Psychiatry and the Law Professor Robert Simon (1934-2016) summarized his seminal findings in a 2002 editorial “Suicide Risk Assessment: What Is the Standard of Care?”Reference Simon 10 He suggested many reasons for the omission of an SSRA based on the feedback from the colleagues, which includes the lack of training, willfully not performing an SSRA, performing but not documenting, and time constraints. His findings also noted the potential hesitancy of psychiatrists to document their complete rationale due to fear of legal repercussions if the assessment turned out to be wrong and the patient dies by suicide. Further reasons for the lack of an SSRA included underestimating the risk and delegating the task to complete SSRA to other members of the team, as well as certain clinical situations such as psychotherapy that do not require a formal suicide risk assessment.Reference Simon 10
Given these emerging trends and complexities, we have reviewed recent advances in the areas of scientific and legal disciplines to assimilative facts for clinicians in making informed decisions.
Methods
A comprehensive search of several databases including PubMed, PsychINFO, Cochrane Library, Google Scholar, Scopus, Medline, and Web of Science from start to the date of the search was conducted. We also searched the databases on PubMed Central (PMC) as the nature of the subject would require book chapters and expert opinions. The search was designed using controlled vocabulary and keywords “suicide*,” “suicidal risk assessment,” “court*,” “law,” “medicolegal,” “legal,” and “documentation.” It was performed in all languages. We also performed a manual search. The inclusion criteria were any published material focusing on documentation of suicidal risk assessment in individuals of all age groups. We specifically focused on published materials that discussed documentation of suicide risk assessment for medicolegal issues in court. We identified 3003 articles, after the removal of duplicates. After reviewing the abstracts, 90 studies met our inclusion criteria. We found various models of documentation, and we chose the latest published material, as mutually agreed upon by the authors. Forty-five articles were used for this review, and 5 articles were later added manually. Reverse citations were also reviewed to update the material. See Figure 1.
The search methodology for the review.
Results and discussion
The scope of the problem and keeping up with evolving evidence
Compared to other specialties, psychiatrists are at an increased risk of disciplinary action,Reference Reich and Schatzberg 11 with patient death by suicide as the number one cause of malpractice lawsuits against psychiatrists.Reference Melonas 12 At any point in their clinical years, 68% of psychiatrists and 28% of psychologists have lost at least one patient due to suicide, with a third of all patient deaths by suicide occurring within 3 months of discharge from an inpatient psychiatric unit.Reference Chung, Ryan and Hadzi-Pavlovic 13 Although there is a huge database of research on suicide, limitations due to multifaceted causation, high false-positive rates among predictive risk factors, and the general unpredictability of suicide prevent optimal predictive value. Research has also failed to address interactions among multiple factors that may lead to a suicidal event. Furthermore, conventional approaches to suicidal risk assessment rely on patients reporting suicidal ideations, but patients may often deny having these thoughts or sharing their plans for suicide.Reference Bolton, Gunnell and Turecki 14 One study suggests nearly 80% of individuals who died by suicide denied suicidal thoughts in their last verbal communication.Reference Maldonado 15
Furthermore, neurobiological factors may also play a large part in the probability of a death by suicide or attempted suicide. Studies have found that interplay among the hypothalamic–pituitary–adrenal axis, serotonergic system, neuroplasticity, neuro-immunological markers, and neurocognitive factors may contribute to the elevated risk of suicidal behaviors. Structurally, volume changes or deficits within the left superior temporal gyrus, rectal gyrus, caudate nucleus, temporal, parietal, and frontal cortices; and three functional areas, for example, right cingulate gyrus, the anterior cingulate, and posterior cingulate, have been found in the brains of those who have attempted suicide, or contribute to suicidal vulnerability.Reference Orsolini, Latini and Pompili 16
Apart from these factors, the American Psychiatry Association suicide practice guidelines have determined that predicting suicide is impossible due to the relative rarity of the event, leading to further statistical challenges.Reference Jacobs and Brewer 17 Because of the unpredictability of suicidal risk factors, many psychiatrists adopt defensive behavioral methods to protect themselves. A particularly well-documented mechanism is that of anchoring or relying on the psychiatrist’s first impression of the patient to dictate how he views the patient in later contexts. When a suicidal patient is not forthright about their suicidal thoughts or plans, a psychiatrist’s judgment based on that impression may become skewed and potentially miss a patient’s true suicidal ideations.Reference Wasserman 18
Thinking through the medicolegal lens
Although logic and perception are the most common tools of a prospective suicidal risk assessment, the court considers these the two most common errors. In court, the acquisition for a psychiatrist accused of negligence focuses only on the thoroughness of the evaluation and whether appropriate measures were taken. Specifically, the court approaches suicide assessments based on a psychiatrist’s ability to elicit a patient’s previous suicidal behavior, current suicidal thoughts and plans, hopelessness, stressors, the presence of mental disorder symptoms, themes of impulsivity and self-control, ready access to highly lethal methods like firearms, and protective factors.Reference Jacobs, Baldessarini and Conwell 19 To be found negligent, a particular set of parameters must be met: that the psychiatrist owed the patient a duty to perform to a standard of conduct, that the psychiatrist was remiss in the breach of that duty by some act of commission or omission, that because of the dereliction, the patient suffered actual damage, and, finally, that the psychiatrist’s conduct was the direct or proximate cause of such damage.Reference Se 20 , Reference Splane 21 The medicolegal risks regarding suicide depend upon the foreseeability of suicide potential and if proper precautions were taken to prevent suicide once this risk was identified.Reference Gutheil and Appelbaum 22
Courts are legally obligated to analyze a psychiatrist’s assessment and management of a patient who attempted or died via suicide. To do so, the legal team first determines the effectiveness of the suicide risk assessment performed and decides whether the patient’s suicide was foreseeable. The court views foreseeability as a critical factor in determining if the suicide was preventable. This is despite differing interpretations of foreseeability by law, which is probabilistic, and medicine, which is scientific.Reference Simon 10 Foreseeability is more clearly defined in contract and tort law which applies the standard of “reasonability.”Reference Perloff 23 This is the process of determining how likely an individual could have anticipated the potential or actual results of a patient’s actions. In clinical contexts, foreseeability, which is a probabilistic construct, is not the same as predictability. Professor Simon writes “However, the law does not require the defendant to ‘foresee events which are merely possible but only those that are reasonably foreseeable’ (Hairston v. Alexander Tank & Equipment Co., 311 S.E.2d 559 (N.C. Ct. App. 2000)). It is only the risk of suicide that can be assessed, and therefore only the risk of suicide that is reasonably foreseeable.” He explains further that despite the most thorough SSRA, suicide is impossible to predict. Proper documentation of an SSRA performed on the patient, including treatment and management, would meet foreseeability criteria.Reference Simon 10
It is imperative to note that a critical aspect of a psychiatric assessment depends on prospective thinking, which includes an indefinite number of inherent uncertainties that are impossible to be addressed. The court reasons only using hindsight, where these inherent uncertainties can be appropriately respected.
The legal system and lessons to be learned from the landmark cases
The relationship between psychiatrists and law dates to the nineteenth century. It first found ground in court requests for mental health professionals to prove that an accused individual was innocent of a crime due to mental illness, a tactic now called the “insanity defense” or famously known as the McNaughton’s Rule.Reference Coleman and Davidson 24 This was followed by lawyers’ increasing involvement in human rights violations, and the de-institutionalization of the mentally ill. The legal system is structured to challenge medical decision-making, and lawyers are trained to find errors in clinical practices. Frequently, lawyers hire psychiatrists for adversarial opinions, as they are aware that conflicting research evidence may be available to win a case. The recent high-profile landmark case of Bell v. Chance, 188 A.3d 930, highlights these issues, and Maryland State Medical Society filed an amicus brief about the standards of care applicable to a physician who discharged a patient from a psychiatric hospital. 25 This difference in thinking can best be understood by several aspects. It is impossible to predict suicide, although sometimes the court believes in foreseeability. It is also important to reconstruct the situation in which the clinical judgment was made, and failure to do so can lead to adverse outcomes for a psychiatrist. The court also makes efforts to address the intrinsic and infinite uncertainties in each clinical scenario. This becomes further complicated if the court assumes unifactorial causation of the suicide and considers these factors to be exclusively in the physician’s control.
Keeping up with growing knowledge of risk and protective factors
There are many theoretical understandings of suicide. The interpersonal theory of suicide suggests that suicidal ideation is developed from a perceived sense of burden and lack of belonging, and, when coupled with the capability to attempt suicide, leads to a decrease in fear of death and an increase in tolerance for physical pain.Reference Fowler 26 Risk factors are divided between static and dynamic and have shown to be a strong association. Static risk factors are those that are unalterable, such as gender, preexisting mental illness, intelligence, and history of military involvement or weapons training (Table 1). Dynamic factors can be modified, and include psychiatric symptoms, substance abuse, or access to dangerous firearms.Reference Conner, Azrael and Miller 27 The history of self-injurious thoughts and behaviors is well established as the strongest predictor of future attempts or death by suicide,Reference Giner, Jaussent and Olié 28 and a meta-analysis suggests a stronger correlation than previously thought, including a study that found 81% of suicidal patients who had attempted suicide later went on to kill themselves within the year.Reference Ribeiro, Franklin and Fox 29 Among the psychiatric diagnoses, depression, bipolar disorder,Reference Aaltonen, Rosenström and Jylhä 30 schizophrenia,Reference Olfson, Stroup and Huang 31 and substance use disordersReference Edwards, Ohlsson and Sundquist 32 carry the highest risk of death by suicide. Interestingly, anorexia nervosa also carries a higher risk due to the preponderance for self-injury, impulsivity, and increased likelihood of comorbid depression and substance abuse.Reference Latzer and Stein 33 , Reference Bodell, Cheng and Wildes 34 Nearly 40% of patients receiving treatment for substance abuse report a history of attempted suicide.Reference Yuodelis-Flores and Ries 35 Patients suffering from personality disorders have a nearly twofold increase in suicide attempts.Reference Chesney, Goodwin and Fazel 36 Multiple co-occurring conditions are a stronger predictor of suicidal behavior than any individual diagnosis.Reference Pagura, Stein and Bolton 37
Summary of Salient Points of Current Use of Suicide Risk Assessment Tools and Medicolegal Risks
The protective factors may offset the risk factors and epigenetic vulnerabilities. Family and community support, skills in problem-solving and conflict resolution, cultural and religious beliefs, and access to clinical care and support have been determined as effective mediators to mitigate the suicidal risks. 38 The APA and the CDC recommend incorporating protective factors which have equal weightage to ascertain the level of risk during comprehensive suicide assessment.Reference Orsolini, Latini and Pompili 16 , Reference Bodell, Cheng and Wildes 34
Establishing the standards
The current practice standards include the knowledge or identification of the risk and protective factors, followed by a comprehensive suicide inquiry and a detailed clinical assessment of the level of risk. These approaches also have many caveats, as they are primarily based on self-reports and therefore make it impossible to predict suicide. Interestingly, a review of 12 studies found that a strong therapeutic alliance is linked to fewer suicidal thoughts. Although the categorical foretelling of low, moderate, or high risk of suicide used in current suicide risk assessments (Figure 2) is taught in training, in truth it is not evidence-based. Pisani et alReference Pisani, Murrie and Silverman 39 proposed a model of risk formulation which synthesized data into four distinct judgments to directly inform intervention plans: (a) risk status (based on the patient’s risk relative to a specified subpopulation), (b) risk state (the patient’s risk compared to baseline or other specified time points), (c) available resources from which the patient can draw in crisis, and (d) foreseeable changes that may exacerbate risk.Reference Pisani, Murrie and Silverman 39
The panoramic overview of suicide risk assessment.
Therefore, given the complex and multifactorial nature of these presentations, it is critical to stay close to the established standard of practice. Since 2019, the Joint Commission mandates the use of validated screening tools for all patients aged 12 or above who are being evaluated or treated for mental health conditions. The appropriate documentation of risk assessment is therefore not just a measure to improve patient safety, but also to mitigate medicolegal risk.
Documentation of the suicide risk assessment
In a court trial, the document is the only evidence of a performed assessment; thus, what is not documented is not done. Therefore, it is imperative to document in proper, clear syntax, and an error-free manner. This not only signifies a better assessment, practice, and care, but also leads to a less vulnerable psychiatrist in a court of law. Another study reviewed clinical notes and concluded that residents with better quality notes ultimately performed better assessments.Reference Sherman, Justesen and Okocha 40 Many recommend extemporaneous assessment at the time of discharge, as a discharge summary written even 1 hour after a patient commits suicide, is considered a worthless document. It is also important to carefully document the thorough process leading to the clinical decision, which helps not only in court cases, but also in the continuity of care.Reference Ribeiro, Gutierrez and Joiner 41 Patient reliability and veracity must also be appropriately addressed. In doubtful cases, never hesitate to ask for second opinions and document these as well; documenting obtained collaterals is as crucial as gathering them. A frequently used factor while assessing suicide is suicidal ideation or intent expressed by the patient during the last human contact before death. According to one study, 38% of patients make some form of healthcare visit within a week of attempting suicide. Sixty-four percent make contact within a month, and nearly 95% of patients contact a healthcare provider within a year of attempting suicide.Reference Ahmedani, Simon and Stewart 42 In the year 2020, the proportion of mental health-related emergency department visits among adolescents aged 12 to 17 years increased 31% compared with that during 2019, and another study found that girls between the ages of 12 to 17 presented to the Emergency Department (ED) for suicide attempts at a 50% higher rate than that compared to 2019.Reference Yard 4 Simply documenting that the patient denies suicidal ideation is not good practice, and the Joint Commission, therefore, recommends that both risk and protective assessment findings should be carefully documented, as well as the immediate plan for intervention and rationale for not choosing alternative interventions in a suicide assessment. 43
The future direction
The growing body of empirical research has been moving toward more objective measurement risks and reducing overreliance on subjective self-reports. There is also an acknowledgment of the gaps in the current process and epidemiological hard data testimonials from many credible voices calling current practices a systematic failure.Reference Cavelti and Kaess 44 The newer research is focusing on using ecological momentary assessment (EMA) and digital phenotyping approaches for identifying real-time markers of change in suicidal ideations and risk factors.Reference Ballard, Gilbert, Wusinich and Zarate 45 The use of cognitive and neurobiological markers is also a paradigm shift in the novel ways of thinking about suicide assessments and are the steps in the right direction.Reference Ballard, Gilbert, Wusinich and Zarate 45 Furthermore, artificial intelligence (AI) has already established itself as a key technology, driving the field of precision medicine for the last decade by analyzing large datasets to develop predictive models.Reference Kessler, Warner and Ivany 46 DelPozo-Banos et alReference DelPozo-Banos, John and Petkov 47 used artificial neural networks to study suicide risk factors in 2604 patients who died by suicide in the United Kingdom. The study linked several factors with high-risk scores to death by suicide, which has potential for the future. Miché et alReference Miché, Studerus and Meyer 48 also utilized similar advancements in AI to conduct a 10-year prospective study. The predictive performance of AI algorithms was measured by the area under the curve, and they all performed well in distinguishing future suicide attempts in a community sample (n = 2797).
There are three main areas where the data are accessed and analyzed: the electronic health records (EHRs), electronic devices measuring EMA, and lastly, social networking sites (SNSs). EHR data have been used to develop algorithms in predicting suicidalityReference Carson, Mullin and Sanchez 49 and likewise, data mining from SNSs could identify individuals at risk of suicide and subsequently provide them with specialized crisis management.Reference Liu, Liu and Sun 50
Limitations
The goal was to review literature specific to the medicolegal aspect of suicidal risk assessment. The narrative literature reviews have an implicit bias of nonprotocol-based methodology, analysis, and conclusions. The inclusion criteria were intentionally broad, and we understand the possibility of selection bias and subjective weighing of the studies included in the review. We also included expert opinions of distinguished authors published in peer-reviewed journals, to best represent multiple viewpoints on this complex topic. Many studies included were of pristine value, but dated. However, we found this field has evolved due to the exceptional work spanning over 30 years, although the most recent published material was also included.
Conclusion
Although studies have accurately identified both risk and protective factors, comorbid conditions, and the importance of a strong therapeutic relationship, much remains to be determined about how to best document suicidal ideation in any given patient population. The documentation of suicidality is imperative not only for accurately assessing the patient’s risk factors, but also within the court, where it is often the only evidence in a suicide case. It is imperative to accurately document these encounters, and yet there is no one risk assessment universally recognized. Thus, the best approach appears to be a complete and detailed assessment, with adequate explanations for assessments, individualized treatment modalities, and patient reliability. Such thorough documentation improves clinical outcomes and is the best defense for medicolegal risks.
Author contributions
Conceptualization: M.G. and M.R.; Data curation: all authors; Writing—original draft: all authors; Writing—review and editing: all authors.
Disclosure
The authors do not have anything to disclose.
