Lawyers make short work of “clinical experience” testimony by defendants and expert witnesses in suicide malpractice cases. Clinical experience, unaided by evidence-based research, can be idiosyncratic, insufficient, uninformed, or just plain wrong when applied to complex, fact-specific suicide cases. Both in the clinical setting and in providing expert witness testimony, clinical experience can be colored by tradition, myths, and conservatism.
Every practitioner’s clinical experience is necessarily limited, yet it may be proffered as the standard of care or even as “best practices.” The question arises: is clinical experience, unaided by evidence-based research, the practice of the average or reasonable, prudent clinician or is evidence-based suicide risk assessment the standard of care? The answer is neither. Most clinicians combine clinical experience with evidence-based research. Substandard suicide risk assessments often rely on clinical experience alone. Expert opinions on the extremes of best practices or unaided clinical experience will be challenged by opposing counsel as not within the legally defined care and treatment ordinarily employed by the average or reasonable, prudent practitioner under same or similar circumstances.1 No single source or authority, however, defines the standard of care in suicide risk assessment.2
Psychiatrists are expected to possess core competencies in suicide risk assessment and in evidence-based psychiatry.3 Acquiring these skills is a current requirement of residency training. Suicide risk assessment identifies acute, high risk suicide factors and available protective factors that inform the treatment and management of suicidal patients.4 Clinical experience alone is usually insufficient to support a competent suicide risk assessment.
Informing Clinical Experience
Evidence-based psychiatry can inform clinical experience in the assessment of suicide risk.5 The research of Fawcett et al.,6 a 10-year prospective study of 954 patients with major affective disorders, identified short-term suicide risk factors that were statistically significant within one year of assessment. The short-term risk factors included panic attacks, psychic anxiety, loss of pleasure and interest, moderate alcohol abuse, depressive turmoil, diminished concentration, and global insomnia. Most short-term risk factors are responsive to anti-anxiety and other medications. Thus, the patient’s suicide risk can be rapidly reduced by treating anxiety and other acute symptoms aggressively, while allowing antidepressants time to work.
Harris and Barraclough,7 in a systematic review and meta-analysis of 249 reports from the medical literature on the mortality of mental disorders, determined the Standard Mortality Ratio (SMR) for psychiatric disorders. They compared the relative risk of suicide for a given psychiatric disorder with the expected suicide rate in the general population (SMR of 1). The highest SMR (23.14) was associated with eating disorders. All psychiatric diagnoses, except mental retardation, had an increased SMR. The SMR underscores the importance of making a correct psychiatric diagnosis in suicide risk assessment.
Anti-suicidal drugs that are diagnosis specific have been identified. There are, however, no drugs that reduce suicide risk for all psychiatric disorders. Baldessarini et al.,8 in a review of 34 reported studies, found that lithium was effective in reducing suicide risk in unipolar depression and in bipolar I and II disorders. Meltzer et al.,9 in a cohort study (n = 980), demonstrated that clozapine reduced suicide attempts and completion rates in patients with schizophrenia and schizoaffective disorder. The evidence-based studies enhance both the treatment and safety management of patients at suicide risk with these mental disorders.
Evaluating protective factors is essential to a balanced assessment of suicide risk. Linehan et al.10 developed The Reason for Living Inventory. The statistically valid inventory measures beliefs that act as protective factors. Coping and survival skills, responsibility to family, child-related concerns, and moral/religious beliefs are identified as protective factors. Unaided clinical experience, however, tends to focus mainly on risk factors.
Suicide risk assessment must take into account unique and distinctive patient risk and protective factors for which no evidence base exists.5 Protective factors may include a cherished animal, rewarding employment, important relationships, a compelling interest or avocation, and other factors that only a thorough knowledge of the patient will reveal. No evidence demonstrates, however, that protective factors can trump acute, high-risk suicide factors in severely ill, suicidal patients. Moreover, there are no evidence-based “imminent” suicide risk factor(s) that can predict when, or even if, patients at acute high risk for suicide will attempt or complete suicide.11
Patients often display prodromal signs and symptoms of suicide risk escalation similar to past suicide crises or actual attempts. Identifying this clinical pattern provides the clinician with a means of appraising a patient’s current level of suicide risk when compared with the patient’s past symptom pattern. While the symptom progression during suicidal crises may be consistent, the method of attempting or completing suicide (method substitution) can be different. In a systematic case review (n = 1,397) of prior suicide attempts, Isometsa et al.12 found that 82 percent used at least two different methods in their suicide attempts. Although patients may rehearse completing suicide when making prior attempts, awareness of the high risk of method substitution should inform the safety management of the suicidal patient.
Trust but Verify
It is important to trust but verify clinical experience. Are the suicide risk factors that a clinician thinks important to assess verifiable by current research? In the clinician’s experience, does psychosis increase a patient’s suicide risk? The Fawcett et al.6 study cited above found no significant difference in suicide between depressed and delusionally depressed patients except for delusions of thought insertion, grandeur, and mind reading. Warman et al.,13 in a cohort study (n = 158), followed a heterogeneous group of psychotic and nonpsychotic suicide attempters for two years. Patients with psychotic disorders attempted suicide at nearly twice the rate of nonpsychotic patients. Coryell et al.,14 however, in a 10-year cohort study (n = 787), found no increase in suicide among patients with psychotic depression versus patients with nonpsychotic depression. Their demonstration that psychotic depression did not increase suicide risk is counterintuitive. Yet the findings arise from a well-designed prospective 10-year study of 787 patients. The researchers observe that “many clinicians intuitively correlate the severity of depressive symptoms with suicide risk and assume, therefore, that patients with psychotic depression are more likely to commit suicide” (Ref. 14, p 488).
Based on these few studies, psychotic patients attempted suicide at a higher rate than do nonpsychotic patients. Specific clinical symptoms such as delusions of thought insertion, grandeur, and mind reading increased suicide risk. Psychotic depression, by itself, was not associated with increased suicide risk. Psychotic depression, however, must be assessed along with other individual patient risk and protective factors for suicide. The clinician may need to review a number of studies critically for answers to questions about specific suicide risk and protective factors. Obviously, it is not necessary to conduct a literature search for every patient at risk for suicide.
There are too many, not too few, suicide studies and articles of varying quality. For example, in the development of the American Psychiatric Association’s “Practice Guideline Assessment and Treatment of Patients with Suicidal Behaviors,”15 17,000 articles were screened. The Guidelines contain 680 references. The Evidence-Based Medicine Resource Center (http://www.ebmny.org) provides the clinician with a methodology for appraising the validity and importance of a study before applying the results.16 Evidence-based research methodologies include: systematic reviews and meta-analyses, randomized controlled double-blind studies (no suicide studies for ethical reasons), cohort studies (prospective), and case-control (retrospective) studies. Gray observes that expert clinical opinion falls at the bottom of the evidence hierarchy for studies of therapy or harm “because such opinion does not necessarily reflect the best evidence found in the current psychiatric literature” (Ref. 16, p 25) (Table 1).
Hierarchy of Evidence for Studies of Therapy or Harm
Conclusion
Professional organizations recognize the importance of developing evidence-based and clinical consensus recommendations that can be applied to the management of various diseases, including behavioral states such as suicide.17 The application of evidence-based risk factors can enhance the clinician’s ability to perform competent suicide risk assessments. Ultimately, suicide risk assessment is an informed judgment call that incorporates information from a number of sources.
Clinical experience is greater than the sum of correct clinical judgments and instructive errors. It reflects the clinician’s knowledge, skill, values, dedication, and openness to change. Clinical experience and judgment are an essential part of suicide risk assessment but should be informed by evidence-based research. Unaided clinical experience is usually insufficient in conducting suicide risk assessment and in providing expert testimony. Moreover, clinical experience is often wrong when it is misguided by unyielding misconceptions, lore, and tradition.
The expert who testifies in a suicide case will have his or her opinion challenged by vigorous cross-examination. Testifying that “it is so because my clinical experience says it is so” will fail to provide the finder-of-fact with useful testimony. Such testimony is unlikely to survive a Daubert18 credibility challenge.
- American Academy of Psychiatry and the Law
