A 52-Year-Old Suicidal Man

DR PARKER: Mr D is a 52-year-old man who, in despair, almost jumped in front of a moving train. He attributes this episode to gambling debts piled up in the prior 6 months and the ensuing conflict with his family. He is a resident of Boston and has health insurance through his blue-collar job.

Mr D describes a history of depression dating to his youth. He grew up in a "tough" family, with a father who gambled and drank and brothers who used drugs and alcohol. He has had suicidal thoughts on many occasions but strongly contemplated suicide only once before when he was briefly held in solitary confinement while in jail for a minor crime.

Despite attending a gambling addiction program, he relapsed and lost several thousand dollars gambling. His problem was magnified because he lost another individual's money as well. A family member was terminally ill at this time, and Mr D got little food or sleep. He considered staging his own accidental death so his family could collect on his life insurance. He denied any access to firearms. His family encouraged him to voluntarily admit himself to the psychiatric ward.

Mr D described his mood as "hopeless," with diminished appetite, poor sleeping, and decreased energy and ability to concentrate. He described feelings of shame and guilt. When asked why he did not jump, he answered that his religious beliefs held him back. In addition, a relative had committed suicide some years ago, and he felt some obligation to "help others prevent that."

Mr D has no prior psychiatric hospitalizations. He denied current alcohol use and admitted to past use of marijuana and LSD as a teenager. He quit smoking cigarettes many years ago. Mr D erratically attends a gambling addiction program and is in treatment with a psychiatrist who monitors his medication use. His past medical history is unremarkable.

Mr D completed high school and some college, has been married for 3 decades, and has children in their early 20s who are "doing well."

He is currently taking fluoxetine hydrochloride, 40 mg daily, and trazodone hydrochloride, 50 mg daily.

On examination, Mr D exhibits a depressed affect with slowed speech. He tends to avoid eye contact, although he became more engaged throughout the interview. There was no psychotic content. He did not know when he would be ready to leave the psychiatric ward.

I'm from a family that doesn't show emotions. A relative hung himself when he was a teenager. I think I was depressed for a long time after that. Then I lost a job, and I had all this leisure time. I have a history of having a little violence in my background. I left home when I was still a kid.

I was what they call a hippie. I'm the opposite now. My drug of choice was LSD. I used to hitchhike all over the place. I lived on the streets. I got picked up, and it was a long weekend. And I had a beef in jail, and I ended up in the hole. I did a lot of thinking in there, and I think I almost killed myself right there. I don't know why I didn't.

I started gambling, and the old saying when you're gambling is, "If I ever get even, I'll quit." Well, a few years ago, I got even and I quit. Then I broke out, and I got even and I said, "I'll quit." You get out of control. It's a compulsive thing. Everywhere I go, my compulsion is gambling. I mean, every member of my family gambles. "You know the number last night? There's $20 million in the jackpot. You going down to the race track?" My mother was a gambler, my father was an alcoholic gambler.

I got into such a financial hole, I was thinking of going out and getting even by pulling an armed robbery or something. And I got scared. That wasn't the way out. It is, and it isn't. I just started over the past month thinking whether to do that, or you know, you're on the train or waiting for the train, and you're looking at it saying, "Should I just jump in front of it?" I came within inches of jumping. I had a [feeling] as if I was seesawing . . . "Should I, shouldn't I?" And then I said, "No, I can't." I started thinking about the family, and I didn't do it. Then I started thinking about ways of doing it, so my wife would get the insurance and stuff. The biggest worry for me right now is shaming members of my family, because we don't do this, you know.

The past 2 years have been very hectic. Even when I wasn't gambling I was getting into these uncontrollable rages. Good friends of mine would come up to me and say, "Gee, pal, you're losing control."

Around our way they have a saying, if you put time between a crime, that will be less time you do. And like, put time between the thoughts then maybe you won't do . . . what I'm looking at.

I hope you translate whatever I'm saying to benefit maybe someone else.

What is the extent of the problem of suicide in the United States for adults? How do we evaluate suicide risk? What are particular warning signs, and how can the primary care physician identify patients in the outpatient setting who may be considering suicide? Do different methods of attempting suicide (eg, shooting, hanging, overdose, or jumping) have different implications? What are the elements of an evaluation of a suicidal patient? What role do medications play? What would you recommend for Mr D?

DR JACOBS: Specific suicide rates vary according to age and sex, but overall, suicide is the eighth leading cause of death nationwide. Of people who complete suicide, 90% to 93% have at least 1 major psychiatric disorder, usually affective disorder, schizophrenia, or alcoholism. Frequently they have comorbid mental illnesses. Men complete suicide more often than women (4:1), but women make more suicide attempts.¹

Approximately 30,000 people commit suicide each year, which represents 1.4% of all deaths in the United States. The surgeon general recently identified suicide as a public health problem and urged the nation to focus on suicide as a national health issue. Statistics driving this national call to action focus on the increase in suicide rates among teenagers, suicide's rise from the ninth to the eighth leading cause of death nationwide within the last year, and the deep secrecy associated with suicide.²

Suicide profoundly affects the survivors, and, in the case of Mr D, this appears to be one of his reasons for participating in the interview. Mr D indicated that the cluster of suicides among teenagers in the mid 1990s,³ which included his relative, was so painful for him that he agreed to take part in this case conference. He hoped his participation in the difficult interview would help physicians better understand the issue of suicide and help them prevent others from committing suicide.

I recommend using a comprehensive suicide assessment protocol (Table 1).¹ Many risk factors have been identified as being associated with suicide (Table 2). However, none of the risk factors, alone or in combination, are specific enough to predict which individuals will complete suicide. Although we cannot predict who will complete suicide, we can assess an individual's level of risk for suicide, and this helps us plan treatment.

Risk factors for suicide fall into 2 categories: predisposing factors and potentiating factors. Other authors have made similar distinctions, dividing risk factors into distal (foundation or root cause) and proximal factors (precipitating or triggering factors).⁵ The predisposing (distal) risk factors include the major psychiatric syndromes of depression or affective illness, schizophrenia, alcoholism or substance abuse, and certain personality disorders. Pathological gambling, present in Mr D's case, is an addiction that increases risk in a manner similar to other addictive illnesses.^{6 - 7}

Specific risk factors are associated with each of the diagnoses mentioned above. The number 1 distal risk factor for suicide is affective disorder (mood disorder), with more than 60% of individuals who complete suicide having some form of affective disorder.⁸ Anxiety and panic symptoms that occur in the context of a major depression also increase suicide risk. This information is relevant for physicians of all disciplines because both depression and the symptoms of anxiety are risk factors that are modifiable through interventions such as medication, psychotherapy, and relaxation techniques.⁹

Based on Mr D's clinical presentation, he would be considered a suicide risk (Table 2). He is male, had a drop in economic status, had a psychiatric diagnosis with comorbidity, was in psychological turmoil, and had suicidal ideation. In addition, he has antisocial traits, which increase suicide risk. His personality traits are likely to make him chronically vulnerable to suicidal thoughts when he feels humiliated or deeply ashamed.

Potentiating risk factors include situational stressors that, when combined with a predisposition to suicide from mental illness, increase the individual's vulnerability to suicide. The potentiating or proximal factors include: physical illness, intoxication, a toxic family or social milieu, access to guns or other methods of suicide, and intense life stresses or crises.¹ In general, the risk factors for suicide are additive such that the more risk factors an individual has, the greater the risk of suicide.⁵

The obvious potentiating risk factors in Mr D's life are the financial difficulties and estrangement from his family that result from his pathological gambling.

Mr D's symptoms would have been detected by a primary care physician using a depression screening tool, even if the physician had not first independently asked about suicidal ideation. In fact, a screening tool would have been a useful avenue for opening such a dialogue.

In the last 10 years, depression screening has proven to be an effective and efficient way of identifying those with undiagnosed depressive illness and is a useful tool for the primary care physician attempting to ascertain the likelihood and severity of depression and the presence of suicidal thoughts.¹⁰ Besides identifying possible serious cases of depression that can lead to suicide, most depression scales include a specific question about suicidal ideation.

A nonprofit organization, Screening for Mental Health, through its annual National Depression Screening Day (NDSD)^{11 - 13} has provided an easy-to-use screening form (the HANDS, Harvard Department of Psychiatry/National Depression Screening Day Scale, Figure 1) designed to minimize physician time by quickly identifying patients who have a positive score or who endorse a suicide question. (The screening form is filled out by the patient in the waiting room and scored by a secretary or other staff. Completed forms are placed in the patient's file for the physician's review.)

The results of this screening effort are striking. Fully 22% of primary care patients had a positive score for depression, including 45% who have received treatment for alcohol abuse, 28% of those with stroke, 19% of those with cancer, 23% with diabetes, 27% with arthritis, and 23% of those with heart disease (unpublished data analyzed by the National Institute of Mental Health from completed screening forms collected by Screening for Mental Health from participating primary care clinicians in 1999). These findings underscore the need for depression screening in primary care settings and the ease with which it can be incorporated. Depression screening is simple, cost-effective, reliable, and potentially money saving because it can identify previously unidentified depression without the time-consuming and costly burden of attempting to diagnose vague medical complaints that often bring the patient into the office for numerous visits.

Certain clinical factors should alert the clinician to the possibility of depression,¹⁰ and screening may be particularly useful in patients who:

By identifying and addressing depression, primary care physicians may also identify suicidal thoughts and behavior.

Multiple screening tools exist, including HANDS.¹⁴ This scale is used in the National Depression Screening Day Primary Care Outreach. It was specifically created to be a brief, easy-to-score, self-report form applicable to a variety of settings that performs at least as well as longer, more complicated scales, and addresses the critical issues of sensitivity and specificity. The HANDS (Figure 1) has been validated for detecting the likelihood of clinical depression based on criteria for a major depressive episode from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Scores of 9 or greater give a sensitivity of at least 95% and indicate that the patient should be evaluated for the likelihood of a major depressive episode. Specificity in the general population is 94%.¹⁴

Primary care providers may worry that asking about suicide will introduce the idea into their patient's minds. My clinical experience is that the opposite is often true. Many patients are relieved when their physician opens a dialogue about suicide; it gives them the opportunity to discuss a frightening and troubling subject and offers the physician the opportunity to both assess the patient and educate him or her about the relationship between suicidal thoughts and depression in particular. By being direct, the physician offers hope, providing a lifeline to a suicidal patient, who, because of feelings of hopelessness, may otherwise become isolated.

People who are experiencing suicidal feelings are almost always ambivalent in those feelings; they have the simultaneous wish to die and wish to live.¹⁵ This ambivalence offers the clinician an opportunity to intervene and ally with the part of the patient that wishes to live.¹⁶ Mr D described his own ambivalence about suicide by saying "I was seesawing."

If gently and tactfully done, asking about suicidal intent can encourage a patient to reveal specific suicidal plans. This can lead to a discussion of alternatives to suicide, which the patient can be encouraged to consider. I believe an inquiry into suicide should be conducted whenever the physician is concerned about suicide risk because of the patient's depression, references to hopelessness or lack of a future, or a positive response to a suicide question on a screening form. The following questions may be used to inquire about a patient's suicidal thoughts.¹⁷

Positive responses to any of the questions should indicate to the primary care physician that the patient is at risk for suicide. The more specific the patient is with regard to a plan and the more lethal the plan is, the more severe the risk. For example, attempts involving shooting or hanging are more likely to be lethal than overdoses or wrist cutting.¹ If a patient has even passive suicidal ideation (eg, "I sometimes wish I would just die in my sleep. I think that my family would be better off without me"), referral for a psychiatric evaluation is the most prudent course. If the patient is actively and imminently suicidal (eg, "Yeah, I've been thinking about dying. I had my gun out this morning. I don't know why I didn't just do it then"), seek emergency psychiatric evaluation.

Mr D experienced suicidal ideation with a very lethal plan. His description of being "inches" from jumping in front of a train is consistent with an aborted suicide attempt, which falls somewhere between suicidal ideation and a suicide attempt in terms of risk severity.¹⁸ If he had presented to his internist's office recently having contemplated jumping in front of a train, his physician would have needed to seek an emergency psychiatric evaluation. The fact that Mr D also was considering alternate methods of suicide increased his suicidal risk. In addition, Mr D demonstrated shame and intolerable rage. Fortunately, the presence of ambivalence and deterrents (primarily his wife and family and his religious beliefs) have positive therapeutic implications.

The fourth component of assessing suicidality entails sorting out disorder-based suicidality from personality-based suicidality. Disorder-based suicidality relates to an Axis I disorder (a major psychiatric syndrome [DSM-IV]), such as depression, and consists of prominent feelings of anguish or pain and a wish to escape. The risk tends to be more acute. There is a compulsive, driven quality. The options for treatment include medication, hospitalization, supportive psychotherapy, or electroconvulsive therapy.

Personality-based suicidality results from feelings of anger, aggression, or vengeance. The risk tends to be more chronic. There is also a complicating impulsive quality to it. When possible, the patient should be accorded as much responsibility as possible. This can be given only in the context of assessing the treatment alliance. Regardless of the basis of suicidality, primary care physicians treating patients who have the ability to carry out a suicide plan should seek emergency psychiatric consultation, voluntary psychiatric admission, or both. If patients are unwilling to be admitted for psychiatric evaluations, the primary care physician should seek the consultation of a psychiatrist in evaluating the patient. States differ with regard to their laws pertaining to involuntary commitment.¹⁹

Studies of completed suicides show that 75% of victims saw a physician within 6 months of their suicide and 60%, within 1 month.²⁰ These findings indicate that persons considering suicide appreciate that something is troubling them and make the effort to see a clinician, but do not or cannot communicate their suicidal thoughts. Primary care physicians can play a pivotal role in recognizing suicide potential in their patients.¹⁷ At least half of the patients receiving mental health care obtain that care through their primary care provider.²¹ In addition to providing psychiatric treatment, primary care providers are often at the point of entry into the health care system and decide when the patient needs psychiatric referral. As responsibility for diagnosing mental illness falls increasingly under the domain of primary care, it becomes more important for these clinicians to be informed about mental health disorders and understand when referral is necessary.²²

Medications have proven effective in the treatment of depressive disorders. Half of people with depression are treated in the medical sector and the other half treated in the mental health care system.²¹ I believe it is important when discussing the risks and benefits of antidepressant medication to educate patients and their families about dosage, the purpose and expected benefits of the medication, risks and adverse effects, and length of time for usual response and to emphasize that improvement can be uneven. The feeling of getting worse again after experiencing some improvement can be devastating to depressed patients, contributing to hopelessness and probably increasing suicide risk. If patients have been warned to anticipate that they will feel better some days and worse on others, then they are less likely to be overwhelmed by apparent setbacks.²³

In recent years, promising reports have been published about the impact of psychopharmacologic agents on reducing suicide risk. For instance, the use of benzodiazepines may modify the risk of suicide by reducing anxiety.⁹ Recent studies have determined that lithium has a strong, and possibly unique, protective effect against suicidal acts in persons with major affective disorders and particularly in bipolar forms of manic-depressive illness.²⁴ Several years ago there was intense media coverage of a possible link between fluoxetine hydrochloride (Prozac), suicidal ideation, and other aggressive acts. This question has been carefully studied and evidence refutes any correlation.^{25 - 27} Clozapine treatment can reduce risk of suicidal acts in patients with neuroleptic-resistant schizophrenia.²⁸ In summary, recent research seems to support hypotheses that treating psychiatric disorders with appropriate medications can reduce suicide risk. Primary care physicians, however, should remember that depressed patients, particularly in the early phase of treatment, can be at risk for suicide, so that attention should be paid to appropriate follow-up, patient education, and prescription size. In the end, the therapeutic patient-caregiver relationship is a critical life-saving treatment component.

In determining the level of intervention for Mr D, his suicidality should be considered as the combination of his depression (disorder based) and personality traits (personality based). The hospitalization he had was indicated to stabilize the crisis and establish a relationship with the therapist or psychiatrist during the hospitalization. It would be important to review pharmacologic interventions, and to educate the patient and family about suicide potential and managing these comorbid illnesses.

In the field of psychiatry, documentation of suicide risk assessment must be incorporated throughout the entire treatment process to ensure that the issue has been addressed for other health care professionals reviewing the medical records and for legal purposes. Useful guideposts for conducting assessments include: (1) the first psychiatric assessment or admission; (2) the occurrence of any suicidal behavior or ideation; (3) any noteworthy clinical change; and (4) any change in level of observation for inpatients, such as increasing privileges or giving passes before discharge. Physicians must also plan the frequency of reassessments. This is critical as it acknowledges that suicidality waxes and wanes. For the primary care physician, documentation of the suicide assessment is also crucial for clinical and risk management purposes to demonstrate that the physician has both inquired about suicide and documented the basis for the treatment decisions. In particular, use of pejorative terms such as "gesture," should be avoided when documenting suicidal behavior. Clinicians use the term gesture to describe patients who perform minor (rather than lethal) self-destructive acts. However, the fact that the self-destructive act is minor does not mean that the patient is not suicidal or that the intent was not lethal.

Mr D has several predisposing risk factors for suicide. Because of his erratic, antisocial personality traits, he needs exterior, sustaining resources more than most people, yet he is likely to alienate the very people he depends on for support. The combination of these factors means Mr D has a long-term, increased risk for suicide, especially an impulsive suicide in response to a crisis.

Mr D's gambling addiction, almost by definition, sets up crises that potentiate his risk for suicide. Every time he gambles away more money than he can afford to lose, he faces financial stress, rejection by his family, and a further drop in self-esteem. Mr D faces other potentiating risk factors as well: he has not been sleeping regularly, and a family member is ill. His high level of risk for suicide was obviously recognized by his physician, who arranged for his admission to the psychiatric ward.

It was appropriate and necessary to ask Mr D specifically about his access to guns. Despite his assurances that he did not have a gun and would never use one on himself, Mr D is obviously at risk for impulsive suicide. In outpatient planning, it would be prudent to talk to his wife and adult children about his access to guns. Also, it would be crucial to document the inquiry about firearms and that Mr D specifically denied having access to a firearm.¹⁹ If a firearm were present, it would be prudent to document instructions to patient and family.

Mr D described in some detail his aborted suicide attempt when he was "inches" from jumping in front of a train. He was, however, less than forthcoming about current or continuing suicidal thoughts. One has to question whether he is keeping an escape plan to himself. Mr D also implied that his wife does not know the full extent of his difficulties. Given that his connection with his wife is probably the single most important factor keeping him alive, this relationship needs to be addressed.

While some of Mr D's suicidality is attributable to depression, it seems likely that most of his suicidality at this time is related to his personality structure and the crises he faces in his personal life. When we see Mr D interviewed, he is already in inpatient treatment. However, if a primary care physician examined Mr D, he should refer Mr D for an immediate psychiatric evaluation. Mr D needs regular suicide assessments until his acute suicidality remits. Once he is discharged, he should be asked about suicide each time he sees his physician until he is stable, and his family should be instructed to call in the event that he starts gambling again or seems more aggressive and threatening.

A central component in managing the chronic nature of Mr D's suicidality is to educate him and his family about his illness and to develop a working alliance with Mr D's health care providers. Prudent use of antidepressant medication will be useful for treating both this depression and reducing his impulsivity. Encouraging Mr D to be engaged with a gambling addiction program is an important part of building an additional support structure in his life. The challenge in caring for a patient like Mr D, who has an addictive disorder and an affective disorder, as well as elements of a character disorder, is to be supportive and nonjudgmental yet stress the need for him to take responsibility.

A PHYSICIAN: Tell us about the decision-making process that brought Mr D into the hospital. I'm still unclear about which patients I should hospitalize and which I should send to a psychiatrist as outpatients.

DR JACOBS: If someone has a major psychiatric disorder with the presence of suicidal ideation and is not involved in a strong therapeutic relationship, then that person should be hospitalized, at least briefly. A man who has considered a lethal method, such as jumping onto a train track, or is thinking of staging an "accident" because of insurance is at very, very high risk. I think we want to believe that our patients want help, that they will tell us how they feel. If someone is feeling suicidal, in my experience, generally they tell us. I've had patients who have talked about serious suicidality in my office, and then I got nervous or anxious saying, maybe we need to do something about this. And one patient said, "Well, Dr Jacobs, if I can't feel suicidal in your office, where else can I feel it?" If individuals are not ambivalent about their suicide, they will not communicate their suicidality. . . . However, the good news is that most patients, even the seriously suicidal, are ambivalent and will communicate their suicidality, thus allowing an opportunity to intervene, whether by increasing visits, altering medications, consultation, referral, or hospitalization.

A PHYSICIAN: You mentioned the high suicide rate in patients with schizophrenia, even higher than in those with depression. But given command hallucinations and that patients with schizophrenia don't always communicate as directly, what does the literature now tell us? How do you sort it out?

DR JACOBS: Eighty percent of suicides are males, and with schizophrenic patients the proportion of males may even be higher. The issue of command hallucinations is somewhat problematic. I would hope that any schizophrenic patient who goes to an emergency department or physician's office having acute command hallucinations will be hospitalized. Part of the problem that schizophrenics have in terms of suicide intervention is their impaired ability to communicate. Because of their repressed affect, they don't appear to be in the kind of internal turmoil that a typical depressed person shows. It's sometimes hard to notice the pain of a schizophrenic person unless the individual is showing some depressive symptoms. Ironically, patients with good premorbid functioning are more aware of their decline, which places them at greater risk for suicide.²⁹ Male sex, chronic relapsing course, and a history of suicide attempts are risk factors for suicide in that population.

A PHYSICIAN: If the ratio of attempts to completions is about 20 to 1, how should we think about these attempts? Are people really that inept at committing suicide, or are they stopping short of a sure thing?

DR JACOBS: Statistically we believe that for every 23 attempts, there is 1 completion. Most attempts are made by females, with a ratio of 3 to 1. Clearly more females are attempting suicide and not completing it. You have to see attempts as part of the ambivalence and as an opportunity to intervene. The good news is that 90% of people who attempt suicide do not complete it.¹

A PHYSICIAN: Can you comment on the needs of families who survive a completed suicide and the needs of clinicians whose patients complete suicide?

DR JACOBS: For the family, the issues are guilt, anger, and self-blame. Try to reach out and understand where the family is. There is also a risk-management perspective here, because they can turn that self-blame onto someone else—sometimes the physician. The myth persists among the public that suicide should not happen, and if it does happen, someone did something wrong.

Physicians need to accept that suicides do happen, despite our best efforts. Suicides are part of the work—just like death in any other branch of medicine. You can learn from a suicide, but you have to be careful to avoid confusing feelings of compassion for the family with acknowledging that you wish you had done something differently. If you feel like that, you should probably consult with a colleague or speak with a risk manager.