A Date With Death: Management of Time-Based and Contingent Suicidal Intent

The assessment of suicidal risk is one of the most common and basic procedures in psychiatry, and many discussions of clinical approaches to this determination have been offered (2–7). Although persons at risk for suicide always raise some anxiety in mental health professionals, the appropriate responses by the treaters to this condition are fairly straightforward: assessment of the imminence and seriousness of the risk, evaluation of available resources, and a clinical disposition that addresses both of these realms. Growing clinical experience may alleviate anxiety in this area. Indeed, experience can even guide the therapist through the tricky terrain of suicidal intent as blackmail or coercion: “Admit me to this hospital or I will kill myself” (8).

One class of patients within the larger population of persons at suicidal risk, however, poses special challenges for the clinician and produces anxiety, not only in the novice but even in the experienced professional. These individuals do not present with immediate or imminent risk through suicidal intent or with an obvious “blackmail” goal (“Do this or else”); rather, they link their suicidal intent to a future point—sometimes even a distant future point—or to a particular life situation on a contingent basis. The following examples illustrate more clearly how the core of such a time-based or contingent dynamic might be posed:

“If I don’t feel better by my 30th birthday, I will kill myself.”

“If my mood is not improved by the third year of therapy, I’ll kill myself.”

“If I am not married by 2 years from today, I’ll kill myself.”

“If my wife leaves me, I’ll kill myself.”

For therapists, patients who take such positions pre­sent significant therapeutic resistances, provoke strong countertransference feelings of helplessness and anger, and invite both under- and overreaction. Moreover, the professional literature provides little guidance on how to deal with the dilemmas posed by such patients.

LITERATURE REVIEW

Acute suicidality is a familiar entity in office and hospital practice (9–12). Although examples are less common, the literature has also addressed patients with chronic suicidality—primarily those patients with personality disorders or fixed delusions who are “never not suicidal” (13–16). One of us suggested (17) that acute suicidality is a problem in the psychology of despair, while chronic suicidality is a problem in the psychology of responsibility—that is, despair and responsibility represent the critical issues in the respective cases. This formulation permits conceptualizing approaches to such patients.

The patients who were our focus were neither acute nor chronic. Their self-destructive intent was not imminent; indeed, according to the patients, it could be years away. Nor was the problem actually chronic. The patients were not always, or even often, suicidal; their lethality was contingent on a time or set of circumstances in the future. Such patients may feel no suicidal intention in the present at all, in part, of course, because the wish has been encapsulated into a particular context set in the future. This suggests that one of the elements of contingent suicidality is its effect as a defense against actual immediate suicide.

Farber (2) expands on this last point as follows:

This special situation is both contingent and chronic. Here, suicide may represent the ultimate control for the patient when all else has failed.

Cavenar et al. (18) discuss reactions to anniversaries, which are often the occasions chosen for time-based suicidal intent. According to their definition, anniversary reactions are time-specific psychological and, at times, physiological reactions that occur or recur at specific anniversaries in response to a trauma that the individual has experienced in the past and has not effectively mastered. Certain patients with obsessional styles appear to accumulate anniversaries, such that almost any day of the year is regarded as the anniversary of some meaningful event and is thus a point of theoretically increased suicidal risk.

The number and variability of all the foregoing factors combined make the use of involuntary commitment—a common intervention in suicidal cases—highly problematic and ethically complex.

THE PRESENT STUDY

We intend here to present some real but disguised and condensed examples of this problem in clinical practice, to analyze their dynamics, and to suggest both clinical and risk-management approaches.

Comment. For Ms. A, as with other patients known to clinicians, death took on the qualities of an object relationship that, paradoxically, permits continued living and provides a sustaining and supportive function. Patients themselves clearly describe how death becomes intrapsychically personified as a kind and helpful healer who “promises” to take away the pain and suffering and to grant blessed relief. The therapist removes or challenges the relationship with such an agent at his or her own peril, since it may be the only relationship in which the patient has any faith. Any interventions made should respect the value of this tie, and the therapist should avoid placing the patient in a position of having to give it up precipitously (e.g., by asserting, “I will put you in the hospital so that you cannot kill yourself”). The therapist should carefully record the rationale for not vigorously opposing this relationship, since opposition might seem to be called for by the standard of care. In addition, the therapist should concretely state that he or she supports the part of the patient that wants to stay alive, even though the patient may feel tempted by death.

Comment. In this case, Ms. B accepted voluntary hospitalization in anticipation of a “date with death,” in part because of a contract with her therapist and widespread ambivalence about suicide. In addition, the therapist terminated her therapy at that time. The outcome of this case raises the question as to whether Ms. B’s therapy with that particular therapist was somehow overstimulating her, since after termination she appeared to improve.

Comment. Mr. C’s case presents a valuable teaching point: clinicians themselves should not be too concrete or literal about the specific deadlines and contingencies proffered by the patient as though they were guarantees of continued life until those set points. The underlying depression may overwhelm the patient on occasions other than the patient’s verbalized dates or times.

Comment. This case further supports the value of refraining from premature or unilateral attempts to take away the patient’s relationship with death.

Comment. Ms. E’s case demonstrates how a suicidal plan can persist despite some clinical improvement and increased insight. Since her suicide plan was child centered, it also appeared changeable in response to a child’s condition. This fact did not seem to decrease the seriousness of her plan, however.

In the Netherlands, where assisted suicide is an accepted medical intervention, examples appear of what must be styled “iatrogenic contingent suicide.” For practitioners, such cases pose different problems from those posed by patient-set contingencies, as the following examples show.

Comment. The foregoing discussion displays the problems with the therapist’s application of contingencies to a patient’s suicide. Clearly, it may be valuable to the patient for the therapist to establish a moratorium on the suicidal act and to separate depressive colorations from the decision making (20). Thus, one might say to a suicidal patient, “I know I can’t stop you from killing yourself, but it would be a pity if you made such an important decision while depression was clouding your judgment. Let’s get you out of the depression and then, if you still want to kill yourself, I can’t stand in your way” (J. Ewalt, personal communication, 1968). Such a negotiated moratorium asserts the truth—that determined persons cannot be stopped. It then goes on not only to postpone the irreversible but also to suggest indirectly a way out of the depression—a future-oriented perspective usually clouded by that very depression.

In contrast, however, clinicians forged alliances with the two Dutch patients described earlier around the promised surcease of dying, indeed, with the psychiatrist’s help, if treatment failed. In the Dutch cases, the authors address neither the possibility of subtle incompetence (21) in the patients nor the possibility of treatment with electroconvulsive therapy in resistant depressions. We could advance a strong clinical counterargument that psychiatrists should never support suicide, but should acknowledge the human impossibility of preventing it.

DISCUSSION

Ethical Dilemmas

These cases raise ethical dilemmas for the treating psychiatrist. When, if ever, is a chronic psychiatric illness analogous to an incurable physical illness? Certainly, psychic pain can be as debilitating as physical pain, and some patients may not respond to treatment. Hendin (22) refers to one type of suicidal patient as the patient who sees himself or herself as already dead. Often such a patient presents with very little affect with which the therapist can work. Another type of patients who are often unresponsive to treatment are those patients with certain personality disorders. Studies of psychological autopsies (postmortem reviews of the apparent psychology of the suicidal patient) suggest that up to one-third of suicide victims had personality disorders (23). Suicidal patients with personality disorders, anxiety disorders, or substance abuse disorders are less likely to experience hopelessness than to experience cognitive distortion (24).

Another ethical concern is to what extent psychiatrists should intervene when a patient’s date with death approaches. Should we respect a rational patient’s right to autonomy and self-determination? How do we reconcile such respect with our duty to protect, to prevent impulsive and irreversible steps (suicide) in response to an often reversible condition (depression)? While we would not take the position of colluding in suicide, as in the example of the Dutch psychiatrist, we recognize that patients who are really intent on suicide will find a way. Absent an immediate crisis that is pushing the patient over the edge, brief hospitalization may not have any deterrent value, as these patients can always reschedule their dates with death.

Studies of therapists’ attitudes towards elective suicide show that a surprising number of them support the notion of rational suicide and believe that people have a right to choose both the quality and duration of life (25, 26). Psychologists in one study, however, were more likely to take action to prevent suicide in the case of a man with psychological pain than in the case of one with physical pain from terminal cancer. Werth and Cobia (25) suggest that when therapists face cases of “rational suicide,” their duty may be more to assess the rationality of the client than to protect him or her. Here, a latent question of competence to make choices appears to be at issue.

Several authors have discussed just what should go into such an assessment. Motto (27) proposed that the intended act must be based on a realistic assessment of one’s life situation and that ambivalence about suicide must be minimal.

Recommendations

Each of the case vignettes noted earlier is followed by specific risk-management suggestions for the issues raised therein. It may yet be helpful to offer some general suggestions about these clinical dilemmas.

First, the patients described in this study were not only planning suicide but were telling the clinicians about it. A patient intent on death might well not disclose this plan; about 60% of the population who commit suicide do so without telling any professional (28). Thus, we may first approach the patient’s date with death as communication: “What are you trying to tell me about your condition by setting this deadline?” Exploration may yield useful clinical data about significant others, lost sources of former pleasure, and similar information; sharing these issues with the therapist may renew the patient’s attachment to life.

From one viewpoint, a date with death is also an attempt to blackmail fate, the world, or other people. Just as patients who manipulate others feel unentitled to ask directly, patients who attempt through blackmail to gain power to control the environment may feel helpless to control anything. Thus, a date with death offers such a patient a feeling of mastery over helplessness. Opening the topic of felt helplessness may enable the therapist to establish empathic contact with a patient whose date with death appears to hold the clinician at a distance, helpless in his or her own right; shared helplessness may provide a pathway to empathic connection. Since suicide may be the patient’s way of communicating unbearable psychic pain, the clinician must acknowledge the pain, letting the patient know that it can be borne by the dyad.

The problem of helplessness constitutes a powerful argument for avoiding ineffectual power struggles with patients about their suicide plans. Indeed, power struggles over suicide may represent a resistance to therapeutic engagement on the issues underlying suicidal feelings, closing off discussion, as in case 5. While involuntary hospitalization of patients for their protection may certainly be necessary at some point, careful pacing of the therapy and the teaching of cognitive and behavioral skills may allow patients to discuss and then to deal effectively with unbearable affects while remaining an outpatient.

Countertransference factors including projection may lead to the clinician’s feeling blackmailed even when this is not the patient’s intent; the patient in a suicidal state is often too caught up in the affect to consider the impact on others.

The assessment of competence is a central consideration in dealing with patients, not only at the very end, when “competence to commit suicide” may be at issue, but also at earlier stages, when patients’ competence to choose to inform or not to inform their clinicians about their status may critically determine joint decision making. Almost every jurisdiction permits competent patients to refuse life-saving interventions by other medical specialties; presumably, a competent patient could refuse life-saving psychiatric intervention as well (29). While not established, this principle has been successfully utilized in the defense of suicide malpractice cases in at least six states. The decisive question is whether the patient suffers from the subtle incompetence created by masked depression (21) or an equally subtle psychosis. Careful assessment can answer this question; equally careful documentation can preserve the answer.

Because the patient is presenting with intent and a plan, though a future-oriented one, clinicians may be able to draw some comfort from the fact that their attempts at involuntary commitment at or near the critical date can almost always be justified, even in ambiguous situations, although obviously the endpoint would be extremely unclear. This freedom, paradoxically, may permit the clinician not to commit the patient unnecessarily. In some jurisdictions, commitment in these cases would be far more difficult, especially in locales where a previous overt act of suicidal intent is required for commitment. Since it would represent the less conservative action, the decision not to commit should be carefully reasoned in the record. Clinicians should seek active, ongoing consultation or supervision to avoid the pitfall of their own anger at being blackmailed—a common and understandable reaction, but one that may threaten clinical judgment. Consultation here (and at earlier points) is valuable in such situations and also prudent in the event that the patient commits suicide, and a suit is brought against the clinician.

CONCLUSIONS

Accepting the patient’s pain and sense of hopelessness is not the same as acceding to his or her wish to commit suicide; the psychiatrist must always hold out hope. At the same time, it may be therapeutic and realistic to let patients know that one can not ultimately prevent their suicides. We believe that a psychiatrist should never condone a patient’s decision to commit suicide; however, there may be cases where the psychiatrist can justify not intervening and accepting the patient’s stated wish.

Under the Damoclean sword of the patient’s date with death, the therapist may wish to follow the above recommendations in the service of maintaining the holding environment despite the pressures brought to bear. While not all suicides can be prevented, the above approaches may at least permit the clinician to remain in the therapeutic relationship at all times while attending to risk-management considerations.

Received July 18, 1997; revisions received Dec. 8, 1997, and March 31, 1998; accepted April 17, 1998. From the Department of Psychiatry, Harvard Medical School, Boston; and the Program in Psychiatry and the Law; Massachusetts Mental Health Center. Address reprint requests to Dr. Gutheil, Massachusetts Mental Health Center, 74 Fenwood Rd., Boston, MA 02115. The authors thank Drs. John T. Maltsberger, Teri Rumpf, Michael Sharpston, and Larry Strasburger for comments and suggestions; and Ms. Ellen Lewy for essential assistance with the manuscript.