The role of no-suicide contracts is but a small tactical piece of the larger strategic approach to the assessment and prevention of suicide. Its many obvious limitations—to some degree in assessment, but primarily in suicide prevention—should have driven serious discussion of no-suicide contracts out of consideration as a practical measure in clinical practice and a legal talking point in the courtroom. Yet the construct, practice, and surrounding discussions of no-suicide contracts survive despite all their many weaknesses and seemingly terminal status. Why? What keeps them alive?
It may be that in the microcosm of suicide prevention, where all rational strategies and predictions dissolve before the impulsiveness, determination, cleverness, and persistence that are the hallmarks of seriously suicidal persons, and where the differentiation between the seriously and imminently suicidal person and the neurotically opportunistic parasuicidal person with hidden agenda is often impossible to divine, the no-suicide contract seems to offer an arena where personal connection and human relatedness make a difference. The no-suicide contract seems to hold the promise that the strength of a relationship of a particular patient to a particular clinician will make the critical difference in keeping a person from taking the final steps in a long chain of visualized and possibly rehearsed scenarios of self-destruction.
Paradoxically, clinicians, especially in those frequent situations in which a strong relationship with the suicidal patient does not exist, instead bank on a peculiar social assumption about human behavior that has its roots in childhood and possibly even in our genes, that promises, commitments, and contracts are binding on each individual even in the face of inconvenience, even when contrary to our strongest urges. It is as if "promises are to be kept" is the 11th commandment by which we guide our lives.
Limitations for suicide assessmentWhat are the limitations of the no-suicide contract? They are vast. If we first examine its use in suicide assessment, it is sometimes said that refusal to commit oneself to a no-suicide contract conveys to the clinician that the person is seriously contemplating suicide and does not want the moral encumbrance of a promise to act as a brake.1 The assumption is that the scruples of the person, however suicidal, would actually interfere with agreeing to a no-suicide contract. Such is the assumed power of keeping one's word, no matter what. Deception is worse than death.
Optimistically, refusal to agree to a no-suicide contract may be intended by the suicidal person as a communication of serious intent in order to give the clinician an opportunity to intervene. In such a case, the clinician would be insensitive and even negligent in not acting on this cue to safeguard the patient. This is the best that can be said about a no-suicide contract as an assessment tool. The problem is, as can readily be appreciated, that refusal to agree to a no-suicide contract can have many other meanings, centered around motives and personality styles of struggle, hostile or dependent engagement, victimization mentality, testing of a relationship, and the need to raise the stakes and create excitement in one's life. Reciprocally, agreement to a no-suicide contract may be a disingenuous attempt to avoid interference with one's serious suicidal intent. The upshot is that agreement with a no-suicide contract does not help in assessment of suicide and refusal of a no-suicide contract still leaves the clinician guessing about the meaning and importance of the refusal.
Furthermore, the attempt during an initial evaluation, especially in situations of high stress and preoccupation with suicide, to obtain a no-suicide promise from a patient whom one has just met has the strong potential to backfire because it lends itself to the perception that the clinician is more interested in legalistic self-protection than in understanding the patient's desperate situation. This again raises the question of whether the no-suicide contract has relevance and benefit only in situations in which some sort of therapeutic relationship already exists. The psychiatric literature is in agreement that a no-suicide contract, if it is employed, cannot substitute for the careful and detailed assessment of suicide risk.2-5
Limitations for suicide preventionIn the arena of suicide prevention, the issues are muddier and more complex rather than simpler. First of all, characteristics of the patient under evaluation, the context and location of the interaction, and the nature of the relationship between patient and clinician are very important. It makes no sense to indiscriminately endorse no-suicide contracts generically as is often done at the time of admission to an inpatient unit. However, collaborating on an initial treatment plan by spelling out what the ideal expectations of patient and clinician are and communicating genuine concern for the patient at both a professional and personal level, even at a first meeting, needs to be distinguished from the formalistic procedure of extracting a promise to forswear suicidal and other self-injurious behaviors.
Some inpatient services include a no-suicide contract in the documents that a patient initials or signs on admission: laundry list, HIPAA confidentiality regulations, personal belongings put into storage, insurance information, smoking policy, and meal schedule. In effect, a tool that might be useful in a family setting in which there is an ongoing relationship with the patient is being used during a nursing evaluation and management of risk of the patient, in a setting in which the parties to the "contract" are essentially strangers to each other.
Given that the average length of inpatient stay has been reduced in the past few decades to 3 to 7 days and that criteria for admission have been constricted to imminent risk of suicide, homicide, or total failure of outpatient treatment programs, clinicians must safeguard and treat an increasingly high-risk, impulsive, often drug-addicted population with strategies that do not allow much development of a therapeutic relationship.
In a retrospective chart study, Drew6 found that patients who had no-suicide contracts were more likely to engage in self-harm, although it is possible that negotiation of a contract reflects staff assessment of high risk. Hospitalization is no longer a process that emphasizes daily involvement with a psychiatrist and a close working relationship with nursing staff. As a corollary, keeping suicidal patients safe on a ward requires well-defined algorithms and protocols for assessing suicidality and implementing whatever procedures, including increased nursing time spent with a patient, are required, with no place for perfunctorily signed no-suicide contracts.7
In an outpatient situation, all the standard limitations of no-suicide contracts apply. Investigators agree that there is no empirical evidence that use of these contracts reduces risk of suicide; there are few studies and no controlled studies. In response to hearing about a clinical psychologist who was censured by the Minnesota Board of Psychology for not obtaining a no-suicide contract with a borderline patient who threatened but did not make a suicide attempt, and curious about the psychology board's rationale for this disciplinary action, I undertook a postcard survey of 514 psychiatrists in Minnesota inquiring about no-suicide contract practices.8
