Q. I would like to know what is considered an appropriate amount of time to discontinue suicide precautions on patients that are admitted with suicidal ideations or attempt. If the patients stabilizes and their mood improves and is ready for discharge, when should discontinued suicide precautions be written by the physician? Would you consider writing the order on the day of discharge appropriate?

A. If you will permit me to put off a direct answer to your question, I would like to suggest that suicide assessment and management is a highly individualized issue that does not lend itself to cut-and-dried policies--even though your hospital probably does have some policies on suicide precautions that might be worth consulting.

A very useful set of general recommendations was prepared in 1996, by the Suicide Risk Advisory Committee of the Risk Management Foundation of the Harvard Medical Institutions. You can find these on the website http://www.rmf.harvard.edu/rmLibrary/clinical-guidelines/suicide/. The guidelines very appropriately note that, "Inpatient treatment of suicidal patients relies upon a progression through a hierarchy of observation levels, supervision levels, privileges, and therapeutic passes. With clinical improvement, suicidality may still persist. Although the ultimate goal is toward a less restrictive environment, the clinical decision must be based on an assessment that the suicide risk has been reduced."

The critical words here are, "...based on an assessment that the suicide risk has been reduced." And notice that the word used is "reduced", not "eliminated"! Thus, it is probably more important that you and your staff have reliable means of assessing suicidal risk, and that you clearly document that these were used, than it is to ponder the appropriate amount of time needed before suicide precautions may be discontinued.

In fact, I personally would avoid writing an order stating, "d/c [discontinue] suicide precautions", since it seems to imply that there is no longer any risk of suicide--and this is very doubtful in a patient who has recently been admitted to the hospital with suicidal ideation or behavior. (An improvement in mood is certainly not a guarantee that the patient is no longer suicidal--in fact, some patients show just such a brightening of mood, immediately prior to making a suicide attempt).

It might be preferable to have a series of levels of observation, such as suggested in the Harvard guidelines:
a. Continuous observation (1:1 or remaining in sight of staff members)
b. Restricting the patient to an area where he or she can be seen at all times by staff
c. Restricting the patient to public areas; not allowing him or her to be alone in room
d. Checks at intervals of 5, 15, or 30 minutes
e. Periodic checks at intervals greater than every 30 minutes

With a system such as this, the clinician can write an order such as, "Level 'e' precautions: check every 45 minutes" rather than "d/c suicide precautions", while carefully documenting the clinical thinking that underlies the order. This, in turn, requires a thorough understanding of risk factors for suicide or attempted suicide; e.g., hopelessness, self-negativity, and poor problem-solving performance in schizophrenic populations (see Hughes SL, Neimeyer RA; Death Stud 1993 Mar-Apr;17(2):103-24).

You might also consider the use of the Modified Suicide Intent Scale and the Hopelessness Scale, which appear to be correlated with immediate suicide risk (see P. Cheung, Aust N Z J Psychiatry 1992 Dec;26(4):592-8). Finally, remember that whereas doctors are not expected to predict suicidal behavior, they are expected to recognize patients at high risk for suicide, and to manage them accordingly.

October 2002

Disclaimer Back to Ask the Expert