How to Write a Suicide Note: Practical Tips for Documenting the Evaluation of a Suicidal Patient

May 1, 2007

Proper suicide assessment is probably the most important part of a clinician's job; appropriately, heavy emphasis is placed on this in our education. Unfortunately, psychiatrists receive comparatively little practical guidance in documenting the history and physical examination (H&P) of a suicidal patient.

Proper suicide assessment is probably the most important part of a clinician's job; appropriately, heavy emphasis is placed on this in our education. Unfortunately, psychiatrists receive comparatively little practical guidance in documenting the history and physical examination (H&P) of a suicidal patient. As a forensic psychiatrist, I have seen reports that fail to convey the rigor, time, and thoughtfulness that went into the work because, although lengthy, the content was not prioritized. The result is an adverse judgment, settlement, or the immense frustration of a lawsuit that could have been avoided with different (and more focused) documentation.

This article is not about how to perform a suicide assessment; it is assumed the clinician is already able to do this, or knows where to go for assistance. This article is about documentation and prioritizing the information contained in the report.

During the patient interview, the clinician will learn far more than he or she can include in the report; how do you decide what is important for posterity? It is unreasonable and disingenuous to assume that every H&P will be a thorough and rigorous journalistic recording of the facts, events, and thought processes that occurred during the interview. Should two thirds of the H&P be devoted to symptomatology-assessing for depression and the like-and one third to the assessment and plan? In addition, within any section, what needs to be written explicitly and what can be left out? In this article I offer 9 maxims to help focus the writing, using the model of an emergency department (ED) setting to illustrate. See the Table for key points addressed here.

  
Conducting an assessment and writing a report are 2 different processes.

Write so the reader does not have to guess what you were thinking

The biggest mistake in medical note writing is spending too much time or using too many words in the "objective" section and not enough in the "assessment" section.

Who is going to read this report? Perhaps insurance companies will read the objective section and use it to bill accordingly. But other doctors, lawyers, and jurors will care about the assessment and plan sections. Certainly, objective facts are important to list, but the strength of the report comes from explicitly describing your thought process while drawing conclusions about the patient. You must write your assessment and plan in a way that makes it completely obvious to others why you did what you did.

A common, and I believe mistaken, approach is to force the reader to infer that the patient was not suicidal by simply listing specific behaviors, actions, or statements in the objective portion without interpreting these behaviors in the assessment section. Consider the following objective: "Patient denied suicidal ideations . . . was joking with staff . . . contracts for safety. . . ."

The problem with this listing of the facts is that the clinician is forcing the reader to connect the dots. Instead, make it obvious what you were thinking. "I was able to conclude that the patient wasn't suicidal because not only was he denying suicidality but his good affect and joking with staff reinforced that he felt better."

Prime real estate should be valued

The maxim "if it wasn't written, it wasn't done" is false. However, if it was written, it shows that it was more important than something that was not. Writing about a patient's future plans demonstrates that it mattered more to you than did, say, the check box near "Appearance: Clean." Or the reverse: not writing about the patient's future plans makes it look as though it mattered less to you than his appearance.

Do not waste too much space on symptom assessment. It is useful in making the diagnosis but the report is really about what the diagnosis was and what you did about it.

Your report should be timeless

Certainly, you need to assess suicidality and explain why you think the patient is not suicidal right now. However, keep in mind that your assessment will last more than 24 hours. You also need to evaluate his future. You assess that he is not suicidal now, but what about next week?

Consider, for example: "Currently he is not suicidal, feels fairly hopeful about the future, and has made some specific future plans like. . . ."

Hopelessness and pessimism about the future is a very important predictor of risk; it should be noted explicitly. "However, given his history of [impulsiveness/drug use/suicide attempts], it is probable that he will attempt suicide again at some point in his life when stressed. Unfortunately, this is a function of his future acute stressors-stressors over which I have no current control-not how he feels right now."

This is key because it sets up the assessment (the plan comes later): admitting that the patient is likely to attempt suicide again but that it has nothing to do with how he feels today.

The report also explains how you cannot be held responsible now for what happens in the future. State why-in your opinion-you did not hospitalize the patient or give him an SSRI or something else.

As an example, consider a specific chronically suicidal patient whose suicidal tendency is related to his catastrophizing stressors. He has had numerous hospitalizations but no decrease in the frequency of his suicidality or presentations to the ED. Assume, in this example, you feel the patient has been stabilized in the ED at this moment and is safe for discharge. What might you add in the report? Do not simply write that you are not going to hospitalize the patient, write why."Hospitalizing Mr X now is not going to alter that future eventuality and thus is not indicated today. In fact, recurrent hospitalization may be detrimental because it seems to have established a pattern of dependency rather than finding better ways to deal with distress."

But what makes the report good (and saves you from a charge of negligence) is how you tie it together. "Given the chronicity of the patient's suicidality, I have to do something that will actually help him in the long term. I believe he is not suicidal now, so my responsibility is to help decrease his suicide risk as best I can. I believe that the best way to help is to refer him for [intensive therapy/day program/psychiatric visit] for long-term follow-up so he can have somewhere to go and someone to manage him as symptoms and stressors develop. We discussed a crisis plan for future suicidality: at the first sign of distress he will call X; if this is not sufficient, he will call Y, and then Z. In addition, person A will stay with him and, if symptoms worsen, A will bring the patient to the ED."

The report should say all this, in your own style and with the contents of your interview. This way, when the lawyer asks, "Why didn't you hospitalize him?" the jury will already be aware of the answer.

The report is a not a news article, it is an op-ed

Note the way the sentences are written in the examples. They are personal and informal. The report is your educated opinion, not a scientific article. You must explain not only what you did but why you did it as opposed to something else (pretend you are writing to a jury if that helps). For example, "His affect changed when I left the room and he was noted to be joking" does not powerfully (enough) convey what you saw and what it meant to you.

Instead, consider: "With me, he was crying, but when I left the room and he thought the evaluation was over, I watched him joking and laughing with one of the nursing assistants, from which I inferred that he was exaggerating some of his symptoms." See how that changes things? Facts and events are the bricks for the assessment house. Don't describe just the bricks-describe how you built the house.

You can be wrong but you cannot be negligent

The report should emphasize judgment; specifically, what the clinician based his judgment on. It should be obvious from the report exactly what you were thinking and, especially, that you were thinking-that you took time to ask questions, observe, assess, and draw conclusions about the patient that were reasonable. You might have been wrong in your assessment but you did a thorough job. That is why the assessment matters.

"This is a 35-year-old man with a history of recurrent depression and near constant suicidal ideations, but rare attempts, who has been in the ED 6 times in the past year, approximately every 2 months, with similar symptoms (see above) brought on by varying stressors. However, this presentation is complicated by the absence of anyone at home-his wife and kids are on a trip-and no other immediate social supports are available. While it may be possible to discharge him and see him in the clinic tomorrow (as we have done in the past and would ordinarily do in the future), he has never been alone for such a long time and I can't be certain how he might act if he becomes even more stressed or does not sleep. Therefore, in this situation, we will hospitalize for further evaluation."

The plan should match the assessment. If the patient has a history of psychosis but is currently malingering and you discharge him, do not write, "Psychosis not otherwise specified; plan: discharge" since the disposition does not follow logically from the diagnosis. (It could be right, but it is not obvious and thus requires more explanation.) The diagnosis should be "Malingering; plan: discharge," which is obvious to the reader.

Clinicians will not always be right. But even lawyers allow doctors to be wrong as long as they were reasonable in their judgment. Think of it like this-the report is not written to win a lawsuit, it is written to prevent one.

Use quotes

In the not too distant future, we will not be writing reports, we will be recording them-in fact, we will be recording all patient interactions. Until then, use as many quotes from the patient as possible and then interpret them.

Call someone

This is not always necessary but it helps reinforce your case. Get another person's opinion and document it. It is one thing for you to say the patient is not suicidal but it is tremendously helpful to have a family member tell you he is not, or that this situation is common, or that this happens whenever he gets upset. If you have made the clinical judgment that the patient is safe to go home and his wife also thinks it is okay, write that down. For example, "Spoke with his wife, who agreed with my plan; she said, 'I didn't think he needed to be hospitalized, but he did need to talk to someone.'"

It is important to note that you are making your own independent assessment, but you are using agreements from other persons to show support for your position.

Conversely, you need to explain clearly why your assessment differs from that of a family member. You do not have to agree with the family member, but you have to explain why you do not. If a wife says her husband needs hospitalization you must have a really good set of reasons for why he does not-so make a specific point of writing them down.

If family and friends are not available (document that you tried-that also shows effort and is above standard of care), get a second doctor, resident, nurse, or another clinician to concur so you can write the next very powerful sentence: "Discussed the situation with X, who also evaluated the patient, and X agreed with me."

Apart from giving you a valuable second opinion, this documenation also helps establish standard of care, loosely defined as how a respectable minority of clinicians in your situation would have proceeded. Two doctors are usually considered to be a respectable minority as far as I am concerned (and have testified as such).

What did you do for the patient that made him safer for discharge?

"The plan is to let him decompress and regroup for an hour in the ED. We will give him lorazepam to help. I'll give him supportive therapy as well as try to teach him some better coping techniques for the future and have a nurse do the same to reinforce them."

Do not hide the fact that the patient disagreed with you, if he did. It shows that you heard him.

"I told him that I did not think hospitalization was the right course to take and I explained my reasons. While he was not happy with this-he wanted to be an inpatient-he at least understood my reasoning and was satisfied that I was actually trying to help him. Of course [note word choice], I called his [family member/friend] who agreed to come pick him up and stay with him continuously, not let him out of sight, and bring him back if things worsened. I explained how and when to give lorazepam, which helped in the ED."

If you are able and it is indicated, add: "I/therapist/nurse will follow up tomorrow to see how the patient did and if he took medication, etc." If you do write this in the report, make sure you actually do it!

Let me be clear. This report is not intended to convince you that the patient is not going to die. You have to already be convinced for yourself that despite his suicidality he will be alive tomorrow. Your clinical judgment comes first. Then, and only then, should you write a statement that supports and explains your decision.

Write the report as if the patient died

Never change your report after the fact. The lawyers already have a copy before you get your subpoena.

But imagine you could change the report. What would you change? You will have wished you had written more about checking this, or the patient denying that. Well, write this all down now. Write what you will someday desperately wish you had written.

It is important to remember that if the patient does commit suicide you will be called to retrospectively address the risk factors, so address them explicitly now. "The main risk factors for suicide in this patient include a history of previous suicidality, a diagnosis of borderline personality disorder, and alcohol abuse; however, he has not actually ever made an attempt, has been abstinent for 2 days (with low risk for withdrawal), is highly motivated to continue rehabilitation, and denies access to weapons (wife corroborates this)."