1. A patient comes to a hospital with recognizable suicide risk, is not admitted, and within a few days kills himself. 
2. A patient is admitted to a hospital, is discharged after just a few days, and then kills himself.
3. A patient is admitted to a hospital and kills himself while he’s there.
4. A patient who is seeing an outpatient doctor has recognizable suicide risk and kills himself at some point during treatment.

Among those four issues, the most common that I see has to do with patients who’ve been hospitalized and then commit suicide after discharge. However, since far more patients are seen in outpatient settings, there are many lawsuits brought against out- patient psychiatrists as well.

TCPR: How can outpatient psychiatrists decrease the chance that their patients will commit suicide, and how can they properly document their efforts?

Dr. Reid: A friend of mine who is a prominent plaintiff’s psychiatric malpractice lawyer told me something very interesting. He said, “If a family calls me and says that they want to sue a doctor because of a suicide, and I get the records and see that the records reflect adequate assessment and good documentation of that assessment, I will very often tell the family that I see no grounds for that lawsuit.” His point is that judges and juries don’t often second-guess a doctor’s good-faith judgment once he or she has gotten as much information about risk as is feasible.

TCPR: What is an adequate evaluation and how do we go about documenting that?

Dr. Reid: First, you need to document that you’ve done a good evaluation, not a 15 minute checklist, but an evaluation that shows that you covered the most important risk factors, sought collaborating information, and thought carefully about your patient’s sui- cide risk. Documentation is part of the standard of care, and it is your friend if a tragedy occurs. The magic words in any chart are “suicide risk assessment,” followed by a few sentences that show that you did a detailed assessment. You cannot simply say, “Patient denies suicidal ideation (SI).” Patients lie about suicidality all the time. If someone comes to your office serious about wanting to kill him- or herself, a part of that patient probably wants to keep the suicide option open, and he or she will say the right things, hide certain information, and downplay the risk to prevent you from taking that option away (eg, to avoid hospitalization).

TCPR: Specifically what are some of the things that we, particularly outpatient psychiatrists, should be documenting in our risk assessments?

Dr. Reid: Write down in some detail your good-faith efforts to assess the level of risk and your efforts to prevent suicide. For example, “Explored patient’s thoughts of death.” “Talked at length about his thoughts of killing himself.” “Discussed the fact that his mother killed herself at about his age.” “Talked about the potential consequences of suicide and patient acknowledged that it would be devastating to his family.” If you find that there is a moderate or greater risk of suicide, document that impression and document what measures you took to protect the patient. For example, “The patient is at moderate risk for suicide, but is not at high enough risk to require immediate inpatient hospitalization. I plan to meet with the patient again tomorrow” or, “I discussed my concerns with his wife who will accompany him to his visit tomorrow” or, “This patient needs the protection and detailed evalu- ation of an inpatient setting.”

TCPR: If we do decide to hospitalize a patient who is actually in the office, how should we approach that, logistically and in terms of documentation?

Dr. Reid: It depends on how imminent you believe the patient’s suicidal ideation is. In many cases, you will want to call the authorities to come directly to your office. Documentation options in that case would include, “Discussed hospitalization with patient and her spouse, and we’ll arrange ambulance transportation to the emergency room.” (Don’t trust high-risk patients to go to the hospital on their own, or to be taken there by family.) Or, “The patient refuses hospital referral even though I’m very con- cerned about suicide risk. Her mother and nurse Jones will sit with her until officers arrive to execute an emergency health hold.” (Don’t attempt to physically keep the patient in your office; that’s a job for law enforcement.)

TCPR: What do you think about having patients sign “no-suicide ‘contracts’”?

Dr. Reid: I strongly discourage them. From a suicide prevention perspective, and from a malpractice liability perspective, so-called no-suicide or no-harm “contracts” are worthless for at-risk patients. Psychiatrists and other clinicians, nurses, and institutions should not rely on them for patient safety. I’m not criticizing using them in a therapeutic context, when a clinician is working closely with a patient on an ongoing basis, but they absolutely must not be relied upon to protect a patient’s life, and they certainly should not take the place of adequate patient monitoring and physical protection.

TCPR: Which of the major suicide risks should we be most concerned with in our assessments?

Dr. Reid: I recommend that clinicians be aware of a standard list of major risk factors that they can refer to during their interviews. Having said that, it’s very important to note that some risk factors are so critical that once they are present, it should be hard for a patient to convince you that he is not at high risk. I’m speaking of factors such as very recent suicide attempts, threats made by psy- chotic or severely depressed patients, and dangerous behavior in very unstable, unpredictable patients (including intoxicated ones). But each patient is different, and assessment is not a “cookie-cutter” proposition.

TCPR: Are there any other major risk factors that get left off of common risk factor lists?

Dr. Reid: Patient instability is a huge one. Both doctors and patients may believe that the patient is currently at lower risk (for example, after a few days of monitoring and respite in a hospital), but one must always remember that the patient’s apparent “improvement” may not be reliable and lasting, and may not stand up to whatever environmental challenges might have precipi- tated the suicidal thoughts in the first place. Be sure to consider whether or not the patient’s apparent improvement is vulnerable to change or breaking down in the days to come.

TCPR: If you can’t predict suicide, what can you do?

Dr. Reid: One thing is sure: Unpredictability doesn’t mean we don’t have to take responsibility for assessing and managing suicide risk. The heart attack analogy is one that juries frequently can identify with. If I have chest pain, the cardiologist cannot predict whether or not I’m going to have a heart attack. However, the standard of care requires that my cardiologist assess me and not let me leave the hospital or clinic until he or she reasonably believes it’s safe for me to leave. If there are things the cardiologist, hospi- tal, or clinic reasonably can do to decrease my risk of a heart attack, they should show that those things have been done. And when I leave the emergency room, they should make sure that I know that I can call them any time, that I can come back in a few days if my symptoms return, etc. That’s how a jury thinks about these things. Analogously, you may not be able to predict whether or not
a patient is going to commit suicide, but you can show that you’ve done all that is reasonably within your power to assess that risk and manage serious risk.

TCPR: You’ve talked a lot about the negative risk factors, but there are also potentially protective factors, such as a patient having children or a supportive home. Should we document those as well?

Dr. Reid: Document them, sure, but unfortunately, those so-called “positive” factors, such as having a supportive family or having a good job, are pretty much worthless when it comes to calculating serious suicide risk. One or two major risk factors, such as severe depression, overwhelm so-called “positive” factors very quickly. Positive factors may be statistically significant when you’re study- ing large groups of people, but people who are truly in danger of dying by their own hand usually don’t care that they have three children, or that they have a job, or that they have a religion that frowns upon suicide. It’s as if they have blinders on: the pain is so great, and the anguish is so all-consuming that they just don’t pay much attention to anything else. When you talk to people who came close to death by suicide, they routinely tell you something such as, “I don’t know what came over me, I wasn’t even thinking of the effect on my kids.” On the other hand, being aware of “positive factors” can be very helpful when you’re doing suicide coun- seling with people at lower risk. Saying things such as, “Do you really want to give your family pain for the next two generations,” (which is often how long a family’s response to suicide lasts), can be quite helpful in a counseling context.

TCPR: One of the most difficult dilemmas for any outpatient psychiatrist is whether or not to commit a patient against his or her will when you think there may be a substantial risk of suicide. These are really the patients that keep us up at night. Can you help us think that issue through?

Dr. Reid: First, let me say that the psychiatrists who stay up at night worrying are the ones who are least likely to be sued. They are caring about their patients and working hard to do the right thing. I want them to document their caring. In terms of committing patients, I’m not suggesting that psychiatrists hospitalize—voluntarily or involuntarily—every person who talks about hurting him- self. On the other hand, every psychiatrist needs to have access to some kind of protective resource, such as an emergency room or a psychiatric hospital. Once you come to a conclusion that there is a higher than moderate risk, you need to talk about hospitaliza- tion. If you choose not to hospitalize, you need to document why you came to that decision.

TCPR: In practical terms, sometimes we will bring up the issue of hospitalization and patients may be very reluctant to be hospitalized. We may believe that there is some degree of risk but it is always a judgment call, and we may choose not to hospitalize in order to maintain the therapeutic alliance.

Dr. Reid: It’s reasonably seen as a judgment call only if the assessment has been adequate and the clinician is trying to do the right thing. It can be a difficult decision. But there are several bad reasons that we sometimes use to decide not to admit patients that you should keep in mind. For example, if the patient doesn’t want to go into a hospital—he thinks he’ll lose his job, or do dam- age to a relationship, or that he can’t afford it—that’s not a good reason to take a chance on his survival. Think about the potential heart attack analogy: Once you make the decision that a person needs around-the-clock protection, don’t be swayed by these kinds of arguments. Another potential pitfall is relying on the family to do the job of a hospital. A family might say, “We’ll keep an eye on him, doctor,” or you might feel tempted to ask the family to do that. But that’s just not reasonable when the risk is high, and that decision is unlikely to assuage your guilt or hold up in court if a tragedy happens. There are two reasons that families can’t take the place of hospitals. First, hospitals have trained staff to monitor and treat patients. Second, in a hospital, a patient can be moni- tored 24-7, but at home, as vigilant as families might be, they are not going to be following patients everywhere they go (such as to the bathroom). Asking them to take over a hospital’s and nursing staff’s role is unfair, and it doesn’t work to protect the patient adequately.

TCPR: You said several times that we need to assess whether the patient is at moderate or greater risk of suicide. Is there an agreed-upon definition of what constitutes more than moderate risk?

Dr. Reid: There really are no agreed-upon definitions. There are various checklists of risk factors that provide a numerical “score” representing moderate or high risk. Being aware of such checklists, or the factors on them, is helpful, provided they are used as a “jumping off place” for a complete, personalized assessment. What I really want to see in my reviews for lawyers and malpractice carriers is some narrative in the chart that indicates personal assessment and judgment on the part of the doctor.

TCPR: Are there any particular examples of poor documentation that you frequently see that you would recommend that we avoid?

Dr. Reid: Sure. One of the worst is a simple “No S/I” without even a few sentences indicating that suicide risk was adequately explored. Such brief shorthand implies—whether it’s true or not—that the clinician didn’t spend much time and the assessment may have been superficial. I also hate the phrase “suicide gesture.” If somebody took six Valium, or superficially cut herself, it is tempting to say that was not a serious event, but just a “gesture.” That’s dangerous. If that happened to your wife or husband, or son or daughter, you would take it very seriously; doctors should as well. In fact, so-called “gestures” often escalate to serious attempts without adequate psychiatric or psychological attention. Another phrase that I don’t like to see downplayed in risk assess- ments is “cry for help.” If a particular patient’s version of a cry for help is to open up a vein, that should be taken very seriously.

TCPR: I think we all know that talking to family members is good practice in psychiatry, particularly in situations where there is a possibility of suicide. However, in this age of HIPAA privacy regulations it sometimes gets confusing as to who we can and cannot talk to. Do you have any words of wisdom about that?

Dr. Reid: Stop worrying about HIPAA or confidentiality when a patient’s life might be in danger and you’re acting in good faith. Talking with relatives who know the patient well, if they can be reached, is often crucial. Juries understand that all bets on confi- dentiality are off when you’re trying to figure out whether or not a patient’s life is in danger. Lawsuits related to suicide are com- mon, but the number of lawsuits related to breaches of confidentiality in situations such as this is so low that I’ve seen only one in over 40 years of clinical work and forensic consultations, and that one was dismissed by the judge. Which do you want to worry about, the patient’s life or whether or not he’ll sue you for talking to his wife?

TCPR: Thank you, Dr. Reid.