Duty to Warn

Confidentiality is one of the pillars of therapy. If there were no expectation of confidentiality, people would be unlikely to be as honest about what's going on with them, as a result of shame/guilt, fear of judgment, or fear that someone else would find out what they said. Therefore, the law has taken steps to protect clients' right to confidentiality, with only a few notable exceptions. For example, therapists are "mandated reporters," meaning that they are required by law to notify the relevant authorities of any suspected cases of abuse or neglect of a child, elderly or disabled person.

Another exception to confidentiality that is more controversial and varies significantly from state to state is the so-called "duty to warn." Laws related to the duty to warn are based on legal precedent set in the case of Tarasoff v. The Regents of the University of California. The court decided (albeit by a narrow margin) that a therapist may be held liable and subject to civil suit if s/he has reason to believe that a client poses an imminent threat to another person and does not take steps to warn that person. In fact, the client in question did not tell his therapist at the University of California the name of the person he planned to kill (Tarasoff), but the court felt that the therapist would have been able to deduce that person's identity from what the client did say.

This decision understandably sent shock-waves through the mental health community: therapists could be held liable if a client posed a threat to someone, even if that someone was not clearly identified! Professional associations and legislative bodies were faced with the challenge of determining how best to respond to this landmark case. Because of the controversial nature of the decision, States have varied significantly in the resulting laws. Many States, including California (where the Tarasoff case took place), and Massachusetts (where I live) passed laws requiring mental health professionals to notify the police and intended victim of a threat (in other words, they established a "duty to warn"). Other states passed laws giving mental health professionals permission to warn, meaning that they could divulge information about a threat without fear of repercussions for violating confidentiality. Some States (such as Texas) only allow the therapist to warn the police (not the intended victim), and some leave it up to the clinical judgment of the therapist.

According to Herbert (2002), "At present, one knows what to do in only half of the jurisdictions in the United States — those that impose a duty to warn. Any 'permission' jurisdiction could, without warning, turn out to be a 'duty' jurisdiction, and in the one-quarter of the states that maintain a legal vacuum on the issue, one can only guess what to do" (p. 423).

Interestingly, New York's newly-passed gun-control law takes the duty to warn in a different direction: it mandates that therapists notify a state official if they believe a client poses a threat to self or others. The client would then be blocked from (legally) purchasing guns, and police would be able to confiscate any guns in the client's possession.

So, what are your thoughts on how therapists should respond to threats against someone's life? Should there be a "duty" to warn, simply "permission" to warn, or neither? Who should be warned: the police, the intended victim, or some other state official? What should the response be?

Does Mental Illness Cause Violent Crime?

Following a month, and year, of multiple mass-shooting incidents around the U.S., questions are being raised about the adequacy of our nation's mental health system. While I'm glad such questions are getting attention, I also want to call attention to the underlying assumption: that those who commit violent acts such as mass shootings must be mentally ill.

It makes sense that people would make such an assumption: we can't imagine how anyone could commit such acts, so we assume there must be something very wrong with those who do. However, is there evidence to support this assumption?

According to a National Institutes of Justice report, past research shows that only about 3% of violent crime can be attributed to mental illness; instead, demographics such as age and gender are better predictors of who perpetrates violence. Indeed, instead of perpetrating crimes, individuals with mental illnesses are much more likely to be victims of crime. 

When individuals with mental illnesses were found to be perpetrators of violent crime, the severity their current psychotic symptoms was a factor, rather than their diagnosis, or the presence of mental illness more generally. Specifically, certain kinds of delusion, and substance abuse, seem to be associated with greater risk of violence. The specific delusions that are most likely to lead to violent behavior are delusions of persecution (believing others want to hurt you), thought insertion (believing an outside force is planting thoughts into your mind), and delusions of control (believing an outside force is controlling your thoughts/feelings/actions). 

I don't know about you, but I can understand how someone experiencing these kinds of delusions may feel like violence is necessary to their self-preservation. I can also understand how such delusions may lead someone to commit suicide after violent behavior: they may feel like the only way to stop the outside force from inserting thoughts, controlling, and/or harming them is through death - particularly if the symptoms are not alleviated by the violence (which seems likely); alternately, they may have some insight into what they've done and feel unable to live with that knowledge. 

Community-based mental health services may help to prevent the severity of psychosis that may trigger violence, by encouraging treatment/medication compliance, monitoring mental status, and intervening when delusions appear. However, people are also more likely to begin refusing treatment and medication, and may withdraw from services, when psychotic symptoms occur. As a society, we highly value personal freedom and self-determination; it is very hard and very messy to attempt to infringe upon anyone's freedom - as it should be. It also becomes challenging to provide for the safety of community-based mental health workers when clients are actively psychotic or substance abusing.

While questions about the mental health system are important, it's also important not to scapegoat individuals with mental illness, when the majority of violent crime is not associated with mental illness. Instead, attention should be paid to identifying and responding to the other 97% of violent crime.

Risk Assessment

If there is one thing we need to be unfailingly conscientious about as therapists, it is risk assessment. While it is of course not possible to prevent all harm to or by our clients, nobody wants to be second-guessing themselves (or have others second-guessing) about whether we did everything we could to ensure our clients' safety. Therefore, it's good to have a system in place for assessing risk. At the minimum, risk should be assessed at the beginning and end of treatment, periodically throughout (e.g., when doing treatment plan updates), and whenever there is a change in a client's mental status.

It takes some practice to get comfortable asking about and talking about risk issues. I made the following "cheat sheet" for my students, and decided to share it here as well. It includes questions to ask to assess five categories of risk: suicidality, non-suicidal self-injury, homicidality, hallucinations, and substance abuse.

Of course, these are just my ways of approaching assessment. Different clinicians may have different styles, and therefore use different wording. If there are things that you've found particularly effective, or if there are other mnemonics you use to remember what to ask, please share a comment below!

Suicidality

• Assess past suicidality
• “Have you ever thought about suicide?”
• “When was that?”
• “What was going on in your life that led you to consider suicide?”
• “Did you ever act on those thoughts?”
• When
• What method
• Did it lead to medical or psychiatric hospitalization?
• “Has anyone in your family ever attempted suicide?”
• Assess current suicidality?
• “Are you having any thoughts of suicide now?”
• “When did you have those thoughts most recently?”
• “When did the thoughts start?”
• “How frequent are they?”
• “Do they feel intrusive or obsessive? Can you stop them or control them?"
• Passive SI = Wanting to die without planning to take action
• Active SI = Plan or intent to end one’s own life
• Assess for a Plan: “Do you have a plan for what you would do?” (acronym: SLAP; P=plan)
• Specificity
• Lethality
• Violent/irreversible methods are most lethal
• Likelihood of rescue
• Accessibility
• Assess Intent
• “Do you think you would act on it?”
• “What will you do if the thoughts become more intense, and you’re feeling more likely to act on them?” (is the patient able to convincingly contract for safety)
• Assess Meaning/Motivation: “What seems appealing about dying?”
• Relief (from distress, psychological or physical pain, unbearable circumstances)
• Revenge (to hurt/punish someone else, or punish oneself)
• Rebirth or Reunion (to escape the current unsatisfactory life through rebirth into a better life; reunion with others who have died)
• High Risk BehaviorPreparation – obtaining supplies to implement plan, writing a suicide note, arranging finances/giving away belongings
• Rehearsal – going through the motions, starting the plan but stopping before the lethal gestureLosses or anticipated losses of relationships or reason for living
• Mood change, including marked brightening of mood

Non-Suicidal Self-Injury

• “Have you ever thought about other ways of hurting yourself or causing yourself pain, like cutting or burning yourself?”
• “Have you ever acted on those thoughts?”
• When, and when most recently
• What did you do
• How often
• Assess potential lethality
• “Did you ever hurt yourself badly enough to need medical attention?”
• Location on body
• Severity (e.g., depth of cut, proximity to blood vessels)
• Assess whether self-injury is:
• Compulsive
• Habitual and repetitive
• Ego-dystonic
• E.g., hair pulling, skin picking, severe nail biting, purging
• Impulsive
• Episodic
• Gratifying
• A reaction to events
• Ego-syntonic
• E.g., cutting, burning, suicide attempts, substance abuse, including laxative/diuretic

Homicidality

• “Have you ever thought about harming someone else?”
• “Have you ever gotten into a physical fight or attacked someone?”
• When, and when most recently
• Frequency
• How
• “Did the other person require medical attention?”
• “Have you ever been arrested?”
• “Have you ever thought about killing someone?”
• Past homicidal ideation:
• When, and when most recently
• “Have you ever acted on those thoughts?”
• How
• “Did the other person require medical attention?”
• Outcome (legal, social, psychological)
• Current homicidal ideation: “Are you having thoughts like that now?”
• Toward whom
• Assess plan (SLAP), intent and ability to contract for safety
• Duty to Warn (Tarasoff): providers must notify the intended victim if s/he is in imminent danger

Hallucinations

• “Have you ever seen or heard something other people couldn’t see or hear?”
• What
• When
• How often
• Is it persistent, intrusive, obsessional
• Assess whether it is a thought or a perceptual disturbance
• “Does the voice seem like it’s inside or outside of your head?”
• “Is it more like picturing something, or seeing a picture?”
• Content
• “Is it one voice, or multiple voices? What does it say?”
• Are there Command Auditory Hallucinations (CAH)?
• Do the voices tell the patient to do something?
• Do the voices command the person to harm self or others?
• Are the visual hallucinations of a violent or scary nature?
• Is the content mood congruent or incongruent?

Substance Use/Abuse

• “Do you currently use any kind of drugs or alcohol?”
• What, how often, how much
• How (method, e.g. smoking, snorting, IV)
• Any negative consequences
• “Have you used any (or any additional) substances in the past?”
• What, how, how often, how much
• Any negative consequences
• “What made you decide to stop using this/these substance(s)?”
• Assess for substance abuse/dependence
• Purpose: “What does (the substance) do for you?” (e.g., social use; lower inhibitions; numbness; euphoria )
• CAGE:
• “Have you ever felt you should CUT DOWN on your use?”
• “Have you ever felt ANNOYED by other people criticizing your use?”
• “Have you ever felt GUILTY about your use?”
• “Have you ever used first thing in the morning (to get over a hangover)?” (EYEOPENER)
• Tolerance: “Have you had to use progressively more of the substance to get the same result?
• Withdrawal: “Have you experienced any kind of side effects when you stopped or cut back on your use?”
• “Have you ever needed detox or other treatment for substance abuse?”

When is a Crisis Evaluation More Than a Crisis Evaluation?

One of my clients, who has OCD, sometimes has intrusive thoughts (i.e., obsessions) about hurting himself or someone else. In the past, he has acted on compulsions to non-suicidal self-injury, but never harmed anyone else, nor attempted suicide. Most of the time, he has the thoughts without acting on them.

Instead, he tends to present to emergency rooms for crisis evaluation and level of care assessment. He went through a period a few years ago, before I met him, of frequent hospitalizations. Then he transitioned to frequent partial hospitalization program admissions. Now, he's usually just referred back to his outpatient providers.

In spite of the steady decrease in his rate of admission to higher levels of care, he's  been getting evaluations with increased frequency over the last few months, creating something of a puzzle for me: If these evaluations are not about being admitted to a higher level of care, then what are they about?

I have a few theories on the matter, or course. For one thing, I have been somewhat less available during this time, with holidays and a new full-time job, so the role of his attachment to me (and related transference) in helping him feel safe could be a factor. He declined a referral to a therapist with more availability (again likely due to attachment), but perhaps occasionally needs a "booster" intervention, which he seeks out via crisis evaluation.

Whether or not my own availability has played a role, it seems clear that the crisis evaluation plays some sort of symbolic role for him. It's not a specific crisis service or clinician, since he presents at various local emergency/crisis locations, so it must be the process itself. One possibility is that the crisis evaluation serves as a kind of transitional object representing the treatment relationship with me and his psychopharmacologist.

Another possibility is that the crisis evaluation itself has become part of the compulsion - perhaps a kind of sublimation of the initial compulsion to harm self/others. In other words, the compulsion to harm is diffused by acting on the compulsion to seek crisis services. Alternatively, getting a crisis evaluation may also be a kind of "safety behavior" diffusing the anxiety in another way: he doesn't trust himself to 1) stay safe, and/or 2) know whether or not he can stay safe. Having a clinician and doctor tell him he's safe to go home provides the reassurance he needs to actually go home and be safe.

Normally, the treatment of choice with safety behaviors is exposure and response prevention - however, the risks of that treatment are considerably higher when the compulsion is about risk. While I'm pretty sure someone can not wash their hands and still come out in one piece, interrupting this safety behavior could result in harm to self or others. I'd rather he get the evaluation than act on a dangerous compulsion!

Instead, I'm thinking that harm-reduction may be the way to go. Since he has needed less and less intensive treatment over time, continuing the work will hopefully lead to less need for evaluations as he trusts himself more to stay safe. If anyone has another idea, though, I'm all ears!

All Therapists are Gamblers

I've never been to a casino, or bought a lottery ticket, but I still gamble. When I work with high-risk clients, I take a gamble every time they walk out the door. I like to think it's not much of a risk, because I'm experienced enough to sense a client's degree of risk/safety - but we all know we can be wrong, and situations can change rapidly.

Because we (as a profession and wider society) value self-determination and are loathe to infringe on anyone's free choice, the threshold is quite high for holding someone against their will on psychiatric grounds. Unless someone is immediately suicidal or homicidal (like, they have a plan and intend to enact it if we let them walk out the door), or behaving so bizarrely that they are at imminent risk of physical harm (wandering down the middle of the street; refusing all food and drink - things like that), they don't have to go to the hospital...and we, by extension, do have to let them walk out the door.

For most clients, most of the time, that's not much of a gamble. But there is a definite gray area where you know the client is at risk, but not enough so to override their self-determination. Sometimes, these clients can be persuaded to be hospitalized voluntarily - but often they refuse.

I find myself with several clients who are hanging out in the gray area, making me a little nervous as they walk out my door.

I have a psychotic client who has reportedly spent most of the week hiding out in her closet, eating only sporadically, and once going to the bathroom in a bucket..."but I only did that one time, Natalie!" She absolutely refused all efforts to convince her to go to the hospital, assuring me her family would be checking on her regularly, and promising to go to the hospital if they found her hiding in the closet again.

I have another psychotic client who missed his last appointment with me (though thankfully he did go to his med evaluation). He has to leave stores and other social settings sometimes because the voices tell him to hurt people.

Another client also keeps to herself because she has impulses to hurt people. She is able to choose not to act on these impulses by thinking about the consequences for herself and her children. But she hopes to someday main or kill the people who sexually assaulted her in the past (including a brother) and does not plan to continue with therapy to work on any of these issues.

Today I saw a woman who has habitual suicidal ideation which she does not intend to act on - but it has gotten markedly worse this week, after a loss and a break-up, and she decided on a method long-side (OD on IV heroin). Even though she assures me she has not intention of acting, I am not terribly comforted by that. She did agree to go to a partial hospital program...but can't start until Tuesday.

Finally, I have a client with OCD who has intrusive, obsessional violent thoughts toward himself and others. These include sexual aggression against children, homicidal ideation, urges to self-injure (because he considers the thoughts of violence toward others signs of his own badness), and thoughts/plans about harming his mother's cats. My worry is assuaged by the fact that he has never tried to harm any other people...but he has engaged in non-suicidal self-injury, and attacked (but not tried to kill) a cat.

So, with all of these clients, I let them walk out the door with some definite trepidation. I'm mostly sure they won't come to harm before I see them next...mostly. I'm wondering how other people would navigate these kinds of risks. How do you manage safety issues on an outpatient basis, and when do you decide that the risk is too high? When and how do you seek consultation on risk issues? In short, what's your gambling strategy?