Key moments, in order
The majority of completed suicides are not from patients reporting suicidality. Most people who complete suicide have never seen a psychiatrist and usually have never mentioned suicidality at all.
A risk factor means different things in different settings. In the ER, 15 prior attempts sounds very high risk but says little about the patient in front of you.
A checklist done blindly is cargo cult psychiatry. A box that stands in for a complex thinking task becomes a symbol that replaces the thinking. It is still standard of care, so understand why the ritual exists.
First rule: do not be afraid. If you are fearful or anxious about the topic, you will show it to the patient, consciously or not.
Rapid-fire screening questions teach patients that answering the expected way gets them discharged.
Being honest with a provider is key to getting better.
Example: a surgeon asked for a suicidality assessment after a patient, following a painful waking procedure, said something like 'if I have to go through that again, I'm going to kill myself.'
If you are starting out, at least read the Columbia Suicide Severity Rating Scale, a gold standard for collecting the history you need. History is not the same as assessment.
The question is not a binary of suicidal or not. Your job is not to predict the future, because you cannot.
Think like an actuary rating a driver for car insurance: how dangerous is this person compared with others, and compared with their own baseline, not will they crash.
Assess chronic risk first, then acute risk on top of it.
Even when the wish seems rational, the fact that the person is in front of a psychiatrist means the pretest probability is high that it is not purely rational.
Stress-based ideation is a response to a stressor. With enough time or resolution of the stressor, we expect it to go away; it is not a baseline feature for that person.
For stress-based ideation, the short-term involuntary detention model used across the United States is warranted.
The right response depends on the formulation. Hospitalization can be the best treatment for adjustment-type suicidality and the wrong one for others.
Intoxicants lower the ability to cope and stressors raise distress. The combination of the two is where involuntary detainment is extremely applicable.
Some people have, at baseline and without any particular stress, thoughts about not wanting to be alive or wanting to hurt or kill themselves most of the time.
For baseline ideation, the indicated treatment may not be hospital admission, and admission can slow overall progress toward getting better.
Do not think "suicidal or depressed, give an SSRI." None of these varieties respond primarily to an antidepressant. Have a formulation, diagnosis and treatment plan before you prescribe.
When ideation comes from bipolar disorder, don't forget lithium; in schizophrenia, if the patient is appropriate for it, don't forget clozapine. They are the two medications with evidence of actually reducing suicidality.
SSRIs have not been shown to specifically reduce suicidality, though they are much safer, with far fewer side effects, than older antidepressants.
Suicidal thoughts can be part of OCD: properly, an ego-dystonic obsession with matching compulsions.
A subset looks like OCD but does not respond well to ERP and SSRIs, because the underlying problem is self and other relating. Suicidality better formulated through borderline is now being recast as ego-dystonic OCD obsessions.
Sum the short-term and long-term risk and protective factors, match them to the kind (or kinds) of suicidal ideation this patient has, then decide on a treatment plan.
Do not catch a whiff of borderline and explain everything else away. Patients with BPD are at elevated risk of completed suicide.
Acute suicidal crisis has five features, and entrapment (feeling trapped, hopeless or stuck) is probably the most important thing you can see in a patient.
Hyperarousal means being irritable and activated. That is why bipolar and agitated depressions are the most dangerous treatable conditions for suicide.
Watch for any patient who evokes a strong pull to rescue or over-involve yourself, or the opposite, to detach and give up on them.
Intense countertransference of any kind is a sign about self and interpersonal functioning, and under distress that can be quite dangerous.
If a patient is not at their usual level of risk, consider changing the plan.
Most states punish the doctor for being lax rather than strict, so you will not go wrong detaining and letting a judge release.
You are liable for a process, not an outcome. Do the assessment within the standard of care and document it.
Escalate care when you see a change from the patient's baseline, or, if you do not know the baseline, when they will not contract for safety or have been planning and preparing.
Each moment is a point made in the episode, linked to the second it was said. The topic links open everything the podcast has said on that topic.