Documenting risk: SI, HI, and safety planning in the note
Documenting suicidal ideation well is one of the highest-stakes pieces of clinical writing you do, and it is also one of the most commonly rushed. When a session surfaces thoughts of death, self-harm, or harm to others, the note has to do two jobs at once: capture what actually happened with enough fidelity to guide the next clinical decision, and stand up later if anyone — a board, an attorney, a treating colleague, a payer — needs to understand what you knew and why you acted as you did. A vague “denies SI/HI” check-box does neither. This post is about what to put in the note when risk is on the table, and why specificity is your friend.
A quick caveat before the detail: risk-documentation standards vary by jurisdiction, license, and payer, and this is editorial guidance rather than legal advice. Confirm specifics with your board, your liability carrier, or an attorney who knows your state.
Why boilerplate fails when it matters most
The phrase “client denies SI/HI” appears in countless notes. The problem is not that it is false — it is that it is empty. It records a conclusion without the clinical reasoning that produced it. If a client dies by suicide weeks after a “denies SI” note, that single phrase tells a reviewer nothing about whether you screened, what you asked, what the client said, or how you weighed it.
Defensible documentation is not about predicting suicide, which no clinician can reliably do. It is about showing that you assessed thoughtfully, reasoned transparently, and responded proportionately to what you found. Reviewers are not looking for a clinician who guessed right. They are looking for a clinician whose process was sound. Specificity is what makes that process visible on the page.
Documenting suicidal ideation: what actually belongs in the note
When risk appears, your note should capture more than a yes or no. The goal is to reconstruct the clinical picture months later from the text alone.
- Screening and how you asked. Note that you assessed and, ideally, how — open-ended inquiry, a structured tool (for example, the C-SSRS or PHQ-9 item 9), or both. Record the result, not just that you “screened.”
- The ideation itself, in specifics. Passive versus active. Frequency, intensity, duration. Presence or absence of a plan, intent, means, and access to means. A note that says “reports passive thoughts of ‘not wanting to wake up,’ two to three times this week, no plan, no intent, denies access to firearms” carries far more weight than a bare “denies SI.”
- Risk factors, stated as present or absent. Prior attempts, recent losses, substance use, acute stressors, command hallucinations, hopelessness. Documenting their absence is as valuable as noting their presence — it shows you looked.
- Protective factors. Reasons for living, social support, engagement in treatment, future orientation, religious or cultural deterrents, dependents. These are central to your judgment and are frequently omitted.
- Client quotes, sparingly. A short verbatim phrase (“‘I think about it but I’d never do it because of my kids’”) grounds the note in the session and is hard to argue with later.
- For HI: the target (identifiable or not), whether a duty-to-protect or duty-to-warn consideration arises, and what you did about it. Tarasoff-type obligations differ by state, so this is an area to confirm with your own board or counsel rather than to treat as settled; document your reasoning regardless.
Make your clinical judgment explicit
This is the step most often skipped, and it is the one that holds the note together. After laying out the data, state your risk formulation in plain language: your estimate of acute and chronic risk, and the reasoning that connects the factors above to that estimate. “Given passive ideation without plan or intent, strong protective factors including childcare responsibilities and active treatment engagement, and no recent change in stressors, acute risk is assessed as low; chronic risk moderate given prior depressive episodes.”
That single sentence does what boilerplate never can: it shows a mind at work. It also makes your next decision — to safety plan rather than hospitalize, say — read as the considered result of a process rather than an arbitrary call. If you want a fuller treatment of how to write reasoning that connects data to conclusions, the approach mirrors what good clinical assessment writing looks like generally; see writing the assessment section.
The safety plan belongs in the record
If risk warrants a safety plan, the plan itself — or a clear reference to it — should live in the note. A defensible entry captures that a collaborative safety plan was developed, what it included, and that the client engaged with it.
| Element | What to record |
|---|---|
| Warning signs | Client’s own identified triggers and early cues |
| Internal coping | Strategies the client can use alone |
| Social contacts and settings | People and places that provide distraction or support |
| People to ask for help | Named supports the client agreed to contact |
| Professional and crisis resources | Crisis line, your protocol, local emergency options |
| Means restriction | Steps discussed to reduce access to lethal means |
Note the client’s level of engagement and any agreement reached. Be cautious about leaning on a “no-harm contract” or “no-suicide contract” as your safety mechanism — the clinical literature gives these little support, and they can read as box-checking. A collaborative safety plan, documented as collaborative, tends to carry more clinical and defensive weight.
A note on tone and tools
Write risk sections in neutral, behavioral, observation-anchored language. Document what the client reported and what you observed, then your reasoning, then your response. Keep follow-up explicit: increased session frequency, a check-in, coordination with a prescriber, the threshold that would change your disposition.
If you work from a structured format, a consistent risk block inside your SOAP note template makes it far less likely you skip an element under time pressure. The point of a template here is not to produce language for you — risk language must always reflect this client in this session — but to prompt the right questions every time. A tool that produces a draft can help you get the session down faster, but the formulation, the judgment, and the signature are, and must remain, yours. A draft you do not read and own is a liability, not a shortcut. CouchNotes produces a draft you review, edit, and sign; the clinical reasoning on a risk note is exactly the part you should never hand off.
When you read back a risk note weeks later and it tells you what you knew, what you weighed, and what you did — without your having to remember the session — it is doing its job. That is the standard worth holding: not a note that protects you because it is defensive, but one that protects you because it is honest, specific, and shows the care you actually took. Documenting risk well takes a few more minutes in the moment, and those minutes are among the best-spent of your clinical week. The client in front of you is the reason the note exists; the reviewer who may one day read it is only the reason it has to be clear. Write for both, in that order, and the rest tends to follow.