Keisha worked at a community mental health center serving a predominantly Black and Latino neighborhood. The agency’s records system came with documentation templates—dropdown menus built to keep charting consistent across providers. Most of the fields were fine. One kept bothering her.

The template had a section called “Barriers to Treatment” with pre-populated checkboxes: “Lacks motivation.” “Poor insight.” “Noncompliant with recommendations.” She was supposed to check all that applied.

Her client, Mr. Rodriguez, was a 52-year-old man referred after a hospitalization for uncontrolled diabetes. He’d missed two appointments. When he did come, he was skeptical about the whole mental health piece—he didn’t think talking would help his diabetes, and he wasn’t interested in antidepressants. The template was prompting Keisha to mark “Lacks motivation,” “Poor insight,” and “Noncompliant.”

But she knew more than the template allowed. Mr. Rodriguez worked two jobs and had missed appointments because his shift schedule changed. He was wary of mental health treatment because his last therapist had brushed off his concerns about side effects. He’d declined antidepressants because he was already taking seven medications and felt like doctors kept adding pills instead of listening. He didn’t lack motivation. He was exhausted, and he felt like the system kept failing him.

The template made him sound like a difficult patient. The reality was more complicated—and Keisha had to find a way to document it without using language that would trail Mr. Rodriguez through every future healthcare encounter.

Observable Behavior vs. Clinical Interpretation

When you write “Client was hostile,” you’re offering an interpretation. Hostile according to whom? Based on what standard? One provider’s “hostile” is another’s “assertive,” or “appropriately angry,” or “activated trauma response.”

When you write “Client raised their voice, stood up abruptly, and said ‘This is bullshit, I’m done talking about this,’” you’re describing behavior. Someone else can read that and form their own clinical impression. You’ve also captured what actually happened—which helps explain whatever you chose to do next.

Here’s a quick test. Write a sentence, then ask yourself, “How do I know that?” If your answer is “It’s obvious” or “Anyone could see it,” you’re probably writing interpretation rather than observation. Push yourself to name what made it obvious. What specifically did you see or hear that led you there?

Your interpretation still matters—it guides your decisions. It just needs to be marked clearly as your professional judgment, not handed to the next reader as objective fact.

Three Ways to Document the Same Session

Maria came in for her third session. She arrived fifteen minutes late, didn’t apologize, sat down heavily, and said, “I don’t even know why I’m here. Nothing’s getting better. This is a waste of time.” When you asked about the thought log you’d assigned, she said she hadn’t done it. You explored what got in the way, and she said everything felt pointless. Her affect was flat, eye contact was minimal, and she answered most questions in one or two words. Near the end, you asked whether she wanted to continue therapy. She said, “I guess.”

Here’s that same session, written three ways.

Judgment-heavy: “Client arrived late without explanation and presented with hostile, resistant attitude. Client refused to complete assigned homework and demonstrated poor motivation throughout session. Client lacks insight into treatment process and shows minimal engagement. Prognosis is poor if client continues to resist treatment recommendations.”

Description-light: “Client arrived for session. Discussed progress and challenges. Client continues to experience low mood. Reviewed homework and discussed barriers. Will continue with current treatment approach.”

Balanced and specific: “Client arrived 15 minutes late (consistent with previous session delays). When asked about the week, client stated ‘Nothing’s getting better’ and questioned the value of continuing therapy. Client did not complete thought log homework, stating ‘everything felt pointless.’ Affect was notably flat with minimal eye contact and brief responses throughout. When I explored what might make therapy more helpful, client said she didn’t know. Asked directly about continuing treatment, client agreed to return next week. Depression symptoms appear to be interfering with treatment engagement. Will reassess suicide risk at next session and consider whether the current approach is meeting client’s needs.”

The first version labels the person and closes the door. The second says almost nothing—it’s so sanitized that a future provider learns nothing useful about what’s happening. The third documents real concern without blaming Maria for her symptoms. She isn’t “resistant.” She’s depressed, and her depression is showing up in how she’s engaging with therapy. That’s clinical information the next reader can actually use.

Notes You Could Read Aloud

Here’s a practice some social workers swear by. Before you finalize a note, imagine reading it aloud to the client. Not because you necessarily will—because it forces a question: would you feel comfortable doing it?

If not, why not? Sometimes the answer is that you’ve written something clinically necessary but hard—a risk assessment that documents serious concern. That’s fine; that’s the job. But sometimes the answer is that you’ve been sloppy, or judgmental, or you reached for a label because it was faster than describing what you saw.

You don’t need to soften difficult content. You need to write with respect. The person you’re documenting has the legal right to read most of what you write, and even when they never do, your words shape how every provider who comes after you understands them.

Keisha eventually skipped the dropdown entirely and wrote a narrative note for Mr. Rodriguez—one that included his perspective, his work schedule, his history with dismissive providers. It took longer than clicking checkboxes. But it was accurate, and it didn’t hand the next clinician a caricature.

The words you choose today might echo through someone’s life for decades. We go deeper on bias, cultural humility, and the specific terms that quietly do the most damage—”noncompliant,” drug-seeking,” “poor historian,” and more—in the full 3-CE course, Ethics in Documentation and Record Keeping, at SWTP CEUs, from which this is drawn.