Part of the Justice-Involved Treatment Mastery Series
The first probation officer I ever met was waiting for me in a hallway, holding a clipboard, looking mildly unsure about the fact that I existed.
I was probably in my first six months of clinical work. All new, all over the place. I had just finished eating an orange before the meeting — the kind of thing you remember years later because your hands were still a little sticky when you reached out to shake his. He looked at my hand, paused, and slowly put his out.
He had no idea what to make of me.
I had no idea what to make of myself yet either. I was in a role where the stakes were real — his clients were also my clients, and my letters were going to end up on his desk, and the decisions I made were going to affect whether someone went to jail or didn’t. But I was twenty-something and I had orange on my fingers and I didn’t know yet that this specific probation officer would end up trusting me for the next decade of my career.
What I learned in the years after that handshake is the thing this article is about. Ethics in justice-involved treatment isn’t made in courtroom testimony or dramatic refusals. It’s made in small, quiet moments where every ethical decision I’ve ever made has been shaped — in real time, in the room, figuring out who I was going to be.
This is the finale of an 8-month series on justice-involved treatment. Over those months we’ve covered harm reduction with mandated clients, the court-clinician communication problem, therapeutic alliance, documentation, motivational interviewing, managing relapse, and navigating a trauma-informed system that often isn’t. Every one of those articles had ethical friction running through it — friction between what the client needed and what the system demanded.
This article is about that friction itself. What it looks like. How you navigate it. And what you can trust to guide you when the answer isn’t obvious.
Part 1: Ethics in Justice-Involved Treatment — The Collisions You’ll Face
Ethics in this work isn’t usually dramatic. It’s not a single courtroom moment where you have to choose between the right thing and the easy thing. It’s a slow accumulation of small decisions, most of them quiet, most of them made before anyone else in the room even realizes a decision is being made.
Here are five of the collisions I’ve lived with.
1. What You Hand Over by Default
Nobody ever asked me to hand over more than I should. That’s not the shape this one takes.
The shape it actually takes is this: you inherit a workflow. You’re told — by a supervisor, a colleague, an agency, a grant requirement — that when a probation officer requests documentation, you send the mental health assessment. That’s how it’s done. That’s what’s expected.
So you do it. Until one day you’re writing an assessment, putting in direct quotes from your client — disclosures about use, criminal history, the things they’ve told you in the room because they trusted the room — and you realize what you’re about to do.
You’re about to hand that piece of paper to someone outside the therapeutic relationship.
I was that clinician. I sat with that discomfort for a while. Then I did what I always do when something doesn’t sit right: I couldn’t let it go until I figured it out. I talked it through with a colleague I trusted. I thought about what probation actually needed from me versus what I had been volunteering.
There’s actual legal backing for pausing here, even though I didn’t know the citation by heart at the time. 42 CFR Part 2— the federal regulation specifically protecting substance use disorder patient records — exists precisely because disclosures made in treatment can and have been used against patients in criminal, administrative, and family proceedings. The ACA Code of Ethics, Section B, says the same thing in different language: counselors “disclose information only to the extent necessary” and “only when essential information is revealed.” The workflow I had inherited wasn’t following either of those principles. It was just… habit.
I drafted a new standard — a detailed letter that would give probation what they legitimately needed without exposing quotes that belonged in the therapeutic context.
Then I just… started doing it that way. No announcement. No explanation. I never got pushback because I never flagged the change. The letters worked. Probation got what they needed. My clients’ words stayed in the room where they were spoken.
Years later, I did the opposite. When I wanted to formalize the practice — to create actual forms for admission, monthly updates, and discharge that other clinicians on my team could use too — I called a meeting. I asked for approval. I invited conversation. Because that wasn’t a personal practice change anymore. That was a system change. (Article 4 in this seriesgoes deep on the documentation philosophy that eventually became those forms.)
The ethical move I didn’t have language for at the time: knowing when to change your own practice quietly and when a change needs a conversation. If what you’re changing only affects what you put on paper, it’s yours to change. If it affects the workflow across a team or an agency, it needs a meeting. Confusing those two categories is how new clinicians either get bogged down asking permission for everything or blow up systems they didn’t realize they were affecting.
Most ethical questions in this work don’t start with a clear violation. They start with you noticing that a routine practice makes you uncomfortable — and deciding that the discomfort is worth investigating.
2. Who Are You Accountable To?
In my first years, I had an abstinence voice in my head that wasn’t mine. It was the system’s. Whenever a client slipped, I felt the dread in my body before I felt anything clinical. Now I have to call the court. Now everything is going to fall apart. I was structured and stern. I was aligned with the posture the system wanted me to have.
I’m honest about this because early-career me would have made decisions I wouldn’t make now. I discharged clients I would keep today. I jumped to the call when I should have slowed down. If a district attorney had questioned my qualifications in my first year the way one did a decade later — I would have folded. I would have changed my recommendation. I wasn’t wrong to be cautious then, but I wasn’t grounded enough to hold my ground when the system’s voice got loud.
The shift didn’t come from a training. It came from time, and from a specific practice.
I started offering to support clients as they made those calls. Not to make the calls for them — to be there while they did it. When a client had to notify their PO about a slip or a positive drug test, I’d invite them to use my office, put the call on speakerphone with me sitting next to them, and help them initiate the conversation. It’s better when it comes from them, with a plan, than from an email or a letter after the fact. They didn’t have to accept the offer — but most did. Beforehand, I’d walk through a simple structure with them so we could stay focused on the call: what happened, what it means, where we go from here. Not a script. Just a way to keep the conversation from spiraling. I’d set the stage at the start of the call, help the client manage their catastrophic thinking, and hold the space through the conversation. (Article 3 on therapeutic alliance covers more of this kind of alliance-building work.)
I also — more rarely — invited POs to meet with me and the client together, to talk through what the client actually needed. What changed things in those meetings wasn’t me. It was the client. Coming in honest. Coming in with a plan already worked out with me. Showing up with some confidence in what they were accountable to. When the client does the work in front of the PO, the whole room shifts.
Over time, specific POs learned to trust me. A few of them moved up, started running things, and when new POs came into those offices, my earlier relationships vouched for me. I didn’t have to prove myself from scratch with every new officer. The relationships compounded.
Somewhere in that process, I stopped treating the court as my second client. I started embracing harm reduction — really embracing it, not just saying the word — and discovered leniency and recovery rather than accountability and punishment.
My client was always my client. I just had to stop letting the court’s voice be louder than hers in my own head.
3. When Your Judgment Conflicts With the Court’s Desired Outcome
Most pushback from the system isn’t dramatic. It’s subtle, relational, and comes from people you’ve already worked with.
The Dramatic Version
The dramatic version does happen. I had a district attorney question my qualifications in open court when I didn’t recommend a higher level of care. I held my ground that time because I’d been doing the work for years and I trusted my clinical reasoning. But I want to be honest: that only worked because I had the experience to back it up. Earlier in my career, the same challenge would have changed my mind. (Article 6 on managing relapse goes deeper on the specific ethics of when to report and when to hold, drawing on SAMHSA TIP 44 and nonmaleficence principles.)
The Everyday Version
The everyday version is different. It’s a trusted PO — someone you have a good working relationship with — sending you an email that says this is happening all the time or asking you in passing, half-joking, aren’t we all getting tired of this?The frustration isn’t hostile. It’s worn down. And that’s harder in some ways because the person putting pressure on you isn’t wrong to be frustrated. The client IS still struggling. The system IS exhausted.
Here’s how I learned to handle it. First, I validate. I don’t dismiss the frustration — it’s real and they’ve been carrying this case too. Then I give my two cents, which usually sounds something like: I hear you, and here’s why I’m still holding off. Here’s what I’m watching for. Here’s what I think has to happen before I’m ready to recommend that.
And then I actually let their frustration sit with me. Not to cave — to check myself. Because sometimes the PO is pointing at something real that I’ve been missing. Their frustration becomes data. That’s not a special move. It’s what holistic, collaborative clinical practice looks like. When I make a decision, I take everyone’s perspective into account. Holding my ground doesn’t mean refusing to re-examine. It means re-examining without abandoning my reasoning just because someone else is tired.
The Consultation Running Both Ways
One more thing I’ve come to realize: the consultation runs in both directions. That PO who’s on the phone with me? They’re also consulting their fellow officers. What do you guys think of Stephanie? Is she really serious? Does she really think that? I’m doing the same thing with my colleague — you wouldn’t believe what they said. Both of us are using our trusted peers to pressure-test our own frustration. Which means the trust that gets built between us over time isn’t one decision. It’s dozens of small moments, both of us getting reality-checked in the background.
That’s what makes the clinical ground holdable over time. Not a single confrontation. A long track record, consulted on both sides, that earns the weight your reasoning needs.
4. When to Terminate Treatment
Termination is one of the hardest clinical decisions in this work. And it’s not one dilemma — it’s two.
Two Shapes of Termination Pressure
The first shape: the PO is ready to successfully close the case so they can get the client off their caseload. The client has done well, technically speaking. And you know what I know: the relapse rates are real, co-occurring issues are common, and clients who struggle often end up back in the system for something else eventually. You want to hold on. Just a little longer.
The second shape: the PO is frustrated and wants the client discharged because they’re not progressing fast enough. You don’t think the client is ready. You want to keep the work going.
The Self-Check
Both shapes require the same self-check. Before I do anything else, I ask myself: Am I holding onto this kid? Not “is the system right?” or “is my gut right?” — something more honest than that. Are my reasons for wanting to continue about the client, or about me?
Because here’s the truth: sometimes we hold on because we’ve invested in the case and want a good outcome, or because we’re afraid of what happens when they go back out, or because the relationship has started to matter to us. Some of those are legitimate clinical reasons. Some of them are countertransference. The question is whether I can tell them apart in a given moment — and whether I’m willing to be honest with myself about what I find.
If the answer is yes, I’m holding on because it’s me — I let them go. Discharge. And if they come back, they come back. That’s not failure. That’s recovery being non-linear.
If the answer is no, my reasons are clinical — I push back. I articulate my hesitation. I give my reasoning.
When the Protective Impulse Is the PO’s
Early in my career, I had a client whose PO was pushing back on my recommendation for successful completion. She wanted me to keep him in treatment longer. She challenged me — you know his experience, he was using needles, you know what he’s been through. And she wasn’t wrong to worry. But I remember saying: I can’t protect him from everything. If he’s doing what he needs to do, I’m not going to keep him from his life.
That was the first time I understood that autonomy means something specific in justice-involved work. The system’s protective impulse — even when it comes from a caring PO who genuinely wants the best for the client — can turn into paternalism if we’re not careful. Clients have often been under system surveillance for years. The ethical move isn’t always more treatment. Sometimes it’s you’ve done the work — go live your life, and if you need to come back, come back.
The Tension Worth Naming
The tension I want to name honestly: research supports longer treatment duration. NIDA’s Principles of Effective Treatment notes that most people with substance use disorders need at least 3 months in treatment to significantly reduce use, and that the best outcomes occur with longer durations. Clinicians who hold on do so partly because the statistics are on their side. And liability is real — the “what if something happens” voice in the back of our heads is the same voice that causes some clinicians to hospitalize for suicidal ideation when it isn’t clinically indicated. Fear of the worst-case outcome drives decisions that look clinical but are actually defensive.
So the discernment matters. “Am I holding onto this kid?” isn’t a rule that says let go every time. It’s a tool that helps you tell the difference between clinical reasoning and your own fear dressed up as clinical reasoning.
The real work doesn’t start until they are out there, living their life authentically. Our job is to prepare them for that — and then, eventually, to get out of their way.
5. When Clients Ask You to Shade the Truth
Probation officers almost never ask me to write something I can’t stand behind.
Clients do. Regularly.
Can you just recommend I meet with you once or twice? Can you just write down that I met with you and that was enough? Can you just tell them I came and we didn’t actually do the assessment? Can you not tell the probation officer about my latest drug test? Can you tell them I stayed the whole time even though I left early? Can you leave out that part?Can you not mention this?
Every one of those requests is a small ethical test. And most of them are coming from one of two places: fear or manipulation.
When it’s fear, the client is scared of what accurate documentation will trigger. They’re not trying to con you — they’re trying to survive a system that has punished them for truth before. When it’s manipulation, the client genuinely wants you to produce false documentation and is testing whether you’ll do it.
The problem is you can’t always tell which one you’re dealing with by the request itself. So I use a move that reveals it. I call it the diagnostic compromise.
It sounds something like this in session:
“I can see your concern.”
“I could possibly say [softer framing] — but is that really better?”
“Or I could say this — but that doesn’t really get the outcome you wanted, I’m hearing.”
“It sounds like this was the safest way to say it.”
I’m not saying no. I’m offering middle-ground versions and asking the client to evaluate them with me. I’m making them do the thinking.
Then one of two things happens.
If the client accepts the softer version — okay, the request was coming from fear. They wanted help framing something hard, not hiding it. I write the softer version. Everyone’s needs are met.
If the client says “no, that’s worse” — now I know. They weren’t trying to edit toward truth. They were trying to edit toward fiction. At that point I let go of the compromise exercise and go back to my original clinical thoughts. I write what I can stand behind. The conversation isn’t adversarial because we’ve walked through the thinking together, and the client has had to articulate — to themselves and to me — that what they actually wanted wasn’t a softer truth. It was a different reality.
I won’t write something I’m not in agreement with. That’s the hard line. But the how you get to that line matters. If you just say no, you’ve made it a power struggle. If you offer the compromise and let the client reveal their motive, you’ve kept the relationship intact and you’ve learned something about what’s really going on underneath.
The Five Collisions at a Glance
Ethics in justice-involved treatment rarely arrives as a clear violation. It usually shows up as one of these.
The framework that runs underneath all five is the same. Part 2 breaks it down.
Part 2: How You Actually Decide
Across all five of those dilemmas, I’m doing the same four things. This is what ethics in justice-involved treatment actually looks like in practice — not a checklist, but a sequence of moves you run whenever something feels off. They’re not written down anywhere. They’re not in my grad school notes. But they’re the moves I run whenever something ethical is at stake — and I think they’re teachable.
Move 1: Notice.
Your discomfort is data.
When a practice, a letter, a recommendation, or a request makes something uneasy in your gut — that’s the signal. Ethics in this work rarely starts with a clear violation. It starts with a feeling that something isn’t sitting right. The clinicians who get in trouble aren’t the ones who feel the discomfort and investigate it. They’re the ones who feel it and override it because the workflow says to proceed.
Pay attention to the moment something makes you pause. That pause is the beginning of ethical practice.
Move 2: Investigate.
Once you’ve noticed, don’t let it go.
I get a little like a dog with a bone when something ethical is bothering me — I have to figure it out. That persistence isn’t neurosis. It’s the second move. Ethical clarity is rarely instant. It comes from staying with the question long enough to see what you’re actually looking at.
If you notice the discomfort and then move on because you’re busy, you’ve bypassed the move. Sit with it. Think about what specifically is off. Name the conflict to yourself in plain language. Investigate until the shape of the problem is clear.
Move 3: Consult.
You don’t decide alone.
Take it to a colleague who knows you well enough to pressure-test your thinking. The goal isn’t permission — it’s a second set of eyes on your instincts. Your colleague might tell you you’re overthinking it. Or they might tell you you’re under-thinking it. Either way, you now have more information than you had.
And here’s something I want new clinicians to understand: the people across the table from you — POs, attorneys, supervisors — are consulting too. They’re asking their trusted colleagues about you in the same way you’re asking yours about them. That’s not weakness. That’s how trust and accuracy get built in any system where the stakes are real. (Article 7 on navigating trauma-informed systems explores this same dynamic from a different angle — the relational work of holding clinical ground in imperfect institutions.)
Move 4: Hold the line, but stay open.
Once you’ve landed somewhere, hold your ground.
Validate what the system is experiencing. Give your clinical reasoning. Let them know where you are and why.
And then — quietly, genuinely — consider whether they might be right.
Ethical work in justice-involved settings isn’t rigid. It’s anchored AND open. You can hold your position with confidence while simultaneously letting the system’s pushback be data. Sometimes they’re seeing something you’ve missed. Most of the time they’re not, but you’ll never know which is which if you don’t stay open to the possibility.
Part 3: The End of a Series
This is the eighth and final article.
Eight months. Eight articles. A lot of clinicians walking this path with me — some of you for years, some of you brand new to justice-involved work, all of you trying to do this well in a system that often doesn’t make it easy.
Here’s what I most want you to take away from all of it:
There isn’t always one way. The system pushes for one — and that one has a long history and a lot of institutional weight behind it. But look at how many places this system is visibly broken. Its authority to dictate the “right way” is questionable. That doesn’t make the system bad. It makes it imperfect, like every system ever built. Which means there’s room — real, ethical, defensible room — to practice differently.
Don’t be afraid to try something new. Harm reduction. Leniency over punishment. Autonomy over paternalism. Clinical judgment over reflexive compliance. You don’t have to reinvent everything on day one. You can change one practice at a time, quietly, over years.
As long as you’re involving others, consulting the people who know this work, moving slowly and methodically — I don’t see why you can’t work this way too.
Ethics in justice-involved treatment is rarely clean. It’s made in small, quiet moments — and over years of those moments, it adds up to the clinician you become.
📁 This Month’s Toolkit: Log in or create a free account to download from the Justice-Involved Mastery Page
Article 8 comes with four downloadable tools for navigating ethics in justice-involved treatment when the systems around you push back:
- Ethical Decision-Making Framework — The four-move sequence for when something feels off. Notice. Investigate. Consult. Hold the line, but stay open. Includes worked examples showing the framework applied to real ethical situations from the article.
- “Stuck on a Call” Ethics Decision Tree — Tactical reference for the moments when a PO, attorney, or supervisor is on the phone pushing you to change your recommendation. Four color-coded actions and the four phrases to memorize for real-time use.
- Ethical Pressure From the Other Direction — A scenario-based toolkit for when you’re asked to compromise your ethical practice (not your clinical approach). Five scripts covering client requests to alter records, subtle PO pressure to soften findings, requests to change a diagnosis, and more. Each scenario includes documentation guidance and red flags.
- 8-Article Cross-Reference Guide — A series integrator that maps all 28 tools, 8 articles, and core concepts across the Justice-Involved Treatment Mastery Series. Find the right article and tool for any clinical situation. Built as the navigation guide for the complete series toolkit dropping May 31.
This article is part of the Justice-Involved Treatment Mastery Series. Previous articles cover harm reduction with mandated clients, abstinence-only courts, therapeutic alliance, documentation, motivational interviewing, managing relapse, and navigating trauma-informed systems. The complete series download — all 8 articles and every resource — releases May 31, 2026.
Previous articles in this series:
- Month 1: Beyond Court Compliance: Harm Reduction with Mandated Clients →
- Month 2: When the Judge Says Abstinence but Evidence Says Harm Reduction →
- Month 3: From Compliance to Care: Why Justice-Involved Clients Need a Different Approach →
- Month 4: Documentation That Satisfies Courts AND Supports Recovery →
- Month 5: Motivational Interviewing for Mandated Clients
- Month 6: Managing Relapse
- Month 7: Navigating a Trauma-Informed System That’s Still Not Trauma-Informed
Browse the full Justice-Involved Treatment Mastery hub →
Not a member yet? Create a free account to access resources →
References
- American Counseling Association. (2014). ACA Code of Ethics. Section B: Confidentiality and Privacy. https://www.counseling.org/docs/default-source/default-document-library/ethics/2014-aca-code-of-ethics.pdf
- Substance Abuse and Mental Health Services Administration. (2024). 42 CFR Part 2 — Confidentiality of Substance Use Disorder Patient Records. Updated final rule effective February 16, 2026. https://www.samhsa.gov/substance-use/treatment/statutes-regulations-guidelines
- Center for Substance Abuse Treatment. (2005). Substance Abuse Treatment for Adults in the Criminal Justice System. Treatment Improvement Protocol (TIP) Series, No. 44. SAMHSA. https://www.ncbi.nlm.nih.gov/books/NBK64137/
- National Institute on Drug Abuse. (2020). Principles of Effective Treatment. In Principles of Drug Addiction Treatment: A Research-Based Guide (Third Edition). https://nida.nih.gov/publications/principles-drug-addiction-treatment-research-based-guide-third-edition/principles-effective-treatment
- Merkt, H., Haesen, S., Eytan, A., Habermeyer, E., Aebi, M. F., Elger, B., & Wangmo, T. (2021). Forensic mental health professionals’ perceptions of their dual loyalty conflict: Findings from a qualitative study. BMC Medical Ethics, 22(123). https://bmcmedethics.biomedcentral.com/articles/10.1186/s12910-021-00688-2
- Klag, S., O’Callaghan, F., & Creed, P. (2005). The use of legal coercion in the treatment of substance abusers: An overview and critical analysis of thirty years of research. Substance Use & Misuse, 40(12), 1777–1795.
- Parhar, K. K., Wormith, J. S., Derkzen, D. M., & Beauregard, A. M. (2008). Offender coercion in treatment: A meta-analysis of effectiveness. Criminal Justice and Behavior, 35(9), 1109–1135.
- McLellan, A. T., Lewis, D. C., O’Brien, C. P., & Kleber, H. D. (2000). Drug dependence, a chronic medical illness: Implications for treatment, insurance, and outcomes evaluation. JAMA, 284(13), 1689–1695. https://jamanetwork.com/journals/jama/fullarticle/193224
- Skeem, J. L., Louden, J. E., Polaschek, D., & Camp, J. (2007). Assessing relationship quality in mandated community treatment: Blending care with control. Psychological Assessment, 19(4), 397–410.
