Ending Therapy Well with Children and Teens

Wednesday at 3:45. A 9-year-old who has been coming since September asks — for the third week — when therapy will be over. His mom has noticed an uptick in irritability the last two appointments and wonders if it's regression. You have two sessions left on the calendar. You haven't told him that yet.

Most clinical training doesn't really prepare you for this conversation. Programs spend weeks on assessment and case conceptualization, days on the early sessions, and a single afternoon on termination. Termination is where the work either consolidates into something the child can carry forward or quietly evaporates.

The numbers nobody tells you about

The largest meta-analysis of dropout in child and adolescent outpatient mental health care, by de Haan and colleagues (2013), looked at 47 studies and found that 28% of children dropped out of efficacy trials and roughly 50% dropped out of real-world effectiveness studies. Real-world: half. Mirabito's 2006 study of adolescent treatment in community settings found that clinicians themselves often categorized terminations as unplanned even when they had foreseen them weeks in advance — the work of preparing the ending hadn't happened, so the ending arrived feeling sudden.

The implication is plain. Most child therapy doesn't get to a planned ending. The ones that do are the exception, and for those cases, doing the ending well is one of the highest-leverage moves a therapist makes. The child's last impression of therapy shapes their relationship with the idea of help — whether they re-engage at 14, at 19, at 32.

So when there is a chance to plan the goodbye, here is what the ending should do.

What a planned ending actually does

A good ending does three things at once, and most of the literature treats them separately when they actually want to be braided.

First, it consolidates skills. The child should leave knowing what they learned, and they should be able to say it back to you in their own words. Not "we did CBT." Something like: "I learned that when my stomach feels tight, I can ask myself if my brain is making a movie that isn't real yet." That sentence is the whole point. If the child can't generate something like it, you haven't finished yet.

Second, it renames the therapeutic relationship. The child has spent months in a particular kind of relationship — intense, intimate, contained, weekly. They need a frame for what it becomes when it stops. For younger kids, the frame that lands is "you outgrew this" — said with respect, not condescension. With teens it's closer to: "you don't need this from me anymore, and that's exactly what we were going for."

Third, it produces a takeaway. Not a metaphor. An actual object the child holds. A certificate, a letter, a "growth book" of drawings, a coping-card deck they decorated. Kids need a tangible thing because their object permanence with abstract experiences is still developing. A certificate that says "Mara mastered five things to do when her worry brain gets loud" — and lists the five — is a transitional object. Children keep these. Clients who return as adolescents or young adults will sometimes mention the letter they got at age nine and still have.

Tapering, not ending

The mistake most often made in the final phase of child therapy is moving from weekly to nothing. Two patterns work better.

The first is session spacing. Weekly becomes every other week becomes once a month becomes a single six-month check-in. Each gap is an exposure trial. The child is testing whether they can do this without you. When they show up to the every-other-week appointment with nothing pressing to report, that's data. They learned the skill.

The second is scheduled boosters. Research on planned booster sessions in adolescent treatment has shown reductions in symptom recurrence after the active phase ends. You don't have to call them booster sessions — most kids hate clinical language. "Three-month check-in" works. So does "tune-up." The fact that you are scheduling the next contact at the moment of saying goodbye changes the affective register of the ending. It is not abandonment. It is graduation with a return address.

When the ending isn't yours to plan

Insurance ends. A family moves. A parent decides the kid is "fine now" and stops scheduling. A teen ages out. Sometimes the planned termination you were building toward gets pre-empted by life.

When this happens — and it will happen more often than the planned version — there are still moves available.

If you have any session left, treat it as the ending session even if no one is calling it that. Do the consolidating questions. Hand over the takeaway. Name what was learned. If the kid is too young to consolidate verbally, draw it with them. Clinical experience and the engagement findings from de Haan both point in the same direction: even a single explicit ending conversation can change how the child relates to help in the future.

If the family disappears mid-treatment with no warning, you still have one move. Send a letter. Brief, warm, written to the child rather than the parent, naming the work you did together and the things the child can take forward. Most of these letters never get a reply. A few do, years later. The cost of writing it is twenty minutes. The cost of not writing it can be the child concluding that the relationship was disposable.

Adolescents are a different animal

Teens are routinely written off as uninterested in termination work. They will say "whatever, it's fine, can we just go" when you bring up the ending. Believe roughly 30% of that.

The adolescent termination move that holds up most reliably is the honesty offer. You tell them: "I know you're going to act like this isn't a thing. I don't actually believe you. So here's what I want to do — I'm going to write down what I think you learned, and you're going to either agree with it or tell me where I got it wrong." Then you write a short list, hand it over, and let them push back. They push back. That's the work.

Teens won't accept a certificate the way an 8-year-old will. A letter, written like an adult would write to another adult, lands. So does a single artifact they choose themselves — a playlist, a printed quote they want, the actual notebook they used in sessions. The point is that something physical leaves with them.

A note on materials

Because this is a post about endings on a site about therapy materials, the honest answer first. There is no material that makes a goodbye work. The ending works because the therapist did the preparing. A nicely designed certificate on the last session in a relationship the child experienced as cold and rule-bound is just decoration. A handwritten note on lined paper at the end of work the child felt seen in is — to that child — priceless.

That said, the part of the goodbye that is concrete is worth taking seriously as a clinical object rather than an afterthought. Print it on real cardstock. Use the child's name. Include the specifics of what they did, not a generic award. "You completed eight brave exposures to feared situations" is meaningful in a way that "Bravery Award" is not. If you want a starting point, the therapy achievement certificate generator lets you write that specificity in directly.

Most of your endings will not be planned. The work of the planned ones is to be the kind of clinician who is ready to do this well when the rarer planned version arrives. That readiness — knowing what you would say, what you would hand them, what you would write — is also what shapes the unplanned endings, when all you have is the last twenty minutes and a child who didn't know it was the last.