# Lesson 7: The Therapeutic Process: Preparation, Dosing Sessions and Integration

*Companion-podcast transcript • Sarah & Kiffer*  
*Office Hours episode to listen to after working through the lesson*

---

**Sarah:** A participant in the first phase three trial of MDMA-assisted therapy for PTSD spent about forty-two hours in sessions with two therapists. Eighteen of those hours were without the drug. So when that trial reports a benefit, how much of it came from those eighteen hours?

**Kiffer:** The trial can't tell us. Both groups had the same preparation, the same session support and the same integration, so the comparison measures what the drug added to that therapy. What the therapy contributed on its own was never put to a test.

**Sarah:** And yet people argue hard about which kind of therapy should go with the drug.

**Kiffer:** They do, and so do we. Sarah and I don't fully agree on what the default should be.

**Sarah:** Welcome back to Office Hours. I'm Sarah.

**Kiffer:** And I'm Kiffer. That gap between the drug and the therapy runs through Lesson seven, The Therapeutic Process: Preparation, Dosing Sessions and Integration, so let's start with it.

**Sarah:** Question one. If both arms of a trial get the same therapy, what would it take to learn what the therapy itself does?

**Kiffer:** It would take a design that varies the therapy while holding the drug constant. Picture a made-up trial with four arms. Every arm gets the same screening, the same two people in the room and the same medical monitoring, because those are safety requirements. What varies is the drug, active or placebo, and the psychotherapy, either a full course like the trials use or a brief version with far fewer hours.

**Sarah:** Give me some numbers to work with, then.

**Kiffer:** These are invented improvements on a depression scale, in points. Placebo with brief support gives five, placebo with full therapy nine, the drug with brief support eleven, and the drug with full therapy twenty. A standard trial gives full therapy to both arms. What would it report as the effect of the drug, and what is the drug's effect when the support is brief? Take a few seconds, and pause the audio if you'd like more time.

*(Pause)*

**Sarah:** The standard trial compares the two full-therapy arms, so twenty minus nine, which is eleven points. With brief support, it's eleven minus five, which is six. So in this made-up trial the drug adds almost twice as much when the full therapy comes with it.

**Kiffer:** Right. Now the tempting next step is to say the therapy adds four points, because nine minus five is four, and four looks small next to eleven.

**Sarah:** Except that four is the therapy's effect with placebo. With the drug, the therapy adds twenty minus eleven, which is nine.

**Kiffer:** Exactly, and the gap between those two figures, five points, is what statisticians call an interaction. The drug and the therapy don't add up as separate parts, because each one changes how much the other does. So the eleven points from the standard trial is an honest estimate of the drug given with that therapy. It says nothing about the drug with less therapy, or with a different therapy.

**Sarah:** Has anyone run a design like that?

**Kiffer:** Writing in 2022, Burton and colleagues stated that no comparative studies of psychotherapy platforms had been conducted in the field of psychedelics. Such a trial would also need a large sample, because an interaction is harder to detect than a main effect.

**Sarah:** The closest thing I can think of is the Roots to Thrive pilot in British Columbia. Fifty-seven health care providers went through the twelve-week group program, and thirty-eight of them chose to add three ketamine sessions. Scores for depression, anxiety and PTSD fell in the whole group, and the reductions were larger with ketamine, though the difference wasn't statistically significant.

**Kiffer:** It shows the problem well. Because people chose their own arm, and the sample was small, the study can't separate the contribution of the group from that of the drug.

**Sarah:** And there's the alliance finding from the psilocybin and escitalopram trial. Among the thirty participants who received psilocybin, a stronger alliance with the guides predicted lower depression scores at the end.

**Kiffer:** That fits the idea that the relationship matters. It's also observational and small, so other differences between participants could account for some of it. To my mind it's a good reason to run the trial, and it can't stand in for one.

**Sarah:** Question two. The therapy costs clinician time. Where can a protocol save that time, and where can't it?

**Kiffer:** Here's a hypothetical protocol for a psilocybin service, a hybrid of the individual and group models. It has three individual preparation sessions of ninety minutes with the same two therapists, and two individual dosing sessions of eight hours with both therapists present. Then come six weekly integration groups of two hours, with six participants in each group and two facilitators. Sarah, how many clinician hours per participant? Everyone else, try it too. Take a few seconds.

*(Pause)*

**Sarah:** Okay, using the clinician hours equation. Non-dosing contact is three times one and a half, which is four and a half hours, plus six times two, which is twelve, so sixteen and a half hours. Two clinicians makes thirty-three. Dosing is two clinicians times two sessions times eight hours, which is thirty-two. Thirty-three plus thirty-two is sixty-five, and each group has six people, so I divide by six. That's about 10.8 clinician hours per participant.

**Kiffer:** That's close to the Roots to Thrive figure from the lesson. Does it make sense for this protocol?

**Sarah:** It doesn't. The dosing sessions are individual. Each participant has two therapists for sixteen hours, so the dosing alone is thirty-two clinician hours per participant, and the total can't be smaller than one of its parts. I divided everything by six, and only the integration groups are shared.

**Kiffer:** Good catch. The equation uses one divisor because it assumes every session is shared the same way, so split the protocol into its parts and try again.

**Sarah:** Preparation is two therapists times four and a half hours, which is nine. Dosing is thirty-two. The groups are two facilitators times twelve hours, divided by six participants, which is four. Nine plus thirty-two plus four is forty-five clinician hours per participant.

**Kiffer:** And thirty-two of the forty-five, about seventy-one percent, sit in the dosing sessions. Individual integration with both therapists would cost twenty-four clinician hours, for a total of sixty-five, the number you had before dividing. So the groups save twenty hours, almost a third. What they can't touch is the dosing day. In Canadian trials, Health Canada expects at least two therapists during administration and a physician readily available on site for the duration of the drug's effects.

**Sarah:** So the floor is set by the staffing of the dosing sessions, and the much lower figure at Roots to Thrive comes partly from dosing in groups.

**Kiffer:** That's how the arithmetic works, and the figure there also leaves out the physician and nurses. The Roots to Thrive evaluations aren't randomized, and the ones reporting outcomes are in health care workers, so a low count of clinician hours describes cost and reach and says nothing about effectiveness.

**Sarah:** Before question three, a short content note, because this part mentions sexual misconduct by therapists in a trial. If you'd like support, you can call or text nine eight eight, the Suicide Crisis Helpline, at any time. The Hope for Wellness Help Line for Indigenous people is at one eight five five, two four two, three three one zero. And for anyone affected by sexual violence in British Columbia, VictimLinkBC is at one eight hundred, five six three, zero eight zero eight. So, question three. Under the lesson's practice standard, consent during a dosing session can be confirmed, narrowed or withdrawn, and the team never widens it. Why hold to that, even when the person asks clearly?

**Kiffer:** Because of what the drug does to the capacities that consent depends on. Health Canada's 2022 expectations for trials list suggestibility and transference among the psychological risks that consent forms should state. Transference is the tendency to feel toward a therapist the way you once felt toward someone important earlier in life. Under those conditions, a request made during the session is less reliable than a decision made beforehand, so the standard treats Health Canada's check in the moment as confirming what was agreed while sober.

**Sarah:** I want to push on that. The MAPS manual lets therapists give touch a participant asks for in a session, if it's appropriate and nonsexual. So say a participant asks for a hug, clearly and more than once. She's an adult. Isn't refusing her paternalistic?

**Kiffer:** It's a fair challenge, and here the standard is stricter than the manual. The person who decided about touch in preparation is the same adult, deciding with full capacity and time to think, and the team is respecting that decision. Nothing is closed off for good, either. The agreement can be revisited in integration and before the next dose, when she can weigh it again while sober.

**Sarah:** And the stakes of getting this wrong are real.

**Kiffer:** They are. In May 2019, MAPS disclosed that two therapists at a Vancouver site of a phase two MDMA trial had committed an ethical violation. One began a sexual relationship with a participant after her active treatment ended in October 2015, and the other reportedly tried unsuccessfully to stop it and did not report it. Lesson nine examines the case. It shows why the prohibition on sexual contact has to be absolute, and why therapists need supervision and a clear duty to report.

**Sarah:** Let's test an agreement, then. This one's hypothetical. A participant in an authorized psilocybin trial agreed while sober to hand holding only. The guides will ask before offering it, with a gentle question if she's distressed and not speaking. The word stop ends any touch, except where touch is needed to protect her from physical harm. She declined hugs. Two hours in, she hasn't spoken for a while, she's crying, and she holds out her hand toward one guide. What does the agreement let the guide do? Take a few seconds.

*(Pause)*

**Kiffer:** The outstretched hand is a strong signal, and the agreement still asks the guide to check first. So the guide asks quietly, "Would you like me to hold your hand?" and waits for a yes or a nod. That's the check in the moment that Health Canada expects, and it stays inside what she agreed.

**Sarah:** Then she grips the hand and says, "Hold me, please." A hug is outside the agreement. The guide answers warmly in words, keeps holding her hand, and the request goes into the record so it can be discussed in integration.

**Kiffer:** Right. Now a harder one. Later she stands up suddenly, unsteady, close to the edge of a table, and a guide steadies her by the arm. She says "stop."

**Sarah:** The stop word ends any touch except touch that protects her from physical harm. So the guide keeps steadying her only as long as it takes to keep her from falling, tells her what's happening, lets go as soon as she's safe, and records the touch and the reason for it.

**Kiffer:** That's how the agreement reads, and Health Canada's own example of the exception is catching someone who's falling. Protecting someone from a fall justifies brief touch. It never justifies continuing touch for comfort once the danger has passed.

**Sarah:** Before question four, another short content note, because this part discusses suicidal thoughts after sessions. You can call or text nine eight eight, the Suicide Crisis Helpline, at any time, and the Hope for Wellness Help Line for Indigenous people is at one eight five five, two four two, three three one zero. Question four. People are told that difficult material may come in waves after a session. How do you tell an expected wave from a reason to refer?

**Kiffer:** Start with the safeguard. Worsening is always assessed, and never assumed to be progress. Reporting on the FDA advisory committee meeting in June 2024, Pharmaceutical Executive noted that a trial participant had told the agency her therapist encouraged her to see worsening symptoms as progress. The MAPS manual notes that suicidal thoughts may increase temporarily after sessions in people with PTSD, and serious adverse events, including suicidal behaviour, have occurred in trials, among them the phase two b trial of the psilocybin product COMP360.

**Sarah:** So here's a worked example with three made-up notes from integration calls. Note one, day two: tearful in the evenings, thinking about her father, sleeping about seven hours, says the sadness eases after she writes. Note two, day five: sleeping three hours a night for four nights, says she has never felt better, has started three new projects, speaking quickly. Note three, day nine: says things have felt heavy, and that sometimes she thinks everyone would be better off without her. Which of these call for referral or escalation? Take a few seconds.

*(Pause)*

**Kiffer:** Note three needs action now. A thought like that calls for a structured assessment with a tool such as the Columbia Suicide Severity Rating Scale, closer contact, the involvement of the responsible physician or psychiatrist, and emergency services if there's immediate danger. Note one reads like an expected wave. It settles, she's sleeping, and she has a practice that helps, so the team keeps the planned contact and records what she said.

**Sarah:** And note two is the one I'd worry about missing, because never having felt better sounds like the outcome everyone hoped for.

**Kiffer:** Yes. An unusually high mood with little need for sleep, a burst of new projects and fast speech are, together, possible signs of mania, which call for psychiatric assessment. A team hoping for a good result is primed to record that call as a success. The same hope can bias a clinic's own ratings of benefit, and MAPP1 used blinded independent raters for its main outcome.

**Sarah:** What about the integration scales? If someone scores high on the Experienced Integration Scale, hasn't integration worked?

**Kiffer:** A high score tells you she feels settled, harmonized and improved. Recovery from depression is a separate question for a clinical scale, and the integration scales are from twenty twenty-two, validated in online samples of a few hundred people in the United States. I'd report them alongside a clinical measure.

**Sarah:** Question five. In the Canadian Substance Use Survey for 2025, about seven percent of people aged fifteen and older, roughly 2.6 million, reported using a psychedelic in the past year, most often psilocybin. And clients sometimes ask counsellors for help before a planned experience. Why might help offered only to make it safer still carry legal risk?

**Kiffer:** This is general legal information about Canadian law as of October 2026. Legal advice applies the law to a particular situation, and anyone facing a real case should get it from a lawyer and consult their college or association. Subsection four one of the Controlled Drugs and Substances Act prohibits possessing a substance in Schedules One to Three except as authorized, and psilocybin is in Schedule Three. Subsection twenty-one one of the Criminal Code makes a person a party to an offence if they do anything for the purpose of aiding someone to commit it, or if they abet it.

**Sarah:** And wanting to reduce harm doesn't change the purpose?

**Kiffer:** The section contains no exception for help given to reduce harm. In the Briscoe decision in 2010, the Supreme Court of Canada held that purpose here is essentially the same as intention, that the helper needn't want the offence committed, and that the helper must know the other person intends to commit it. So a professional who holds preparation sessions for a planned use, or books integration as part of the plan, may be at risk of being treated as having aided or abetted the possession, depending on the facts.

**Sarah:** And help after an experience that's already over, with no plan to use again, is different.

**Kiffer:** In one respect, yes. Help given after a possession has ended does not appear to be given for the purpose of aiding it, and helping people with distress after past substance use is a familiar part of mental health care. That's a reading of the words of the section that a court could see differently, and colleges and associations may set their own expectations.

**Sarah:** Here's a harder case, and it's hypothetical. A client sees a counsellor in British Columbia after a psilocybin experience last spring that left her shaken. They work on it over several sessions. Then she says, "Next time, I want to be better prepared. Can we plan for that?"

**Kiffer:** That's the moment the analysis changes, because preparation for a planned unauthorized use is where a professional may be at legal risk. The counsellor can say so plainly, keep supporting her mental health, explore what she hopes psilocybin would do for her, and talk about treatments, including whether a physician might seek a trial place or a Special Access Program authorization for her. Before going further, the counsellor should get legal advice and consult their professional body.

**Sarah:** And British Columbia's decriminalization exemption doesn't change this. It ran from January 2023 to the thirty-first of January 2026 and never covered psilocybin.

**Kiffer:** Professional guidance points the same way. In 2024, the British Columbia Association of Clinical Counsellors said that trained counsellors may prepare people, support sessions under medical supervision and facilitate integration, and should offer those services only within sanctioned trials or programs.

**Sarah:** Which brings us to the authorized routes.

**Kiffer:** There are four: clinical trials authorized by Health Canada, the Special Access Program, section fifty-six exemptions, and lawfully prescribed ketamine. Since the first of October 2026, the Special Access Program has operated under the Controlled Substances Regulations, which allow a practitioner named in a letter of authorization to administer the drug and the patient to possess it. Lesson eleven covers these routes in detail.

**Sarah:** Okay. Now the part where we disagree, which takes us back to the start. What kind of therapy should go with the drug?

**Kiffer:** You go first, since you're the one who would change the default.

**Sarah:** I would. Yaden and colleagues argued in 2022 that cognitive behavioural approaches, including acceptance and commitment therapy, should be the default. They have the largest evidence base outside psychedelic therapy, they make few speculative assumptions about the mind, and their claims are easier to test. Brennan and Belser pointed out that basic support models leave interventions loosely specified and may rely on an untested inner healing intelligence. And at the FDA advisory committee in 2024, members cited minimal evidence for the benefit of the psychological approach. A specified therapy at least gives us something definite to test.

**Kiffer:** Those are good arguments, and the point about specification is the strongest. Where I'd push back is on calling anything a default yet. Burton and colleagues replied that no approach can claim default status before the evidence is collected, and that psychoanalytic therapy has supporting evidence too. Brennan and Belser also disclosed that they developed a competing model while working for a psychedelic drug company.

**Sarah:** Fair, although a conflict of interest doesn't make the critique wrong.

**Kiffer:** Agreed. Here's where I land. During the dosing session, while suggestibility is heightened, I think a mostly non-directive stance makes sense, because it keeps the therapist's interpretations out of the experience. Structured techniques have a strong case in integration, where the person is sober and working to turn insight into action, a point Yaden and colleagues make about behavioural activation. I'd hold all of that loosely until someone compares models with the drug held constant.

**Sarah:** I'd still make the structured model the starting point, so that the first comparison has a clearly specified arm. We agree on the practical rule, though. Whatever the model, write down the interventions, check adherence with the session records, and describe any benefit as the effect of the drug given with that therapy, until a trial varies the therapy and holds the drug constant.

**Kiffer:** Let's pull it together with three things to take away.

**Sarah:** First, when both arms get the same therapy, a trial estimates what the drug adds given that therapy. Learning what the therapy contributes, or whether drug and therapy interact, takes a design that varies the therapy while holding the drug constant.

**Kiffer:** Second, consent to touch and recording is settled while the person is sober. During the session it can be confirmed, narrowed or withdrawn, the team never widens it, and the exception for preventing physical harm stays narrow.

**Sarah:** And third, worsening after a session is always assessed, and the signs that call for referral include some that sound like good news, such as an unusually high mood with little need for sleep.

**Kiffer:** If you'd like more practice, rework the hybrid protocol from question two with integration groups of eight, and check that your total never falls below the clinician hours in the dosing sessions.

**Sarah:** Next time, it's Lesson eight, Psilocybin, Ketamine and Depression, Anxiety and End-of-Life Distress, where we turn from the process to its results and how to read them.

**Kiffer:** Take care, everyone.

**Sarah:** See you in Lesson eight.
