MDMA-assisted therapy with mindfulness skills

1. Introduction

Version followed: MAPS-sponsored study MAA-1 (NCT02008396), MDMA-assisted therapy for social anxiety in autistic adults, Harbor-UCLA / LA BioMed, run 2014–2017 by Charles S. Grob and Alicia L. Danforth. The therapy is described in protocol version 2 (21 May 2013) and the results paper (Danforth et al., 2018). Therapist conduct is filled in from the MAPS teaching manual (Ruse, Mithoefer, Jerome, Doblin & Gibson, revised 23 Oct 2010), which MAA-1 drew on, and each such point is labelled. [source: Danforth et al. 2018; MAA-1 protocol v2; MAPS manual 2010]

Source status. No named protocol has been published for this modality, and MAA-1 is the closest published model. The MAA-1 inclusion criteria come from protocol v2, pp. 21–22, and the 2018 paper, which also gives the wording on mindfulness/DBT and the after-session arrangements. Regulatory events come from press and trade reports (see 2), and Australia from the TGA. The meditation adverse effects come from Farias et al. 2020 and Lindahl et al. 2017. The Danforth 2016 treatment-model paper is known only through a published summary, the 2018 supplementary criteria (ESM 1–2) are not used here, and no source used names the DBT skills taught or gives a touch policy for MAA-1. Those parts stay [synthesis] or [uncertain].

What it is, plainly. No stand-alone, named protocol called "MDMA-assisted therapy with mindfulness skills training" has been published. The only published model that matches this guide's topic is MAA-1. Its preparatory phase delivered "standardized mindfulness-based therapy adapted from dialectical behavioral therapy (DBT)", alongside the usual MAPS frame: preparation, all-day medicine sessions with a therapist pair, then integration. [source: Danforth et al. 2018] Protocol version 2 (21 May 2013) does not mention mindfulness or DBT in pp. 1–35. It describes the therapy as adapted from social-skills-focused techniques used with adults with an Asperger's diagnosis. Mindfulness first appears in the 2018 report, so it was added or specified after version 2. [source: MAA-1 protocol v2 pp. 10, 17, 31; Danforth et al. 2018] Neither source says which DBT skills were taught, in what order or in how many sessions. The 2016 treatment-model paper may do so; it is not used here. Section 4 therefore gives the published frame and marks the content of the skills training as [synthesis]. [uncertain]

A related trial, SAMATI (NCT05138068, Portland Psychotherapy, Luoma and colleagues), treats social anxiety disorder in non-autistic adults. It cites the Danforth mindfulness approach but bases its own manual on ACT and Emotion-Focused Therapy, not on DBT. [source: Lear et al. 2023]

How change is thought to happen. In the MAPS model, MDMA is a catalyst, not the therapy. It is said to reduce fear in response to emotional threat, so the person can stay in an "optimal arousal zone": engaged with feelings but not dissociating or flooding. [source: MAPS manual 2010, pp. 3, 5] In MAA-1 the mindfulness preparation gave participants "vocabulary and skills" for entering the altered state and for communicating during it. Therapists encouraged "self-directed healing and meaning-making". Observers also saw latent social skills (starting and sustaining conversation) emerge once participants relaxed. The investigators did not teach social skills directly. [source: Danforth et al. 2018]

Legal and regulatory status (as of the sources, Sept 2025 – Oct 2026).

USA: MDMA is Schedule I. The FDA declined Lykos Therapeutics' application for MDMA-assisted therapy for PTSD in a Complete Response Letter, which Lykos announced on 9 August 2024. [source: Lykos press release, 9 Aug 2024, as reported by Psychiatric Times and HCPLive] The letter itself (NDA 215455) is dated 8 August 2024, the day before Lykos's announcement. [source: HCPLive, Sept 2025, citing the FDA letter; The Microdose, 5 Sept 2025] The FDA asked for an additional Phase 3 trial. The letter cited study-design concerns, durability of effect and participants' prior MDMA use. [source: Psychiatric Times 2025] The FDA published the letter on 4 September 2025. In August 2026, Resilient Pharmaceuticals (formerly Lykos) resubmitted the application without running a new Phase 3 trial. The resubmission relies on the MAPS-funded programme completed in 2022, on third-party data including a small VA study, and on a new Phase 1 safety study in healthy volunteers. The company has not commented publicly. As of the sources' date (12 Aug 2026), no FDA acceptance, goal date or decision had been reported. [source: Psychedelic Alpha, 12 Aug 2026; HCPLive, Aug 2026] Check the current status before advising anyone. [uncertain after Aug 2026] Access is limited to trials and Expanded Access. [source: Psychedelic Journal guide 2026]
Australia: Since 1 July 2023, psychiatrists approved under the TGA Authorised Prescriber scheme may prescribe MDMA for PTSD. The indication is not social anxiety or autism. [source: TGA, MDMA and psilocybin page]
Canada: Restricted, case-by-case access through Health Canada's Special Access Program. UK and Netherlands: research only. Israel: limited compassionate use for PTSD. [source: Psychedelic Journal guide 2026, secondary] [uncertain]
Implication: outside an approved trial or a lawful prescriber programme, a psychotherapist cannot deliver the drug part of this model. What you can do today: (a) work as a trained therapist inside an approved trial or programme; (b) teach the non-drug mindfulness and preparation skills; (c) give harm-reduction and integration support to clients who use MDMA elsewhere (see 4.22). [synthesis]

2. Framework

[source: MAA-1 protocol v2 pp. 28–36; Danforth et al. 2018; for comparison, SAMATI (Lear et al. 2023)]

Phase	Goals	Sessions and format (MAA-1)	Criteria for moving on
0. Screening and consent	Confirm diagnosis and eligibility, medical and psychiatric screening, informed consent	Clinical interview plus SCID, C-SSRS, LSAS, ADOS-2 Module 4; medical history and tests by study physicians	Meets inclusion criteria and no exclusions; psychiatric medicines tapered with the prescriber under medical supervision; consent judged adequate
1. Preparation	Rapport and alliance; review past and current issues; set goals; DBT-adapted mindfulness skills; orientation to setting and staff; support partner agreed	3 non-drug sessions of 60–90 min (two participants needed a 4th)	Rapport established; participant knows the session structure and has the skills vocabulary; support partner consented and briefed; study rules reviewed
2. Experimental (medicine) sessions	Self-directed processing in a safe, sensory-adapted setting; structured rapport and social-perception periods alternating with free-choice periods	2 blinded all-day sessions (about 8 h), co-therapist pair, video recorded with consent; medical monitoring by study staff	Stable at closing (distress and suicidality checks); support partner present for transport and overnight
3. Integration	Process what emerged; plan the return to daily life; carry "acceptance" and "reduced fear" into everyday social situations	Next-morning session (about 60 min), then 2 more integrative sessions over about 3 weeks (60–90 min) after each medicine session; daily phone calls for 7 days	Emotionally stable; second medicine session only after review and with participant agreement
4. Follow-up	Durability, closure	Primary endpoint 1 month after the 2nd medicine session; 6-month follow-up; optional extra visit	Exit plan; referral on if needed [synthesis]

Notes:

The 2013 protocol specifies three preparatory sessions but gives conflicting lengths: "hour-long" (p. 13), about 60 min (p. 30) and longer visit estimates in Table 2 (p. 19). The 2018 paper says 60–90 min, and this guide follows the paper. The paper is also inconsistent on integration. Its Methods give three 60–90 min integrative sessions over 3 weeks after each medicine session, while its later description gives a next-morning session plus two in-person sessions at 2-week intervals. [source: MAA-1 protocol v2; Danforth et al. 2018]
Corrected: the PTSD wording is in the inclusion list itself, not only in the summary. Protocol v2 (section 4.2.1, p. 21) requires at least one unsuccessful attempt at treatment for PTSD, with talk therapy or medication. Stopping a treatment because it could not be tolerated also counts. The other inclusion criteria nearby are a diagnosis of autism spectrum disorder and at least moderate to severe anxiety. The 2018 paper's own criteria do not include this item, and its participants were not described as having PTSD (83% had prior psychotherapy, 67% prior medication). The item is probably carried over from the MAPS PTSD template [uncertain: inferred], so this guide follows the 2018 paper. [source: MAA-1 protocol v2 p. 21; Danforth et al. 2018]
Published exclusions (MAA-1): pregnancy or nursing, or no effective contraception; weight below a protocol threshold; current illicit drug abuse; inability to give informed consent; exclusionary medicines; any current or past diagnosis judged to make participation risky; prior MDMA use (participants had to be MDMA-naive). [source: MAA-1 protocol v2 p. 22] The 2018 paper adds these criteria: age 21 or over; MDMA-naive by self-report; "physically healthy, and psychologically stable"; LSAS score of 60 or more; autism confirmed on ADOS-2 Module 4; and not under conservatorship. [source: Danforth et al. 2018] The full list is in the paper's supplementary files (ESM 1–2), which the article page names but does not link, and they are not drawn on here. [uncertain]
SAMATI, a newer adaptation, adds explicit cardiovascular, QT-prolongation, liver, hyponatraemia/hyperthermia, seizure, serious risk to others, unstable housing and pending-litigation exclusions. [source: Lear et al. 2023]

3. Guiding principles

The medicine is a tool; the therapy is the relationship and the participant's own process. [source: MAPS manual 2010 p. 5] In practice: your job is to keep conditions safe, not to produce breakthroughs.
Trust the participant's own healing capacity; follow, don't lead. The manual's phrase is "trust your own healing intelligence" (p. 22), and it warns "Don't get ahead of the medicine" (p. 26). In practice: follow the participant's material and pace, and do not steer toward a planned topic.
Largely non-directive, with "gentle guidance" only when needed. [source: MAA-1 protocol v2 p. 31] In practice: offer choices, not instructions, especially when the participant keeps avoiding contact.
Encourage self-directed healing and meaning-making. [source: Danforth et al. 2018] In practice: ask what the experience means to them and do not interpret. Attendants were told "not to interpret the participant's experiences". [source: MAPS manual 2010 p. 15]
Stay in the optimal arousal zone. [source: MAPS manual 2010 p. 5] In practice: use breath, body awareness and grounding to bring the person back from flooding or shutdown, and encourage them to stay with difficult feelings.
Mindfulness gives a shared language. Skills taught before the medicine sessions help participants enter the altered state and describe it. [source: Danforth et al. 2018] In practice: agree words and signals in preparation, then use exactly those words during sessions.
Adapt the setting to autistic sensory needs, in consultation with autistic people. The room was den-like, softly lit and noise-reduced, with natural elements, fidget objects and a private bathroom. [source: Danforth et al. 2018] In practice: treat stimming as self-regulation and accommodate it, and keep the environment predictable.
Rapport first. Conventional therapy can stall for autistic adults because rapport is hard to establish. [source: MAA-1 protocol v2 p. 7] In practice: spend preparation time on the relationship, not only on information.
No social-skills training is imposed. The investigators gave no social-skills psychoeducation. [source: Danforth et al. 2018] In practice: notice and reflect social capacities as they appear, rather than coaching them.
Accept and respect unusual experiences. Multiplicity of the psyche, and experiences the participant calls transpersonal, are treated as normal and respected. Hold "a posture of acceptance rather than analysis". [source: MAPS manual 2010 pp. 10–11, 34] In practice: you do not need to share or endorse the participant's metaphysical framing. [synthesis]
Wellbeing comes before study goals, and participation is voluntary. [source: MAPS manual 2010 pp. 14–15] In practice: stopping, postponing or skipping a second session is always a legitimate choice.
Clear touch and sexual boundaries. Touch is agreed in advance, used only with permission and ends whenever the participant says "stop". Therapists state that "at no time will they engage in any form of sexual contact". [source: MAPS manual 2010 p. 16] The MAA-1 protocol v2 (pp. 1–35 checked) and the 2018 paper say nothing about touch. The MAPS manual policy is given here as the closest published standard, not as MAA-1 practice. [uncertain] In practice: get the touch agreement in writing during preparation and re-check consent in the moment.
Two-therapist team, supervised. MAA-1 used a co-therapist pair (Grob and Danforth). The sponsor reviewed session video and gave ongoing feedback. [source: MAA-1 protocol v2 pp. 17, 28] In practice: never run a session alone, and keep recordings for supervision and adherence review.

4. Intervention inventory

All example language is illustrative unless it is quoted and cited. Items 4.1–4.21 apply only inside an approved trial or a lawful prescriber programme. Medical screening, medical monitoring and any medical rescue are the responsibility of the study physicians and are not described technically here. Item 4.22 is for any clinician.

Phase 1: Screening and preparation

4.1 Eligibility and safety screening (psychotherapist's part)

Purpose / when: before enrolment, to confirm the diagnosis and catch risks. [source: Danforth et al. 2018; MAA-1 protocol v2 pp. 21–22]
Steps: (1) Confirm autism with a structured observational assessment (MAA-1 used ADOS-2 Module 4) and social anxiety with a clinician-rated scale (LSAS; the trial required a high threshold score). (2) Run a structured diagnostic interview (MAA-1 used SCID-I-RV) and a structured suicide assessment (C-SSRS). (3) Get a signed release so you can contact the client's own therapist and doctors. (4) Ask about prior MDMA and other drug use (the trial required MDMA-naive participants). (5) Refer for medical screening by study physicians. (6) Coordinate any taper of psychiatric medicines with the prescriber; this is medical, so you do not manage it. (7) Confirm practicalities: a support partner, willingness to stay nearby overnight, no self-driving, phone contact, and recording of sessions.
Example language (illustrative): "Before we decide anything, I need a full picture of your health, your history and anything that would make this unsafe for you. Saying no at any point is completely fine."
Watch for: under-reported substance use, which the trial could not verify; conservatorship or impaired capacity, both exclusionary; current suicide risk; housing or support gaps. Working if: you have a clear yes or no decision that you, the medical team and the client all understand.

4.2 Informed consent and orientation

Purpose / when: preparation sessions 1–3. [source: MAPS manual 2010 pp. 14–15; MAA-1 protocol v2 pp. 29–32]
Steps: (1) Explain the design. In a blinded trial, say plainly that they may get placebo. (2) Describe what a medicine day is like: length, recording, monitoring by medical staff, the overnight and next-morning plan. (3) Explain that heightened openness also brings vulnerability. (4) State that participation is voluntary and wellbeing comes before study goals. (5) Review study rules about pre-session restrictions and other substances. (6) Check understanding with a teach-back, not a yes/no question. [synthesis]
Example language (illustrative): "Can you tell me in your own words what will happen on the day, and what you can do if you want to stop?"
Watch for: agreement given to please you, or literal misreading of what you said. Re-explain using written or visual summaries. [synthesis] Working if: the client can describe the day and their rights accurately.

4.3 Rapport and alliance building

Purpose / when: the main aim of every preparatory session. [source: Danforth et al. 2018; MAA-1 protocol v2 p. 30; MAPS manual 2010 pp. 11–12]
Steps: (1) Take the history of salient past and current issues at the client's pace. (2) Listen for their own concerns about the treatment and answer each one. (3) Ask about any prior non-ordinary states (meditation, intense experiences) and how they managed them. (4) Give feedback on screening results. (5) Both therapists attend, so the client knows the whole team before medicine day. [synthesis: rationale]
Example language (illustrative): "What would help you feel comfortable with us? Some people like to know exactly what we'll do; some prefer quiet. Tell us what works for you."
Watch for: masking. A calm presentation can hide high anxiety, so ask directly and offer numeric ratings, which autistic participants in the trial often preferred. [source: Danforth et al. 2018 on quantifying responses] Working if: the client starts raising concerns spontaneously.

4.4 Mindfulness-based skills training adapted from DBT

Purpose / when: in preparation, to give the vocabulary and skills to move into the altered state and to communicate during it. DBT was chosen because it targets interpersonal relationships, emotion regulation and distress tolerance, which are areas often hard for autistic adults with social anxiety. [source: Danforth et al. 2018]
What the sources do NOT say: which exercises, how many, how long, or whether there was homework. The 2018 paper names no specific DBT skills, protocol v2 does not mention mindfulness or DBT, and a published summary of the 2016 treatment-model paper speaks only of "core mindfulness skills training" in preparation, without naming exercises. No published step-by-step protocol was found, so everything below is the writer's [synthesis]. It is general mindfulness practice, not attributed to the trial or to DBT's developers. [uncertain]
Steps [synthesis]: (1) Noticing: a short practice of attending to breath, sounds or a held object, then reporting what was noticed without judging it. (2) Naming: build a personal list of words for body sensations and emotions (useful where alexithymia is present); a written list can be kept for medicine day. (3) Stepping back: practise noticing a thought or urge as a passing event. (4) Riding a wave of distress: rate distress 0–10, stay with it while breathing slowly, and notice the rating change. (5) Signalling: agree simple ways to communicate during the session: a word, a gesture, a number, or a typed message (the protocol allowed text-to-speech). (6) Practise each skill in session, and offer optional home practice.
Example language (illustrative): "Let's practise noticing without changing anything. When you're ready, tell me one thing you noticed in your body, or give it a number."
Watch for: literal or concrete difficulty with metaphor-heavy instructions (make instructions concrete); sensory discomfort with eyes-closed or body-scan practice (offer eyes-open or object-based options); anxiety, depersonalisation or re-emerging trauma during meditation. Stop, ground the client and adapt. These are documented adverse effects. A systematic review of 83 studies found adverse events in about 8% of meditators, most often anxiety, depression and cognitive anomalies. Some occurred in people with no previous mental-health history. [source: Farias et al. 2020] Interviews with Western Buddhist meditators found fear, anxiety, panic, hypersensitivity to light or sound, insomnia, involuntary movements and distressing changes in the sense of self. Some lasted from days to more than a decade, and practice intensity and the relationship with the teacher shaped them. [source: Lindahl et al. 2017] The screening and adaptation steps here are [synthesis]. Working if: the client can name a state, rate it and signal it, and later uses the same words during the medicine session.

4.5 Breathing methods for relaxation and for staying with difficult feelings

Purpose / when: taught in preparation; used early in medicine sessions to settle, and later to "breathe into" difficult material. [source: MAPS manual 2010 pp. 17–18; MAA-1 itself is not documented as using this]
Steps: (1) Teach a slow, comfortable breathing pattern for settling. (2) Separately, teach directing breath and attention toward a felt sensation or emotion, rather than away from it. (3) Practise both in session until they feel familiar.
Example language (illustrative): "If something hard comes up, see if you can let your breath go toward it, just a little, instead of pushing it away."
Watch for: fast breathing or dizziness (slow down and return to normal breathing); a client who uses breathing only to avoid feelings. Working if: the client uses breath on their own to regulate.

4.6 Touch agreement

Purpose / when: preparation; required before any touch on medicine day. [source: MAPS manual 2010 p. 16] Neither MAA-1 protocol v2 (pp. 1–35) nor the 2018 paper describes touch, so check your own protocol. [uncertain]
Steps: (1) Explain that touch is optional and the client decides. Nurturing touch such as hand-holding can be requested, and the manual describes focused bodywork separately. (2) Agree what kinds of touch, if any, are acceptable, and record this. (3) Agree that "stop" ends touch immediately, except where touch is needed for physical safety. (4) State the no-sexual-contact rule explicitly. (5) Discuss physical proximity as well as touch. For autistic clients, also cover sensory aversions to pressure, textures and proximity. [synthesis]
Example language (illustrative): "You may want no touch at all, and that's completely fine. If you'd like, say, a hand to hold, we'll only offer it when you ask or say yes."
Watch for: consent given under the medicine's influence that goes beyond what was agreed. Stick to the prior agreement. [synthesis] Working if: the agreement is clear and recorded, and the client can repeat it back.

4.7 Setting preparation and tour (sensory adaptation)

Purpose / when: preparation session 2. A predictable, low-stimulation environment. [source: MAA-1 protocol v2 p. 31; Danforth et al. 2018]
Steps: (1) Give a full tour of the treatment room and introduce all staff. (2) Show the features: soft lighting, noise reduction, natural elements, fidget objects, private bathroom, comfortable furniture. (3) Ask what to change, such as light level, temperature or seating. (4) Show the optional eyeshades and headphones, and let the client try them.
Example language (illustrative): "This is the room you'll be in. Is there anything about the light, sound or furniture you'd like different?"
Watch for: sensory triggers the client did not mention. Ask directly about smells, fabrics and sudden sounds. [synthesis] Working if: the client says the room feels manageable.

4.8 Support partner briefing

Purpose / when: before the first medicine session. [source: Danforth et al. 2018; MAA-1 protocol v2 pp. 27, 31]
Steps: (1) The client chooses any trusted adult. (2) Meet the partner, explain the role, and get their consent (MAA-1 used an addendum to the consent form). (3) Role: drive the client to and from sessions (no self-driving), stay with or near the client through the next morning, act as "a nonintrusive supportive presence", and be available for 24 h. (4) Give the partner the team's emergency contact.
Example language (illustrative, to the partner): "Your job is to be nearby and kind, not to ask about or analyse the experience. If you're worried, call us."
Watch for: a partner who probes, criticises or is themselves part of the client's difficulty. Consider choosing another partner. [synthesis] Working if: the partner can state their role and the contact plan.

4.9 Intentions and expectations

Purpose / when: preparation session 3. [source: MAA-1 protocol v2 p. 31; MAPS manual 2010 pp. 19, 32]
Steps: (1) Discuss goals for treatment. (2) Frame intentions as held loosely, since the experience may go elsewhere. (3) Prepare for unfamiliar or challenging experiences. The manual's framing: "nothing is going to come up that is not already there" (p. 19). (4) For a second session, build on the first session's intentions.
Example language (illustrative): "What would you like to be different socially? Let's write it down, then hold it lightly."
Watch for: rigid or performance-based goals ("I must be cured"). Reframe them as curiosity. [synthesis] Working if: the intention is written down, personally meaningful and flexible.

Phase 2: Experimental (medicine) sessions

4.10 Opening review and guided progressive muscle relaxation

Purpose / when: the start of each medicine day, to settle the client and re-establish the frame. [source: Danforth et al. 2018 (guided progressive muscle relaxation); MAPS manual 2010 pp. 19–20 (opening review)]
Steps: (1) Check in on sleep, worries and current state, and take a distress rating. (2) Restate the approach: largely non-directive, the client can talk any time, the touch agreement, and the agreed signals. (3) Lead a guided progressive muscle relaxation: tense and release muscle groups in turn, with slow breathing. Adapt the pace and script to the client; the trial's script is not published. [synthesis]
Example language (illustrative): "Let's start by letting the body settle. Gently tighten your hands... and let them go. Notice the difference."
Watch for: discomfort with interoceptive focus; offer an eyes-open version. Working if: the client's distress rating falls or stays steady, and they look settled.

4.11 Non-directive, inner-directed holding

Purpose / when: the default stance throughout the session. [source: MAA-1 protocol v2 p. 31; MAPS manual 2010 pp. 20, 26–27]
Steps: (1) Sit nearby, attentive and quiet. (2) Let the client lead. (3) When feelings arise, encourage them to stay present with difficult experiences rather than distract. (4) Reflect briefly, without interpreting. (5) Support emotional processing through to some resolution, at the client's pace.
Example language (illustrative): "You don't have to do anything with it. Just let it be here, and we're here with you."
Watch for: your own urge to fix things, explain, or "get ahead of" the process. Working if: the client deepens into their material with your support, without you steering.

4.12 Alternating structured and unstructured periods

Purpose / when: a feature of MAA-1 that allows relational work and free exploration in the same session. [source: MAA-1 protocol v2 pp. 16, 31–32]
Steps: (1) Structured periods: therapist-client interactions aimed at building the relationship, plus a social-perception task (the trial used the TASIT, a licensed video-based test of social inference; items are not reproduced here). (2) Unstructured periods: the client chooses an activity: art supplies, preselected music, journaling, silent introspection, or talking with the therapists. (3) Move between the two flexibly, according to the client's state. [synthesis: flexibility]
Example language (illustrative): "Would you like some time on your own with music or drawing, or would you rather talk with us for a while?"
Watch for: structured tasks feeling like a test, which can raise social anxiety. Present them as optional and low-stakes. [synthesis] Working if: the client initiates contact or conversation more easily as the day goes on. Danforth et al. report latent social skills emerging during these sessions.

4.13 Optional inward focus: eyeshades and music

Purpose / when: offered whenever the client wishes to turn inward. [source: MAA-1 protocol v2 p. 33; MAPS manual 2010 pp. 16, 20]
Steps: (1) Offer eyeshades and headphones; the client decides. (2) Use preselected music, mostly without lyrics, that supports the experience without being intrusive. (3) The client can remove them at any time to talk.
Example language (illustrative): "The eyeshades are here if you'd like them, or not. You can take them off any time."
Watch for: sensory overload from the music or headphones; lower the volume or turn it off. [synthesis] Working if: the client uses inward periods productively and comes out of them willing to share.

4.14 Check-ins

Purpose / when: when the client has been silent for a long time. In MAA-1, a check-in was made after the first hour if the participant had not spoken. [source: MAA-1 protocol v2 p. 33; MAPS manual 2010 p. 22]
Steps: (1) Ask quietly about what the experience is like. (2) Depending on the answer, either encourage a return to inner focus or invite sharing. (3) Take regular distress ratings (MAA-1 used a subjective distress scale at regular intervals) and bring in extra support if they rise.
Example language (illustrative): "Just checking in. Would you like to say anything about what's happening, or go back inside?"
Watch for: too-frequent check-ins that interrupt the process. Working if: the client can answer briefly and choose where to go next.

4.15 Gentle redirection of interpersonal avoidance

Purpose / when: when the client repeatedly avoids interpersonal contact. [source: MAA-1 protocol v2 p. 31; MAPS manual 2010 pp. 25, 27]
Steps: (1) First, accept the direction they have taken. (2) If avoidance keeps recurring, offer "gentle guidance or redirection as a choice", framed as joint exploration. (3) If the client intellectualises or distracts, offer gentle options, not instructions.
Example language (illustrative): "I notice we keep drifting away from talking together. Would you be willing to explore what happens when we stay in contact for a moment? It's your choice."
Watch for: the redirection turning into pressure; withdrawal can also be sensory self-protection. [synthesis] Working if: the client re-engages by choice, or explains why they need distance.

4.16 Stepwise response to anxiety and distress

Purpose / when: anxiety (the most common in-session reaction in the trial), fear or a sense of losing control. [source: Danforth et al. 2018; MAPS manual 2010 pp. 28–30]
Steps: follow the manual's escalation order: (1) reassure the client that they are safe; (2) invite body awareness; (3) invite breath, including breathing into the feeling; (4) offer gentle touch only within the prior agreement and with permission in the moment; (5) explore the content of what is frightening; (6) invite a return to inner focus; (7) consider focused bodywork only if needed and only if agreed. If distress becomes severe, inform the study physician, who holds the protocol's medical options. [source: MAA-1 protocol v2 p. 16]
Example language (illustrative): "You're safe here. We're right beside you. Can you feel your feet? Let's breathe together."
Watch for: dissociation, panic, or suicidal statements (structured suicide assessment is repeated during each medicine day). Working if: distress ratings fall, and the client returns to engagement within the optimal arousal zone.

4.17 Accommodating stimming and self-regulation

Purpose / when: throughout the session. [source: Danforth et al. 2018]
Steps: (1) Have fidget objects in reach. (2) Treat repetitive movement as self-regulation and do not interrupt it. (3) Allow typed or text-to-speech communication. [source: MAA-1 protocol v2 p. 21]
Example language (illustrative): "Use whatever helps you feel steady. You can type to us if talking is hard."
Watch for: your own urge to read stimming as distress; ask the client instead. [synthesis] Working if: the client regulates without needing you to step in.

4.18 Session closing and handover

Purpose / when: the end of each medicine day. [source: Danforth et al. 2018; MAA-1 protocol v2 pp. 27, 34]
Steps: (1) Hold a brief closing: what stood out, and how they are now. (2) Repeat the suicide assessment and confirm the client is stable with the medical team. (3) Hand over to the support partner. Protocol v2 required willingness to stay overnight at the study site. The 2018 paper reports that the support partner drove the participant away from the centre and stayed with or near them until the next-morning visit. [source: MAA-1 protocol v2 p. 22; Danforth et al. 2018] Follow your own protocol. (4) Give the 24-hour contact details. (5) Encourage rest and quiet reflection, and practise self-soothing before the client leaves. [source: MAPS manual 2010 p. 31]
Example language (illustrative): "Today might keep unfolding overnight. Rest, jot down anything you want to remember, and call us if you need to."
Watch for: low mood after the session. Depressed mood was reported in some MDMA participants. [source: Danforth et al. 2018] Working if: the client leaves settled, with a clear safety plan.

Phase 3: Integration

4.19 Next-morning integrative session

Purpose / when: the morning after each medicine session, about 60 min. [source: MAA-1 protocol v2 pp. 34–35; Danforth et al. 2018]
Steps: (1) Review the previous day: what emerged and what it meant to the client. (2) Assess emotional stability and repeat the suicide assessment. (3) Plan the return to daily life. (4) Begin linking insights to everyday social situations, for example carrying "acceptance" and "reduced fear" into specific settings. (5) Discuss how and with whom to share the experience, using discretion. [source: MAPS manual 2010 p. 18]
Example language (illustrative): "What from yesterday would you most like to keep? Where in your week might that show up?"
Watch for: over-reliance on the medicine-day state ("I was only OK because of the drug"). Point to the capacities that were already there. [synthesis] Working if: the client names one or two concrete situations in which to try something different.

4.20 Daily telephone contacts

Purpose / when: daily for 7 days after each medicine session, usually 5–15 min, extended if the client needs to process more. [source: MAA-1 protocol v2 pp. 35–36; Danforth et al. 2018]
Steps: (1) Ask about wellbeing, sleep and mood. (2) Screen for safety. (3) Briefly support processing. (4) Record any adverse reactions.
Example language (illustrative): "How have the last 24 hours been? Anything that's worried you?"
Watch for: new or worsening symptoms, or suicidal thinking; escalate to an in-person session. Working if: the client is stable and engaged.

4.21 Later integrative sessions and expressive integration

Purpose / when: two further in-person sessions after each medicine session (about 2 weeks apart in the 2018 report), plus 6-month follow-up and an optional extra visit. [source: Danforth et al. 2018; MAPS manual 2010 pp. 31–32]
Steps: (1) Encourage writing, art and dream records between sessions. (2) Review real social encounters since the session and the client's experience of them. (3) Reinforce the mindfulness skills (4.4) as a bridge back to daily life. [synthesis] (4) Before a second medicine session, review the first and update the intentions. (5) At the end, make an exit plan and refer for ongoing therapy if needed. [synthesis]
Example language (illustrative): "You mentioned it was easier to start a conversation with your sister. What did you notice in yourself when that happened?"
Watch for: disappointment when the sense of ease fades; normalise it and rehearse the skills. [synthesis] Working if: the client reports, and their support partner notices, improved interactions. Partners described this in the trial.

Outside trials

4.22 Harm-reduction and integration support for clients using MDMA outside legal settings [synthesis]

Purpose / when: a client tells you they plan, or have had, an MDMA experience outside a lawful programme. You do not facilitate, supply or guide use. [synthesis]
Steps: (1) Assess without judgement: history, mental state, suicide risk, cardiovascular and other medical issues, and current medicines. (2) Explain in general terms that medicine interactions and medical conditions can make use dangerous, and that the trials excluded people with significant heart, liver, seizure or psychiatric conditions and those on certain medicines. Advise a medical review. Do not give dosing or sourcing information. (3) Make a safety plan: a trusted sober person present, emergency services contacts, and an agreement to seek help if they become unwell. (4) Use the mindfulness skills (4.4) for grounding afterwards. (5) Offer integration sessions along the lines of 4.19–4.21. (6) Refer to lawful trials or programmes if appropriate.
Example language (illustrative): "I can't help you get or use it, but I can help you think about the risks and make sense of whatever happens. Your safety is my priority."
Watch for: acute distress, mania, psychosis or suicidal thoughts afterwards; refer urgently. Working if: the client is safer, better informed, and stays engaged in therapy.

5. Handling problems

Stuck client: little effect, or nothing "happens". In MAA-1, one MDMA participant showed few of the expected effects. That participant had stopped an SSRI some weeks earlier. [source: Danforth et al. 2018] Response: don't push for an experience. Use the unstructured options (art, journaling, quiet) and treat the day as rapport time. Leave any explanation of pharmacology to the physician. In a blinded trial, remind the client that a placebo day is still a therapy day. [synthesis]
Repeated avoidance or intellectualising. Response: accept it at first. If it recurs, offer gentle choices and joint exploration rather than directives (4.15). [source: MAA-1 protocol v2 p. 31; MAPS manual 2010 pp. 25, 27]
Resistance in preparation (doubts, ambivalence). Response: say clearly that participation is voluntary, and explore the concerns as valid information. Add a preparatory session if needed; two MAA-1 participants had one for clinical reasons. [source: Danforth et al. 2018; MAPS manual 2010 pp. 14–15]
Overwhelm, panic or fear of losing control. Response: follow the stepwise sequence in 4.16. If distress is severe and does not respond, involve the study physician. After a very difficult session, offer extended support, including daily meetings. In extreme cases, hospitalisation is considered. Decide on a second session only after assessment and with the client's agreement. [source: MAPS manual 2010 pp. 29–31]
Sensory overload (autism-specific). Response: lower light and sound, remove the headphones, offer fidget objects and a break in the private bathroom, and switch to eyes-open grounding. Plan likely triggers during preparation. [source: Danforth et al. 2018 (setting); responses synthesis]
Communication breakdown during altered states. Response: fall back on the signals agreed in preparation: numbers, single words, typing. Ask yes/no or rating questions. [synthesis, building on the skills rationale in Danforth et al. 2018]
Suicidal ideation. In MAA-1, ideation occurred equally in both arms and was often pre-existing. One participant with a history of suicidal behaviour had brief ideation during a personal crisis, which resolved. [source: Danforth et al. 2018] Response: use the structured suicide assessment at every contact and on medicine days. Make a written safety plan with the support partner, and keep 24-hour access to the team. Lethal-means counselling covers safe storage and removal only. Escalate to emergency care if risk is imminent. [synthesis per standard practice]
Low mood or a "crash" after the session. Depressed mood was reported in a quarter of MDMA participants. [source: Danforth et al. 2018] Response: normalise it, increase phone contact, and screen for depression. Bring forward an integrative session. [synthesis]
Non-adherence to study rules (e.g., other substances, restarting medicines without telling the team, driving after a session). Response: review the rules again in a non-shaming way. Tell the medical team, who decide whether it is safe to continue. Rules exist for safety, not as a test of commitment. [synthesis; rules from MAA-1 protocol v2 pp. 21–22, 31–32]
Rupture with a therapist (e.g., the client feels misunderstood, or a check-in felt intrusive). Response: name it, apologise for your part, and ask what would help. Use the two-therapist team: the other therapist may be easier for the client to approach. Review the video in supervision. [synthesis; supervision via video from MAA-1 protocol v2 p. 17]
Boundary pressure around touch or closeness. Response: stick to the prior written agreement. "Stop" always ends touch. There is never sexual contact. Discuss what happened in integration and in supervision. [source: MAPS manual 2010 p. 16]
Inappropriate support partner. Response: the client may choose any trusted adult. If the partner is intrusive, critical or unreliable, help the client choose someone else before the session. [source: Danforth et al. 2018 (choice of partner); response synthesis]
Unmasking of latent social ability, followed by disappointment. Response: frame the skills seen on medicine day as already present. Help the client find small real-world steps, and keep practising the mindfulness skills. [synthesis, building on Danforth et al. 2018]
Adverse effects of mindfulness practice (anxiety, depersonalisation, trauma re-experiencing). Response: use shorter practices with eyes open, external anchors and choice. Stop if symptoms escalate. Screen for a history of trauma, dissociation, psychosis or panic before starting, and keep practice intensity low. [synthesis] Adverse events, most often anxiety, depression and cognitive anomalies, occur in about 8% of meditators, including some with no psychiatric history. [source: Farias et al. 2020] Effects can last long after practice, and how intensely someone practises and the teacher relationship both matter. [source: Lindahl et al. 2017]
Client using MDMA outside legal settings, or asking you to guide a session. Response: decline to facilitate. Offer harm-reduction and integration support (4.22). Know your professional body's rules and local law on illegal drug use and confidentiality. [synthesis]
Over-generalisation. MAA-1 was a 12-person pilot in autistic adults with social anxiety. SAMATI adapts the general frame for non-autistic adults with social anxiety disorder, using ACT and EFT. Response: don't present this model as established outside its research context. [source: Danforth et al. 2018; Lear et al. 2023]

Sources

Sources used:

Further sources:

Also cited:

Danforth AL, Struble CM, Yazar-Klosinski B, Grob CS (2016). MDMA-assisted therapy: a new treatment model for social anxiety in autistic adults. Progress in Neuro-Psychopharmacology & Biological Psychiatry 64:237–249 (online March 2015). This is the treatment-model paper. A Blossom research summary describes preparatory psychotherapy with minimal psychodynamic work and an emphasis on core mindfulness skills training, a male–female co-therapist team in a sensory-adapted room, and music, art, journaling and silent introspection in the medicine sessions; it does not mention touch or DBT. [source: Blossom summary; uncertain: the full text]
The MAA-1 supplementary inclusion and exclusion criteria (Danforth et al. 2018, ESM 1–2) are not summarised here. [uncertain]

← Previous · VII-058MDMA with cognitive-behavioural conjoint therapy Next · VII-060 →Ayahuasca research