“Parents should not be in the room with the therapists during OT and ABA therapies”

Many clinics tell caregivers that sessions go better without them, because children look to a parent instead of the therapist and behave differently when a parent is watching. That observation is real for some children in some sessions, and it has hardened into a general rule that parents belong outside the therapy room.

• Parent-mediated intervention, where the caregiver is coached in the room and then practises with their child, is one of the most rigorously tested approaches in early autism support. The PACT randomised controlled trial of 152 children (Green et al., Lancet, 2010) found improvements in social communication, and the follow-up nearly six years later (Pickles et al., Lancet, 2016) found the relative gains were sustained, which is unusual in this field.
• The Cochrane review of parent-mediated early intervention (Oono, Honey and McConachie, 2013, 17 randomised trials) is more cautious. It found gains in parent-child interaction and some evidence on language comprehension and autism severity, but results across most outcomes were inconsistent and the quality of the evidence was low. Involving parents is well supported as good practice; it is not a guaranteed result.
• A systematic review and meta-analysis of 30 randomised trials with 1,934 children (Frontiers in Psychiatry, 2021) compared parent-mediated intervention with no such intervention on adaptive functioning and other outcomes, and found effects that were generally small and rated low-certainty under GRADE.
• On the occupational therapy side, an AOTA evidence-based practice systematic review of family-centred interventions for autistic children (Watling, Benevides and Robertson, AJOT, 2023) found emerging evidence supporting caregiver coaching for both child and parent outcomes, including the child's occupational performance and the caregiver's sense of competence.
• A broader systematic review of paediatric occupational therapy (Novak and Honan, 2019) reported that parent-delivered intervention appears about as effective as therapist-delivered intervention, and named collaboration with parents as a key point for practice.
• No study was found that tested excluding caregivers from the room as an intervention and showed it produced better outcomes. The case for keeping parents out rests on clinical observation about individual children, not on trial evidence.
• The BACB Ethics Code for Behavior Analysts (Standard 2.09) requires behaviour analysts to make reasonable efforts to involve caregivers in treatment, and many insurers require documented caregiver training. A blanket no-parents policy sits awkwardly against that requirement.

The grain of truth here is genuine. Some children do behave differently when a parent is in the room: they turn to the parent for help instead of attempting the task, or a hard-won routine falls apart because the safest, most rewarding option is sitting on the chair behind them. In those cases a therapist may reasonably ask you to step out of a specific session, or to watch from behind glass or on video, while your child builds a working relationship with the therapist and learns that the skill belongs to them and not only to you.

What makes the statement misleading is the jump from that specific, time-limited clinical judgement to a standing rule. The strongest evidence in early autism support does not point away from caregivers. It points at them. PACT, the trial with the longest sustained effect in this field, worked by coaching parents directly and having them practise with their child every day. Occupational therapy has moved in the same direction, towards coaching families and building strategies into home routines, because that is where a child actually needs the skill. A skill that only appears in a therapy room with one particular therapist has not finished being learned. Somebody has to carry it into mornings, mealtimes and school, and that somebody is usually you.

It is also worth being honest about what the evidence does not show. The Cochrane review and the later meta-analysis both found effects that were often small, inconsistent and low-certainty, so involving caregivers is not a switch that guarantees progress. But the asymmetry matters: there is real trial evidence supporting caregiver involvement, and none at all supporting the opposite policy. And there is a second reason that has nothing to do with effect sizes. Being able to see what happens to your child is how you notice whether the approach is respectful, whether your child is distressed, and whether the goals being worked on are the ones you agreed to. Concerns raised about ABA in particular, including accounts from autistic adults of lasting distress, are an argument for more caregiver visibility, not less.

• A therapist asking you to step out of a particular session, for a stated reason and for a defined period, is normal practice and worth trying. A service that will not let you in at all, ever, is a different thing, and it is fair to ask why.
• Ask what the plan is for teaching you the strategies, not just whether you can watch. Scheduled caregiver coaching, video review, or a session where the therapist coaches you while you work with your child are all reasonable answers.
• If your child needs to work without you present for a while, ask what would have to be true for that to change, and when it will be reviewed. Fading your presence should have an endpoint.
• Watching from behind glass, from a video feed, or from the back of the room are middle options. Not sitting in the chair beside your child does not have to mean not seeing the session.
• For ABA specifically, caregiver involvement is an expectation in the profession's own ethics code, so it is reasonable to raise that if you are being shut out entirely.

• This is about how services are usually organised, not about any individual child. A therapist who knows your child may have a good, specific reason for asking you to step out of a session, and that judgement deserves a hearing.
• Most of the trial evidence here concerns early intervention with young children. How much it tells us about a twelve-year-old in a weekly OT session is genuinely unclear.
• The evidence for parent-mediated intervention is supportive but not strong across the board. The Cochrane review rated the quality of the evidence as low and found inconsistent results on several outcomes.
• Practical constraints are real. Small therapy rooms, siblings, work schedules and clinic rules all shape these policies, and not every no-parents rule is a clinical claim in disguise.

• Parent-mediated social communication therapy for young children with autism (PACT): long-term follow-up of a randomised controlled trial — Pickles A, Le Couteur A, Leadbitter K, et al., 2016, The Lancet 388(10059):2501-2509
• Parent-mediated early intervention for young children with autism spectrum disorders (ASD) — Oono IP, Honey EJ, McConachie H, 2013, Cochrane Database of Systematic Reviews, CD009774
• Parent-Mediated Interventions for Children and Adolescents With Autism Spectrum Disorders: A Systematic Review and Meta-Analysis — Frontiers in Psychiatry, 2021, 30 randomised controlled trials, 1,934 participants
• Family-Centered Interventions for Children on the Autism Spectrum (2013-2021) — Watling R, Benevides T, Robertson SM, 2023, American Journal of Occupational Therapy 77(Suppl 1)
• Effectiveness of paediatric occupational therapy for children with disabilities: A systematic review — Novak I, Honan I, 2019, Australian Occupational Therapy Journal 66(3):258-273
• Ethics Code for Behavior Analysts (Standard 2.09, Involving Clients and Stakeholders) — Behavior Analyst Certification Board, 2020, effective 2022
• Building Collaborative Partnerships between Behavior Analysts and Families — Behavior Analysis in Practice, 2024

• Does the Order of OT and ABA Therapy Matter? — General (Summary)
• We shouldn't do brushing in OT to kids with epilepsy and seizure history — Sensory (Myth Check)
• the fastest development in children can happen during 3 to 5 years of age — General (Myth Check)