Not yet established care
Research-stage treatments
This page cuts across the others. Whatever a treatment is made of — cells, light, peptides or a named physiotherapy approach — what matters to a family weighing it up is how far the evidence has actually got. Everything here is still being worked out.
These are options that may reasonably be considered for a child who has not responded adequately to conventional treatment, or whose progress has plateaued. Research-stage does not mean worthless and it does not mean proven; it means the question is open. A treatment in registered clinical trials is being tested properly and an answer is coming; a treatment with only early research behind it is being used ahead of its evidence; and a few have been tested and not supported. These are three different positions and are set out separately below.
An unproven treatment carries the risk that the answer, when it arrives, is negative. That risk is real and should be stated. It should also be put in proportion: licensed medicines are not free of risk either, and serious adverse effects are often identified years after authorisation. The difference is not that newer treatments carry risk and established ones do not — it is that for the newer ones the information has not yet been collected. What is being weighed is uncertainty, not the absence of danger on one side and its presence on the other.
In clinical trials
Under formal investigation, with registered protocols and published results. If your child fits the entry criteria, a trial is usually the best way to access these — it is supervised, it is monitored, and it is free.
Stem cell therapy
In clinical trialsMagnetic stimulation
In clinical trialsTechnology-assisted rehabilitation
In clinical trialsNamed physiotherapy approaches
In clinical trialsSensory therapies
In clinical trialsBobath / NDT
In clinical trialsConductive education
Early research only
Laboratory work, animal studies, small uncontrolled series or case reports. This is the group most often sold to families as though it were treatment. The honest position on all of it is that nobody yet knows whether it changes a child’s outcome.
Exosomes
The cell-free next step after stem cells, with genuinely interesting laboratory science — and, in children, almost no controlled clinical evidence at all.
Extracellular vesicles, MSC-derived EVs
Early research onlyMuse cells
A distinct, well-characterised cell type with an unusual biology — it appears to find damaged tissue on its own — supported by a substantial laboratory literature and an early but real clinical programme. Controlled paediatric efficacy data are not yet available.
Dezawa MuseCells, Multilineage-differentiating Stress Enduring cells, SSEA-3 positive cells, CL2020
Early research onlyPhotobiomodulation
Non-invasive, painless and simple to deliver compared with cell therapies, with a literature that is growing quickly and is mostly positive. The studies are still small and short, so the open question is not whether anything happens but how large the effect is and how long it lasts.
Transcranial near-infrared light, low-level laser therapy, PBM
Early research onlyPeptide preparations
Two quite different things share this word: prescription neuropeptide preparations used routinely in some countries, and an unregulated wellness trade. Neither has good evidence in children.
Cerebrolysin, Cortexin, Semax, and “peptide therapy” sold by wellness clinics
Early research onlyMedicinal mushrooms and nootropics
A research band worth watching, and a supplement shelf to approach carefully. Some of these compounds have real laboratory interest; almost none has been tested in children with neurological conditions.
Lion's mane (Hericium erinaceus), reishi, cordyceps, citicoline, piracetam, psilocybin
Early research onlyProbiotics, prebiotics and the gut–brain axis
Genuinely interesting science with a few solid, specific uses in children — and a large gap between those uses and what is sold to families of autistic and neurologically disabled children.
Lactobacillus, Bifidobacterium, synbiotics, fibre supplements, faecal microbiota transfer
Early research onlyMEDEK / CME
A deliberately provocative, gravity-based approach that draws strong loyalty from families and has almost no controlled evidence behind it.
Cuevas MEDEK Exercises, Dynamic Method of Kinetic Stimulation
Early research onlyMAES therapy
Theoretically the most modern of the named approaches — it is built on current motor-learning thinking — and the one with the least published evidence, because it is also the newest.
Mobilisation of Awareness, Exploration and Strategies
Early research onlyAnat Baniel Method
Gentle, attention-focused movement work with a clear conceptual framework, a devoted following, and no controlled trial evidence of functional benefit in children.
NeuroMovement, ABM, Feldenkrais-derived
Early research onlyVojta therapy
Pressure applied at defined body points to trigger whole-body movement patterns. Long-established in some countries, little tested, and the child's distress is a serious part of the discussion.
Reflex locomotion
Tested and not supported
These have had their chance. Where a treatment has been studied and the evidence came back negative, that is a result in itself — and a reason to put the time elsewhere.
Chelation therapy
The idea behind it was tested and failed. There is no trial evidence that it helps an autistic child, there is a documented record of children harmed, and at least one child has died. This is the clearest no on this site.
DMSA (succimer), DMPS, EDTA, “heavy metal detox”, provoked urine testing
Not supported by evidencePatterning
The one approach on this site that has been tested, rejected, and formally advised against — repeatedly, over more than fifty years.
Doman–Delacato method, Institutes for the Achievement of Human Potential
How to judge a therapy sets out the questions that separate a treatment worth trying from one worth declining, and Safety covers what can actually go wrong.