This page is the reference background for the Autism and digestive symptoms page. It goes through all six papers: what they measured, what they found, and where no effect was demonstrated. For parents and treating physicians alike.
What follows summarises the studies of microbiota transfer therapy — commonly called faecal transplantation — carried out in autism. Six papers deal with this procedure, which is based on altering the gut microbiota; we go through them all, together with their limitations.
The data are strongest in relation to the digestive symptoms. In the open-label study of an Arizona research group, the severity of gastrointestinal symptoms changed as follows compared with baseline:
Change compared with the baseline GSRS score[G], the same group of 18
Open-label study, 18 participants (7–16 years), with two-year follow-up, without placebo control. These are the same 18 children measured at two time points 122, 123, not a pooling of several independent studies. The 2025 systematic review found symptom reductions of 35–82% in the uncontrolled studies.126
On the behavioural symptoms: the primary endpoint of both controlled studies was negative. In the only double-blind, placebo-controlled trial to date (103 children: 52 treated, 51 placebo), on the primary endpoint — the social responsiveness scale — there was no significant difference between the two groups. On one secondary scale (the Vineland-3 socialisation domain) a difference did appear in favour of the treated group.124 In a controlled observational study, the autism score based on structured clinical observation likewise showed no significant between-group difference, while the parent- and physician-rated scales and the digestive measures did.125
The more striking figures — the autism severity score 47% lower at two years in the 18-participant study, and the fact that 44% of the participants fell below the diagnostic threshold 123 — come from an uncontrolled, open-label study. The authors themselves note that 12 of the 18 participants changed their medication, their food supplements or their diet over the two years, so the improvement cannot be attributed unambiguously to the treatment.
The 2025 systematic review describes the same pattern: all the before-and-after studies reported improvement, whereas on the primary endpoints the controlled studies did not confirm the effect.126
The data on the safety of using FMT are extremely reassuring. The 2025 systematic review pools 9 studies, the data of 394 children and adolescent participants (2–17 years) in total; of the nine, six examined safety, and none of them classified an adverse event as serious (three studies did not address the question). The complaints reported were mild and intermittent — transient diarrhoea, abdominal discomfort, slightly raised temperature, skin rash and, in some studies, irritability, restlessness and sleep disturbance.126 In the double-blind trial, adverse events were more frequent in the FMT arm (23% vs. 14%); the most common event was fever, which occurred at a similar rate in both arms, and the difference was mainly due to the neurological and psychiatric events recorded (8 vs. 2). No serious event occurred in either arm.124
The facts and figures on this page come from six papers in all: two open-label studies and the two-year follow-up of one of them, one double-blind, placebo-controlled trial, one controlled observational study and two systematic reviews. In total this covers the data of a few hundred children — the largest review pools 394 participants.
What this literature does support: improvement in the digestive symptoms appears consistently across the studies, and the safety data on the treatment are reassuring — no serious adverse event has so far been reported in children. What it does not support: a conclusion about the behavioural symptoms. On the primary endpoint of both controlled studies to date, no significant difference was demonstrated; improvement appears more consistently on the parent- and clinician-rated scales than on structured clinical observation or the standardised primary instruments. The uncontrolled studies show a more favourable picture than this, but they carry a high risk of bias.
The data listed relate to the MTT (FMT) method as such, not to a specific product. The studies cited were, without exception, carried out in children living with autism spectrum disorder; clinical data on ADHD are essentially unavailable.
[122] Kang DW, Adams JB, Gregory AC, Borody T, Chittick L, Fasano A, Khoruts A, Geis E, Maldonado J, McDonough-Means S, Pollard EL, Roux S, Sadowsky MJ, Lipson KS, Sullivan MB, Caporaso JG, Krajmalnik-Brown R. Microbiota Transfer Therapy alters gut ecosystem and improves gastrointestinal and autism symptoms: an open-label study. Microbiome 2017;5(1):10. doi:10.1186/s40168-016-0225-7
Open-label, uncontrolled clinical trial in 18 children aged 7–16 living with autism spectrum disorder. The protocol comprised two weeks of antibiotic pre-treatment, a bowel cleanse, a stomach-acid suppressant and then extended FMT: a high initial dose followed by daily maintenance dosing for 7–8 weeks. Gastrointestinal symptoms (GSRS) fell by roughly 80% by the end of treatment, and the improvement persisted 8 weeks later; behavioural symptoms also improved significantly, and bacterial diversity increased. Limitations: no placebo group and no blinding, n = 18, the scales are largely parent-reported, and since the protocol also included antibiotics and a bowel cleanse, the effect cannot be attributed to FMT alone.Source: references-v2.bib · ref-122
[123] Kang DW, Adams JB, Coleman DM, Pollard EL, Maldonado J, McDonough-Means S, Caporaso JG, Krajmalnik-Brown R. Long-term benefit of Microbiota Transfer Therapy on autism symptoms and gut microbiota. Sci Rep 2019;9(1):5821. doi:10.1038/s41598-019-42183-0
The [122] study: a two-year follow-up in the same 18 participants. Most of the gastrointestinal improvement was retained, and the autism-related symptoms have, since the end of treatment, improved further. The microbiota changes — higher bacterial diversity, greater proportions of Bifidobacterium and Prevotella — were still present at follow-up. Limitations: this is not a new study but the follow-up of the same 18 people; there is no control group and no blinding, and the children's natural development over two years and the effect of concurrent therapies cannot be separated out — the authors themselves urge a double-blind, placebo-controlled trial.Source: references-v2.bib · ref-123
[124] Wan L, Wang H, Liang Y, Zhang X, Yao X, Zhu G, Cai J, Liu G, Liu X, Niu Q, Li S, Zhang B, Gao J, Wang J, et al. Effect of oral faecal microbiota transplantation intervention for children with autism spectrum disorder: A randomised, double-blind, placebo-controlled trial. Clin Transl Med 2024;14(9):e70006. doi:10.1002/ctm2.70006
Randomised, double-blind, placebo-controlled trial of orally administered FMT in 103 children living with autism spectrum disorder (52 treated, 51 placebo). In this field it is the only trial of this design to date — all the other papers are open-label or non-randomised. On the primary endpoint — the social responsiveness scale — there was no significant difference between the groups (p=0.362), nor on the ABC (p=0.569); the trial did not measure any digestive outcome. On one secondary scale, the Vineland-3 socialisation domain, a difference appeared in favour of the treated group (p=0.037). Adverse events were more frequent in the FMT arm (23% vs. 14%); fever occurred at a similar rate in both arms, and the difference was mainly due to the neurological and psychiatric events recorded (8 vs. 2). No serious event occurred in either arm.Source: references-v2.bib · ref-124
[125] Varnas D, Kunevičius A, Burokas A, Urbonas V. Fecal Microbiota Transplantation for Autism Spectrum Disorder in Children: Results from a Prospective Open-Label Controlled Observational Study. Medicina (Kaunas) 2026;62(1):65. doi:10.3390/medicina62010065
Prospective, single-centre, open-label, controlled — but not randomised study of colonoscopic FMT in 30 children (15 in the intervention group, 15 controls). The key result: at week 8, on the prespecified primary endpoint — the change in ADOS score — there was no significant difference between the groups. On the secondary measures there were differences: CARS (p<0.001), overall parental impression (p<0.001), CBCL internalising problems (p=0.001) and GSRS (p=0.037); the CARS and PGI-R improvements persisted at 6 months. No serious adverse event occurred, and three mild, self-limiting events were recorded; the authors regard the findings as exploratory.Source: references-v2.bib · ref-125
[126] Liber A, Więch M. The Impact of Fecal Microbiota Transplantation on Gastrointestinal and Behavioral Symptoms in Children and Adolescents with Autism Spectrum Disorder: A Systematic Review. Nutrients 2025;17(13):2250. doi:10.3390/nu17132250
A systematic review conducted according to PRISMA 2020 and registered in PROSPERO, on the effect of FMT on the digestive and behavioural symptoms of children and adolescents living with autism. Nine studies were included: 2 randomised controlled trials and 7 before-and-after studies. The before-and-after studies all reported improvement, whereas the RCT results were inconsistent — the former showed a high risk of bias, the latter a low one. The authors conclude that, because of the methodological limitations, no clear conclusion can be drawn about the effectiveness of FMT — this entry is therefore a cautionary source, not a citation supporting efficacy.Source: references-v2.bib · ref-126
[127] Taha H, Issa A, Muhanna Z, Al-Shehab M, Wadi T, Awamleh S, Ateiwi YA, Abusido M, Berggren V. Microbiota-based interventions for autism spectrum disorder: a systematic review of efficacy and clinical potential. Front Microbiol 2025;16:1648118. doi:10.3389/fmicb.2025.1648118
A systematic review conducted according to PRISMA and registered in PROSPERO (CRD42024615043) on the full range of microbiota-based interventions: probiotics, prebiotics, synbiotics and FMT. 33 studies were included (16 RCTs), of which 9 assessed FMT. It was FMT that showed the most consistent and most sustained improvement on both the behavioural and the digestive outcomes; adverse events were minimal, predominantly transient digestive symptoms. The authors urge standardised protocols and larger controlled trials. Worth reading together with review [126]: the two works strike a different balance on the same literature.Source: references-v2.bib · ref-127
If you want to find out whether your child is suitable, what happens when and what it costs, you will find that on the service page.
The MicroBiome Bank service is not a substitute for professional medical advice, diagnosis or treatment. The decision to use it rests in every case with the treating physician.