Effect of nutritional supplements on gut microbiome in individuals with neurodevelopmental disorders: a systematic review and narrative synthesis

Abstract

Background

Neurodevelopmental disorders (NDDs) encompass a range of disruptive conditions with varying prevalence rates and multiple contributing factors. Recent studies have suggested a potential connection between NDDs and the gut-brain axis. Furthermore, there is evidence indicating that nutritional supplements might have an impact on gastrointestinal (GI) and behavioral symptoms. This study aimed to explore the effects of nutritional supplements on the gut microbiota and behavioral symptoms in individuals with NDDs.

Methods

A systematic search of databases such as PubMed, Scopus, Web of Science, Embase, and APA PsycINFO was conducted, utilizing relevant keywords until February 2025. In addition, the search for gray literature was carried out on Google Scholar and ProQuest. The risk of bias was assessed using the ROBINS-I tool for non-randomized studies and the RoB-1 tool for randomized controlled trials. Due to the heterogeneity of the studies, a Synthesis without Meta-analysis (SWiM) approach was employed.

Results

The overall findings from the studies indicated positive effects of supplementation in reducing the Gastrointestinal Severity Index (GIS) score and alleviating GI symptoms. Supplementation with probiotics and vitamins increased good microbiomes (GM) and decrease in bad microbiomes (BM) among individuals with autism spectrum disorder (ASD). Moreover, the Firmicutes to Bacteroidetes ratio (F/R ratio) exhibited significant changes after supplementation. Additionally, improvements were observed in various assessment scores, including ATEC, ABC, CARS, and PGI-2.

Conclusions

Nutritional supplementation in individuals with NDDs can have a positive influence by modulating the microbiome, reducing dysbiosis, and enhancing gut barrier integrity. Shifting in the F/R ratio can be considered as the reason for improving gastrointestinal and behavioral symptoms by influencing neurotransmitter activity and neuroinflammation. Targeting the gut-brain axis with interventions that focus on gut microbiota offers a promising adjunct therapy for the management of NDD.

Registration of the review protocol.

PROSPERO registration no. CRD42023460449.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40795-025-01043-z.

Background

According to the latest edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), neurodevelopmental disorders (NDDs) are a group of conditions with developmental deficits that accompany impairments in social, cognitive, and emotional functioning [1]. These intellectual disabilities include autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), specific learning disorders (SLD), communication disorders (CDs), and motor disorders (MDs) [2]. The prevalence rates of these conditions vary, with ADHD affecting 5–11% of the population, ASD 0.7–3%, SLD 3–10%, CD 1–3.4%, and MD 0.8–17% [3].

The etiology of NDDs is multifactorial. Several environmental, familial, and genetic factors are likely to be responsible for NDD development, including exposure to trauma and toxic chemicals such as lead, mercury, and organophosphate pesticides [4]. Additionally, maternal exposure to recreational drugs, alcohol, certain medications, stress during pregnancy, neonatal prematurity, and low birth weight have been identified as potential risk factors [4–6]. More recently, the gut-brain axis has emerged as a potential key player in the etiology and clinical symptoms of NDDs [7]. This axis composed of a complicated crosstalk between brain and bowel microbiota, integrates various neurologic pathways including the brain, spinal cord, hypothalamus-pituitary axis, autonomic nervous system, and enteric nervous system [8]. Evidence strongly suggests that gut microbiota (GM) by regulating brain chemistry as well as the neuroendocrine system, can affect both brain health and the gastrointestinal (GI) system [9, 10]. For instance, studies have reported that the gut-brain axis plays a role in developing irritable bowel syndrome (IBS), inflammatory bowel disease (IBD), epilepsy, migraine headaches, cognitive disorders, and also NDDs [11, 12].

Moreover, up to half of individuals with NDDs experience GI symptoms [13]. These symptoms include constipation, diarrhea, bloating, abdominal pain, nausea, and gastroesophageal reflux, which can impact quality of life and may even exacerbate behavioral and cognitive difficulties [13, 14]. Although the etiology of the frequent GI symptoms in individuals with NDDs is not well understood, they have been proposed to be linked to the gut-brain axis and therefore GM too [14].

It is worth mentioning that the majority of the GM effects appear to be strain-specific [9]. This suggests that dietary habits and certain nutrients influence GM strains, the gut-brain axis, and the development of certain diseases or clinical outcomes [15, 16]. Probiotics consumption and micronutrient intake have been associated with improvements in both GI symptoms and behavioral problems in children with ASD [17, 18]. Furthermore, research has demonstrated that even short-term consumption of special diets consisting of animal vs. plant products can alter GM community structure, leading to inter-individual differences in microbial gene expression [19]. Interestingly, an increase in the activity of Bilophila wadsworthia, which is abundant in animal-based diets, has been shown to support the association between dietary fat, bile acids, and the growth of microbiotas capable of triggering IBD. In addition, adopting a Western diet has been linked to a decrease in Bacteroidetes levels and an increase in Firmicutes and Proteobacteria [20–22].

There is growing evidence regarding the beneficial effects of altering GM in NDDs. Some clinical trials reported the effect of probiotics supplementation in improving GI symptoms in ADHD patients [17, 23]. Furthermore, several systematic reviews have investigated the role of gluten- and casein-free diets [24], ketogenic diets [25], and specific carbohydrate diets in improving GI symptoms in ADHD patients [26]. Moreover, the role of some vitamins in improving behavioral symptoms in individuals with NDDs has been discussed [27, 28].

Despite these advances, existing research largely focuses on the effect of supplementation either on the gut microbiota or clinical outcomes in isolation, and to our knowledge, there has been no study to systematically review the existing evidence on the effect of nutritional supplements on both GM and clinical outcomes in NDDs. Despite the growing evidence on the role of nutritional supplements in modulating gut microbiota and improving clinical outcomes in individuals with NDDs, a gap remains in systematically evaluating their combined effects. Therefore, this study aims to systematically review and narratively synthesize data from studies assessing the role of nutritional supplementation on behavioral and gastrointestinal symptoms of individuals with NDDs via modifications of gut microbiota.

Methods

The current systematic review which was registered in PROSPERO (CRD42023460449), has been written based on the guidelines provided in the Cochrane Handbook for Systematic Reviews of Interventions [29] as well as the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) [30].

A thorough search was done up to February 2025 by two independent authors (ED and KK) in PubMed, Scopus, Web of Science, Embase, and APA PsycINFO to identify the eligible papers. In addition to electronic database searches, the gray literature was explored using Google Scholar, where the first 100 results were screened, and ProQuest dissertations and theses. A detailed search strategy was developed based on the population-intervention-comparison-outcome-study design (PICOS) framework to ensure the comprehensive identification of relevant studies. The population (P) comprised individuals with neurodevelopmental disorders. The intervention (I) involved nutritional supplements. Comparators (C) included standard of care, non-intervention comparators, placebo, or other interventions. The outcomes (O) evaluated changes in gut microbiome composition or gastrointestinal symptoms. Eligible study designs (S) included randomized controlled trials (RCTs), single-arm studies, open-label trials, and non-randomized trials. Additional details on the PICOS framework are provided in Supplementary File 1. There was no constraint regarding the article’s language or publishing date. At the end of the search process in the mentioned databases, the references list of the relevant papers was also checked to make sure no relevant publications were going to be missed. To manage and accelerate the review process, all studies were moved to Endnote (version X9.3.3). Duplicate references were identified and removed using the built-in duplicate detection feature of the Endnote software.

Selection of studies

Studies were chosen if they met the following criteria: (a) to be a clinical trial including RCTs and non-RCTs (b) to investigate neurodevelopmental disorders as their population of interest, (c) to be an original study looking for any short- or long-term effects imposed by any type of nutritional supplement, and (d) to report sufficient data about GM as a primary or secondary outcome. Studies were excluded if they were (a) in vitro, in silico, or in vivo animal studies, and (b) ecological/ cross-sectional/ case–control studies, systematic reviews, and meta-analyses. After the initial search, the endnote software was used to screen all recorded items retrieved from the electronic search (version X9.3.3). Study titles and abstracts were reviewed according to the inclusion criteria separately by two reviewers (ED and KK). Studies that met the eligibility criteria based on the title and abstract screening were chosen for full-text review. Discrepancies between reviewers during the screening process were resolved through discussion. In cases where consensus could not be reached, a third reviewer (AHM) was consulted to make the final decision.

Outcome measures

The outcomes investigated in this study are classified into primary and secondary outcomes. The primary outcomes consist of the Gastrointestinal Severity Index (GIS), Gastrointestinal Symptoms (GI), Good Microbiome, Bad Microbiome, and Firmicute to Bacteroidetes (F/R ratio) in individuals with NDDs. In addition, secondary outcomes are also divided into two categories: social skills and symptom severity. Autism Behavior Checklist (ABC), Aberrant Behavior Checklist-Second Edition (ABC-2), and Social Responsiveness Scale (SRS) were tools related to social skill assessment in the included studies. Moreover, tools regarding the assessment of symptom severity were ADHD-IV-RS (ADHD rating scale), ATEC (Autism Treatment Evaluation Checklist), CARS (Childhood autism rating scale), CGI (Clinical Global Imprisonment), and PGI-2 (Parent Global Impressions — Revised-2) in the final selected studies.

Data extraction

Two separate researchers extracted the data from each selected article (KK and ED). The following information was extracted from each study: first author's name, year of publication, country, study design, duration of intervention, subjects' age and gender, type of intervention (probiotic, vitamin, and other nutritional supplementation (ONS)), number of participants, type of neurodevelopmental disorders, comparison group, and primary/secondary outcomes.

Risk of bias assessment

Risk of bias in the selected studies was independently assessed by two authors (ED and SA) using the ROBINS-I tool (Risk of Bias In Non-randomized Studies–of Interventions) for non-RCT studies and the RoB-1 tool (Risk of Bias for randomized trials) for RCTs. RoB-1 assesses bias in seven distinct domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other sources of bias. The judgments within each domain lead to an overall risk of bias of: “low risk of bias,” “unclear risk of bias,” or “high risk of bias” [31]. In addition, ROBINS-I views each study as an attempt to emulate a hypothetical pragmatic randomized trial and assesses seven domains through which bias might be introduced: bias due to confounding, bias in the selection of participants into the study, bias in classification of interventions, bias due to deviations from intended interventions, bias due to missing data, bias in the measurement of outcomes, and bias in the selection of the reported result. The judgments within each domain carry forward to an overall risk of biases of: “Low risk of bias”, “Moderate risk of bias”, “Serious risk of bias”, “Critical risk of bias” or “No information” [32]. In addition, the results of these assessments were drawn using the robvis (visualization tool) online platform.

Data analysis

According to the Synthesis without Meta-analysis (SWiM) criteria, the findings of the included articles were narratively synthesized and reported [33]. SWiM was created to promote transparency in the reporting of narrative synthesis-based systematic reviews that do not include meta-analysis. In this systematic review, a meta-analysis could not be performed due to the substantial heterogeneity in different supplement interventions and populations, the random-effects model of the outcome measures in the statistical analyses, and the study design. We summarized quantitative and qualitative results using the effect estimates, the direction of effects, the p-value of each study, and combined P-values by Fisher’s method using the Corbi packages in R software for each outcome which is shown in Table 2. It is also worth noting that the combined P-value was calculated only for studies that had reported a significance level. We grouped the outcomes into five main categories: Gastrointestinal Severity Index (GIS), Gastrointestinal (GI) Symptoms, Good Microbiome, Bad Microbiome, and F/B Ratio (Firmicutes to Bacteroidetes). Additionally, we classified the interventions into three categories based on the type of supplement (i.e. probiotic, vitamin, and ONS). We qualitatively presented the main characteristics of the studies in the corresponding Table 1 and tabulated the quantitative results separately.

Table 2.

Narrative synthesis for determining effect estimates, direction of effects, and combining P values

Study	Supplement	Effect estimate range	Positive effect (count)	Negative effect (count)	No effect (count)	p-value of each study	Combine p-value for each outcome
GIS
Adams and Holloway. 2004	VS	NA	1	0	0	0.03	< .0001
Shaaban et al. 2018	PS	-3.66	1	0	0	< 0.0001
Sanctuary et al. 2019	PS	95% CI: -1.573 to -0.177	1	0	0	0.021
Santucci et al. 2020	PS	95% CI: -8.00 to -2.00	1	0	0	0.041
Wang et al. 2020	PS	95% CI: -4.50 to -2.38	1	0	0	< 0.001
Abele et al. 2021	ONS	NA	1	0	0	0.01
Sherman et al. 2022	PS	NA	1	0	0	< 0.05
Billeci et al. 2023	PS	95% CI: 2.1–4.3	1	0	0	< 0.05
Phan et al. 2024	PS	NA	1	0	0	NA
Wang et al. 2024	PS	NA	1	0	0	< 0.05
Palmer et al. 2024	PS	NA	1	0	0	0.05
Mitchell et al. 2024	PS	NA	1	0	0	< 0.001
GI symptoms
Capra and Hannan-Jones. 1992	ONS	NA	0	0	1	> 0.05	< .0001
Adams and Holloway. 2004	VS	+ 1.0	1	0	0	0.03
Shaaban et al. 2018	PS	-0.44	1	0	0	0.002
Sanctuary et al. 2019	PS	95% CI: -1.843 to -0.032	1	0	0	0.04
Santucci et al. 2020	PS	95% CI: -2.00 to 0.00	1	0	0	< 0.001
Wang et al. 2020	PS	95% CI: -1.45 to -0.61	1	0	0	0.008
Abele et al. 2021	ONS	NA	1	0	0	0.01
Sherman et al. 2022	PS	NA	1	0	0	< 0.05
Lawrence et al. 2022	ONS	95% CI: -0.33 to 0.44	0	0	1	0.06
Billeci et al. 2023	PS	95% CI: 1.8–3.7	1	0	0	< 0.05
Phan et al. 2024	PS	NA	1	0	0	< 0.01
Hrnciarova et al. 2024	ONS	NA	1	0	0	< 0.05
Wang et al. 2024	PS	NA	1	0	0	< 0.05
Palmer et al. 2024	PS	NA	1	0	0	0.05
Mitchell et al. 2024	PS	NA	1	0	0	0.05
Good microbiome
Pärtty et al. 2015	PS	NA	1	0	0	0.03	< .0001
Liu et al. 2017	VS	NA	1	0	0	< 0.05
Shaaban et al. 2018	PS	NA	1	0	0	< 0.0001
Sanctuary et al. 2019	PS	NA	1	0	0	< 0.05
Stevens et al. 2019	VS	NA	1	0	0	0.018
Wang et al. 2020	PS	NA	1	0	0	< 0.05
wang et al. 2022	PS	NA	1	0	0	< 0.05
Sherman et al. 2022	PS	NA	1	0	0	< 0.05
Lawrence et al. 2022	ONS	NA	1	0	0	< 0.05
Raghavan et al. 2023	ONS	NA	1	0	0	0.04
Billeci et al. 2023	PS	95% CI: 10.2–14.7%	1	0	0	< 0.05
Phan et al. 2024	PS	NA	1	0	0	< 0.05
Hrnciarova et al. 2024	ONS	NA	1	0	0	< 0.05
Wang et al. 2024	PS	NA	1	0	0	< 0.05
Palmer et al. 2024	PS	NA	1	0	0	0.001
Mitchell et al. 2024	PS	NA	1	0	0	< 0.05
Trezzi et al. 2025	PS	NA	1	0	0	< 0.05
Bad microbiome
Pärtty et al. 2015	PS	NA	1	0	0	0.04	< .0001
Liu et al. 2017	VS	NA	1	0	0	< 0.05
Shaaban et al. 2018	PS	NA	1	0	0	< 0.0001
Stevens et al. 2019	VS	NA	1	0	0	0.01
Sanctuary et al. 2019	PS	NA	0	0	1	> 0.05
Wang et al. 2020	PS	-8.7%	1	0	0	< 0.05
wang et al. 2022	PS	NA	1	0	0	0.0012
Sherman et al. 2022	PS	NA	1	0	0	< 0.05
Raghavan et al. 2023	ONS	NA	1	0	0	0.02
Billeci et al. 2023	PS	95% CI: -12.0% to -6.2%	1	0	0	< 0.05
Phan et al. 2024	PS	NA	1	0	0	< 0.05
Hrnciarova et al. 2024	ONS	NA	1	0	0	0.01
Wang et al. 2024	PS	NA	1	0	0	0.01
Palmer et al. 2024	PS	NA	0	0	1	> 0.05
Mitchell et al. 2024	PS	NA	0	0	1	> 0.05
Trezzi et al. 2025	PS	NA	1	0	0	< 0.05
Firmicutes to Bacteroidetes ratio (F/B ratio)
Liu et al. 2017	VS	NA	1	0	0	< 0.05	< .0001
Stevens et al. 2019	VS	NA	1	0	0	0.018
Wang et al. 2020	PS	NA	0	0	1	> 0.05
wang et al. 2022	PS	NA	1	0	0	0.0012
Billeci et al. 2023	PS	NA	1	0	0	< 0.05
Phan et al. 2024	PS	NA	0	0	1	NA
Hrnciarova et al. 2024	ONS	NA	0	0	1	> 0.05
Wang et al. 2024	PS	NA	1	0	0	< 0.05
Palmer et al. 2024	PS	NA	0	0	1	0.36

Abbreviations:VS Vitamin supplementation, GI Gastrointestinal, GIS Gastrointestinal severity index, NA Not Available, ONS Other Nutritional Supplementation, PS Probiotic Supplementation

Table 1.

Summary characteristics of included studies in the review

Study, year Country	Disorder	Study design and duration	Intervention	Intervention group	Comparison group	Primary outcome	Mechanism
Probiotic supplementation
Pärtty et al. 2015 Finland	ADHD and ASD	Double-blind, prospective RCT 6 months	Lactobacillus rhamnosus GG	40 children 0–13 y M/F M = 24	35 children 0–13 y M/F M = 16	↓ ADHD/AS risk in probiotic group (0% vs. 17.1%, P = 0.008), ↓ Early-life Bifidobacterium linked to ↑ ADHD/AS risk	• Enhance gut-brain communication via immune modulation and vagus nerve signaling [54, 56] • Reduce pathogenic bacteria like Clostridium, Desulfovibrio, and Collinsella [55, 56] • Stimulate SCFA production (butyrate, acetate, propionate) for gut barrier improvement [54, 55] • Increase neurotransmitter levels (serotonin, dopamine, GABA) for cognitive enhancement [12, 58, 61] • Strengthen gut integrity, reducing permeability and systemic inflammation [12, 62]
Shaaban et al. 2017 Egypt	ASD	Open-label prospective trial 3 months	Lactobacillus acid-ophilus + Lactobacillus rhamnosus + ZBifidobacteria longum	30 children 5 -9 y Mean = 84.77 month M/F M = 63.3%	–	↑ Bifidobacteria & Lactobacillus (P < 0.0001), ↑ GI symptom improvement, ↓ GIs-6 total score
Sanctuary et al. 2019 USA	ASD	Double-blind Crossover RCT 5 weeks	BCP + Bifidobacterium infantis and BCP only	8 children 3.9–10.9 y Mean = 6.8 M/F M = 87.5%	Combination arm: 5 BCP only: 4 M/F 3.9–10.9 y Mean = 6.8 M = 87.5%	↑ GI symptom improvement (BCP-only: 87.5%, Combo: 100%), ↑ Normal stools, ↓ Hard/soft stools, No consistent effect on enterotype or metabolite profiles
Wang et al. 2020 China	ASD	Double-blind trial 6-months	Probiotics strains + fructo-oligosaccharide	16 children 2–8 y Mean = 4.3 M/F	10 children 2–8 y Mean = 4.3 M/F	↑ Bifidobacteria & B. longum in intervention group, ↓ Clostridium & Ruminococcus (ASD + probiotics + FOS), ↓ Acetic, butyric, & propionic acid pre-intervention (ASD)
Santucci et al. 2020 Italy	ASD	Double-blind Parallel, factorial, efficacy RCT 6-month	Probiotic supplement (Marketed as Vivo mix®)	31 children Mean = 4.16 M/F M = 80%	32 children Mean = 4.13 M/F M = 86%	↓ 6-GSI score in intervention group (− 4.33 vs. − 2.28, 6 months), ↑ Stool smell & flatulence improvement vs. placebo
wang et al. 2022 Taiwan	ADHD	Open-label, single-arm trial 8 weeks	Bifidobacterium bifidum Bf-688	30 children 4–16 y Mean = 6.9 M/F M = 70%	–	↓ Firmicutes/Bacteroidetes ratio (P = 0.0012, week 8), ↓ B. ovatus (ADHD, probiotics), ↓ Bacteroidota, ↑ Shigella post-intervention, Bf-688 significantly altered gut microbiota
Sherman et al. 2022 USA	ASD	Post-hoc exploratory analysis from RCT 16 weeks	Lactobacillus plantarum probiotic (6 × 1010 CFUs)	18 children 3–20 y Mean = 10.26 M/F M = 75%	17 children 3–20 y Mean = 9.45 M/F M = 71%	↑ GSI improvement correlated with baseline GFAP, Significant LDA score differences in microbiota at baseline & post-intervention
Billeci et al. 2023 Italy	ASD	Double-blind Parallel, factorial, efficacy RCT 6 months	Probiotic supplement	26children 2.2- 6.1 y Mean = 4.4 M/F M = 76.9%	20 children 2.6 -5.6 y Mean = 3.8 M/F M = 75%	↑ 6-GSI score improvement in intervention group after 6 months
Phan et al. 2024 USA, Europe, Canada	ASD	Open-label, pilot study 3 months	probiotics + prebiotics based on the gut microbiome profile	170 participants 2.5—75 y Mean = 10.41 M/F M = 79.7%	123 participants 2.5—75 y Mean = 10.74 M/F M = 52.0%	↑ Microbiome diversity, ↓ GI symptoms, ↑ ASD-related behaviors (language, cognition, social) in > 50% of participants
Wang et al. 2024 Taiwan	ADHD	Double-Blind, RCT 12-week	Bifidobacterium bifidum (Bf-688) (5 × 10⁹ CFUs daily) + MPH	51 children 6–12 y Mean = 9.1 M/F M = 42	51 children 6–12 y Mean = 9.1 M/F M = 45	↑ CPT & CATA performance, ↑ Firmicutes/Bacteroidetes ratio (Bf-688), No significant ADHD symptom changes vs. placebo
Palmer et al. 2024 Australia	ASD	Double-Blind, RCT 6-week	Prebiotic supplementation (2.4 g/d GOSYAN® GOS)	22 children 4–10 y Mean = 7.18 M/F M = 16	19 children 4–10 y Mean = 6.75 M/F M = 11	no major social/mealtime behavior changes, moderate effect (d = 0.44) on parental quality of life & ↓ severe GI symptoms
Mitchell et al. 2024 Australia	ASD	Open-Label, Randomized, Pilot Study 24 weeks	Prebiotic + Probiotic ± GDH	15 children 5–10.99 y (SYN group) Mean = 7.07 M/F M = 11 16 children (SYN + GDH group) Mean = 8.39 M/F M = 7	–	↓ GI symptoms (SYN & SYN + GDH, p < 0.001), ↓ anxiety (p = 0.002) & irritability (p < 0.001, SYN + GDH only), ↑ Bifidobacterium animalis & Dialister in both groups
Trezzi et al. 2025 Italy	ADHD	Double-Blind, RCT 3-month	Lactobacillus plantarum + Lactobacillus acidophilus + Bifidobacterium animalis + prebiotic acacia fiber	20 children 6–16 y Mean = 11.95 M/F M = 14	21 children 6–16 y Mean = 12.19 M/F M = 19	No significant difference in fatty acid levels or ADHD symptoms, comparison group showed better fNIRS improvement, both groups had partial gut microbiota modulation
Vitamin supplementation
Adams and Holloway. 2004 USA	ASD	Double-blind, RCT 3 months	Vitamin B6, vitamin C, and alpha lipoic acid	11 children 3–8 y Mean = 5.3 M/F M = 77%	9 children 3.4–9 y Mean = 5.1 M/F M = 64%	↑ Sleep & GI symptom improvements in intervention vs. placebo	• Influence microbial composition by promoting beneficial bacteria like Lactobacillus and Bifidobacterium [60] • Modulate immune responses, reducing inflammation and oxidative stress [59, 61] • Help maintain intestinal permeability and reduce gut dysbiosis [60, 61] • Support neurotransmitter synthesis and cognitive function [59, 60]
Liu et al. 2017 China	ASD	Single-blind, non-randomized trial 6 months	Vitamin A supplement	20 children 1–8 y Mean = 5.6 M/F M = 85%	–	↑ Bacteroidetes (43.2% → 62.9%), ↓ Firmicutes (43.5% → 31.2%), ↑ Bacteroidetes/Firmicutes ratio (1.0 → 2.0), ↓ Proteobacteria (10.1% → 4.5%), ↓ Actinobacteria (2.8% → 0.5%), Post-Vitamin A Intervention: ↑ Bacteroidales (order), Bacteroida (class), Bacteroidetes (phylum)
Stevens et al. 2019 New Zealand	ADHD	Double-blind, RCT 10 weeks	Blend of vitamins, minerals, amino acids and antioxidants	10 children 7–12 y Mean = 10.29 M	7 children 7–12 y Mean = 9.3 M	↑ Microbiome diversity, ↓ Actinobacteria (Bifidobacterium), no significant gut microbiome shift or ADHD symptom change
ONS
Capra and Hannan-Jones. 1992 Australia	ID	Controlled Trial 6 weeks	High-fiber dietary supplementation (7g of additional fiber per day)	18 participantes 20 – 59y Mean = 36.1 M/F M = 55.6%	19 participants 19 – 51y Mean = 32.8 M/F M = 47.4%	Fiber supplementation prevented constipation worsening, control group had ↓ bowel movements & stool texture deterioration	• Enhance beneficial gut bacteria growth through prebiotic interventions [54, 57, 63] • Reduce neuroinflammation and oxidative stress via omega-3 and polyphenol intake [61] • Influence gut microbial composition and serotonin synthesis with amino acids [12, 54] • Modify gut microbiota to improve cognitive and behavioral outcomes [62, 63]
Abele et al. 2021 Latvia	ASD	Non-blinded, non-randomized, pilot, trial 3 months	Specific Carbohydrate Diet	10 children 2–18 y M/F	7 children 2–18 y M/F	↓ ATEC score by 23% (p = 0.05), ↑ Socializing (p = 0.03) & Health/Behavior (p = 0.02), GI symptoms improved (p = 0.01 intervention, p = 0.02 control)
Lawrence et al. 2022 UK	ADHD	Non-randomized, feasibility study 6 weeks	Dietary plan of macronutrient	9 children 8–13 y Mean = 10.4 M/F M = 6	–	No improvement in cognitive function, sleep, or GI symptoms, modest ↑ in alpha diversity after diet plan
Raghavan et al. 2023 India	ASD	Open-label, pilot clinical trial 3 months	Nichi Glucan food supplement	9 children 3–18 y M/F	4 children 3–18 y M/F	No significant genus-level changes (Shannon index), ↓ Enterobacter (~ 0) in supplementation group, ↓ E. coli (intervention vs. baseline), ↓ Lactobacillus (intervention vs. placebo), ↑ Akkermansia muciniphila & Clostridium bolteae (intervention), ↓ in placebo
Hrnciarova et al. 2024 Czech Republic	ASD	Double-Blinded, RCT, Pilot Study 3 months	Juvenil	16 children 3–7 y Mean = 6 M/F M = 13	12 children 3–7 Mean = 5 M/F M = 8	Juvenil shifted gut microbiota toward neurotypical profile, ↑ motor skills, visual response, nonverbal communication, activity level, no significant overall ASD symptom improvement

Abbreviations: ADHD Attention Deficit Hyperactivity Disorder, AS Asperger Syndrome, ASD Autism Spectrum Disorder, ATEC Autism Treatment Evaluation Checklist, BCP Bovine Colostrum Product, CATA Conners’ Auditory Test of Attention, CFUs Colony Forming Units, CGI Clinical Global Imprison, CPT Continuous Performance Test, F Female, FOs Fructo- Oligosaccharide, fNIRS focused attention and neurophysiological activation, ONS Other Nutritional Supplementation, g gram, GABA Gamma-aminobutyric acid, GDH Gut-Directed Hypnotherapy, GFAP Glial Fibrillary Acidic Protein, GIS Gastrointestinal Severity Index, GIS-6 6 Items Gastrointestinal Severity Index, ID Intellectual Disorder, LDA Linear Discriminant Analysis, M Male, MPH methylphenidate, PCR Polymerase Chain Reaction, RCT Randomized Control Trial, SCFA Short-chain fatty acids, SYN symbiotic, USA United States of America, UK United Kingdom, VAI Vitamin A Intervention, y year

Results

Study selection and characteristics

In the initial search and after removing duplicate articles, 1173 research articles were screened for titles and abstracts and 1124 articles were removed from the library at this stage. Then, in the next step, 49 publications remained for further full-text evaluation. Finally, 21 eligible clinical trials remained to be included in the current systematic review [23, 34–53]. The reason for the exclusion of the 28 papers and the study selection process is illustrated in Fig. 1 based on the PRISM 2020 statement. The included articles were eleven RCTs [23, 36–39, 41, 44, 45, 49, 51, 52], five open-label trials [35, 40, 42, 43, 46], three non-randomized trials [47, 48, 53], one double-blind trial [34], and one controlled Trial [50]. These trials were published between 1992 and 2025 and were from the USA [39, 41, 43, 52], Europe [23, 37, 44, 48, 49, 51, 53], Asia [34–36, 42, 47], New Zealand [38], Egypt [40], and Australia [45, 46, 50]. One study was exclusively performed on male subjects [38] and the remaining studies were on both genders. The sample size of included studies varied from 9 to 293 participants, resulting in a total sample size of 884 individuals. In addition, the age group of participants were children and adults. The final selected studies were performed on individuals with ASD [23, 34, 39–43, 45–47, 49, 51–53], ADHD [35–38, 48], both ASD and ADHD [44], and ID [50]. Furthermore, among included studies, the type of supplements were Probiotics [23, 34–37, 39–41, 43–46, 51], Vitamins [38, 47, 52], and ONS [42, 48–50, 53]. Intervention protocols varied across studies in terms of supplementation type and duration. Refer to Table 1 for study-specific methodologies.

Fig. 1.

Flow diagram of the selection process

Outcomes

Primary outcomes

In regard to primary outcomes, out of all 21 trials, twelve studies assessed Gastrointestinal severity index (GIS) [23, 34, 36, 39–41, 43, 45, 46, 51–53], fifteen studies assessed Gastrointestinal (GI) symptoms [23, 34, 36, 39–41, 43, 45, 46, 48–53], seventeen studies evaluated Good microbiome [34–49, 51], sixteen studies evaluated Bad microbiome [34–47, 49, 51], and nine studies evaluated Firmicutes to Bacteroidetes ratio (F/B ratio) [34–36, 38, 43, 45, 47, 49, 51]. For statistical values and individual study results, refer to Table 1.

Gastrointestinal Severity Index (GIS)

Twelve out of a total of 21 studies [23, 34, 36, 39–41, 43, 45, 46, 51–53], investigated the effect of nutritional supplements on the GIS in NDDs. Regarding the type of supplementation, one out of the twelve studies administered an ONS [53], one with vitamin supplementation [52], and the remaining studies used probiotic supplements as their intervention of interest. The target population of all studies investigating the effect of probiotic supplements [23, 34, 39–41, 43, 45, 46, 51], was ASDs except one study whose target population was ADHD [36]. One study assessed the effect of ONS and the other study evaluated the impact of vitamin supplementation on ASD. All of these studies showed positive effects of intervention on reducing GIS scores after supplementation. The p-values of these studies were in the range (< 0.0001 to 0.05) and the combined p-value (< 0.0001) indicates the strong significance of such effects. See Table 1 and Table 2 for statistical significance details.

Gastrointestinal symptoms

In addition, 15 out of total 21 studies reported gastrointestinal (GI) symptoms as their primary outcomes [20, 31, 33, 36–38, 40, 42, 43, 45–50]. Nine out of these 15 studies [23, 34, 39–41, 43, 45, 46, 51], examined probiotic supplementation in ASDs. They represented those probiotic supplements positively affect GI in individuals with ASD. In addition, one of these studies evaluated the impact of probiotics on ADHD children [36]. Also, four other studies investigated the effect of ONS on GI symptoms in ASD [49, 53], ADHD [48], and ID [50] individuals. According to their findings, ONS could significantly reduce GI symptoms in the mentioned populations except in two studies [48, 50]. Moreover, the remaining trial in this category studied the effect of vitamin supplementation on GI symptoms of individuals with ASD [52]. They also showed that consuming nutritional supplementation can significantly improve GI symptoms in individuals with ASD and ADHD. Based on the combined p-value (< 0.0001) calculated from the reported p-values (fourteen studies), nutritional supplementation could have a significant positive effect on GI symptoms in the NDD population. For specific bacterial abundance shifts and p-values, refer to Table 1 and Table 2.

Good microbiome

Seventeen out of total 21 studies [34–49, 51], investigated the effect of supplementation (probiotic, vitamin, and ONS) on the diversity and abundance of the GM in individuals with ASD and ADHD populations. Eleven of these studies on ASD [34, 39–43, 45–47, 49, 51], five on ADHD [35–38, 48], and one on ASD as well as ADHD [44]. They found that probiotic supplements could increase the abundance of good microbiomes (GM) in the gut meaningfully (< 0.0001), such as Bifidobacterium and Lactobacillus when compared with pre-intervention status in ASD individuals. In addition, vitamin supplements influenced microbiome composition, with significant increases in beneficial bacterial strains. For specific microbial shifts and mechanisms[12, 54–63], refer to Table 1.

Bad microbiome

A total of sixteen studies out of a total 21 studies [34–47, 49, 51], demonstrated a positive effect of taking supplements on reducing the number of bad microbiome (BM) to P < 0.0001 in comparison to baseline frequency. However, in three studies the interventions did not change BM significantly in ASD populations [41, 45, 46]. Six out of these sixteen studies [34, 35, 39, 40, 43, 51] reported a reduction of BM in the gut after taking probiotics in ASD children, and two studies [42, 49] reported the effect of ONS on reducing BM in the gut in ASD children. Also, two studies in this category reported a positive reduction in the number of BM following taking vitamins in children with ASD and ADHD [38, 47]. The combined p-values calculated from these studies were strongly significant (< 0.0001). See Table 2 for statistical significance details.

Firmicutes to Bacteroidetes ratio

Nine out of 21 total selected studies [31–33, 35, 40, 42, 44, 46, 48], measured Firmicutes to Bacteroidetes ratio (F/R ratio) (a measure of the relative abundance of two dominant phyla in the gut microbiota). According to the results of these five studies, the F/R ratio changed significantly in ASD and ADHD individuals after supplementation with probiotics and vitamins [35, 36, 38, 47, 51] but in the other four studies, no significant changes were shown. The combined p-value for the results in this category was significant like other primary outcomes (< 0.0001). Table 2 shows more details.

Secondary outcomes

Symptom severity index

The results of a study by Shaaban et al. [40], showed significant improvement in ASD symptoms following three months of supplementation with three probiotic strains in children aged 5–9 years. In addition, another study by Wang et al., reported that total ATEC scores significantly decreased in children with ASD aged 2–4 years after 6 months of probiotics consumption as well as FOS supplementation [34]. Moreover, the result of post-hoc exploratory analysis of a RCT study which was carried out in 35 children with ASD, showed a significant change in CGI-I following a 16-weeks period with probiotic supplement [39]. Furthermore, Stevens et al. revealed a significant association between higher relative abundance of Bifidobacterium and lower ADHD-IV-RS1 score after 10 weeks of supplementation with a vitamin-base supplement in individuals with ADHD. In addition, lower Actinobacteria abundance before and after the intervention was also correlated with lower scores of ADHD-IV-RS after supplementation [38]. Another study on 17 children with ASD also reported a significant decrease in ATEC score (23%) in those who consumed a special carbohydrate diet (such as: Gut and Psychology Syndrome diet (SCD/GAPS) and omega-3 essential fatty acids, ascorbic-palmitate, vitamin D, and vitamin C) for three months [53]. Additionally, the findings of a recent study by Hrnciarova et al. [49] demonstrated that supplementation with Juvenil can decrease ATEC scores in children with ASD significantly, further reinforcing the role of nutritional interventions in symptom severity improvement. Finally, one open-label trial which assessed Nichi Glucan food supplement on 13 ASD children, reported that all children who received this nutritional supplement for three months showed improvement in CARS scores evaluation [42]. Moreover, another study showed a significant decline in ABC scores after consumption of a symbiotic-based intervention, highlighting the potential of gut microbiota modulation in alleviating behavioral symptoms in people who suffer from NDDs [46].

Social skill assessment

The results of the study by Sherman et al., on 35 ASD children aged 3–20 years old, showed a significant change in SRS motivation sub-score, ABC-2 total score, and ABC-2 stereotypic behavior, in those who consumed probiotic supplements for 16 weeks in comparison to controls [39]. A recent study by Phan et al. [43] reported meaningful enhancement in SRS2 scores in ASD children after consumption of a 3-month symbiotic intervention. Furthermore, another study on 30 children with ADHD aged 4–16 years old reported that inattention and hyperactivity/impulsivity symptoms improved significantly following eight months after an eight-week probiotic supplementation[34]. In addition, Abele et al., in their study on 17 children with ASD aged 2–10 years old [53] reported a significant decrease in ABC score (29%) in those who consumed for three months a special carbohydrate diet (such as Gut and Psychology Syndrome diet (SCD/GAPS) and mega-3 essential fatty acids, ascorbic-palmitate, vitamin D, and vitamin C). No serious side effects or adverse events were reported in these included studies.

Quality assessment

The ROBINS-I and RoB-1 tools were administered for assessing risk of bias in non-RCT and RCTs studies respectively. Three out of twelve RCT studies [23, 37, 45] could be considered a high-quality study with a totally low risk of bias for all domains of ROB-1. In addition, six RCTs [36, 38, 41, 49, 51, 52] were considered of unclear risk of bias in which one or more domains had an unclear risk of bias. The remaining RCTs had low quality since they had a high risk of bias for one or more domains [39, 44, 50]. Moreover, regarding non-randomized trials two out of six studies had a low risk of bias [42, 48], four had a moderate risk of bias [34, 40, 47, 53], and three remaining studies had a serious risk of bias [42, 43, 46] for all domains of the ROBINS-I. Refer to Figs. 2 and 3 for detailed risk of bias evaluations.

Fig. 2.

Risk of Bias of Included randomized-controlled trials using RoB-1 (n = 12(

Fig. 3.

Risk of Bias of Included non-randomized studies of interventions using ROBINS-I (n = 9)

Discussion

In this systematic review, we investigated the effect of nutritional supplements on the gut microbiome, gastrointestinal issues, and behavioral symptoms among individuals with neurodevelopmental disorders (NDDs). While prior reviews have examined the effects of nutritional supplements on gastrointestinal or psychiatric symptoms and the gut microbiome in NDDs separately, our study represents the first systematic review of clinical trial studies to comprehensively evaluate the effects of supplementation on both the gut microbiome and clinical outcomes in individuals with NDDs.

Our results showed that nutritional supplementation effectively increased good microbiomes such as Bifidobacterium and Bacteroidetes while reducing bad microbiomes and decreasing the Firmicutes-to-Bacteroidetes (F/B) ratio. Additionally, we found positive effects of nutritional supplements on improving GIS and gastrointestinal symptoms, symptom severity, and social skills among individuals with NDDs.

These microbiome changes through these interventions are clinically relevant, as they contribute to reduced gut inflammation and enhanced gut barrier integrity, thereby improving GIS and GI symptoms [64, 65]. Since gastrointestinal discomfort can exacerbate behavioral symptoms in NDDs, alleviating these issues may also lead to behavioral improvements [66–68]. Furthermore, positive changes in the intestinal flora can also contribute to modulating neurotransmitter activity, reducing neuroinflammation, and regulating the gut-brain axis, all of which are implicated in NDD symptomatology [68]. Therefore, our results suggest to clinicians that nutritional supplements, especially probiotics and vitamins, have a potential therapeutic effect in improving both gut health and behavioral symptoms in individuals with NDDs by targeting these mechanisms.

Gut microbiota acts in different interconnected pathways to improve the gut-brain axis. This axis operates through three main pathways: the immune, neuroendocrine, and nerve pathways [7]. Through its residence and the production of microbial-derived metabolites and bioactive molecules, the intestinal flora modulates local immunity within the gut, subsequently impacting the central nervous system via systemic circulation. For instance, short-chain fatty acids seem to play a role in strengthening intestinal immunological barriers and reducing gut permeability [7]. Conversely, dysbiosis of the gut microbiota can compromise the gut barrier, leading to increased permeability and the translocation of pathogens, potentially exacerbating NDDs. Additionally, the gut microbiota influences neurodevelopment through various mechanisms, including its modulation of neuroactive molecules (such as taurine, 5-amino valeric acid, and 4-ethyl phenyl (sulfate)), ultimately affecting behaviors associated with NDDs. Moreover, bidirectional gut-brain communication pathways, such as the vagus nerve and the enteric nervous system, serve as conduits for signaling between the gut and the brain, further highlighting the relationship between the gut microbiome and neurodevelopment in individuals with NDDs [7].

Given these mechanistic pathways, previous research has extensively investigated the relationship between gut microbiota and the gut-brain axis on neurological function and behavior in individuals with NDDs [69, 70]. In a systematic review of 31 studies with a total of 3002 and 84 participants for ASD and ADHD, respectively, dysbiosis was reported in 28 studies. Additionally, an increase in the Firmicutes/Bacteroidetes ratio was reported as a consistent result, suggesting a potential biomarker of dysbiosis in neurodevelopmental disorders [71]. Other reviews have also shown alterations in the gut microbial composition in individuals with NDDs compared to healthy controls [72–75]. However, the variability and complexity observed in microbial composition across studies require further research to elucidate specific microbial alterations associated with NDDs. In addition, probiotics as the most repetitive nutritional supplement among included studies demonstrate promise in modulating the intestinal flora, with notable increases in beneficial bacteria and potential reductions in pathogenic strains [76–78]. Hence, dysbiosis can potentially be inversed through the beneficial effects of nutritional supplementation.

Regarding GI symptoms, several studies have explored the impact of these supplements on gastrointestinal symptoms in individuals with NDDs. Only one meta-analysis showed no difference in the severity of GI problems in ASD patients between the probiotics and/or prebiotics group and the placebo group; while it should be noted that the total sample size of this meta-analysis was small [79]. Nevertheless, other reviews have supported the beneficial effects of probiotics and prebiotics in improving GI symptoms among individuals with ASD [54, 77, 78, 80]. Further research with larger sample sizes and robust methodologies is needed to determine the precise effectiveness of probiotics in alleviating GI symptoms in NDDs.

Moreover, previous reviews and meta-analyses have evaluated the effects of nutritional supplements on behavioral symptoms linked to NDDs and potential approaches for treatment. For instance, a systematic review demonstrated the favorable use of vitamin and mineral supplementation as a combination approach for the treatment of symptoms associated with ADHD and ASD [81]. However, the effectiveness of specific supplements remains variable. Meta-analytical findings on Iron supplementation in ADHD management revealed large effect sizes in reducing both hyperactivity and inattentive symptom severity despite non-significant differences between the supplement and placebo groups [82]. Zinc supplementation has also been linked to improvements in ADHD severity [83–85]. Omega-3 fatty acids have been investigated for their potential to improve social communication difficulties among individuals with ASD. Still, the effect sizes remain small, and the supporting evidence is low quality [86]. Similarly, a meta-analysis found improvement in total ADHD symptoms with omega-3 supplementation compared to placebo but with a modest effect size [87]. While omega-3 and polyunsaturated fatty acids (PUFAs) may not show significant behavioral improvements across all studies, they could still be beneficial in cases of documented deficiency [84, 88–90]. Vitamin D supplementation has shown a small but statistically significant effect on the improvement in inattention, hyperactivity, and overall ADHD symptoms [27]. Conversely, a review of randomized clinical trials on Vitamin B6/Magnesium supplementation did not demonstrate efficacy in alleviating ASD symptoms [88]. There is also inconsistency in findings regarding the efficacy of supplements like N-acetylcysteine, D-cycloserine, and pyridoxine-magnesium, although some studies have suggested that N-acetylcysteine may help reduce irritability in individuals with ASD [91]. The inconsistencies across reviews might stem from relatively small sample sizes throughout studies, heterogeneity in intervention protocols, and differences in baseline microbial composition. Other factors such as dietary habits, concurrent medication use, and genetic predisposition could also influence the inter-individual differences in response to nutritional supplementation [7, 92, 93].

Regarding probiotics, findings remain controversial. Consistent with our findings, some reviews support the effectiveness of probiotics in improving behavior among individuals with ASD and ADHD [76, 77, 94]. For example, a review of five RCTs found overall support for probiotics in altering behavior among children diagnosed with ADHD. However, one study within the review found no significant differences between placebo and treatment groups, which may be attributed to variations in sample characteristics, bacterial strains, and participants' baseline gut microflora [94]. On the other hand, there are contradictory studies exist. The results of a meta-analysis showed no difference in the severity of the overall ASD symptoms between the probiotics and/or prebiotics group and the placebo group. However, only three clinical controlled trials were included in the meta-analysis, resulting in a small number of cases [79]. Variability in treatment duration with probiotics and concurrent interventions across studies may also influence the outcomes [77]. Future studies should address these limitations by implementing standardized probiotic formulations, controlling for baseline microbiome composition, and extending follow-up periods.

Further research is needed to continue to explore the mechanisms underlying the interaction between supplementation, gut microbiome, and symptomatology in NDDs to optimize treatment strategies and improve outcomes for affected individuals. Also, more rigorous research should be employed in randomized controlled trials with standardized outcome measures, stratification by baseline gut microbiota composition, and long follow-up periods are needed to determine optimal supplementation strategies, long-term efficacy, and interactions with existing pharmacological treatments.

Limitations

This study is not without limitations. Our focus on both primary and secondary outcomes of nutritional supplements in individuals with NDDs contributes to a comprehensive understanding of the effect of nutritional supplements on this population. However, the included studies were highly heterogeneous in duration, type of intervention, methodology, and the outcomes assessed. Thus, we use the SWiM method to make a structured comparison to address this heterogeneity. In addition, to enhance comparability among various nutritional supplement protocols throughout the included studies, we categorized nutritional supplements into three major groups: probiotics, vitamins, and other nutritional supplements. The relatively small number of included trials and participants restricted the robustness and generalizability of our findings. Thus, we used the combined p-value calculated using Fisher’s method to enhance statistical synthesis. Most studies had a high or unclear risk of bias, resulting in an overall low quality of evidence. We made efforts to use robust risk of bias assessment tools and provided detailed data on each study's risk of bias items. Furthermore, the short duration of follow-up in most studies restricted our ability to assess the long-term effects of nutritional supplements on gut microbiomes and clinical outcomes. Additionally, several studies lacked a placebo-controlled group, which could affect the reliability of the results. Future randomized clinical trials should address these limitations by including larger sample sizes, multiple centers, standardized nutritional supplement protocols, longer follow-up periods, and placebo-controlled groups.

Conclusion

This systematic review and narrative synthesis study fill a critical gap in clinical evidence by evaluating the association between the impact of nutritional supplements on the gut microbiome and the subsequent gastrointestinal and behavioral outcomes in individuals with NDDs. Nutritional supplements improve gut health by affecting the gut microbiome, reducing dysbiosis, and enhancing gut barrier integrity. Additionally, improvement in the microbial composition by nutritional supplementation modulates neurotransmitter activity and neuroinflammation linked to behavioral symptoms. These effects result in better gastrointestinal and behavioral outcomes. These findings have important clinical implications by indicating the gut-brain axis as a therapeutic target for individuals with NDDs. Clinicians should incorporate gut microbiota-targeted interventions, such as nutritional supplementation, as a promising adjunct therapy for addressing both gastrointestinal and behavioral symptoms in individuals with NDDs. To implement targeted and individualized supplementation strategies, clinicians should consider screening for gut dysbiosis and nutrient deficiencies in individuals with NDDs.

Abbreviations

NDD

Neurodevelopmental disorder

GI

Gastrointestinal

GIS

Gastrointestinal severity index

GM

Good microbe

BM

Bad microbe

SLD

Specific learning disorder

CD

Communication Disorder

MD

Motor Disorder

IBD

Inflammatory bowel disease

ONS

Other nutritional supplements

Authors’ contributions

E.D conceptualize the project, wrote the original draft, and extracted the data; K.K wrote the original draft, did the formal analysis and devised the study methodology; S.A collected the data and wrote the manuscript; M.A revised and wrote the manuscript; R.R revised the manuscript; M.SH critically revised the draft and provide the final manuscript; AH.M conceptualize and supervised the project, critically reviewed the draft and verified the final manuscript. All authors have seen and approved the manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

Data is provided within the manuscript or supplementary files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.