Gut microbiome changes with micronutrient supplementation in children with attention-deficit/hyperactivity disorder: the MADDY study

Abstract

Micronutrients have demonstrated promise in managing inattention and emotional dysregulation in children with attention-deficit/hyperactivity disorder (ADHD). One plausible pathway by which micronutrients improve symptoms is the gut microbiome. This study examines changes in fecal microbial composition and diversity after micronutrient supplementation in children with ADHD (N = 44) and highlights potential mechanisms responsible for the behavioral improvement, as determined by blinded clinician-rated global improvement response to micronutrients. Participants represent a sub-group of the Micronutrients for ADHD in Youth (MADDY) study, a double blind randomized controlled trial in which participants received micronutrients or placebo for 8 weeks, followed by an 8-week open extension. Stool samples collected at baseline, week 8, and week 16 were analyzed using 16S rRNA amplicon sequencing targeting the V4 hypervariable region. Pairwise compositional analyses investigated changes in fecal microbial composition between micronutrients versus placebo and responders versus non-responders. A significant change in microbial evenness, as measured by alpha diversity, and beta-diversity, as measured by Bray-Curtis, was observed following micronutrients supplementation. The phylum Actinobacteriota decreased in the micronutrients group compared to placebo. Two butyrate-producing bacterial families: Rikenellaceae and Oscillospiraceae, exhibited a significant increase in change following micronutrients between responders versus non-responders. These findings suggest that micronutrients modulated the composition of the fecal microbiota and identified specific bacterial changes associated with micronutrient responders.

Keywords: ADHD; children; microbiome; micronutrients.

Figure 1.

Graphic study design for all analyses of microbiota composition. a) analysis 1 compares children who received micronutrients to those who received placebo from baseline to week 8. b) analysis 2 compares micronutrient responders and non-responders. c) analysis 3 investigates placebo’s potential influence on the fecal microbiome by comparing micronutrients alone versus placebo followed by micronutrients. Created in BioRender. Ast, H. (2025) https://BioRender.com/f93m447.

Figure 2.

Enrollment, micronutrients exposure, and micronutrients response flowchart categorized and structured based on strengthening the organization and reporting of microbiome studies (STORMS) flowchart recommendations. The flow chart does not include analysis of placebo’s impact on microbiome (Figure 1c).

Figure 3.

Distribution of change in alpha diversity and abundance of specific taxa baseline to week 8 by micronutrients and placebo. A) the distribution of change in alpha diversity (within sample community diversity) during intervention for each intervention group at the family level. See table S3 for median [IQR] values. The Y-axes represent the change in each respective alpha diversity metric during intervention (value after intervention – value before intervention = change during intervention). B) the distribution of change in relative abundance during intervention for each intervention group, for select taxa that exhibited a significant difference between intervention groups. Faecalibacterium and bifidobacterium are genera, bifidobacteraceae is a family, and Actinobacteriota and Verrucomicrobiota are phyla. The Y-axes represents the change in relative abundance during intervention (relative abundance after intervention – relative abundance before intervention = change in relative abundance during intervention). In both figures, the X-axis represents the response group for which each distribution belongs to. Those subjects in the “placebo” intervention group received a placebo during the timeframe of this analysis, whereas subjects in the “micronutrients” intervention group received micronutrients. Zero (indicating no change) is indicated by a red dashed line. P-values here were not corrected for multiple testing and correspond to the significance of the difference in distribution of change in alpha diversity, between intervention groups, as calculated using a Wilcoxon rank-sum test for significance.

Figure 4.

Stacked relative abundance taxa bar plot demonstrating compositional breakdown of each subject’s gut microbiome, before and after micronutrients baseline to week 8. The Y-axis represents the percent abundance of each member of the gut microbial community and totals to 1. The X-axis represents samples, collected before and after micronutrients, for each subject and is ordered by the relative abundance of Firmicutes in samples collected before micronutrients, in descending order. Each color represents a distinct microbial phylum, whereas each shade represents a distinct microbial family.

Figure 5.

Summary of bacteria significantly changed through 8 weeks of micronutrients or placebo. Arrows indicate significant change within group in taxa abundance and a weighted outline indicates significant between group change in taxa abundance. For example, Actinobacteriota has decreased within group change in abundance within the micronutrients group and significant between group change in micronutrients compared to placebo. Bacteroidaceae had increased within group change in abundance in both micronutrients and placebo groups. Verrubomicrobiota changed significantly between groups. Significance at p < 0.05. Created in BioRender. Ast, H. (2025) https://BioRender.com/d23w192

Figure 6.

Distribution of between group change in relative abundance following micronutrients by response group. The Y-axis represents the change in relative abundance during micronutrient intervention (change in relative abundance during micronutrient intervention = relative abundance after micronutrients – relative abundance before micronutrients). The X-axis represents the response group. Subjects in the “responder” response group exhibited a positive response to micronutrient intervention, whereas subjects in the “non-responder” response group did not. P-values were not corrected for multiple testing and correspond to the significance of the difference in distribution of change in abundance of the select taxa, between response groups, as calculated using a Wilcoxon rank-sum test for significance.

Figure 7.

Significant bacterial changes identified in responder vs. non-responder analysis based on taxonomic rank. Arrows indicate significant change within group in taxa abundance and a weighted outline indicates significant between group change in taxa abundance. For example, Rikenellaceae had increased within group change in abundance in responders and significant between group change between responders and non-responders, and Bifidobacterium had decreased within-group change in abundance in responders and non-responders. Significance indicated at p < 0.05. Created in BioRender. Ast, H. (2025) https://BioRender.com/k06g491

Figure 8.

Distribution of change in beta diversity by intervention and response group. Children who received micronutrients during the RCT and responded (n = 17) or did not respond (n = 16) are compared to children who received placebo first and responded (n = 7) or did not respond (n = 4) to micronutrients during week 8 to week 16. The y-axis represents the beta-diversity distance between each individuals’ sample before and after micronutrient supplementation. The x-axis represents the intervention group with responders and non-responders depicted by color.