Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Perturbations in Gut Microbiota Composition in Psychiatric Disorders

Perturbations in Gut Microbiota Composition in Psychiatric Disorders Research JAMA Psychiatry | Original Investigation A Review and Meta-analysis Viktoriya L. Nikolova, MRes; Megan R. B. Hall, BSc; Lindsay J. Hall, PhD; Anthony J. Cleare, MBBS, PhD; James M. Stone, MBBS, PhD; Allan H. Young, MD, PhD Multimedia IMPORTANCE Evidence of gut microbiota perturbations has accumulated for multiple Supplemental content psychiatric disorders, with microbiota signatures proposed as potential biomarkers. However, no attempts have been made to evaluate the specificity of these across the range of psychiatric conditions. OBJECTIVE To conduct an umbrella and updated meta-analysis of gut microbiota alterations in general adult psychiatric populations and perform a within- and between-diagnostic comparison. DATA SOURCES Cochrane Library, PubMed, PsycINFO, and Embase were searched up to February 2, 2021, for systematic reviews, meta-analyses, and original evidence. STUDY SELECTION A total of 59 case-control studies evaluating diversity or abundance of gut microbes in adult populations with major depressive disorder, bipolar disorder, psychosis and schizophrenia, anorexia nervosa, anxiety, obsessive compulsive disorder, posttraumatic stress disorder, or attention-deficit/hyperactivity disorder were included. DATA EXTRACTION AND SYNTHESIS Between-group comparisons of relative abundance of gut microbes and beta diversity indices were extracted and summarized qualitatively. Random-effects meta-analyses on standardized mean difference (SMD) were performed for alpha diversity indices. MAIN OUTCOMES AND MEASURES Alpha and beta diversity and relative abundance of gut microbes. RESULTS A total of 34 studies provided data and were included in alpha diversity meta-analyses (n = 1519 patients, n = 1429 control participants). Significant decrease in microbial richness in patients compared with control participants were found (observed species SMD = −0.26; 95% CI, −0.47 to −0.06; Chao1 SMD = −0.5; 95% CI, −0.79 to −0.21); however, this was consistently decreased only in bipolar disorder when individual diagnoses were examined. There was a small decrease in phylogenetic diversity (SMD = −0.24; 95% CI, −0.47 to −0.001) and no significant differences in Shannon and Simpson indices. Differences in beta diversity were consistently observed only for major depressive disorder and psychosis and schizophrenia. Regarding relative abundance, little evidence of disorder specificity was found. Instead, a transdiagnostic pattern of microbiota signatures was found. Depleted levels of Faecalibacterium and Coprococcus and enriched levels of Eggerthella were consistently shared between major depressive disorder, bipolar disorder, psychosis and schizophrenia, and anxiety, suggesting these disorders are characterized by a reduction of anti-inflammatory butyrate-producing bacteria, while pro-inflammatory genera are enriched. The confounding associations of region and medication were also evaluated. CONCLUSIONS AND RELEVANCE This systematic review and meta-analysis found that gut microbiota perturbations were associated with a transdiagnostic pattern with a depletion of certain anti-inflammatory butyrate-producing bacteria and an enrichment of pro-inflammatory bacteria in patients with depression, bipolar disorder, schizophrenia, and Author Affiliations: Author anxiety. affiliations are listed at the end of this article. Corresponding Author: Viktoriya L. Nikolova, MRes, Centre for Affective Disorders, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, De Crespigny Park, JAMA Psychiatry. 2021;78(12):1343-1354. doi:10.1001/jamapsychiatry.2021.2573 London SE5 8AF, United Kingdom Published online September 15, 2021. Last corrected on December 1, 2021. (viktoriya.nikolova@kcl.ac.uk). (Reprinted) 1343 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders espite evidence that probiotic formulations can im- prove mental health dating back to the early 20th Key Points 1,2 D century, it was only following advances in DNA/ Question Do psychiatric disorders present with distinct or shared RNA sequencing technologies that the involvement of the gut gut microbial alterations? microbiota in the pathophysiology of psychiatric disorders was Findings This review and meta-analysis of 59 case-control studies recognized. Preclinical studies have consistently demon- found that gut microbiota perturbations were associated with a strated that fecal microbiota transplants from patients with a transdiagnostic pattern with a depletion of certain wide range of psychiatric conditions result in the develop- anti-inflammatory butyrate-producing bacteria and an enrichment ment of the behavioral and physiological profile of the condi- of pro-inflammatory bacteria in depression, bipolar disorder, 3-7 tion in germ-free mice. This suggests that psychiatric dis- schizophrenia, and anxiety. orders may be associated with a distinct pattern of microbial Meaning These findings are in line with genetic and inflammatory perturbations, which may serve as a biomarker. marker studies and support the transdiagnostic dimensional Attempts to characterize the composition of the micro- model of psychiatric disorders by highlighting the gut microbiota biota in psychiatric populations have yielded plentiful yet con- as an additional dimensional component. tradictory results. Nevertheless, systematic reviews in indi- vidual disorders have been able to identify patterns that may 8-10 be promising biomarker targets. Indeed, the addition of such Selection Criteria biomarkers can improve diagnostic accuracy, guide treat- Systematic reviews and meta-analyses were considered eli- ment, and assist the monitoring of treatment response. For the gible if they followed established guidelines and included at definition of a biomarker to be met, ie, “substance, structure or least 1 eligible original study. Original studies were eligible if process that can be measured in the body and influence or pre- they (1) applied an observational case-control design, (2) per- dict the incidence of outcome or disease,” the specificity and formed gut microbiota analysis and reported diversity or abun- reproducibility of the alteration needs to be demonstrated. dance measures, and (3) sampled a general adult population Therefore,itiscrucialtocomparemicrobialperturbationsacross (age 18-65 years) with a psychiatric diagnosis of interest. In- the wider range of psychiatric conditions. terventional or longitudinal comparisons in the absence of a We performed an umbrella and updated review and meta- control group were excluded. Records were screened by 2 au- analysis of gut microbiota studies in adults with major depres- thors (V.L.N. and M.R.B.S) and discrepancies resolved via dis- sive disorder (MDD), bipolar disorder, psychosis and schizo- cussion and consultation with a third author (A.H.Y.). phrenia, anxiety disorders, obsessive compulsive disorder (OCD), eating disorders (anorexia nervosa and bulimia Data Extraction nervosa), autism spectrum disorder, attention-deficit/ Information was extracted using a predesigned template by 2 hyperactivity disorder (ADHD), and posttraumatic stress dis- authors (V.L.N. and M.R.B.S) and cross-checked. From sys- order (PTSD) to evaluate the specificity and reproducibility of tematic reviews and original studies, we extracted publica- gut microbiota alterations and delineate those with potential tion details, participant demographic and clinical character- to become biomarkers. istics, and methodological information. As primary outcomes of interest, we extracted community-level measures of gut mi- crobiota composition (alpha and beta diversity) and taxo- nomic findings at the phylum, family, and genus levels (rela- Methods tive abundance). Alpha diversity provides a summary of the The protocol for this review was preregistered with microbial community in individual samples and can be com- PROSPERO (CRD42021224342). We followed Preferred pared across groups to evaluate the role of a particular factor Reporting Items for Systematic Reviews and Meta-analyses (in this case psychiatric diagnosis) on the richness (number of (PRISMA) reporting guideline as well as Cochrane guidance species) and evenness (how well each species is represented) 14,15 10,16 for umbrella and updated reviews. in the sample. Beta diversity is a measure of interindi- vidual (between samples) diversity that assesses similarity of communities compared with the other samples analyzed. Search Details We searched Cochrane Library, PubMed, Embase, and This analysis allows us to see whether patient samples clus- PsycINFOonJanuary27,2021.Thesearchstringsusedareavail- ter significantly differently (ie, with little or no overlap) com- able in eAppendix 1 in the Supplement. This search was lim- pared with control participant samples or whether they over- ited to systematic reviews and meta-analyses in English, in- lap, thus suggesting the 2 groups are not distinct. Control cluding human studies, published since 2005. After reviewing samples were defined as individuals without the relevant the results, we realized that a large body of recent literature condition. was missed, as numerous studies have become available fol- lowing the publication of the latest reviews. To ensure thor- Quality Assessment ough coverage, we performed an updated search for each dis- We performed quality assessment of the systematic reviews order on February 2, 2021, from the search date recorded in using the ROBIS tool and of the original studies not covered the latest available high-quality review for that disorder (eAp- in any review with the Joanna Briggs Institute Critical pendix 1 in the Supplement). Appraisal Checklist for Case-Control Studies. No studies were 1344 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research Table. Summary Characteristics of the Identified Reviews and Original Studies by Psychiatric Disorder No. Mean patient Female, a c Disorder Reviews Studies Total patients Region of studies age, y mean % MDD 8 21 930 East: n = 14; west: n = 7 35 60 Schizophrenia and psychosis 5 11 699 East: n = 9; west: n = 2 36 45 Bipolar disorder 3 9 465 East: n = 5; west: n = 4 38 55 Anorexia nervosa 3 10 211 East: n = 2; west: n = 8 26 99 Anxiety 2 3 84 East: n = 2; west: n = 1 40 77 OCD 0 2 59 West: n = 2 36 54 PTSD 0 1 18 Africa: n = 1 42 14 ADHD 1 1 19 West: n = 1 20 32 MDD + anxiety NA 2 60 West: n = 2 39 82 MDD + bipolar disorder NA 2 98 East: n = 1; west: n = 1 37 69 Total 16 59 2643 East: n = 32; west: n = 24; NA NA Africa: n = 1 Abbreviations: ADHD, attention-deficit/hyperactivity disorder; MDD, major Some include >1 disorder. depressive disorder; NA, not applicable; OCD, obsessive compulsive disorder; c West region includes US, Canada, Europe, Australia, and New Zealand. East PTSD, posttraumatic stress disorder. region includes China, Japan, and Taiwan. Africa includes South Africa. a 21,22 Studies that examined combined cohorts (MDD + bipolar disorder or d Adult populations only. 23,24 MDD + anxiety ) are presented separately. excluded owing to quality concerns. The detailed assessment of psychiatric medication. All analyses were completed in R is available as eAppendix 2 in the Supplement. version 4.17-0 (meta package; R Foundation). Two-sided P values were statistically significant at less than .05. Qualitative Synthesis For the relative abundance of microbial taxa, we performed a qualitative synthesis owing to the large number and limited Results overlap of findings. Owing to the significant likelihood of false positives noted in previous meta-analyses, results reported Search Results only by a single study were excluded. Further, results re- We identified 16 systematic reviews (eAppendices 4 and 5 in ported only by 1 research group were also excluded because the Supplement for PRISMA flowcharts and details of the sys- these were considered potentially methodology or popula- tematic reviews) containing 39 eligible studies. There were no tion specific. To identify disease-specific and shared altera- reviews capturing OCD, PTSD, or autism spectrum disorder in tions, we performed a within- and between-diagnostic com- adults. In the second search, a further 20 studies were iden- parison.First,wesummarizedwithin-disorderfindingsforeach tified, resulting in 59 studies across 8 disorders. The most re- taxon reported in at least 2 studies and labeled those in- searched disorder was MDD, followed by psychosis and schizo- creased, decreased, or not consistent. Not consistent was any phrenia, bipolar disorder, and anorexia nervosa (Table). finding with less than 75% agreement between studies report- ing this taxon. A consistent finding by 2 studies was consid- Characteristics of Included Studies ered worth noting for future validation, whereas a finding by The 59 studies provided 64 case-control comparisons captur- 3 or more studies (from ≥2 research groups) was considered ing 2643 patients and 2336 controls (eAppendix 6 in the potentially associated with the disorder. A taxon was consid- Supplement provides a detailed summary of study character- ered a candidate for disease-specific response if it was istics). Most studies (32 [54.2%]) were conducted in East Asia altered (in a consistent direction) in a single disorder only. (China, Japan, and Taiwan), 24 (40.7%) in westernized popu- Alternatively, if a shift was replicated in several disorders with lations (US, Canada, Europe, Australia, and New Zealand; known symptomatic and pathophysiological overlap, this was grouped according to typical diet and lifestyle), and 1 (1.7%) in considered a transdiagnostic alteration. Taxa similarly al- Africa(SouthAfrica).Moststudieshadsmalltomoderatesample tered across all/multiple unrelated diagnostic categories were sizes (median, 62), ranging between 4 and 156 per group (eAp- interpreted as general disease response. pendix 6 in the Supplement). Studies were similar in exclusion criteria; however, few attempted to minimize dietary changes Quantitative Synthesis or control dietary intake (12 of 59 [20.3%]) or smoking status (8 Meta-analysis was performed on differences in alpha diver- of 59 [13.6%]). Use of psychiatric medication also varied sub- sity between patients and controls for indices with data re- stantially,with11of59studies(18.6%)conductedinmedication- ported in 10 or more studies. Detailed methods of data trans- free or drug-naive groups, 5 of 59 (8.5%) in groups undergoing formation and interpretation thresholds are available in treatment and the remainder not controlling this, resulting in eAppendix 3 in the Supplement. Publication bias was evalu- anywhere between 20% and 96% of patients taking medica- ated with funnel plots and Egger test. Preplanned subgroup tion. Methodology of stool processing (eAppendix 7 in the analyses were disorder, region of study (east/west), and use Supplement) and composition analysis (eAppendix 6 in the jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1345 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders Supplement) also varied widely, with 16S ribosomal RNA se- be noted that shotgun metagenomics showed increased quencing being most common (44 of 59 studies [74.6%]) fol- Shannon diversity in patients (4 studies) in comparison with lowed by 9 studies (15.2%) using quantitative polymerase chain 16SrRNA V3-V4 sequencing, which showed an overall de- reactionorreal-timequantitativepolymerasechainreactionand crease (12 studies). This could be because the shotgun ap- 7 (11.9%) using shotgun metagenomics. proach quantifies all genomic DNA (including mycobiome and virome) rather than just specific regions of bacterial DNA. Fur- Alpha Diversity ther studies using shotgun metagenomics or comparing the 2 Of 44 studies reporting alpha diversity, 34 provided data and methodologies on the same population are needed. were included in meta-analyses (1519 patients and 1429 con- trols). Eleven indices were used to assess alpha diversity, in- Beta Diversity cluding estimates of richness (observed species, Chao1, abun- Beta diversity comparison between patients and controls was dance coverage estimator, and incidence coverage estimator), reported in 43 studies, with 1 study reporting on 3 separate ), using a variety evenness, richness/evenness (Shannon, Simpson, inverse groups (MDD, anxiety, and MDD + anxiety Simpson, Fisher), biodiversity (Faith phylogenetic diversity), of measures (eAppendix 9 in the Supplement). Consistent non- and 1 newly developed index (eAppendix 6 in the Supple- significant differences were reported by 16 studies, and a fur- ment). The most widely used were observed species, Chao1, ther 3 reported conflicting results between the measures used. Shannon, Simpson, and phylogenetic diversity. There was no Patients’ samples clustered differently from controls in 12 of evidence of publication bias in any of the analyses (eAppen- 15 studies in MDD, 7 of 9 in psychosis and schizophrenia, 3 of dix 8 in the Supplement). 6 in bipolar disorder, 3 of 6 in anorexia nervosa, 2 of 3 in anxi- Regarding richness, 20 studies provided data on ob- ety, 0 of 2 in OCD, and 0 of 1 in PTSD (eAppendix 9 in the served species in patients (n = 897) vs controls (n = 789). The Supplement). One of 2 combined MDD + bipolar disorder co- pooled estimate showed a significant decrease in patients hort was also significantly different from controls, whereas the with a small effect size (standardized mean difference MDD + anxiety cohort was not. Although, while Mason et al [SMD] = −0.26; 95% CI, −0.47 to −0.06; P = .01) and high found no differences when looking at diagnostic categories, 2 3,4,22,26-42 heterogeneity (I =75%) (Figure 1A). Within diag- they found a significant difference when clustering partici- nostic categories, there was a significant decrease only in bi- pantsaccordingtoself-reportedsymptoms.Thesefindingssug- polar disorder (SMD = −0.61; 95% CI, −1.19 to −0.03; P = .04; gest there is reliable evidence for differences in the shared phy- I = 80%). Twenty-six studies provided data on Chao1 in pa- logenetic structure in MDD and psychosis and schizophrenia tients (n = 956) vs controls (n = 961). The pooled estimate compared with controls; however, method of measurement showed a significant decrease in patients with a medium ef- and method of patient classification (symptom vs diagnosis fect size (SMD = −0.5; 95% CI, −0.79 to −0.21; P = .001; based) may affect findings. I = 88%). Regarding individual diagnoses, there was a signifi- cant decrease only in bipolar disorder and anorexia nervosa Differentially Abundant Microbial Taxa (SMD = −0.53; 95% CI, −1.01 to −0.05; P = .03; I = 62% and All studies assessed the relative abundance of gut microbes and SMD = −0.86; 95% CI, −1.52 to −0.21; P = .01; I = 80%, respec- 57 of 59 (96.6%) identified significant differences between pa- 4,21,25,27,29,31-33,35-49 tively) (Figure 1B). tients and controls at phylum, family, or genus levels. Overall, Regarding diversity, 29 studies reported the Shannon in- in MDD (21 comparisons), 94 taxa were differentially abundant; dex in patients (n = 1176) vs controls (n = 1172). The pooled es- in psychosis and schizophrenia (11 comparisons), 136; in bipo- timate demonstrated a nonsignificant difference between lar disorder (9 comparisons), 60; in anxiety (2 comparisons), 36; groups (SMD = −0.12; 95% CI, −0.27 to 0.03; P = .11) inanorexianervosa(10comparisons),32;inOCD(2comparisons), 3,4,21,22,25,27,28,31-35,38-46,48,50-53 (Figure 2A). Simpson index data 15; and in ADHD and PTSD (1 study each), 9 and 3, respectively. were provided by 11 studies (n = 418 patients; n = 377 con- Afterremovalofnonreplicatedfindings,thedifferencesspanned trols). There was a nonsignificant difference between groups 7 phyla, 28 families, and 67 genera. Study-level findings are pre- (SMD = 0.04; 95% CI, −0.13 to 0.21; P = .66), with nonsignifi- sented in eAppendix 10 in the Supplement. 21,26,27,31-33,40,43,52 cant heterogeneity (Figure 2B). Finally, 10 Figure 3providesthesummaryofthewithin-andbetween- studies provided phylogenetic diversity data in patients disorder comparison for the disorders with sufficient studies (n = 412) vs controls (n = 454). The pooled estimate showed a (anorexia nervosa, MDD, bipolar disorder, anxiety, and psy- significant decrease in patients with a small effect size chosis and schizophrenia). There was high within-disorder in- (SMD = −0.24; 95% CI, −0.47 to −0.0012; P = .049; 64%) consistency and the majority of consistent within-disorder 3,4,28,32-34,39,40,42,44 (Figure 2C). changes were replicated by only 2 studies and thus require fur- To explore sources of interstudy heterogeneity, sub- ther investigation. Considerably fewer were replicated by more group analyses and meta-regressions were performed for the than 2 studies from different research groups. analyses with sufficient studies (observed species, Chao1, Shannon). Body mass index, age, sex, smoking, region (east/ Limited Evidence of Disorder Specificity west), psychiatric medication use, subgrouping of psychosis Disorder specificity was observed for the enrichment of gen- and schizophrenia into first episode, and chronic and sequenc- era Holdemania and Olsenella and the depletion of genera ing method (including hypervariable region sequenced) did not Fusicatenibacter, Dialister, and Sutterella in MDD (Figure 3C). have a significant association with findings. However, it should However, these findings were weakly reproduced (3 to 4 of 21 1346 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research Figure 1. Forest Plots of Alpha Diversity Richness Estimators in the Gut Microbiota of Patients With Psychiatric Disorders Compared With Healthy Controls A Observed species B Chao1 Source SMD (95% CI) Decreased Increased Source SMD (95% CI) Decreased Increased MDD MDD 26 43 Naseribafrouei et al, 2014 0.44 (–0.13 to 1.01) Jiang et al, 2015 1.32 (0.75 to 1.89) 4 4 Kelly et al, 2016 –0.90 (–1.40 to –0.39) Kelly et al, 2016 –0.82 (–1.32 to –0.32) 3 44 Zheng et al, 2016 0.32 (–0.03 to 0.68) Huang et al, 2018 –1.15 (–1.73 to –0.57) 27 25 Chen et al, 2021 0.14 (–0.24 to 0.53) Rong et al, 2019 –1.05 (–1.60 to –0.51) 28 27 Liu et al, 2020 –0.17 (–0.59 to 0.24) Chen et al, 2021 0.21 (–0.17 to 0.60) 22 21 Vinberg et al, 2019 –0.80 (–1.27 to –0.33) Jiang et al, 2020 –0.57 (–1.30 to 0.16) Total –0.16 (–0.58 to 0.27) Total –0.34 (–1.08 to 0.40) 2 2 2 2 Heterogeneity: χ = 28.71; P <.001; I = 83% Heterogeneity: χ = 58.52; P <.001; I = 91 % 5 5 Bipolar disorder Bipolar disorder 29 21 Painold et al, 2019 –0.59 (–1.31 to 0.13) Jiang et al, 2020 0.25 (–0.54 to 1.05) 30 29 Coello et al, 2019 –0.22 (–0.51 to 0.07) Painold et al, 2019 –0.33 (–1.04 to 0.39) 31 25 Hu et al, 2019 –1.05 (–1.47 to –0.62) Rong et al, 2019 –0.77 (–1.30 to –0.25) Total Hu et al, 2019 –0.61 (–1.19 to –0.03) –0.94 (–1.36 to –0.52) 2 2 Heterogeneity: χ = 9.9; P =.007; I = 80% Total –0.53 (–1.01 to –0.05) 2 2 Schizophrenia and psychosis Heterogeneity: χ = 7.8; P =.05; I = 62% Shen et al, 2018 –0.27 (–0.64 to 0.09) Schizophrenia and psychosis 33 32 Pan et al, 2020 0.41 (–0.11 to 0.93) Shen et al, 2018 –0.26 (–0.63 to 0.10) 34 45 Li et al, 2020 –0.03 (–0.34 to 0.27) Zheng et al, 2019 –0.18 (–0.52 to 0.16) 35 33 Zhang et al, 2020 –0.21 (–1.00 to 0.58) Pan et al, 2020 0.13 (–0.39 to 0.65) Total Ma et al, 2020 (FEP) –0.04 (–0.31 to 0.24) –0.04 (–0.43 to 0.35) 2 2 46 Heterogeneity: χ = 4.61; P =.20; I = 35% Ma et al, 2020 (schizophrenia) –0.63 (–0.96 to –0.31) Anxiety Zhang et al, 2020 –0.36 (–1.16 to 0.44) 36 47 Jiang et al, 2018 –0.55 (–1.01 to –0.09) Xu et al, 2020 –2.71 (–3.13 to –2.29) Total Total –0.55 (–1.01 to –0.09) –0.58 (–1.29 to 0.12) 2 2 Heterogeneity: NA Heterogeneity: χ = 121.69; P <.001; I = 95% Anorexia nervosa Anxiety 37 36 Kleiman et al, 2015 –1.97 (–2.90 to –1.03) Jiang et al, 2018 –0.56 (–1.02 to –0.10) Mack et al, 2016 Total –0.11 (–0.48 to 0.26) –0.56 (–1.02 to –0.10) Total –0.99 (–2.80 to 0.83) Heterogeneity NA 2 2 Heterogeneity: χ = 13.13; P <.001; I = 92% Anorexia nervosa PTSD Kleiman et al, 2015 –1.90 (–2.82 to –0.98) 39 38 Hemmings et al, 2017 –0.30 (–1.04 to 0.43) Mack et al, 2016 –0.15 (–0.53 to 0.22) –0.30 (–1.04 to 0.43) –0.92 (–1.49 to –0.35) Total Hanachi et al, 2019 Heterogeneity: NA Monteleone et al, 2021 –0.83 (–1.47 to –0.19) OCD Total –0.86 (–1.52 to –0.21) 40 2 2 Domènech et al, 2020 –0.53 (–1.00 to –0.05) Heterogeneity: χ = 14.81; P =.002; I = 80% Turna et al, 2020 0.09 (–0.51 to 0.68) PTSD Total –0.25 (–0.85 to 0.35) Hemmings et al, 2017 –0.33 (–1.07 to 0.40) 2 2 Heterogeneity: χ = 2.48; P =.12; I = 60% Total –0.33 (–1.07 to 0.40) ADHD Heterogeneity NA Aarts et al, 2017 0.25 (–0.25 to 0.76) OCD Total 0.25 (–0.25 to 0.76) Domènech et al, 2020 –0.53 (–1.00 to –0.05) Heterogeneity NA Turna et al, 2020 –0.15 (–0.75 to 0.45) Total –0.26 (–0.47 to –0.06) Total –0.38 (–0.75 to –0.01) 2 2 95% PI (–1.12 to 0.59) Heterogeneity: χ = .96; P =.33; I = 0% 2 2 Heterogeneity: χ = 74.75; P <.001; I = 75% ADHD –3 –2 –1 0 1 2 3 Aarts et al, 2017 0.10 (–0.40 to 0.60) SMD (95% CI) Total 0.10 (–0.40 to 0.60) Heterogeneity NA Total –0.50 (–0.79 to –0.21) 95% PI (–1.96 to 0.96) 2 2 Heterogeneity: χ = 217.38; P <.001; I = 88% –3 –2 –1 0 1 2 3 SMD (95% CI) ADHD indicates attention-deficit/hyperactivity disorder; FEP, first episode psychosis; MDD, major depressive disorder; NA, not applicable; OCD, obsessive compulsive disorder; PI, prediction interval; PTSD, posttraumatic stress disorder; SMD, standardized mean difference. studies). The archaeon Methanobrevibacter and genus estingly, an alteration in the same direction was also reported Anaerotruncus may also be candidates for disorder specific- in 2 studies from the other disorder, which could not be ex- ity because they were consistently associated with anorexia plained by apparent demographic, clinical, or methodologi- nervosa and psychosis and schizophrenia, respectively. Inter- cal factors. Nevertheless, specificity in anorexia nervosa jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1347 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders Figure 2. Forest Plots of Alpha Diversity in the Gut Microbiota of Patients With Psychiatric Disorders Compared With Healthy Controls A Shannon index B Simpson index Source SMD (95% CI) Decreased Increased Source SMD (95% CI) Decreased Increased MDD MDD 43 26 Jiang et al, 2015 Naseribafrouei et al, 2014 0.55 (0.03 to 1.08) 0.29 (–0.27 to 0.86) 4 43 Kelly et al, 2016 –1.03 (–1.54 to –0.52) Jiang et al, 2015 –0.35 (–0.86 to 0.16) 50 3 Liu et al, 2016 –1.56 (–2.34 to –0.79) Zheng et al, 2016 0.00 (–0.35 to 0.36) 3 27 Zheng et al, 2016 –0.17 (–0.53 to 0.19) Chen et al, 2021 0.35 (–0.03 to 0.74) 44 21 Huang et al, 2018 –0.78 (–1.34 to –0.23) Jiang et al, 2020 0.60 (–0.14 to 1.33) Lai et al, 2019 0.45 (–0.08 to 0.99) Total 0.14 (–0.14 to 0.43) 25 2 2 Rong et al, 2019 0.53 (0.01 to 1.04) Heterogeneity: χ = 6.95; P =.14; I = 42% Chen et al, 2021 –0.21 (–0.60 to 0.17) Bipolar disorder 21 21 Jiang et al, 2020 –0.59 (–1.32 to 0.15) Jiang et al, 2020 0.00 (–0.79 to 0.79) 28 31 Liu et al, 2020 –0.11 (–0.53 to 0.30) Hu et al, 2019 –0.22 (–0.62 to 0.18) 22 52 Vinberg et al, 2019 –0.50 (–0.96 to –0.04) Lai et al, 2021 0.28 (–0.26 to 0.82) Total –0.28 (–0.62 to 0.06) Total –0.03 (–0.34 to 0.28) 2 2 2 2 Heterogeneity: χ = 50.55; P <.001; I = 80% Heterogeneity: χ = 2.14; P =.34; I = 6% 10 2 Bipolar disorder Schizophrenia and psychosis 21 32 Jiang et al, 2020 –0.17 (–0.96 to 0.63) Shen et al, 2018 0.20 (–0.17 to 0.56) 25 33 Rong et al, 2019 Pan et al, 2020 –0.41 (–0.93 to 0.11) 0.46 (–0.05 to 0.97) Hu et al, 2019 –0.44 (–0.84 to –0.03) Total –0.08 (–0.68 to 0.52) 52 2 2 Lai et al, 2021 0.53 (–0.02 to 1.08) Heterogeneity: χ = 3.56; P =.06; I = 72% Total 0.09 (–0.43 to 0.62) OCD 2 2 40 Heterogeneity: χ = 11.19; P =.01; I = 73% Domènech et al, 2020 –0.12 (–0.58 to 0.35) –0.12 (–0.58 to 0.35) Schizophrenia and psychosis Total Shen et al, 2018 –0.16 (–0.52 to 0.21) Heterogeneity NA Zheng et al, 2019 –0.22 (–0.57 to 0.12) Total 0.04 (–0.13 to 0.21) Pan et al, 2020 0.36 (–0.16 to 0.88) 95% PI (–0.36 to 0.45) 53 2 2 Zhu et al, 2020 0.30 (–0.01 to 0.60) Heterogeneity: χ = 14.19; P =.16; I = 30% Li et al, 2020 –0.12 (–0.43 to 0.19) –2 –1 0 1 2 Ma et al, 2020 (FEP) 0.11 (–0.28 to 0.50) SMD (95% CI) Ma et al, 2020 (schizophrenia) –0.33 (–0.65 to –0.01) Zhang et al, 2020 0.29 (–0.51 to 1.08) Phylogenetic diversity Total –0.02 (–0.20 to 0.17) Source SMD (95% CI) Decreased Increased 2 2 Heterogeneity: χ = 13.23; P =.07; I = 47% MDD Anorexia nervosa Kelly et al, 2016 –1.16 (–1.68 to –0.64) Mack et al, 2016 0.13 (–0.24 to 0.51) Zheng et al, 2016 0.15 (–0.20 to 0.51) Hanachi et al, 2019 –0.40 (–0.94 to 0.15) Huang et al, 2018 –0.54 (–1.09 to 0.00) Total –0.09 (–0.61 to 0.42) 2 2 Liu et al, 2020 –0.22 (–0.63 to 0.20) Heterogeneity: χ = 2.48; P =.12; I = 60% Total –0.42 (–0.96 to 0.13) PTSD 2 2 Heterogeneity: χ = 17.6; P <.001; I = 83% Hemmings et al, 2017 0.15 (–0.58 to 0.88) Schizophrenia and psychosis Total 0.15 (–0.58 to 0.88) Shen et al, 2018 –0.07 (–0.43 to 0.29) Heterogeneity NA Pan et al, 2020 0.29 (–0.23 to 0.81) OCD Li et al, 2020 –0.07 (–0.38 to 0.23) Domènech et al, 2020 –0.31 (–0.78 to 0.15) 41 Total –0.01 (–0.22 to 0.20) Turna et al, 2020 –0.53 (–1.14 to 0.08) 2 2 Heterogeneity: χ = 1.54; P =.46; I = 0% Total –0.40 (–0.77 to –0.02) 2 PTSD 2 2 Heterogeneity: χ = 0.31; P =.58; I = 0% Hemmings et al, 2017 –0.28 (–1.01 to 0.46) ADHD Total –0.28 (–1.01 to 0.46) Aarts et al, 2017 –0.10 (–0.60 to 0.40) Heterogeneity NA Total –0.10 (–0.60 to 0.40) OCD Heterogeneity NA Domènech et al, 2020 –0.51 (–0.99 to –0.04) Total –0.12 (–0.27 to 0.03) Total –0.51 (–0.99 to –0.04) 95% PI (–0.81 to 0.58) Heterogeneity NA 2 2 Heterogeneity: χ = 85.63; P <.001; I = 67% ADHD –3 –2 –1 0 1 2 3 Aarts et al, 2017 –0.20 (–0.70 to 0.31) SMD (95% CI) Total –0.20 (–0.70 to 0.31) Heterogeneity NA Total –0.24 (–0.47 to 0.00) 95% PI (–0.97 to 0.50) 2 2 Heterogeneity: χ = 24.66; P =.003; I = 64% –2 –1 0 1 2 SMD (95% CI) ADHD indicates attention-deficit/hyperactivity disorder; FEP, first episode psychosis; MDD, major depressive disorder; NA, not applicable; OCD, obsessive compulsive disorder; PI, prediction interval; PTSD, posttraumatic stress disorder; SMD, standardized mean difference. 1348 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research Figure 3. Changes in Relative Abundance of Microbial Taxa Reported by at Least 2 Studies From a Diagnostic Category A Level: phylum B Level: family Increased Decreased Not consistent AN BD MDD ANX SCZ C Level: genus: phylum firmicutes AN BD MDD ANX SCZ Level: genus: all other phyla AN BD MDD ANX SCZ Gray cells indicate not examined, not reported, or not replicated. bipolar disorder (BD), 9; major depressive disorder (MDD), 21; anxiety (ANX), 2; psychosis and schizophrenia (SCZ), 11. Most replicated findings are indicated here, all of which have been reported by more than 1 research group. Number of studies: anorexia nervosa (AN), 10; cannot be assessed here because no studies in other eating dis- of 10 studies). Further, Atopobium was enriched in bipolar dis- orders were identified, and conditions such as obesity were be- order and MDD (5 of 5 studies), while Veillonella was en- yond the scope. No distinct disorder-specific alterations were riched in psychosis and schizophrenia and MDD (5 of 6 stud- observed for the remaining taxa. ies). There was also evidence for the increase of the pathogen Escherichia-Shigella in bipolar disorder, anxiety, and psycho- sis and schizophrenia (6 of 7 studies) but not MDD. The Transdiagnostic Alterations Our findings indicate an overlap between certain disorders: bi- Bifidobacterium and Bacteroides genera were reported fre- polar disorder, psychosis and schizophrenia, and anxiety were quently but inconsistently across these disorders (14 and 16 associated with MDD. The most consistent changes were deple- studies, respectively). tion of Faecalibacterium (in 15 of 17 studies reporting this ge- nus) and Coprococcus (10 of 10 studies) and the enrichment of Exploring Confounders: Region and Psychiatric Medication Eggerthella (in 10 of 11 studies) (eAppendix 10 in the Supple- We explored the association of study region (east/west) with ment). These were followed by enriched Lactobacillus (10 of microbial alterations. Owing to the limited overlap in find- 13 studies), Enterococcus (8 of 9 studies), and Streptococcus (8 ings and the imbalanced availability of studies by region (eg, jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1349 Actinomyces Anaerostipes Actinobacteria Bifidobacterium Blautia Bacteroidetes Atopobium Coprococcus Firmicutes Collinsella Dorea Proteobacteria Eggerthella Roseburia Fusobacteria Olsenella Lachnoclostridium Alistipes Fusicatenibacter Actinomycetaceae Bacteroides Anaerotruncus Bifidobacteriaceae Odoribacter Faecalibacterium Coriobacteriaceae Parabacteroides Ruminococcus Bacteroidaceae Paraprevotella Subdoligranulum Prevotellaceae Prevotella Oscillibacter Rikenellaceae Bilophila Gemmiger Alcaligenaceae Desulfovibrio Ruminiclostridium 9 Desulfovibrionaceae Escherichia-Shigella Butyricicoccus Enterobacteriaceae Citrobacter Clostridium Pasteurellaceae Enterobacter Clostridium cl. IV Succinivibrionaceae Klebsiella Clostridium cl.XI Sutterellaceae Parasutterella Clostridium cl. XIVa Acidaminococcaceae Sutterella Coprobacillus Clostridiaceae Haemophilus Erysipelotrichaceae is. Enterococcaceae Succinivibrio Holdemania Eubacteriaceae Methanobrevibacter Turicibacter Lachnospiraceae Eubacterium Lactobacillaceae Eubacterium ventriosum Oscillospiraceae Dialister Peptostreptococcaceae Megasphaera Ruminococcaceae Veillonella Streptococcaceae Turicibacteraceae Lactobacillus Acidaminococcus Veillonellaceae Enterococcus Parvimonas Flavonifractor Phascolarctobacterium Streptococcus Megamonas Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders MDD and psychosis and schizophrenia were largely investi- altered taxa, suggesting these likely harbor transdiagnostic gated in the east, while anorexia nervosa and OCD were alterations associated with overlapping pathophysiology as investigated in the west), this analysis should be considered has previously been seen in analyses of inflammatory mark- preliminary. Clustering according to region identified several ers, neutrophil-lymphocyte ratios, and genome-wide asso- 12,56,57 taxa that were altered only in studies from Eastern coun- ciation studies. tries: Acidaminococcus (increased), Blautia (not consistent), Most consistently, the genus Eggerthella was enriched in Megamonas (decreased), Megasphaera (increased), MDD, bipolar disorder, and psychosis and schizophrenia, while Atopobium (increased), and Bacteroides (not consistent). the genera Faecalibacterium and Coprococcus were decreased These differences were driven entirely by studies from in all. Eggerthella is associated with gastrointestinal 10,58 China, highlighting the need to distinguish the Chinese inflammation, while Faecalibacterium has known anti- microbiome from other East Asian nations as more evidence inflammatory properties and is depleted in immune- 58,60 becomes available. mediated inflammatory diseases. These associations are There is evidence that psychiatric medication can affect likely mediated by short-chain fatty acid butyrate, as Faecali- 16,54 61 microbiota composition. To investigate this, we com- bacterium and Coprococcus are involved in its production, pared results from medication-free studies (n = 11) with while Eggerthella has been associated with its depletion. those in which 80% or more of patients were taking medica- Butyrate has a key role in maintaining mucosal integrity and tion (n = 21). We found that increases in the family Lacto- reducing inflammation via macrophage function and de- bacillaceae (although not member genus Lactobacillus) and crease in proinflammatory cytokines, while increasing anti- 61-63 the genera Clostridium, Klebsiella, and Megasphaera were inflammatory mediators. Further, Faecalibacterium was in- only reported in medicated groups, while Dialister was versely associated with depression severity in 2 MDD studies, 28,43,64,65 decreased in medicated and increased in medication-free 1 bipolar disorder study, and 1 anorexia nervosa study, groups. Further, 6 of 8 studies in treated patients reported suggesting depletion of this genus may be characteristic of the increases in Streptococcus, which was not reported in drug- depressive state, irrespective of diagnosis. Therefore, clinical free studies. features and underlying pathophysiology that manifest across diagnoses may be better suited to explain the observed mi- crobial alterations than distinct diagnostic categories. The mer- its of incorporating the gut microbiota as a dimensional com- Discussion ponent to the Research Domain Criteria have previously been To our knowledge, this is the first review to assess gut micro- discussed and our results reinforce this by demonstrating that biota perturbations across a spectrum of psychiatric disor- while gut microbiota abnormalities were ubiquitously ob- ders with the aim of evaluating the reproducibility and speci- served, these do not seem to congregate according to distinct ficity of potential gut microbial biomarkers. The pattern of diagnoses but instead exhibit a transdiagnostic pattern. alterations observed suggests an increased magnitude and Interestingly, the family Lactobacillaceae and member ge- complexity of microbial disorganization for some disorders nus Lactobacillus, strains from which are components of pro- compared with others. For example, the highest number of dif- biotic supplements and linked to positive health outcomes, ferentially abundant taxa was in psychosis and schizophre- were enriched in MDD, psychosis and schizophrenia, and bi- nia (136 taxa; 11 studies), despite almost twice as many stud- polar disorder. A possible explanation could be that species ies in MDD (94 taxa; 21 studies). Conversely, anorexia nervosa from this genus have differential effect. For example, 1 study was associated with fewer differences (32 taxa; 10 studies), de- identified the increase in psychosis and schizophrenia to be spite the larger number of studies compared with anxiety (36 in subspecies not typically present in the healthy gut. Alter- taxa; 2 studies) and bipolar disorder (60 taxa; 9 studies). This natively, increased Lactobacillus has previously been associ- is reminiscent of genome-wide association studies’ findings, ated with antipsychotic use. This finding was somewhat cor- in which the highest number of loci have been associated with roborated here, as 4 psychosis and schizophrenia studies psychosis and schizophrenia followed by MDD and bipolar dis- reporting increased Lactobacillus were conducted in medi- 32,34,46,70 order, and fewer have been associated with anorexia ner- cated groups, while the one that reported decreased 55 71 vosa, PTSD, and ADHD. This increased complexity, also re- Lactobacillus was in a treatment-naive group. In our explor- flected in the microbiota, is consistent with the wider spectrum atory analyses, the family Lactobacillaceae was significantly of clinical presentations associated with the former com- increased only in medicated groups. This suggests that psy- pared with the latter set of disorders. chotropic medication may be exacerbating the presence of Overall, we did not find evidence for disorder specificity: illness-associated Lactobacilli species. whenever microbial alterations merited specificity, these Measures of alpha diversity (within sample) were widely were weakly reproduced, suggesting they may instead used, following the general assumption that higher diversity reflect specific population characteristics (eg, depression is more beneficial to the host and thus expected to be subtype) and thus need further verification. Instead, our decreased in psychiatric patients, as has previously been findings indicated that certain disorders share similar pat- observed for various diseases. However, our meta-analysis terns of microbial changes. Specifically, we observed an demonstrated a nonsignificant association with diversity overlap between psychosis and schizophrenia, bipolar disor- indices and small to medium decrease in richness, suggest- der, anxiety, and MDD in consistently and inconsistently ing that while richness is somewhat compromised (although 1350 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research the clinical significance of this decrease is unclear), diver- Limitations sity is overall preserved. The high residual heterogeneity Although there were insufficient studies to perform in-depth following subgrouping according to disorder type suggests analyses of OCD, PTSD, and anxiety, we believe the inclusion that diagnosis is not a good discriminator of alpha diversity. of these disorders provides a comprehensive overview of cur- Regarding beta diversity (between samples), patients with rentevidence.Therewerenostudiesinadultswithautismspec- MDD and psychosis and schizophrenia consistently clus- trumdisorderandonly1studyinADHD,thusprecludingusfrom tered differently from controls. However, it is yet unknown comparing the association of neurodevelopmental disorders whether psychiatric disorders cluster differently from one with the microbiota in adulthood. The decision to exclude stud- another, thus questioning the suitability of diversity mea- ies in children and elderly individuals was dictated by an ap- sures as biomarkers. From the studies summarized, only 2 preciation of the specialist nature of these populations and the studies compared beta diversity cross-diagnostically and substantialage-relateddifferencesinthemicrobiota. Next,we 21,25 neither found a significant difference. acknowledge that the division into Eastern and Western coun- Among the numerous clinical and demographic factors tries is a crude approach to controlling for geographical differ- that may have contributed to the widespread inconsisten- ences in diet and genetics and does not allow detection of re- cies between studies, current evidence allowed us to gional variations in the microbiome, which might also explain explore 2 key characteristics: geographical region and psy- why we found no alterations specific to Western populations. chiatric medication. Geographical region and the associated As more studies become available, more nuanced analyses will factor of diet can profoundly affect the composition of the bepossible.Additionally,moststudieshadmodestsamplesizes, 74,75 microbiota. Our analysis suggested that some of the suggesting our analyses may still be underpowered and pre- observed perturbations may be specific to Chinese popula- liminary.Similarly,asmoststudiesincludedbothmedicatedand tions (eg, increased Acidaminococcus), others may be owing unmedicated patients, our analyses of the confounding ef- to the effect of psychiatric medication (eg, increased Klebsi- fects of medication require further verification in larger strati- ella and decreased Dialister), while third may be influenced fied populations. Our summary may also suffer from the use of by a combination of both, such as the genus Megasphaera, different reference databases between studies, as inconsisten- which was enriched only in Chinese populations undergo- cies in assigning taxonomy have been described. Finally, the ing treatment. Future studies should be encouraged to aim of this review was to evaluate gut microbial composition, report findings (even nonsignificant) on all dominant taxa rather than function. Early evidence has suggested that func- to help delineate the effect of confounders from true dis- tionalpotentialsassociatedwithpsychiatricillnessincludeshort- ease effects. Additionally, more studies will be needed in chain fatty acid synthesis, tryptophan metabolism, and neu- 52,53,83,84 currently underrepresented populations from low- to rotransmitter synthesis/degradation. Given the noted middle-income countries, as mental health problems functional redundancy, functional analysis will be key in un- become an increasing concern. derstanding the role of host-microbiome interactions in neu- For brevity, we have not discussed methodological differ- ropsychiatric disorders. ences that may have contributed to inconsistent findings such as processing, sequencing, or analysis pipelines because these 9,10,74,77,78 have been extensively reviewed by others. Further, Conclusions some have suggested that the current approaches to microbi- ome analyses may be unreliable owing to inappropriate han- This review suggests a transdiagnostic commonality of micro- dling of inherently compositional data. The lack of power cal- bial disturbances in MDD, bipolar disorder, anxiety, and culationsisasignificantdeterrentinthefield.Tomoveforward, psychosis and schizophrenia, characterized by depleted anti- the reporting of quantitative effect sizes of abundance find- inflammatory butyrate-producing bacteria and enriched pro- ings in addition to P values is needed to enable meta- inflammatory bacteria. The effect of key confounders such as analyses and the evaluation of potentially relevant biological psychiatric medication and diet should be carefully considered. effects. Even then, technical and clinical variation between Researchersshouldinterprettheirfindingswithinthelargercon- studies may make it difficult to compare effect sizes, which re- text of psychiatric disorders to prevent unmerited claims of dis- inforces the need of harmonizing methodologies and encour- orderspecificityofgutmicrobialbiomarkers.Theevidencesum- aging data sharing with sufficient metadata. marized here is a good starting point for such comparisons. ARTICLE INFORMATION Author Affiliations: Centre for Affective Disorders, Life Sciences, ZIEL–Institute for Food & Health, Institute of Psychiatry, Psychology & Neuroscience, Technical University of Munich, Freising, Germany Accepted for Publication: July 21, 2021. King’s College London, London, United Kingdom (L. J. Hall); National Institute for Health Research Published Online: September 15, 2021. (Nikolova, Cleare, Stone, Young); Department of Biomedical Research Centre at South London and doi:10.1001/jamapsychiatry.2021.2573 Psychosis Studies, Institute of Psychiatry, Maudsley NHS Foundation Trust, King’s College Correction: This article was corrected on December Psychology and Neuroscience, King’s College of London, London, United Kingdom (Cleare, Young); 1, 2021, to add indication of the open access license. London, London, United Kingdom (M. R. B. Hall); South London and Maudsley NHS Foundation Trust, Quadram Institute Bioscience, Norwich Research Bethlem Royal Hospital, Beckenham, United Open Access: This is an open access article Park, Norwich, United Kingdom (L. J. Hall); Norwich Kingdom (Cleare, Young); Brighton and Sussex distributed under the terms of the CC-BY License. Medical School, University of East Anglia, Norwich Medical School, Brighton, United Kingdom (Stone). © 2021 Nikolova VL et al. JAMA Psychiatry. Research Park, Norwich, United Kingdom (L. J. Hall); Chair of Intestinal Microbiome, School of jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1351 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders Author Contributions: Ms Nikolova had full access induces neurobehavioural changes in the rat. systematic reviews was developed. J Clin Epidemiol. to all of the data in the study and takes J Psychiatr Res. 2016;82:109-118. doi:10.1016/ 2016;69:225-234. doi:10.1016/j.jclinepi.2015.06.005 responsibility for the integrity of the data and the j.jpsychires.2016.07.019 18. Moola S, Munn Z, Tufanaru C, et al Systematic accuracy of the data analysis. 5. Zhu F, Guo R, Wang W, et al. Transplantation of reviews of etiology and risk. In: Aromataris E, Munn Concept and design: Nikolova, Hall, Cleare, Stone, microbiota from drug-free patients with Z, eds. Joanna Briggs Institute Reviewer’s Manual. The Young. schizophrenia causes schizophrenia-like abnormal Joanna Briggs Institute; 2017. Acquisition, analysis, or interpretation of data: behaviors and dysregulated kynurenine 19. Duvallet C, Gibbons SM, Gurry T, Irizarry RA, All authors. metabolism in mice. Mol Psychiatry. 2020;25(11): Alm EJ. Meta-analysis of gut microbiome studies Drafting of the manuscript: Nikolova, Stone, Young. 2905-2918. doi:10.1038/s41380-019-0475-4 identifies disease-specific and shared responses. Critical revision of the manuscript for important 6. Li N, Wang Q, Wang Y, et al. Fecal microbiota Nat Commun. 2017;8(1):1784. doi:10.1038/s41467- intellectual content: All authors. transplantation from chronic unpredictable mild 017-01973-8 Statistical analysis: Nikolova. stress mice donors affects anxiety-like and Obtained funding: Hall, Cleare, Young. 20. Balduzzi S, Rücker G, Schwarzer G. How to depression-like behavior in recipient mice via the Administrative, technical, or material support: perform a meta-analysis with R: a practical tutorial. gut microbiota-inflammation-brain axis. Stress. Smith, Young. Evid Based Ment Health. 2019;22(4):153-160. 2019;22(5):592-602. doi:10.1080/10253890.2019. Supervision: Hall, Cleare, Stone, Young. doi:10.1136/ebmental-2019-300117 Conflict of Interest Disclosures: Dr Cleare 21. Jiang HY, Pan LY, Zhang X, Zhang Z, Zhou YY, 7. Sharon G, Cruz NJ, Kang D-W, et al. Human gut reported grants from Protexin Probiotics Ruan B. Altered gut bacterial-fungal interkingdom microbiota from autism spectrum disorder promote International (industrial partner of the Medical networks in patients with current depressive behavioral symptoms in mice. Cell. 2019;177(6): Research Council studentship that Ms Nikolova is episode. Brain Behav. 2020;10(8):e01677. 1600-1618.e17. doi:10.1016/j.cell.2019.05.004 funded by) outside the submitted work; received doi:10.1002/brb3.1677 honoraria for educational activities from Lundbeck 8. Sanada K, Nakajima S, Kurokawa S, et al. Gut 22. Vinberg M, Ottesen NM, Meluken I, et al. and Janssen in the last 3 years; honoraria for microbiota and major depressive disorder: Remitted affective disorders and high familial risk of consulting from Allergan, Livanova, and Janssen; a systematic review and meta-analysis. J Affect affective disorders associate with aberrant and sponsorship for conference attendance from Disord. 2020;266:1-13. doi:10.1016/j.jad.2020.01.102 intestinal microbiota. Acta Psychiatr Scand. 2019; Janssen. Ms Nikolova reported personal fees from 9. Di Lodovico L, Mondot S, Doré J, Mack I, Hanachi 139(2):174-184. doi:10.1111/acps.12976 Janssen. Dr Stone reported grants from Protexin M, Gorwood P. Anorexia nervosa and gut 23. Mason BL, Li Q, Minhajuddin A, et al. Reduced Probiotics International, personal fees from microbiota: a systematic review and quantitative anti-inflammatory gut microbiota are associated Janssen, and grants from Takeda outside the synthesis of pooled microbiological data. Prog with depression and anhedonia. J Affect Disord. submitted work. Dr Young has received honoraria Neuropsychopharmacol Biol Psychiatry. 2021;106: 2020;266:394-401. doi:10.1016/j.jad.2020.01.137 for speaking from AstraZeneca, Lundbeck, Eli Lilly 110114. doi:10.1016/j.pnpbp.2020.110114 and Company, and Sunovion; honoraria for 24. Stevens BR, Goel R, Seungbum K, et al. 10. Simpson CA, Diaz-Arteche C, Eliby D, Schwartz consulting from Allergan, Livanova, Lundbeck, Increased human intestinal barrier permeability OS, Simmons JG, Cowan CSM. The gut microbiota in Sunovion, and Janssen; and research grants from plasma biomarkers zonulin and FABP2 correlated anxiety and depression: a systematic review. Clin Janssen, Compass, and Protexin Probiotics with plasma LPS and altered gut microbiome in Psychol Rev. 2021;83:101943. doi:10.1016/j.cpr. International in the last 3 years. No other anxiety or depression. Gut. 2018;67(8):1555-1557. 2020.101943 disclosures were reported. doi:10.1136/gutjnl-2017-314759 11. IPCS INCHEM. Environmental health criteria Funding/Support: Ms Nikolova is funded by a 25. Rong H, Xie XH, Zhao J, et al. Similarly in 222: Biomarkers in risk assessment: validity and Medical Research Council PhD Studentship. This depression, nuances of gut microbiota: evidences validation. Accessed February 8, 2021. http://www. article represents independent research partly from a shotgun metagenomics sequencing study on inchem.org/documents/ehc/ehc/ehc222.htm funded by the National Institute for Health major depressive disorder versus bipolar disorder Research Biomedical Research Centre at South 12. Yuan N, Chen Y, Xia Y, Dai J, Liu C. with current major depressive episode patients. London and Maudsley National Health Service Inflammation-related biomarkers in major J Psychiatr Res. 2019;113:90-99. doi:10.1016/ Foundation Trust and King’s College London. psychiatric disorders: a cross-disorder assessment j.jpsychires.2019.03.017 of reproducibility and specificity in 43 Role of the Funder/Sponsor: The funders had 26. Naseribafrouei A, Hestad K, Avershina E, et al. meta-analyses. Transl Psychiatry. 2019;9(1):233. no role in the design and conduct of the study; Correlation between the human fecal microbiota doi:10.1038/s41398-019-0570-y collection, management, analysis, and and depression. Neurogastroenterol Motil. 2014;26 interpretation of the data; preparation, review, or 13. Hutton B, Salanti G, Caldwell DM, et al. The (8):1155-1162. doi:10.1111/nmo.12378 approval of the manuscript; and decision to submit PRISMA extension statement for reporting of 27. Chen YH, Xue F, Yu SF, et al. Gut microbiota the manuscript for publication. systematic reviews incorporating network dysbiosis in depressed women: the association of meta-analyses of health care interventions: Disclaimer: The views expressed are those of the symptom severity and microbiota function. J Affect checklist and explanations. Ann Intern Med. 2015; authors and not necessarily those of the UK Disord. 2021;282:391-400. doi:10.1016/j.jad.2020. 162(11):777-784. doi:10.7326/M14-2385 National Health Service, the National Institute for 12.143 Health Research, or Department of Health. 14. Pollock M, Fernandes RM, Becker LA, Pieper D, 28. Liu RT, Rowan-Nash AD, Sheehan AE, et al. Hartling L. Chapter V: overviews of reviews. In: Reductions in anti-inflammatory gut bacteria are REFERENCES Higgins JPT, Thomas J, Chandler J, Cumpston M, Li associated with depression in a sample of young T, Page MJ, Welch VA, eds. Cochrane Handbook for 1. Phillips JGP. The treatment of melancholia by the adults. Brain Behav Immun. 2020;88:308-324. Systematic Reviews of Interventions, version 6.1. lactic acid bacillus. J Mental Sci. 1910;56(234):422- doi:10.1016/j.bbi.2020.03.026 Cochrane; 2020. 430. doi:10.1192/bjp.56.234.422 29. Painold A, Mörkl S, Kashofer K, et al. A step 15. Cumpston M, Changler J. Chapter IV: updating a 2. Bested AC, Logan AC, Selhub EM. Intestinal ahead: exploring the gut microbiota in inpatients review. In: Higgins JPT, Thomas J, Chandler J, microbiota, probiotics and mental health: from with bipolar disorder during a depressive episode. Cumpston M, Li T, Page MJ, Welch VA, eds. Metchnikoff to modern advances: part I: Bipolar Disord. 2019;21(1):40-49. doi:10.1111/bdi.12682 Cochrane Handbook for Systematic Reviews of autointoxication revisited. Gut Pathog. 2013;5(1):5. 30. Coello K, Hansen TH, Sørensen N, et al. Gut Interventions, version 6.1. Cochrane; 2020. doi:10.1186/1757-4749-5-5 microbiota composition in patients with newly 16. Ait Chait Y, Mottawea W, Tompkins TA, 3. Zheng P, Zeng B, Zhou C, et al. Gut microbiome diagnosed bipolar disorder and their unaffected Hammami R. Unravelling the antimicrobial action of remodeling induces depressive-like behaviors first-degree relatives. Brain Behav Immun. 2019;75: antidepressants on gut commensal microbes. Sci Rep. through a pathway mediated by the host’s 112-118. doi:10.1016/j.bbi.2018.09.026 2020;10(1):17878. doi:10.1038/s41598-020- metabolism. Mol Psychiatry. 2016;21(6):786-796. 31. Hu S, Li A, Huang T, et al. Gut microbiota 74934-9 doi:10.1038/mp.2016.44 changes in patients with bipolar depression. Adv Sci 17. Whiting P, Savović J, Higgins JPT, et al; ROBIS 4. Kelly JR, Borre Y, O’ Brien C, et al. Transferring (Weinh). 2019;6(14):1900752. doi:10.1002/advs. group. ROBIS: a new tool to assess risk of bias in the blues: depression-associated gut microbiota 201900752 1352 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research 32. Shen Y, Xu J, Li Z, et al. Analysis of gut 46. Ma X, Asif H, Dai L, et al. Alteration of the gut common dysbiosis exist? Microbiome. 2018;6(1):221. microbiota diversity and auxiliary diagnosis as a microbiome in first-episode drug-naïve and chronic doi:10.1186/s40168-018-0603-4 biomarker in patients with schizophrenia: medicated schizophrenia correlate with regional 59. Qiu X, Zhang M, Yang X, Hong N, Yu C. a cross-sectional study. Schizophr Res. 2018;197: brain volumes. J Psychiatr Res. 2020;123:136-144. Faecalibacterium prausnitzii upregulates regulatory 470-477. doi:10.1016/j.schres.2018.01.002 doi:10.1016/j.jpsychires.2020.02.005 T cells and anti-inflammatory cytokines in treating 33. Pan R, Zhang X, Gao J, Yi W, Wei Q, Su H. 47. Xu R, Wu B, Liang J, et al. Altered gut TNBS-induced colitis. J Crohns Colitis. 2013;7(11): Analysis of the diversity of intestinal microbiome microbiota and mucosal immunity in patients with e558-e568. doi:10.1016/j.crohns.2013.04.002 and its potential value as a biomarker in patients schizophrenia. Brain Behav Immun. 2020;85:120-127. 60. Sokol H, Seksik P, Furet JP, et al. Low counts of with schizophrenia: a cohort study. Psychiatry Res. doi:10.1016/j.bbi.2019.06.039 Faecalibacterium prausnitzii in colitis microbiota. 2020;291:113260. doi:10.1016/j.psychres.2020. 48. Hanachi M, Manichanh C, Schoenenberger A, Inflamm Bowel Dis. 2009;15(8):1183-1189. et al. Altered host-gut microbes symbiosis in doi:10.1002/ibd.20903 34. Li S, Zhuo M, Huang X, et al. Altered gut severely malnourished anorexia nervosa (AN) 61. Miquel S, Martín R, Bridonneau C, et al. Ecology microbiota associated with symptom severity in patients undergoing enteral nutrition: an and metabolism of the beneficial intestinal schizophrenia. PeerJ. 2020;8:e9574. doi:10.7717/ explicative factor of functional intestinal disorders? commensal bacterium Faecalibacterium prausnitzii. peerj.9574 Clin Nutr. 2019;38(5):2304-2310. doi:10.1016/ Gut Microbes. 2014;5(2):146-151. doi:10.4161/gmic. j.clnu.2018.10.004 35. Zhang X, Pan L-Y, Zhang Z, Zhou Y-Y, Jiang H-Y, 27651 Ruan B. Analysis of gut mycobiota in first-episode, 49. Monteleone AM, Troisi J, Fasano A, et al. 62. Chang PV, Hao L, Offermanns S, Medzhitov R. drug-naïve Chinese patients with schizophrenia: Multi-omics data integration in anorexia nervosa The microbial metabolite butyrate regulates a pilot study. Behav Brain Res. 2020;379:112374. patients before and after weight regain: intestinal macrophage function via histone doi:10.1016/j.bbr.2019.112374 A microbiome-metabolomics investigation. Clin Nutr. deacetylase inhibition. Proc Natl Acad SciUSA. 2021;40(3):1137-1146. doi:10.1016/j.clnu.2020.07. 36. Jiang H-Y, Zhang X, Yu Z-H, et al. Altered gut 2014;111(6):2247-2252. doi:10.1073/pnas.1322269111 microbiota profile in patients with generalized 63. Sublette ME, Cheung S, Lieberman E, et al. anxiety disorder. J Psychiatr Res. 2018;104:130-136. 50. Liu Y, Zhang L, Wang X, et al. Similar fecal Bipolar disorder and the gut microbiome: doi:10.1016/j.jpsychires.2018.07.007 microbiota signatures in patients with a systematic review. Bipolar Disord. Published online diarrhea-predominant irritable bowel syndrome 37. Kleiman SC, Watson HJ, Bulik-Sullivan EC, et al. January 29, 2021. doi:10.1111/bdi.13049 and patients with depression. Clin Gastroenterol The intestinal microbiota in acute anorexia nervosa 64. Evans SJ, Bassis CM, Hein R, et al. The gut Hepatol. 2016;14(11):1602-1611.e5. doi:10.1016/j.cgh. and during renourishment: relationship to microbiome composition associates with bipolar 2016.05.033 depression, anxiety, and eating disorder disorder and illness severity. J Psychiatr Res.2017; psychopathology. Psychosom Med. 2015;77(9): 51. Lai W-T, Deng W-F, Xu S-X, et al. Shotgun 87:23-29. doi:10.1016/j.jpsychires.2016.12.007 969-981. doi:10.1097/PSY.0000000000000247 metagenomics reveals both taxonomic and 65. Chen Y-H, Bai J, Wu D, et al. Association tryptophan pathway differences of gut microbiota 38. Mack I, Cuntz U, Grämer C, et al. Weight gain in between fecal microbiota and generalized anxiety in major depressive disorder patients. Psychol Med. anorexia nervosa does not ameliorate the faecal disorder: severity and early treatment response. 2021:51(1):90-101. doi:10.1017/S0033291719003027 microbiota, branched chain fatty acid profiles, and J Affect Disord. 2019;259:56-66. doi:10.1016/j.jad. gastrointestinal complaints. Sci Rep. 2016;6:26752. 52. Lai WT, Zhao J, Xu SX, et al. Shotgun 2019.08.014 doi:10.1038/srep26752 metagenomics reveals both taxonomic and 66. Insel T, Cuthbert B, Garvey M, et al. Research tryptophan pathway differences of gut microbiota 39. Hemmings SMJ, Malan-Müller S, van den domain criteria (RDoC): toward a new classification in bipolar disorder with current major depressive Heuvel LL, et al. The microbiome in posttraumatic framework for research on mental disorders. Am J episode patients. J Affect Disord. 2021;278:311-319. stress disorder and trauma-exposed controls: an Psychiatry. 2010;167(7):748-751. doi:10.1176/ doi:10.1016/j.jad.2020.09.010 exploratory study. Psychosom Med. 2017;79(8): appi.ajp.2010.09091379 936-946. doi:10.1097/PSY.0000000000000512 53. Zhu F, Ju Y, Wang W, et al. Metagenome-wide 67. Kelly JR, Clarke G, Cryan JF, Dinan TG. association of gut microbiome features for 40. Domènech L, Willis J, Alemany M, et al. Dimensional thinking in psychiatry in the era of the schizophrenia. Nat Commun. 2020;11(1):1612. Changes in the stool and oropharyngeal Research Domain Criteria (RDoC). Ir J Psychol Med. doi:10.1038/s41467-020-15457-9 microbiome in obsessive-compulsive disorder. 2018;35(2):89-94. doi:10.1017/ipm.2017.7 medRxiv. Preprint posted online May 28, 2020. 54. Macedo D, Filho AJMC, Soares de Sousa CN, 68. Yong SJ, Tong T, Chew J, Lim WL. doi:10.1101/2020.05.26.20113779 et al. Antidepressants, antimicrobials or both? gut Antidepressive mechanisms of probiotics and their microbiota dysbiosis in depression and possible 41. Turna J, Grosman Kaplan K, Anglin R, et al. The therapeutic potential. Front Neurosci. 2020;13:1361. implications of the antimicrobial effects of gut microbiome and inflammation in doi:10.3389/fnins.2019.01361 antidepressant drugs for antidepressant obsessive-compulsive disorder patients compared effectiveness. J Affect Disord. 2017;208:22-32. 69. Bahr SM, Tyler BC, Wooldridge N, et al. Use of to age- and sex-matched controls: a pilot study. doi:10.1016/j.jad.2016.09.012 the second-generation antipsychotic, risperidone, Acta Psychiatr Scand. 2020;142(4):337-347. and secondary weight gain are associated with an doi:10.1111/acps.13175 55. Ikeda M, Saito T, Kanazawa T, Iwata N. altered gut microbiota in children. Transl Psychiatry. Polygenic risk score as clinical utility in psychiatry: 42. Aarts E, Ederveen THA, Naaijen J, et al. Gut 2015;5(10):e652-e652. doi:10.1038/tp.2015.135 a clinical viewpoint. J Hum Genet. 2021;66(1):53-60. microbiome in ADHD and its relation to neural doi:10.1038/s10038-020-0814-y 70. Schwarz E, Maukonen J, Hyytiäinen T, et al. reward anticipation. PLoS One. 2017;12(9):e0183509. Analysis of microbiota in first episode psychosis doi:10.1371/journal.pone.0183509 56. Lee SH, Ripke S, Neale BM, et al; identifies preliminary associations with symptom Cross-Disorder Group of the Psychiatric Genomics 43. Jiang H, Ling Z, Zhang Y, et al. Altered fecal severity and treatment response. Schizophr Res. Consortium; International Inflammatory Bowel microbiota composition in patients with major 2018;192:398-403. doi:10.1016/j.schres.2017.04.017 Disease Genetics Consortium (IIBDGC). Genetic depressive disorder. Brain Behav Immun. 2015;48: relationship between five psychiatric disorders 71. Yuan X, Zhang P, Wang Y, et al. Changes in 186-194. doi:10.1016/j.bbi.2015.03.016 estimated from genome-wide SNPs. Nat Genet. metabolism and microbiota after 24-week 44. Huang Y, Shi X, Li Z, et al. Possible association 2013;45(9):984-994. doi:10.1038/ng.2711 risperidone treatment in drug naïve, normal weight of Firmicutes in the gut microbiota of patients with patients with first episode schizophrenia. Schizophr 57. Brinn A, Stone J. Neutrophil-lymphocyte ratio major depressive disorder. Neuropsychiatr Dis Treat. Res. 2018;201:299-306. doi:10.1016/j.schres.2018. across psychiatric diagnoses: a cross-sectional 2018;14:3329-3337. doi:10.2147/NDT.S188340 05.017 study using electronic health records. BMJ Open. 45. Zheng P, Zeng B, Liu M, et al. The gut 2020;10(7):e036859. doi:10.1136/bmjopen-2020- 72. Shade A. Diversity is the question, not the microbiome from patients with schizophrenia 036859 answer. ISME J. 2017;11(1):1-6. doi:10.1038/ismej. modulates the glutamate-glutamine-GABA cycle 2016.118 58. Forbes JD, Chen C-Y, Knox NC, et al. and schizophrenia-relevant behaviors in mice. Sci Adv. A comparative study of the gut microbiota in 73. Gerritsen J, Smidt H, Rijkers GT, de Vos WM. 2019;5(2):eaau8317. doi:10.1126/sciadv.aau8317 immune-mediated inflammatory diseases-does a Intestinal microbiota in human health and disease: jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1353 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders the impact of probiotics. Genes Nutr. 2011;6(3): 78. Bastiaanssen TFS, Cowan CSM, Claesson MJ, 82. Balvočiūtė M, Huson DH. SILVA, RDP, 209-240. doi:10.1007/s12263-011-0229-7 Dinan TG, Cryan JF. Making sense of…the Greengenes, NCBI and OTT: how do these microbiome in psychiatry. Int J taxonomies compare? BMC Genomics. 2017;18(2) 74. Cheung SG, Goldenthal AR, Uhlemann A-C, Neuropsychopharmacol. 2019;22(1):37-52. (suppl 2):114. doi:10.1186/s12864-017-3501-4 Mann JJ, Miller JM, Sublette ME. Systematic review doi:10.1093/ijnp/pyy067 of gut microbiota and major depression. Front 83. Valles-Colomer M, Falony G, Darzi Y, et al. The Psychiatry. 2019;10:34. doi:10.3389/fpsyt.2019. 79. Gloor GB, Macklaim JM, Pawlowsky-Glahn V, neuroactive potential of the human gut microbiota 00034 Egozcue JJ. Microbiome datasets are in quality of life and depression. Nat Microbiol. compositional: and this is not optional. Front 2019;4(4):623-632. doi:10.1038/s41564-018- 75. Simpson HL, Campbell BJ. Review article: Microbiol. 2017;8:2224. doi:10.3389/fmicb.2017. 0337-x dietary fibre-microbiota interactions. Aliment Pharmacol Ther. 2015;42(2):158-179. doi:10.1111/apt. 84. Yang J, Zheng P, Li Y, et al. Landscapes of 13248 80. Debelius J, Song SJ, Vazquez-Baeza Y, Xu ZZ, bacterial and metabolic signatures and their Gonzalez A, Knight R. Tiny microbes, enormous interaction in major depressive disorders. Sci Adv. 76. Sankoh O, Sevalie S, Weston M. Mental health impacts: what matters in gut microbiome studies? 2020;6(49):eaba8555. doi:10.1126/sciadv.aba8555 in Africa. Lancet Glob Health. 2018;6(9):e954-e955. Genome Biol. 2016;17(1):217. doi:10.1186/s13059- doi:10.1016/S2214-109X(18)30303-6 85. Heintz-Buschart A, Wilmes P. Human gut 016-1086-x microbiome: function matters. Trends Microbiol. 77. Vindegaard N, Speyer H, Nordentoft M, 81. Odamaki T, Kato K, Sugahara H, et al. 2018;26(7):563-574. doi:10.1016/j.tim.2017.11.002 Rasmussen S, Benros ME. Gut microbial changes of Age-related changes in gut microbiota composition patients with psychotic and affective disorders: from newborn to centenarian: a cross-sectional a systematic review. Schizophr Res. 2020;S0920- study. BMC Microbiol. 2016;16(1):90. doi:10.1186/ 9964(19)30584-5. doi:10.1016/j.schres.2019.12.014 s12866-016-0708-5 1354 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Supplemental Online Content Nikolova VL, Smith MRB, Hall LJ, Cleare AJ, Stone JM, Young AH. Perturbations in gut microbiota composition in psychiatric disorders: a systematic review and meta-analysis. JAMA Psychiatry. Published online September 15, 2021. doi:10.1001/jamapsychiatry.2021.2573 eAppendix 1. Systematic search details eAppendix 2. Quality assessment eAppendix 3. Detailed methods of the meta-analysis performed eAppendix 4. PRISMA flowcharts for the umbrella review search and the updated review searches eAppendix 5. Details of the identified systematic reviews eAppendix 6. Detailed characteristics of the included studies eAppendix 7. Stool sample processing methods in the included studies eAppendix 8. Publication bias assessment for the alpha diversity meta-analyses eAppendix 9. Beta diversity eAppendix 10. Figures for study-level findings of relative abundance of microbial taxa This supplemental material has been provided by the authors to give readers additional information about their work. © 2021 American Medical Association. All rights reserved. eAppendix 1. Systematic search details 1. Systematic reviews and meta-analysis search – performed on 27 Jan 2021 1.1. Cochrane Library Search Hits Cochrane Reviews 8498 Topic Mental Health 637 AND gut OR gastrointestinal OR intestinal OR feacal OR fecal OR stool 56 AND microbiome OR microbiota OR ecosystem OR bacteria OR flora OR microflora OR dysbiosis 0 1.2. PubMed* Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 3699 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((depression) OR (depress*) OR (mdd) OR (trd) OR (bipolar) OR (mania) OR (bipolar depression) OR (anxiety) OR (psychosis) OR (schizophrenia) OR (obsessive compulsive disorder) OR (ocd) OR (ptsd) OR (post-traumatic stress disorder) OR (adhd) OR (attention deficit hyperactivity disorder) OR (autism) OR (autism spectrum disorder) OR (ASD) OR (eating disorder) OR (anorexia) OR (bulimia)) AND (systematicreview[Filter]) 80 Limits: 2005-current 79 Limits: human 47 Limits: English 45 1.3. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138475 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612345 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370239 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87064 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60712 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42743 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68233 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111701 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464415 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56856 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14943 © 2021 American Medical Association. All rights reserved. 14. depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1041348 15. depress*.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1193818 16. MDD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 32383 17. TRD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 3628 18. bipolar.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 159279 19. mania.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42988 20. bipolar depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 10336 21. anxiety.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 626469 22. psychosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 186882 23. schizophrenia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 341812 24. obsessive compulsive disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 50116 mh] 25. OCD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 26104 26. PTSD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 72988 27. post-traumatic stress disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 28176 mh] 28. autism.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 134153 29. Autism spectrum disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42153 30. pervasive developmental disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, 4354 tm, mh] 31. attention deficit hyperactivity disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, 61349 id, tm, mh] 32. ADHD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 69132 33. eating disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 49763 34. anorexia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108326 35. bulimia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 28823 36. systematic review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 386059 37. meta-analysis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 338776 38. 1 or 2 or 3 or 4 or 5 or 6 1066368 39. 7 or 8 or 9 or 10 or 11 or 12 or 13 693756 40. 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 2248227 30 or 31 or 32 or 33 or 34 or 35 41. 36 or 37 557086 42. 38 AND 39 AND 40 AND 41 252 43. limit 42 to english language 241 44. limit 43 to human 236 45. limit 44 to yr="2005 -Current" 234 © 2021 American Medical Association. All rights reserved. 2. Supplementary searches – all performed on 02 Feb 2021 2.1. PTSD & OCD – no date restriction as no systematic reviews were available 2.1.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 45 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((obsessive compulsive disorder) OR (ocd) OR (ptsd) OR (post-traumatic stress disorder) Limits: human, English 20 2.1.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. obsessive compulsive disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 50157 mh] 15. OCD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 26128 16. PTSD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 73062 17. post-traumatic stress disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 28199 mh] 18. 1 or 2 or 3 or 4 or 5 or 6 1067443 19. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 20. 14 or 15 or 16 or 17 133639 21. 18 AND 19 AND 20 98 22. limit 42 to english language 95 23. limit 43 to human 66 2.2. Anxiety & Depression - last review search of both disorders done in March 2020 (Simpson et al., 2021) 2.2.1. Pubmed © 2021 American Medical Association. All rights reserved. Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 1714 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((depression) OR (depress*) OR (mdd) OR (trd) OR (anxiety)) AND NOT (review) Limits: human, English, start date: February 2020 42 2.2.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1047721 15. depress*.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1200820 16. MDD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 32605 17. TRD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 3663 18. anxiety.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 630521 19. 1 or 2 or 3 or 4 or 5 or 6 1067443 20. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 21. 14 or 15 or 16 or 17 or 18 1511690 22. 19 AND 20 AND 21 2774 23. limit 22 to english language 2687 24. limit 23 to human 1894 25. limit 24 to yr="2020 -Current" 509 26. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 27. 25 NOT 26 278 © 2021 American Medical Association. All rights reserved. 2.3. Bipolar Disorder & Psychosis/Schizophrenia - last review search of both disorders done on 17 Jan 2019 (Vindegaard et al., 2020) 2.3.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 199 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((bipolar) OR (mania) OR (bipolar depression) OR (psychosis) OR (schizophrenia)) AND NOT (review) Limits: human, English, start date: January 2019 84 2.3.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. bipolar.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 160089 15. mania.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 43186 16. bipolar depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 10387 17. psychosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 187899 18. schizophrenia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 343421 19. 1 or 2 or 3 or 4 or 5 or 6 1067443 20. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 21. 14 or 15 or 16 or 17 or 18 577348 22. 19 AND 20 AND 21 661 23. limit 22 to 6English language 631 24. limit 23 to human 551 25. limit 24 to yr="2019 -Current" 291 26. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 © 2021 American Medical Association. All rights reserved. 27. 25 NOT 26 148 2.4. ADHD & ASD: last review search of both disorders done on 31Aug2018 (Jurek et al., 2020). 2.4.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 415 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((adhd) OR (attention deficit hyperactivity disorder) OR (autism) OR (autism spectrum disorder) OR (ASD)) AND NOT (review) Limits: human, English, start date: August 2018 20 2.4.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. attention deficit hyperactivity disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, 61699 id, tm, mh] 15. adhd.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 69524 16. autism.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 135286 17. autism spectrum disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42650 18. ASD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56686 19. 1 or 2 or 3 or 4 or 5 or 6 1067443 20. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 21. 14 or 15 or 16 or 17 or 18 221434 22. 19 AND 20 AND 21 1209 23. limit 22 to english language 1159 © 2021 American Medical Association. All rights reserved. 24. limit 23 to human 962 25. limit 24 to yr="2018 -Current" 531 26. adult. mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 8881500 27. 25 AND 26 63 28. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 29. 27 NOT 28 46 2.5. Eating disorders: last review search done in June 2020 (Di Ludovico et al., 2021). 2.5.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 489 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((eating disorder) OR (anorexia) OR (bulimia)) AND NOT (review) Limits: human, English, start date: May 2020 13 2.5.2. Embase & PsychINFO (via Ovid) Search Hits 30. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 31. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 32. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 33. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 34. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 35. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 36. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 37. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 38. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 39. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 40. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 41. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 42. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 43. eating disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 50029 44. anorexia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108993 45. bulimia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 28918 46. 1 or 2 or 3 or 4 or 5 or 6 1067443 47. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 48. 14 or 15 or 16 152111 © 2021 American Medical Association. All rights reserved. 49. 17 AND 18 AND 19 550 50. limit 20 to english language 512 51. limit 21 to human 392 52. limit 22 to yr="2020 -Current" 83 53. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 54. 23 NOT 24 37 © 2021 American Medical Association. All rights reserved. eAppendix 2. Quality assessment Two authors (VLN and MRBS) performed the rating independently and resolved discrepancies via discussion. Summary: The overall risk of bias in the systematic reviews was low (12/16) (Table S5.1), however, only 3/16 had a pre-registered protocol (Table S3.1). The most frequent concerns were the lack of clarity how evidence was collected and evaluated for quality and the thoroughness of the search strategy (Figure S5.1). However, due to the significant overlap, a recent high-quality review was available for each disorder. Regarding original studies published after the reviews, quality was high as assessed with the JBI tool for case-control studies. The primary concern was incomplete consideration of confounders with 7/20 either not identifying these and/or not accounting for them in the analyses. We did not penalize studies that considered only some factors (e.g. age, gender, psychiatric medication) as other, particularly lifestyle factors such as diet, are difficult to fully control. Table e2.1. Quality assessment of the included systematic reviews using the ROBIS tool. Review Phase 1 Phase 2 Disorder First Author Year 1. Study 2. 3. Data 4. RISK OF eligibility Identificatio collection Synthesis BIAS criteria n and and study and IN THE selection of appraisal findings REVIEW* studies low low low low low AN Di Lodovico 2021 low high unclear high low AN Schalla 2019 unclear high low low low AN Schwensen 2018 low low low low low MDD, ANX Simpson 2021 high high unclear unclear high MDD, SCZ Fond 2020 unclear high unclear high high MDD Li 2020 low low low low low MDD, ANX Simpson 2020 low low low low low MDD Sanada 2020 MDD, BD, low low low low low Vindegaard 2020 SCZ low high high unclear unclear MDD Cheung 2019 low low low high low MDD Barandouzi 2020 low low unclear low low BD, SCZ Nguyen 2019 high unclear unclear low low BD, SCZ Nguyen 2018 high high high high high SCZ Cuomo 2018 low low low low low SCZ Kraeuter 2020 low low low low low ASD, ADHD Jurek 2020 Low = low risk of bias, high = high risk of bias, unclear = insufficient information to assess risk of bias; *this criterion is scored by answering the following questions: A. Did the interpretation of findings address all of the concerns identified the Phase 2 assessment?; B. Was the relevance of identified studies to the review's research question appropriately considered?; and C. Did the reviewers avoid emphasizing results on the basis of their statistical significance?. © 2021 American Medical Association. All rights reserved. Table e2.2. Quality assessment of the included original studies published after the systematic reviews using the Joanna Briggs Institute Critical Appraisal Checklist for Case Control Studies. © 2021 American Medical Association. All rights reserved. RISK OF BIAS IN THE REVIEW High Low 4. Synthesis and findings Unclear 3. Data collection and study appraisal 2. Identification and selection of studies 1. Study eligibility criteria 0% 20% 40% 60% 80% 100% Figure e5.1. Quality assessment of the included systematic reviews using the ROBIS tool. © 2021 American Medical Association. All rights reserved. eAppendix 3: Detailed methods of the meta-analysis performed Medians and inter-quartile ranges were transformed to means (M) and standard deviations (SD) using a web-based tool (http://www.math.hkbu.edu.hk/~tongt/papers/median2mean.html). For significantly skewed data, an alternative validated procedure was followed . Where necessary, numerical data were extracted from graphs using WebPlotDigitizer (v.4.4 ) and Adobe Acrobat's inbuilt measuring tool (Adobe Systems, California, USA), as previously done by others . A random-effects meta-analysis on Hedge’s g standardised mean difference (SMD) was performed applying the inverse-variance method. Effect size was categorized as small (SMD 0.2), moderate (SMD=0.5), or large (SMD=0.8). Inter-study heterogeneity was quantified using the DerSimonian–Laird estimator, reported with the I2 statistic and interpreted according to convention (25% - low, 50% - moderate, and 75% - high) . Publication bias was evaluated with funnel plots and Egger's regression test. Pre-planned subgroup analyses were disorder, region of study (east/west) and use of psychiatric medication. All analyses were completed with the meta package (v.4.17-0 ) in R. References: 1. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol. 2014;14. doi:10.1186/1471-2288-14-135 2. Rohatgi A. WebPlotDigitizer.; 2020. https://automeris.io/WebPlotDigitizer 3. Safadi JM, Quinton AMG, Lennox BR, Burnet PWJ, Minichino A. Gut dysbiosis in severe mental illness and chronic fatigue: a novel trans-diagnostic construct? A systematic review and meta-analysis. Molecular Psychiatry. Published online February 8, 2021:1- 13. doi:10.1038/s41380-021-01032-1 4. Higgins DJPT. Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons; 2008. Accessed December 17, 2020. 5. Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. Evid Based Ment Health. 2019;22(4):153-160. doi:10.1136/ebmental-2019- © 2021 American Medical Association. All rights reserved. eAppendix 4: PRISMA flowcharts for the umbrella review search and the updated review searches Records identified through Additional records identified through database searching other sources (n = 279) (n = 0) Records after duplicates removed (n = 228) Records excluded (n = 196), reasons: Clearly irrelevant (n = 136) Records screened Review of intervention (diet, (n = 228) probiotics, etc.) (n = 21) Not systematic review (n = 13) Gut microbiota not assessed (n = 8) Not disorder of interest (n = 7) Poster/Abstract (n = 8) Not adult (n = 2) Full-text articles assessed for eligibility (n = 32) Full-text articles excluded (n = 16), reasons: Not adult (n = 8) gut microbiota not assessed (n = 5) Eligible reviews and meta- Not disorder of interest (n = 2) analyses (n = 16) Not systematic review (n = 1) Eligible original cross- sectional studies (n = 39) Figure e4.1. PRISMA flowchart of the umbrella review search © 2021 American Medical Association. All rights reserved. Included Eligibility Screening Identification Table e4.1. PRISMA charts for the updated systematic review searches. OCD & MDD & ANX BD & SCZ ASD & ADHD Eating disorders PTSD Records 86 293 209 66 50 identified Reviewed 86 293 209 66 50 Excluded at 83 282 193 62 47 Title/Abstract Excluded at full 0 5 (n=1 gut 6 (n=2 gut 4 (n=2 non- 2 (n=1 secondary text (with microbiota not microbiota not adult, n=1 non- analysis from reasons) assessed; n=2 assessed, n=1 human, n=1 gut included study, outcome of interest outcome of microbiota not n=1 gut not reported; n=2 interest not assessed) microbiota not no control group) assessed; n=3 analysed) conference abstract) Included 3 (2 OCD, 6 (5 MDD, 1 MDD + 10 (6 SCZ, 4 BD) 0 1 (AN) 1 PTSD) BD) © 2021 American Medical Association. All rights reserved. eAppendix 5. Details of the identified systematic reviews Table e5.1. Details of the identified systematic reviews. Pre- First Author Year Ref. Disorder Studies included (eligible only)* registered? Di Lodovico 2021 [1] anorexia N = 9 no Armougoum 2009; Million 2013; Kleiman 2015; Morita 2015; Mack 2016; Mörkl 2017; Borgo 2017; Hanachi 2018; Hata 2019 Schalla 2019 [2] anorexia N = 5 no Morita et al., 2015; Mörkl et al., 2017; Kleiman et al., 2015; Borgo et al., 2017; Mack et al., 2016 Schwensen 2018 [3] anorexia N = 7 Yes Mörkl et al., 2017; Borgo et al 2017; Kleiman et al 2015; Morita et al 2015; Amougom et al 2009; Million et al 2013; Mack et al 2016 Simpson 2021 [4] depression N = 18 no Aizawa et al (2016); Chahwan et al. (2019); Chen, Li et al. (2018); Chen, Zheng et al. (2018); Chen et al. (2019); Chung et al. (2019); Huang et al. (2018); Jiang et al. (2015); Kelly et al. (2016); Lai et al., (2019); Lin et al. (2017); Liu et al. (2016); Mason et al., (2020); Naseribafrouei et al. (2014); Rong et al. (2019); Valles- Colomer et al. (2019); Vinberg et al. (2019); Zheng et al. (2016) anxiety N = 3 Mason et al., (2020); Jiang et al. (2018); Chen et al. (2019) Fond 2020 [5] depression N = 7 no Chen et al. 2018; Kelly et al. 2016; Liu et al. 2016; Jiang et al. 2015; Madan et al. 2020; Mason et al. 2020; Naseribafrouei et al., 2014 schizophrenia N = 0 Li 2020 [6] depression N= 9 no Naseribafrouei et al., 2014; Jiang et al., 2015; Kelly et al., 2016; Zheng et al., 2016; Huang et al., 2018; Chen et al., 2018b; Chung et al., 2019; Chen et al., 2018a; Rong et al., 2019 Simpson 2020 [7] depression N = 2 no Liu et al. (2016), Aizawa et al. (2016) Sanada 2020 [8] depression N = 10 no Chen 2018a; Chen 2018b; Huang 2018; Lin 2017; Aizawa 2016; Kelly 2016; Liu 2016; Zheng 2016; Jiang 2015; Naseribafrouei 2014 © 2021 American Medical Association. All rights reserved. Vindegaard 2020 [9] depression N = 9 yes Chen et al., 2018; Huang et al., 2018; Stevens et al., 2018; Lin et al., 2017; Aizawa et al., 2016; Kelly et al., 2016; Zheng et al., 2016; Jiang et al., 2015; Naseribafrouei et al., 2014 bipolar N = 4 disorder Coello et al., 2019; Aizawa et al., 2018; Painold et al., 2018; Evans et al., 2017 psychosis N = 4 Nguyen et al., 2018; Schwarz et al., 2018; Shen et al., 2018; Yuan et al., 2018 Cheung 2019 [10] depression N = 6 no Naseribafrouei et al 2014; Jiang et al 2015; Aizawa et al 2016; Zheng et al 2016; Lin et al 2017; Chen et al 2018 Barandouzi 2020 [11] depression N = 9 no Chen et al 2018; Zheng et al 2016; Liu et al 2016; Chen et al 2018; Jiang et al 2015; Naserbafrouei et al 2014; Lin et al 2017; Aizawa et al 2016; Kelly et al 2016; Nguyen 2019 [12] bipolar N = 4 no disorder Painold et al., 2019; Evans et al 2017; Coello et al., 2019; Aizawa et al., 2018 psychosis N = 5 Schwarz et al., 2018; Yuan et al., 2018; Nagamine et al 2018; Nguyen et al 2019; Shen et al 2018 Nguyen 2018 [13] bipolar N = 1 no disorder Evans et al 2017 psychosis N = 1 Schwarz et al. (2017) Cuomo 2018 [14] psychosis N = 3 no Schwarz et al. (2018); Shen et al (2018); Yuan et al (2018) Kraeuter 2020 [15] psychosis N = 5 no Schwarz et al 2018; Shen et al 2018; Yuan et al 2018; Nguyen et al 2018; Zheng 2019 Jurek 2020 [16] autism, N = 1 yes ADHD Aarts et al. 2017 *after duplicate studies from these systematic reviews were removed, the total number of studies added through this route was 39. eReferences: 1. Di Lodovico L, Mondot S, Doré J, Mack I, Hanachi M, Gorwood P. Anorexia nervosa and gut microbiota: A systematic review and quantitative synthesis of pooled microbiological data. Prog Neuropsychopharmacol Biol Psychiatry. 2020 Sep 22;110114. 2. Schalla MA, Stengel A. Gastrointestinal alterations in anorexia nervosa - A systematic review. Eur Eat Disord Rev. 2019;27(5):447–61. 3. Schwensen HF, Kan C, Treasure J, Høiby N, Sjögren M. A systematic review of studies on the faecal microbiota in anorexia nervosa: future research may need to include microbiota from the small intestine. Eat Weight Disord. 2018 Aug;23(4):399–418. © 2021 American Medical Association. All rights reserved. 4. Simpson CA, Mu A, Haslam N, Schwartz OS, Simmons JG. Feeling down? A systematic review of the gut microbiota in anxiety/depression and irritable bowel syndrome. Journal of Affective Disorders [Internet]. 2020 Apr [cited 2020 Jun 2];266:429–46. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0165032719325777 5. Fond GB, Lagier JC, Honore S, Lancon C, Korchia T, De Verville PLS, Llorca PM, Auquier P, Guedj E, Boyer L. Microbiota-orientated treatments for major depression and schizophrenia. Nutrients [Internet]. 2020;12(4). 6. Li S, Hua D, Wang Q, Yang L, Wang X, Luo A, Yang C. The Role of Bacteria and Its Derived Metabolites in Chronic Pain and Depression: Recent Findings and Research Progress. Int J Neuropsychopharmacol. 2020 10;23(1):26–41. 7. Simpson CA, Diaz-Arteche C, Eliby D, Schwartz OS, Simmons JG, Cowan CSM. The gut microbiota in anxiety and depression - A systematic review. Clin Psychol Rev. 2020 Oct 29;83:101943. 8. Sanada K, Nakajima S, Kurokawa S, Barceló-Soler A, Ikuse D, Hirata A, Yoshizawa A, Tomizawa Y, Salas-Valero M, Noda Y, Mimura M, Iwanami A, Kishimoto T. Gut microbiota and major depressive disorder: A systematic review and meta-analysis. J Affect Disord. 2020 01;266:1–13. 9. Vindegaard N, Speyer H, Nordentoft M, Rasmussen S, Benros ME. Gut microbial changes of patients with psychotic and affective disorders: A systematic review. Schizophr Res. 2020 Jan 14; 10. Cheung SG, Goldenthal AR, Uhlemann A-C, Mann JJ, Miller JM, Sublette ME. Systematic Review of Gut Microbiota and Major Depression. Front Psychiatry [Internet]. 2019 Feb 11 [cited 2020 Jun 2];10:34. Available from: https://www.frontiersin.org/article/10.3389/fpsyt.2019.00034/full 11. Barandouzi ZA, Starkweather AR, Henderson WA, Gyamfi A, Cong XS. Altered Composition of Gut Microbiota in Depression: A Systematic Review. Front Psychiatry. 2020;11:541. 12. Nguyen TT, Hathaway H, Kosciolek T, Knight R, Jeste DV. Gut microbiome in serious mental illnesses: A systematic review and critical evaluation. Schizophr Res. 2019 Sep 5; 13. Nguyen TT, Kosciolek T, Eyler LT, Knight R, Jeste DV. Overview and systematic review of studies of microbiome in schizophrenia and bipolar disorder. J Psychiatr Res. 2018;99:50–61. 14. Cuomo A, Maina G, Rosso G, Beccarini Crescenzi B, Bolognesi S, Di Muro A, Giordano N, Goracci A, Neal SM, Nitti M, Pieraccini F, Fagiolini A. The Microbiome: A New Target for Research and Treatment of Schizophrenia and its Resistant Presentations? A Systematic Literature Search and Review. Front Pharmacol [Internet]. 2018 Oct 15 [cited 2021 Mar 17];9. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6196757/ 15. Kraeuter A-K, Phillips R, Sarnyai Z. The Gut Microbiome in Psychosis From Mice to Men: A Systematic Review of Preclinical and Clinical Studies. Front Psychiatry. 2020;11:799. 16. Jurek L, Sevil M, Jay A, Schröder C, Baghdadli A, Héry-Arnaud G, Geoffray M-M. Is there a dysbiosis in individuals with a neurodevelopmental disorder compared to controls over the course of development? A systematic review. Eur Child Adolesc Psychiatry. 2020 May 8; © 2021 American Medical Association. All rights reserved. eAppendix 6. Detailed characteristics of the included studies The 59 studies provided 64 case-control comparisons capturing 2643 patients and 2336 controls (Table 6.1). Most studies (n=32) were conducted in East Asia (China, Taiwan or Japan), 24 in westernised populations (grouped on the basis of typical diet and lifestyle: USA, Canada, Europe, Australia, New Zealand) and one in Africa. All but one used formal diagnostic criteria to define their population. Studies were similar in exclusion criteria such as major medical and gastrointestinal conditions, pregnancy and recent consumption of antibiotics (except four studies in which 17,25,28,33 antibiotics weren’t mentioned ). Recent probiotic consumption was excluded in 35/59 studies, was not mentioned in 21, and three studies 3,27,50 included a small number of participants taking probiotics . Few studies imposed restrictions on diet such as no major changes in the months 10,19,35,36 14,15,18,20,31 39,44 preceding enrolment or no weight loss, vegetarian or vegan diets . Two studies matched groups according to diet and one 55 60 controlled dietary intake . Despite its known impact on microbial communities , smoking was generally not controlled: only three studies 18,20,38 19,23,28,45,51 excluded smokers and five controlled for it during analyses . Amplicon 16S rRNA sequencing was used in 44 studies, although choice of hypervariable region (V1-V9) varied, seven studies used shotgun metagenomics to sample all microbial genes, nine studies used either qPCR or RT-qPCR to target a pre-specified range of microbial taxa, and one study employed metaproteomic analysis (Table 6.1). Supplementary Table 6.1. Key sample and methodology characteristics of case-control comparisons of the gut microbiome by disorder. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Smokers size n Age Female BMI Stage medication MDD Naseribafrou Norway ICD-10 P: 37 P: 49.2 56% P: 25.9 nr Nr 16S rRNA : Observed sp., Simpsons ei et al. 2014 HC: 18 HC: 46.1 HC: 24.7 nr : not measured MDD Jiang et al. China DSM-IV P: 29 P: 25.3 38% P: 20.3 P: 10% most, total 16S rRNA : Chao1, ACE, Shannon, Simpson, 2015 HC: 30 HC: 26.8 50% HC: 19.6 HC: 7% % nr V1-V3 evenness; : UniFrac (unweighted) MDD Aizawa et al. Japan DSM-IV P: 43 P: 39.4 41% P:23.2 nr 65% RT-qPCR : not measured 2016 HC: 57 HC: 42.8 61% HC: 22.3 16S rRNA : not measured MDD Kelly et al. Ireland DSM-IV P: 34 P: 45.8 38% P: 26.2 P: 13% 96% 16S rRNA : Observed sp., Chao1, Shannon, PD 2016 HC: 33 HC: 45.8 HC: 24.6 HC: 3% nr : UniFrac (weighted & unweighted), Bray-Curtis MDD Liu et al. China DSM-IV P: 15 P: 44.8 69% P: 22.0 nr 0% 16S rRNA : Observed sp., Chao1, Shannon, PD 2016 HC: 20 HC: 43.9 HC: 22.0 V1-V3 : measured, nr © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication MDD Zheng et al. China DSM-IV P: 58 P: 40.6 63% P: 22.0 P: 16% 67% 16S rRNA : Observed sp., Shannon, Simpson, PD 2016 HC: 63 HC: 41.8 HC: 22.6 H: 21% V3-V5 : Bray-Curtis MDD Lin et al. China DSM-IV P: 10 P: 36.2 60% P: 23.8 P:40% 100% 16S rRNA : measured, nr 2017 HC: 10 HC: 38.1 HC: 24.2 HC:30% V3-V4 : UniFrac (weighted) MDD Chen et al. China HAMD P: 44 P: 40.9 55% P: 22.1 nr 0% 16S rRNA : Observed sp., Shannon, Simpson, PD 2018a HC: 44 HC: 43.4 HC: 22.6 (drug- seq : UniFrac (nr), PLS-DA naïve) V3-V5 MDD Chen et al. China DSM-IV P: 10 P: 43.9 50% P: 23.5 nr 20% Meta- : not measured 2018b HC: 10 HC:39.6 HC: 22.6 proteomics : not measured MDD Huang et al. China ICD-10 P: 27 P: 48.7 74% P: 23.8 nr Nr 16S rRNA : Chao1, ACE, Shannon, PD 2018 HC: 27 HC: 42.3 HC: 23.4 V3-V4 : UniFrac (weighted & unweighted) MDD Chahwan et Australi M.I.N.I P: 68 P: 36.1 70% nr P: 25% 0% 16S rRNA : Observed sp., Chao1, Shannon al. 2019 a HC: 20 HC: 40.0 HC: V3-V4 : UniFrac (weighted) 11% MDD Valles- Belgium GP & P: 80 50.9 55% 24.9 nr 50% 16S rRNA : not measured Colomer / NL self-report HC: 70 nr : not measured MDD Chung et al. Taiwan DSM-5 P: 36 P: 45.8 70% P: 22.8 P: 19% 86% 16S rRNA : Shannon 2019 HC: 37 HC: 41.2 HC: 24 HC: 3% V3-V4 : UniFrac (weighted) MDD Lai et al. China DSM-5 P: 26 P: 43.7 P: P: 27.2 nr 81% Shotgun : Shannon, Fisher 2019 HC: 29 HC: 39.4 69% HC: 21.1 Metagenomic : Bray–Curtis HC: s 55% MDD Rong et al. China DSM-5 P: 31 P: 41.6 P: P: 21.5 nr 74% Shotgun : Chao 1, Shannon, Inverse Simpson, 2019 HC: 30 HC: 39.5 71% HC: 22.0 Metagenomic Gm coefficient; : Bray-Curtis HC:53 s MDD Mason et al. USA DSM-IV P: 14 P: 41.9 P:79% P: 31.0 nr 64% 16S rRNA : Shannon 2020 HC: 10 HC: 33.0 HC: HC: 25.6 V4 : UniFrac (weighted) 60% MDD Chen et al. China DSM-IV Young Young 72% Young P: nr Young: 16SrRNA : Chao1, ACE 2020 P: 25 P: 24.0 22.1 28% V3-V5 : OPLS-DA HC: 27 HC: 25.0 HC: 21.5 Mid-age: Mid-age Mid-age Mid-age 31% P: 45 P: 45.0 P:22.6 © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication HC: 44 HC: 47.2 HC: 23.2 MDD Chen et al. China DSM-5 P: 62 P: 39.6 100% P: 22.0 0% 0% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2021 HC: 46 HC: 34.0 HC: 22.2 V3-V4 Simpson; : UniFrac (weighted & unweighted) MDD Yang et al. China DSM-IV P: 156 P: 29.6 P: 36 P: 22.3 nr 24% Shotgun : Chao1, Shannon, Inv. Shannon 2020 HC: 155 HC: 29.1 % HC: 22.4 Metagenomic : Bray-Curtis HC: s 54% MDD Liu et al. USA DSM-5 P: 43 P: 21.9 P: nr 0% 65% 16SrRNA : Observed sp., Shannon, PD 2020 HC: 47 HC: 22.1 88% V4 : UniFrac (weighted & unweighted), HC: Bray-Curtis 72% MDD Stevens et USA DSM-IV P: 20 P & HC: P: nr nr 75% nr : Chao1, Shannon al. 2020 HC: 20 34 50% : Bray-Curtis HC: 70% MDD + Stevens et USA DSM-5 P: 22 nr nr nr nr Nr nr : not measured ANX al. 2018 HC: 28 : not measured MDD & Mason et al. USA DSM-IV P: 38 P:39.2 P: P: 20.4 nr 42% 16S rRNA : Shannon ANX 2020 HC: 10 HC:33.0 82% HC: 25.6 V4 : UniFrac (weighted) HC: 60% MDD & Vinberg et al. Denmar ICD-8 & P: 74 P: 37.7 77% P: 26.5 P > HC, 61% 16SrRNA : Observed sp., Shannon BD 2019 k ICD-10 HC: 25 HC: 37.0 HC: 24.5 % nr V3-V4 : generalized UniFrac remission MDD & Jiang et al. China DSM-IV MDD:14 P: 37.2 45% P: 23.6 P: 8% most, total 16SrRNA : Chao1, ACE, Shannon, Simpson BD 2020 BD: 10 HC: 35.8 HC: 22.3 HC: 6% % nr V1-V3 : UniFrac (weighted & unweighted), HC: 16 Bray-Curtis BD Evans et al. USA DSM-IV P:115 P: 50.2 73% P: 29.3 nr most, total 16SrRNA : not measured 2017 HC: 64 HC: 48.6 HC: 26.0 % nr V4 : Yue & Clayton BD Painold et al. Austria DSM-IV, P: 32 P: 41.3 44% P: 28.4 nr 100% 16SrRNA : Observed sp., Chao1, Shannon, 2018 current HC: 10 HC: 31.4 HC: 24.3 V1-V2 Simpson : UniFrac (weighted & unweighted) © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication depression or euthymia BD Aizawa et al. Japan DSM-IV, P: 39 P: 40.3 56% P: 23.9 nr 92% RT qPCR for : not measured 2019 any HC: 58 HC: 43.1 HC: 22.4 16S/23S : not measured episode rRNA BD Rong et al. China DSM-5, P: 30 P: 38.4 52% P: 21.9 nr 83% Shotgun : Chao 1, Shannon, Inv. Simpson, Gm 2019 current HC: 30 HC: 39.5 HC: 22.0 Metagenomic coefficient depression s : Bray-Curtis BD Coello et al. Denmar ICD-10, P: 113 P: 31 62% P: 24.8 P: 36% 88% 16SrRNA : Observed sp., Shannon 2019 k HC: 77 HC: 29 HC:24.2 HC: V3-V4 : UniFrac (weighted & unweighted) any 11% episode BD McIntyre et Canada DSM-5, P: 23 P: 45 70% P: 30 HC: P: 21% nr 16SrRNA : Observed sp., Shannon, Inv. Simpson al. 2019 current HC: 23 HC: 43.8 26 HC: 9% V3 : Bray-Curtis depression BD Hu et al. China DSM-IV- P: 52 P: 24.2 48% P: 21.6 nr 0% 16SrRNA : Observed sp., Chao1, Shannon, 2019 TR, current HC: 45 HC:36.3 HC: 22.4 V3-V4 Simpson, Inv. Simpson, ICE depression : UniFrac (weighted & unweighted) BD Lai et al. China DSM-5, P: 25 P: 36.9 48% P: 22.1 nr 80% Shotgun : Shannon, Simpson, Fisher 2021 current HC: 28 HC: 39.2 HC:21.1 Metagenomic : Bray-Curtis depression s BD Lu et al. China DSM-IV- P: 36 P: 32.6 43% P: 22.2 0% 0% qPCR : not measured 2019 TR, current HC: 27 HC: 28.9 HC: 21.8 : not measured depression SCZ Nguyen et al. USA DSM-IV-TR P: 25 P: 52.9 44% P: 31.8 P: 56% 100% 16SrRNA : Observed sp., Shannon, Faith's PD 2019 HC: 25 HC: 54.7 HC:28.9 HC: 4% V4 : UniFrac (unweighted), Bray-Curtis SCZ Schwarz et Finland DSM-IV; P: 28 P: 25.9 43% P: 23.8 nr 93% qPCR for : not measured al. 2018 FEP HC: 16 HC: 27.1 HC:23.9 16S rRNA : not measured primers, Metagenomic SCZ Shen et al. China ICD-10 P: 64 P: 42 44% P:23.5 P: 19% 100% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2018 HC: 53 HC: 39 HC:23.1 HC: V3-V4 Simpson, Faith's PD 23% : UniFrac (unweighted) © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication SCZ Yuan et al. China DSM-IV; P: 41 P: 23.1 44% P:20.5 P: 5% 0% qPCR for : not measured 2018 FEP HC: 41 HC: 24.7 HC: 20.8 HC: 6% 16S rRNA : not measured primers SCZ Zheng et al. China DSM-IV P: 63 P: 43.5 P:33% P: 22.9 nr 92% 16SrRNA : Chao 1, Shannon 2019 HC: 69 HC: 40.0 HC: HC: 23.2 V3-V4 : PLS-DA 48% SCZ Pan et al. China DSM-IV P: 29 P: 34.9 34% P: 23.7 0% 90% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2020 HC: 29 HC: 34.8 HC: 23.5 V3-V4 Simpson, Faith's PD : UniFrac (unweighted) SCZ Zhu et al. China DSM-IV P: 90 P: 28.6 49% P: 20.6 P: 30% 0% Shotgun : Shannon 2020 HC: 81 HC: 32.3 HC: 21.7 HC: Metagenomic : Bray-Curtis 25% s SCZ Li et al. China DSM-IV-TR P: 82 P: 42.2 47% P:24.5 P:21% 91% 16SrRNA : Observed Sp., Evenness, Shannon, 2020 HC: 80 HC: 41.0 HC: 23.0 HC: 5% V4 Faith’s PD; : Bray-Curtis SCZ Ma et al. China DSM-IV; FEP: 40 P: 24.2 46% nr nr FEP: 0% 16SrRNA : Chao1, Shannon 2020 FEP & SCZ SCZ: 85 HC:23.1 (drug- V4 : UniFrac (weighted & unweighted) HC: 69 naïve) SCZ: 100% SCZ Zhang et al. China DSM-IV; P:10 P: 37.6 42% P: 23.3 P: 10% 0% 16SrRNA : Observed sp., Chao1, Shannon, 2020 FEP HC:16 HC: 35.8 HC:22.3 HC: (drug nr Simpson; : UniFrac (weighted & 6.3% naïve) unweighted), Bray-Curtis SCZ Xu et al., China DSM-5 P: 84 P: 35.0 43% P:22 nr 98% 16SrRNA : Chao1 2020 HC: 84 HC: 35.0 HC:23.1 V4 & : non-metric multidimensional Shotgun scaling Metagenomic ANX Jiang et al. China DSM-IV P: 40 P: 33.4 P: P:21.7 P: 2.5% 70% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2018 HC: 36 HC: 35.6 75% HC: 21.4 HC: 3% V3-V4 Simpson; : UniFrac unweighted HC: 64% ANX Chen et al. China DSM-5 P: 36 P: 46.1 57% P: 23.1 P: nr 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2019 HC: 24 HC: 41.8 HC: 22.5 19.4% V3-V4 Simpson HC: : UniFrac (weighted & unweighted) 16.7% © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication ANX Mason et al. USA DSM-IV P: 8 P: 40.0 P: P: 33.3 nr 62% 16SrRNA : Shannon 2020 HC: 10 HC: 33.0 100% HC: 25.6 V4 : UniFrac weighted HC: 60% AN Armougom et France DSM-IV P: 9 P: 19-36 nr P: 12.7 nr nr RT qPCR : not measured al. 2009 HC: 20 HC: 13-68 HC: 20.7 : not measured AN Million et al. France DSM-IV P: 15 P: 27.3 P: P: 13.5 nr nr qPCR : not measured 2013 HC: 76 HC: 49.5 93% HC: 22.4 : not measured HC:43 AN Kleiman et USA DSM-IV-TR P: 16 P: 28.0 100% P: 16.2 nr nr 16SrRNA : Observed sp., Chao1 al. 2015 HC: 12 HC: 29.8 HC: 21.5 V1-V3 : UniFrac (weighted & unweighted) AN Morita et al. Japan DSM-IV-TR P: 25 P: 30.0 100% P: 12.8 nr nr qPCR for : not measured 2015 HC: 21 HC: 31.5 HC: 20.5 16S/23S : not measured rRNA AN Mack et al. German ‘Diagnosis’ P: 55 P: 23.8 100% P: 15.3 nr nr 16SrRNA : Observed sp., Chao 1, Shannon 2016 y (not HC: 55 HC: 23.7 HC: 21.6 V4 : UniFrac (weighted & unweighted), specified) Bray-Curtis AN Mörkl et al. Austria ICD-10 P: 18 P: 22.4 100% P: 15. 25% nr 16SrRNA : Observed sp., Chao 1, Shannon 2017 HC: 26 HC: 24.9 HC: 21.9 V1-V2 : UniFrac (weighted & unweighted) AN Borgo et al. Italy DSM-5 P: 15 P: 25.6 100% P: 13.9 nr nr 16SrRNA : measured, nr 2017 HC: 15 HC: 24.4 HC: 22.1 V3-V4 : measured, nr AN Hanachi et France DSM-IV-TR P: 33 P: 32 100% P: 11.7 nr nr 16SrRNA : Chao 1, Shannon al. 2018 HC: 22 HC: 36 HC: 21.0 V3-V4 : UniFrac (weighted & unweighted) AN Hata et al. Japan DSM-IV-TR P: 4 P: 23.0 100% P:13.7 nr nr 16SrRNA : Observed sp., Chao 1, Shannon 2019 restrictive HC: 4 HC: 25.3 HC:21.6 V3-V4 : UniFrac (weighted & unweighted) only AN Monteleone Italy DSM-5 P:21 P: 21.7 100% P: 14.6 nr nr 16SrRNA : Chao1, Fisher et al. 2021 HC: 20 HC:23.0 HC: 20.3 V4 : non-metric multidimensional scaling OCD Domenech et Spain DSM-IV P: 38 P: 40.2 53% nr nr nr 16SrRNA : Observed sp., Chao 1, Shannon, al. pre-print HC: 33 HC: 36.0 V3-V4 Simpson, Inv. Simpson, Faith's PD : UniFrac (weighted & unweighted), Bray-Curtis, Jensen-Shannon, Canberra © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication OCD Turna et al. Canada DSM-5 P: 21 P: 31.0 54% P: 24.6 nr 0% 16SrRNA : Observed sp., Chao 1, Inv. Simpson, 2020 HC: 22 HC:29.3 HC: 23.2 V3 Shannon; : UniFrac (weighted & unweighted), Bray-Curtis, Jaccard PTSD Hemmings et South DSM-5 P: 18 P: 42.0 P: P:28.5 P: 50% 33% 16SrRNA : Observed sp., Chao1, Shannon, al. 2017 Africa HC: 22 HC: 38.7 14% HC:28.6 HC:42% V3-V4 Faith's PD HC: : UniFrac (weighted & unweighted), 7% Bray Curtis ADHD Aarts et al. NL DSM-IV P: 19 P: 19.5 P:32% P: 23.8 nr nr 16SrRNA : Observed sp., Chao1, Shannon, 2017 HC:77 HC: 27.1 HC:47 HC: 23.0 V3-V4 Faith's PD % : not measured ADHD attention deficit hyperactivity disorder, AN anorexia nervosa, ANX anxiety, BD bipolar disorder, OCD obsessive compulsive disorder, MDD major depressive disorder, PTSD post-traumatic stress disorder, SCZ schizophrenia and psychosis, BMI body mass index, P patient, HC healthy control, NL Netehrlands, ICD International Classification of Diseases, DSM Diagnostic and Statistical Manual of Mental Disorders, MINI Mini-International Neuropsychiatric Interview, FEP first episode psychosis, seq sequencing, (RT) qPCR (real time) quantitative polymerase chain reaction, ACE abundance-based coverage estimator, ICE incidence-based estimator, Faith’s PD Faith’s phylogenetic diversity, PLS-DA partial least squares discriminant analysis, OPLS-DA orthogonal projections to latent structures discriminant analysis, n number, nr not reported eReferences: 1. Naseribafrouei A, Hestad K, Avershina E, et al. Correlation between the human fecal microbiota and depression. Neurogastroenterol Motil Off J Eur Gastrointest Motil Soc. 2014;26(8):1155-1162. doi:10.1111/nmo.12378 2. Jiang H, Ling Z, Zhang Y, et al. Altered fecal microbiota composition in patients with major depressive disorder. Brain Behav Immun. 2015;48:186-194. doi:10.1016/j.bbi.2015.03.016 3. Aizawa E, Tsuji H, Asahara T, et al. Possible association of Bifidobacterium and Lactobacillus in the gut microbiota of patients with major depressive disorder. J Affect Disord. 2016;202:254-257. doi:10.1016/j.jad.2016.05.038 4. Kelly JR, Borre Y, O’ Brien C, et al. Transferring the blues: Depression-associated gut microbiota induces neurobehavioural changes in the rat. J Psychiatr Res. 2016;82:109-118. doi:10.1016/j.jpsychires.2016.07.019 © 2021 American Medical Association. All rights reserved. 5. Liu Y, Zhang L, Wang X, et al. Similar Fecal Microbiota Signatures in Patients With Diarrhea-Predominant Irritable Bowel Syndrome and Patients With Depression. Clin Gastroenterol Hepatol. 2016;14(11):1602-1611.e5. doi:10.1016/j.cgh.2016.05.033 6. Zheng P, Zeng B, Zhou C, et al. Gut microbiome remodeling induces depressive-like behaviors through a pathway mediated by the host’s metabolism. Mol Psychiatry. 2016;21(6):786-796. doi:10.1038/mp.2016.44 7. Lin P, Ding B, Feng C, et al. Prevotella and Klebsiella proportions in fecal microbial communities are potential characteristic parameters for patients with major depressive disorder. J Affect Disord. 2017;207:300-304. doi:10.1016/j.jad.2016.09.051 8. Chen J, Zheng P, Liu Y, et al. Sex differences in gut microbiota in patients with major depressive disorder. Neuropsychiatr Dis Treat. 2018;14:647-655. doi:10.2147/NDT.S159322 9. Chen Z, Li J, Gui S, et al. Comparative metaproteomics analysis shows altered fecal microbiota signatures in patients with major depressive disorder. Neuroreport. 2018;29(5):417-425. doi:10.1097/WNR.0000000000000985 10. Huang Y, Shi X, Li Z, et al. Possible association of Firmicutes in the gut microbiota of patients with major depressive disorder. Neuropsychiatr Dis Treat. 2018;14:3329-3337. doi:10.2147/NDT.S188340 11. Chahwan B, Kwan S, Isik A, van Hemert S, Burke C, Roberts L. Gut feelings: A randomised, triple-blind, placebo-controlled trial of probiotics for depressive symptoms. J Affect Disord. 2019;253:317-326. doi:10.1016/j.jad.2019.04.097 12. Valles-Colomer M, Falony G, Darzi Y, et al. The neuroactive potential of the human gut microbiota in quality of life and depression. Nat Microbiol. 2019;4(4):623-632. doi:10.1038/s41564-018-0337-x 13. Chung Y-CE, Chen H-C, Chou H-CL, et al. Exploration of microbiota targets for major depressive disorder and mood related traits. J Psychiatr Res. 2019;111:74-82. doi:10.1016/j.jpsychires.2019.01.016 14. Lai W-T, Deng W-F, Xu S-X, et al. Shotgun metagenomics reveals both taxonomic and tryptophan pathway differences of gut microbiota in major depressive disorder patients. Psychol Med. Published online November 5, 2019:1-12. doi:10.1017/S0033291719003027 15. Rong H, Xie X, Zhao J, et al. Similarly in depression, nuances of gut microbiota: Evidences from a shotgun metagenomics sequencing study on major depressive disorder versus bipolar disorder with current major depressive episode patients. J Psychiatr Res. 2019;113:90-99. doi:10.1016/j.jpsychires.2019.03.017 © 2021 American Medical Association. All rights reserved. 16. Mason BL, Li Q, Minhajuddin A, et al. Reduced anti-inflammatory gut microbiota are associated with depression and anhedonia. J Affect Disord. 2020;266:394-401. doi:10.1016/j.jad.2020.01.137 17. Chen J-J, He S, Fang L, et al. Age-specific differential changes on gut microbiota composition in patients with major depressive disorder. Aging. 2020;12(3):2764-2776. doi:10.18632/aging.102775 18. Chen Y, Xue F, Yu S, et al. Gut microbiota dysbiosis in depressed women: The association of symptom severity and microbiota function. J Affect Disord. 2021;282:391-400. doi:10.1016/j.jad.2020.12.143 19. Yang J, Zheng P, Li Y, et al. Landscapes of bacterial and metabolic signatures and their interaction in major depressive disorders. Sci Adv. 2020;6(49):eaba8555. doi:10.1126/sciadv.aba8555 20. Liu RT, Rowan-Nash AD, Sheehan AE, et al. Reductions in anti-inflammatory gut bacteria are associated with depression in a sample of young adults. Brain Behav Immun. 2020;88:308-324. doi:10.1016/j.bbi.2020.03.026 21. Stevens BR, Roesch L, Thiago P, et al. Depression phenotype identified by using single nucleotide exact amplicon sequence variants of the human gut microbiome. Mol Psychiatry. Published online January 27, 2020:1-11. doi:10.1038/s41380-020-0652-5 22. Stevens BR, Goel R, Seungbum K, et al. Increased human intestinal barrier permeability plasma biomarkers zonulin and FABP2 correlated with plasma LPS and altered gut microbiome in anxiety or depression. Gut. 2018;67(8):1555-1557. doi:10.1136/gutjnl-2017-314759 23. Vinberg M, Ottesen NM, Meluken I, et al. Remitted affective disorders and high familial risk of affective disorders associate with aberrant intestinal microbiota. Acta Psychiatr Scand. 2019;139(2):174-184. doi:https://doi.org/10.1111/acps.12976 24. Jiang H, Pan L, Zhang X, Zhang Z, Zhou Y, Ruan B. Altered gut bacterial–fungal interkingdom networks in patients with current depressive episode. Brain Behav. 2020;10(8):e01677. doi:https://doi.org/10.1002/brb3.1677 25. Evans SJ, Bassis CM, Hein R, et al. The gut microbiome composition associates with bipolar disorder and illness severity. J Psychiatr Res. 2017;87:23-29. doi:10.1016/j.jpsychires.2016.12.007 26. Painold A, Mörkl S, Kashofer K, et al. A step ahead: Exploring the gut microbiota in inpatients with bipolar disorder during a depressive episode. Bipolar Disord. 2019;21(1):40-49. doi:10.1111/bdi.12682 27. Aizawa E, Tsuji H, Asahara T, et al. Bifidobacterium and Lactobacillus Counts in the Gut Microbiota of Patients With Bipolar Disorder and Healthy Controls. Front Psychiatry. 2019;9:730. doi:10.3389/fpsyt.2018.00730 © 2021 American Medical Association. All rights reserved. 28. Coello K, Hansen TH, Sørensen N, et al. Gut microbiota composition in patients with newly diagnosed bipolar disorder and their unaffected first-degree relatives. Brain Behav Immun. 2019;75:112-118. doi:10.1016/j.bbi.2018.09.026 29. McIntyre RS, Subramaniapillai M, Shekotikhina M, et al. Characterizing the gut microbiota in adults with bipolar disorder: a pilot study. Nutr Neurosci. Published online May 28, 2019:1-8. doi:10.1080/1028415X.2019.1612555 30. Hu S, Li A, Huang T, et al. Gut Microbiota Changes in Patients with Bipolar Depression. Adv Sci. 2019;6(14):1900752. doi:https://doi.org/10.1002/advs.201900752 31. Lai W, Zhao J, Xu S, et al. Shotgun metagenomics reveals both taxonomic and tryptophan pathway differences of gut microbiota in bipolar disorder with current major depressive episode patients. J Affect Disord. 2021;278:311-319. doi:10.1016/j.jad.2020.09.010 32. Lu Q, Lai J, Lu H, et al. Gut Microbiota in Bipolar Depression and Its Relationship to Brain Function: An Advanced Exploration. Front Psychiatry. 2019;10. doi:10.3389/fpsyt.2019.00784 33. Nguyen TT, Hathaway H, Kosciolek T, Knight R, Jeste DV. Gut microbiome in serious mental illnesses: A systematic review and critical evaluation. Schizophr Res. Published online September 5, 2019. doi:10.1016/j.schres.2019.08.026 34. Schwarz E, Maukonen J, Hyytiäinen T, et al. Analysis of microbiota in first episode psychosis identifies preliminary associations with symptom severity and treatment response. Schizophr Res. 2018;192:398-403. doi:10.1016/j.schres.2017.04.017 35. Shen Y, Xu J, Li Z, et al. Analysis of gut microbiota diversity and auxiliary diagnosis as a biomarker in patients with schizophrenia: A cross-sectional study. Schizophr Res. 2018;197:470-477. doi:10.1016/j.schres.2018.01.002 36. Yuan X, Zhang P, Wang Y, et al. Changes in metabolism and microbiota after 24-week risperidone treatment in drug naïve, normal weight patients with first episode schizophrenia. Schizophr Res. 2018;201:299-306. doi:10.1016/j.schres.2018.05.017 37. Zheng P, Zeng B, Liu M, et al. The gut microbiome from patients with schizophrenia modulates the glutamate-glutamine-GABA cycle and schizophrenia-relevant behaviors in mice. Sci Adv. 2019;5(2):eaau8317. doi:10.1126/sciadv.aau8317 38. Pan R, Zhang X, Gao J, Yi W, Wei Q, Su H. Analysis of the diversity of intestinal microbiome and its potential value as a biomarker in patients with schizophrenia: A cohort study. Psychiatry Res. 2020;291:113260. doi:10.1016/j.psychres.2020.113260 39. Zhu F, Ju Y, Wang W, et al. Metagenome-wide association of gut microbiome features for schizophrenia. Nat Commun. 2020;11(1):1612. doi:10.1038/s41467-020-15457-9 © 2021 American Medical Association. All rights reserved. 40. Li S, Zhuo M, Huang X, et al. Altered gut microbiota associated with symptom severity in schizophrenia. PeerJ. 2020;8. doi:10.7717/peerj.9574 41. Ma X, Asif H, Dai L, et al. Alteration of the gut microbiome in first-episode drug-naïve and chronic medicated schizophrenia correlate with regional brain volumes. J Psychiatr Res. 2020;123:136-144. doi:10.1016/j.jpsychires.2020.02.005 42. Zhang X, Pan L-Y, Zhang Z, Zhou Y-Y, Jiang H-Y, Ruan B. Analysis of gut mycobiota in first-episode, drug-naïve Chinese patients with schizophrenia: A pilot study. Behav Brain Res. 2020;379:112374. doi:10.1016/j.bbr.2019.112374 43. Xu R, Wu B, Liang J, et al. Altered gut microbiota and mucosal immunity in patients with schizophrenia. Brain Behav Immun. 2020;85:120-127. doi:10.1016/j.bbi.2019.06.039 44. Jiang H-Y, Zhang X, Yu Z-H, et al. Altered gut microbiota profile in patients with generalized anxiety disorder. J Psychiatr Res. 2018;104:130-136. doi:10.1016/j.jpsychires.2018.07.007 45. Chen Y-H, Bai J, Wu D, et al. Association between fecal microbiota and generalized anxiety disorder: Severity and early treatment response. J Affect Disord. 2019;259:56-66. doi:10.1016/j.jad.2019.08.014 46. Armougom F, Henry M, Vialettes B, Raccah D, Raoult D. Monitoring Bacterial Community of Human Gut Microbiota Reveals an Increase in Lactobacillus in Obese Patients and Methanogens in Anorexic Patients. PLOS ONE. 2009;4(9):e7125. doi:10.1371/journal.pone.0007125 47. Million M, Angelakis E, Maraninchi M, et al. Correlation between body mass index and gut concentrations of Lactobacillus reuteri , Bifidobacterium animalis , Methanobrevibacter smithii and Escherichia coli. Int J Obes. 2013;37(11):1460-1466. doi:10.1038/ijo.2013.20 48. Kleiman SC, Watson HJ, Bulik-Sullivan EC, et al. The Intestinal Microbiota in Acute Anorexia Nervosa and During Renourishment: Relationship to Depression, Anxiety, and Eating Disorder Psychopathology. Psychosom Med. 2015;77(9):969–981. doi:10.1097/PSY.0000000000000247 49. Morita C, Tsuji H, Hata T, et al. Gut Dysbiosis in Patients with Anorexia Nervosa. PLOS ONE. 2015;10(12):e0145274. doi:10.1371/journal.pone.0145274 50. Mack I, Cuntz U, Grämer C, et al. Weight gain in anorexia nervosa does not ameliorate the faecal microbiota, branched chain fatty acid profiles, and gastrointestinal complaints. Sci Rep. 2016;6:26752. doi:10.1038/srep26752 © 2021 American Medical Association. All rights reserved. 51. Mörkl S, Lackner S, Müller W, et al. Gut microbiota and body composition in anorexia nervosa inpatients in comparison to athletes, overweight, obese, and normal weight controls. Int J Eat Disord. 2017;50(12):1421-1431. doi:10.1002/eat.22801 52. Borgo F, Riva A, Benetti A, et al. Microbiota in anorexia nervosa: The triangle between bacterial species, metabolites and psychological tests. Sanz Y, ed. PLOS ONE. 2017;12(6):e0179739. doi:10.1371/journal.pone.0179739 53. Hanachi M, Manichanh C, Schoenenberger A, et al. Altered host-gut microbes symbiosis in severely malnourished anorexia nervosa (AN) patients undergoing enteral nutrition: An explicative factor of functional intestinal disorders? Clin Nutr Edinb Scotl. 2019;38(5):2304- 2310. doi:10.1016/j.clnu.2018.10.004 54. Hata T, Miyata N, Takakura S, et al. The Gut Microbiome Derived From Anorexia Nervosa Patients Impairs Weight Gain and Behavioral Performance in Female Mice. Endocrinology. 2019;160(10):2441-2452. doi:10.1210/en.2019-00408 55. Monteleone AM, Troisi J, Fasano A, et al. Multi-omics data integration in anorexia nervosa patients before and after weight regain: A microbiome-metabolomics investigation. Clin Nutr Edinb Scotl. 2021;40(3):1137-1146. doi:10.1016/j.clnu.2020.07.021 56. Domènech L, Willis J, Alemany M, et al. Changes in the stool and oropharyngeal microbiome in obsessive-compulsive disorder. medRxiv. Published online May 28, 2020:2020.05.26.20113779. doi:10.1101/2020.05.26.20113779 57. Turna J, Grosman Kaplan K, Anglin R, et al. The gut microbiome and inflammation in obsessive-compulsive disorder patients compared to age- and sex-matched controls: a pilot study. Acta Psychiatr Scand. 2020;142(4):337-347. doi:10.1111/acps.13175 58. Hemmings SMJ, Malan-Müller S, van den Heuvel LL, et al. The Microbiome in Posttraumatic Stress Disorder and Trauma-Exposed Controls: An Exploratory Study. Psychosom Med. 2017;79(8):936-946. doi:10.1097/PSY.0000000000000512 59. Aarts E, Ederveen THA, Naaijen J, et al. Gut microbiome in ADHD and its relation to neural reward anticipation. PloS One. 2017;12(9):e0183509. doi:10.1371/journal.pone.0183509 60. Savin Z, Kivity S, Yonath H, Yehuda S. Smoking and the intestinal microbiome. Arch Microbiol. 2018;200(5):677-684. doi:10.1007/s00203-018-1506-2 © 2021 American Medical Association. All rights reserved. eAppendix 7. Stool sample processing methods in the included studies Table e7.1. Stool sample collection, storage and DNA extraction procedures of the included studies. Study Collection & handling by participant Long-term storage DNA extraction method Naseribafrouei et al. Outpatient samples frozen at 20 °C in home at 70 °C until use MagTM mini kit (LGC), 2014 freezer upon collection and transported at below following the manufacturer recommendations zero. Inpatients stored directly at -70 °C. Jiang et al. Sterile plastic cups were used to collect samples at 80 °C within 15 min of QIAamp® DNA Stool Mini Kit (QIAGEN), following 2015 and were kept in an icebox collection, until use manufacturer instructions, with additional glass-bead beating steps on a Mini-beadbeater (FastPrep; Thermo Electron Corp) Aizawa et al. 2016 collected with RNA stabilizer an stored at room - Total RNA fractions were extracted (Yakult Central temperature or at 4°C until sent to lab Institute), method not reported Kelly et al. Collected in plastic containers containing an Homogenized, aliquoted and QIAamp DNA Stool Mini Kit (QIAGEN) 2016 anaerobic generator AnaeroGen Compact Oxoid stored at 80 °C until further use sachet Liu et al. - Immediately stored at 80 °C until PowerSoil DNA Isolation Kit (MoBio), following Human 2016 use Microbiome Project recommendations Zheng et al. 2016 - Immediately stored at 80 °C until PowerSoil DNA Isolation Kit (MoBio), following use standard protocols Lin et al. - Immediately stored at 70 °C until Tiagen DNA Stool Mini Kit (Tiagen Biotech), following 2017 use manufacturer protocols Chen et al. 2018a - at 80 °C until use PowerSoil DNA kit following standard protocol Chen et al. 2018b Sterile plastic cups were used to collect samples at 80 °C until use Tandem mass spectra were extracted and analyzed using Mascot (Matrix Science) against a combined Swiss prot-human (20151226) and TrEMBL bacteria database Huang et al. 2018 sterile containers were used to collect samples Immediately stored at 80 °C until PowerSoil DNA Kit (Missouri Biotechnology use Association) Chahwan et al. kept on ice or refrigerated before delivery to study Samples placed at 4 °C and PowerFecal DNA Isolation Kit (MoBio), following 2019 staff aliquots stored at 80 °C within manufacturer instuctions several days © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method Valles-Colomer Frozen at -18°C at home and cool transported to stored at -18°C until transport on PowerMicrobiome RNA Isolation kit (MoBio 2019 collection point dry ice to the research facility for - Laboratories) 80°C storage Chung et al. 2019 delivered in 4 °C to staff and 80 °C until use QIAamp DNA Stool Mini Kit (QIAGEN) or phenol– chloroform extraction method Lai et al. 2019 - immediately stored at 80 °C until StoolGen DNA kit (CWBiotech Co) use Rong et al. 2019 - immediately stored at 80 °C until StoolGen DNA kit (CWBiotech Co) shipped to lab Chen et al. 2020 - - standard PowerSoil kit protocol Chen et al. 2021 Sterile plastic cups were used to collect samples stored at 80 °C within 30 min of Qiagen QIAamp DNA Stool Mini Kit (Qiagen) according collection until use to manufacturer’s instructions Yang et al. 2020 - - E.Z.N.A. Soil DNA Kit (Omega Bio-tek, Norcross, GA, USA) according to manufacturer’s instructions Liu et al. 2020 OMNIgene•GUT stool collection kits at 80 °C until use ZymoBIOMICS 96 DNA Kit (Zymo Research) according to manufacturer’s instructions Stevens et al. 2020 collected with OMNIgeneGUT fecal at 80 °C until use Stool Extraction Kit following the manufacturer’s collection kits (DNAGenotek, Ontario, Canada) instructions (Omega Bio-tek, Doraville, CA). Stevens et al. 2018 [ - at 80 °C until use - Vinberg et al. 2019 stool collection kits (Sarsted Kit sterile) at 80 °C within 24-72 hours, until NucleoSpin 96 Soil kit (Macherey-Nagel) use Jiang et al. 2020 collected in a sterile plastic cup at the stored at 80 °C within 30 min of FastDNATM SPIN Kit for Feces (MP Biomedicals) hospital and refrigerated collection until use according to manufacturer's instructions Evans et al. 2017 Home stool collection kits (DNA Genotek, Ontario at 80 °C until use PowerMag soil isolation kit (MoBio) CA) Painold et al. 2018 - PowerLyzer PowerSoil DNA Isolation Kit (MoBio) - 20 C until use Aizawa et al. 2019 collected with RNA stabilizer and stored at 4°C at at 4°C until use Total RNA fractions were extracted (Yakult Central home Institute), method not reported Coello et al. 2019 collected with OMNIgeneGUT fecal at 80 °C until use NucleoSpin ® 96 Soil kit collection kits (DNAGenotek, Ontario, Canada) © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method McIntyre et al. 2019 Collected in sterile screw-capped sample jar and Processed upon receipt; back-ups Using in-house protocol on a MagMax™ robot frozen in home freezer stored at 80 °C (Thermo Fischer Scientific,WalthamMass) Hu et al. 2019 at -80°C within 30 min of PSP Spin Stool DNA Plus Kit (Stratec, Germany) collection, until use according to the manufacturer’s instructions. Lai et al. 2021 - At -80 until use StoolGen DNA kit (CWBiotech Co., Beijing, China) Lu et al. 2019 - at -80°C within 30 min of Qiagen Stool Kit (Qiagen, Hilden, Germany), according collection, until use to a modified protocol for cell lysis Nguyen et al. 2019 home stool collection kits (BD SWUBE Dual Swab at 80 °C until use Earth Microbiome Project (EMP), modified from Collection System; BD Worldwide MagAttract® PowerSoil® DNA KF Kit Schwarz et al. 2018 Collected in a larger sampling bowl, from which at 80 °C until use FastDNA Spin Kit for Soil (QBIOgene duplicate samples were transferred to smaller tubes 80 °C until use Shen et al. 2018 - at PowerSoil DNA kit (MoBio) Yuan et al. 2018 - - QIAamp Fast DNA Stool Mini Kit (QIAGEN) Zheng et al. 2019 - immediately stored at 80 °C until QIAamp DNA Stool Mini Kit (QIAGEN, Hilden, use Germany). Pan et al. 2020 Collected and immediately transported to lab with at 80 °C until use E.Z.N.A. soil kit (Omega Bio-tek, Norcross, GA, U.S.) ice packs according to manufacturer's instructions Zhu et al. 2020 - - - Li et al. 2020 - at 80 °C until use MOBIO PowerSoil DNA Isolation Kit 12,888–100 protocol Ma et al. 2020 collected in sterile plastic containers immediately stored at 80 °C until QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany) use according to the manufacturer's instructions Zhang et al. 2020 Sterile plastic cups were used to collect samples at 80 °C within 30 min of FastDNA™ SPIN Kit for Feces (MP Biomedical Inc., and were kept in an icebox collection, until use Santa Ana, CA, USA) according to the manufacturer’s instructions with additional glass-bead beating steps Xu et al., 2020 Collected in disposable sterile potty and Same or next day moved to 80 StoolGen fecal DNA extraction kit (CWBiotech, Beijing, transferred to tubes then frozen at -20°C °C until use China) Jiang et al. 2018 Sterile plastic cups were used to collect samples at 80 °C within 30 min of QIAamp DNA Stool Mini Kit (QIAGEN), following and were kept in an icebox collection, until use manufacturer instructions, with the addition of a glass- © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method bead beating step on a Mini-beadbeater (FastPrep; Thermo Electron Corp) Chen et al. 2019 Sterile plastic cups were used to collect samples at 80 °C within 15 min of QIAamp DNA Stool Mini Kit (QIAGEN), following collection, until use manufacturer instructions Mason et al. 2020 Frozen in home freezer after collection at 80 °C until use Crude DNA extracts were treated with RNAseA (QIAGEN) and column-purified (PCR Purification Kit, QIAGEN) Armougom et al. - - NucleoSpinH Tissue Mini Kit according to 2009 manufacturer’s instructions Million et al. 2013 collected using sterile plastic containers and at -80°C until use NucleoSpin® Tissue Mini Kit (information taken from transported “as soon as possible” to the lab cross-referenced source [1] Kleiman et al. 2015 collected by nurses trained in collection protocols at -80°C until use Qiagen DNeasy® Blood and Tissue extraction kit (Qiagen, Valencia, CA, USA) Morita et al. 2015 Collected in tubes containing the RNA stabilizer at 4°C until use Total RNA fractions were extracted (Yakult Central Institute), method not reported Mack et al. 2016 Collected with a stool-collecting kit (Süsse immediately stored at 80 °C until PSP Spin Stool Kit (Stratec Molecular, Berlin, Labortechnik, Gudensberg, Germany) from eight use Germany) according different sites of the stool to the manufacturers’ instructions Mörkl et al. 2017 Collected with the PSP spin stool DNA stool immediately stored at 20 °C until PowerLyzer PowerSoil DNA Isolation Kit (MO BIO collection use Laboratories, CA) according to the manufacturer’s kit (Stratec, Birkenfeld, Germany) instructions. Borgo et al. 2017 - at -80°C until use Spin stool DNA kit (Stratec Molecular, Berlin, Germany), according to the manufacturer's instructions Hanachi et al. 2018 - - Standard Operating Procedure 07 of the IHMS Hata et al. 2019 Collected and sealed in a plastic bag containing a - - disposable oxygen-absorbing and carbon dioxide– generating agent and transported on ice to the laboratory within several hours. Monteleone et al. - at -80°C until use PowerSoil DNA isolation kit (Qiagen, Germantown, 2021 MD, USA) Domenech et al. collected in Stool Collection Tubes (Stratec At -20°C until use PSP Spin Stool 15 DNA Basic Kit (Stratec Molecular) pre-print 14 Molecular) © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method Turna et al. 2020 collected and transferred into a sterile screw - - capped sample jar and placed in a household freezer (for up to 1 week) Hemmings et al. - - PSP® Spin Stool DNA Plus Kit (STRATEC Molecular, 2017 Birkenfeld, Germany) according to the manufacturer’s protocol 2 (“Isolation of total DNA from 1.4 ml stabilized stool homogenate with enrichment of bacterial DNA”). Aarts et al. 2017 pea-sized amount stored it in a 50ml Falcon tube, At -80°C within 24hrs, until use DNeasy1Blood and Tissue Kit (Qiagen, Venlo, The then Netherlands) stored at 4°C until delivery to site – information not available eReferences not included elsewhere: 1. Dridi B, Henry M, El Khéchine A, et al. High Prevalence of Methanobrevibacter smithii and Methanosphaera stadtmanae Detected in the Human Gut Using an Improved DNA Detection Protocol. PLoS One; 4. Epub ahead of print 17 September 2009. DOI: 10.1371/journal.pone.0007063. © 2021 American Medical Association. All rights reserved. eAppendix 8. Publication bias assessment for the alpha diversity meta-analyses Figure e8.1. Funnel plots assessing publication bias in the meta-analyses of alpha diversity. A. Chao1, B. Observed species, C. Phylogenetic diversity, D. Shannon, E. Simpson index. © 2021 American Medical Association. All rights reserved. eAppendix 9. Beta diversity Table e9.1. Methodology and findings of the included studies assessing beta diversity for the patient vs. control group comparison. Disorder Study Year Metric Analysis Finding MDD Jiang 2015 Unweighted Unifrac - no sig. difference MDD Kelly 2016 Bray-Curtis PCoA, Adonis sig. different Unweighted Unifrac PERMANOVA sig. different Weighted Unifrac sig. different MDD Zheng 2016 Bray-Curtis PCoA sig. different MDD Lin 2017 Weighted Unifrac PCoA sig. different MDD Chen 2018a Unifrac (nr) PCoA, PLS-DA sig. different MDD Huang 2018 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different MDD Chawan 2019 Weighted Unifrac PCoA, no sig. difference PERMANOVA MDD Chung 2019 Weighted Unifrac PERMANOVA sig. different MDD Lai 2019 Bray-Curtis PCoA, sig. different PERMANOVA MDD Chen 2020 - OPLS-DA sig. different MDD Chen 2021 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different MDD Yang 2020 Bray-Curtis PERMANOVA sig. different MDD Liu 2020 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different Bray - Curtis sig. different MDD Stevens 2020 Bray-Curtis (unfiltered PERMANOVA no sig. difference data) sig. different Bray-Curtis (filtered data) MDD, ANX, Mason 2020 Weighted Unifrac PCoA no sig. difference MDD +ANX MDD, Jiang 2020 Unweighted Unifrac PCoA no sig. difference BD Weighted Unifrac no sig. difference Bray - Curtis sig. different MDD, BD Rong 2019 Bray-Curtis PCoA no sig. difference MDD+BD Vinberg 2019 Generalized UniFrac PCoA, no sig. difference PERMANOVA BD Evans 2017 Yue & Clayton distance PCoA, AMOVA sig. different BD Painold 2018 Unweighted Unifrac PCoA no sig. difference Weighted Unifrac BD Coello 2019 Weighted Unifrac - no sig. difference Unweighted Unifrac sig. different BD Mcintyre 2019 Bray–Curtis PCoA no sig. difference BD Hu 2019 Unweighted Unifrac PCoA sig. different Weighted Unifrac BD Lai 2021 Bray-Curtis PERMANOVA sig. different SCZ Nguyen 2019 Unweighted Unifrac PCoA sig. different Bray-Curtis sig. different SCZ Shen 2018 Unweighted Unifrac PCoA, ANOSIM sig. different SCZ Zheng 2019 - PLS-DA sig. different SCZ Pan 2020 Unweighted Unifrac PCoA, ANOSIM no sig. difference © 2021 American Medical Association. All rights reserved. SCZ Zhu 2020 Bray-Curtis - sig. different SCZ Li 2020 Bray-Curtis PCoA, sig. different PERMANOVA SCZ Ma 2020 Unweighted Unifrac PCoA, sig. different Weighted Unifrac PERMANOVA no difference SCZ Zhang 2020 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different Bray-Curtis sig. different SCZ Xu 2020 - NMDS sig. different ANX Jiang 2018 UniFrac unweighted PCoA, sig. different PERMANOVA ANX Chen 2019 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different AN Kleiman 2015 Unweighted Unifrac PCoA no sig. difference Weighted Unifrac no sig. difference AN Mack 2016 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different Bray Curtis sig. different AN Mörkl 2017 Unweighted Unifrac PCoA, ANOSIM, sig. different Weighted Unifrac Adonis sig. different AN Borgo 2017 - RDA, ANOSIM no sig. difference AN Hanachi 2018 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different AN Montelione 2020 - NMDS, no sig. difference PERMANOVA PTSD Hemmings 2017 Unweighted Unifrac PCoA, ANOSIM no sig. difference Weighted Unifrac no sig. difference Bray-Curtis no sig. difference OCD Domenech pre- Unweighted Unifrac PCoA, no sig. difference print Weighted Unifrac PERMANOVA no sig. difference Bray-Curtis no sig. difference Jensen-Shannon no sig. difference Canberra no sig. difference OCD Turna 2020 Unweighted Unifrac PCoA, no sig. difference Weighted Unifrac PERMANOVA no sig. difference Bray-Curtis no sig. difference Jaccard no sig. difference PCoA = principal coordinates analysis; PERMANOVA = permutational analysis of variance; PLS-DA = principal least squares discriminant analysis; OPLS-DA = orthogonal principal least squares discriminant analysis; NMDS = non-metric multidimensional scaling © 2021 American Medical Association. All rights reserved. eAppendix 10. Figures for study-level findings of relative abundance of microbial taxa Figure e10.1. Study-level findings of relative abundance of microbial taxa in patients with psychiatric disorders compared to healthy controls at the: A. Phylum level, B. Family level, C1-2. Genus level. ADHD= attention deficit hyperactivity disorder, AN= anorexia nevrosa, ANX= anxiety, BD= bipolar disorder, MDD= depression, OCD= obsessive compulsive disorder, PTSD= post-traumatic stress disorder, SCZ= psychosis & schizophrenia © 2021 American Medical Association. All rights reserved. Figure e10.1. Continued © 2021 American Medical Association. All rights reserved. Figure e10.1. Continued © 2021 American Medical Association. All rights reserved. Figure e10.1. Continued © 2021 American Medical Association. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JAMA Psychiatry American Medical Association

Perturbations in Gut Microbiota Composition in Psychiatric Disorders

Loading next page...
 
/lp/american-medical-association/perturbations-in-gut-microbiota-composition-in-psychiatric-disorders-5BN0w5qUoT
Publisher
American Medical Association
Copyright
Copyright 2021 Nikolova VL et al. JAMA Psychiatry.
ISSN
2168-622X
eISSN
2168-6238
DOI
10.1001/jamapsychiatry.2021.2573
Publisher site
See Article on Publisher Site

Abstract

Research JAMA Psychiatry | Original Investigation A Review and Meta-analysis Viktoriya L. Nikolova, MRes; Megan R. B. Hall, BSc; Lindsay J. Hall, PhD; Anthony J. Cleare, MBBS, PhD; James M. Stone, MBBS, PhD; Allan H. Young, MD, PhD Multimedia IMPORTANCE Evidence of gut microbiota perturbations has accumulated for multiple Supplemental content psychiatric disorders, with microbiota signatures proposed as potential biomarkers. However, no attempts have been made to evaluate the specificity of these across the range of psychiatric conditions. OBJECTIVE To conduct an umbrella and updated meta-analysis of gut microbiota alterations in general adult psychiatric populations and perform a within- and between-diagnostic comparison. DATA SOURCES Cochrane Library, PubMed, PsycINFO, and Embase were searched up to February 2, 2021, for systematic reviews, meta-analyses, and original evidence. STUDY SELECTION A total of 59 case-control studies evaluating diversity or abundance of gut microbes in adult populations with major depressive disorder, bipolar disorder, psychosis and schizophrenia, anorexia nervosa, anxiety, obsessive compulsive disorder, posttraumatic stress disorder, or attention-deficit/hyperactivity disorder were included. DATA EXTRACTION AND SYNTHESIS Between-group comparisons of relative abundance of gut microbes and beta diversity indices were extracted and summarized qualitatively. Random-effects meta-analyses on standardized mean difference (SMD) were performed for alpha diversity indices. MAIN OUTCOMES AND MEASURES Alpha and beta diversity and relative abundance of gut microbes. RESULTS A total of 34 studies provided data and were included in alpha diversity meta-analyses (n = 1519 patients, n = 1429 control participants). Significant decrease in microbial richness in patients compared with control participants were found (observed species SMD = −0.26; 95% CI, −0.47 to −0.06; Chao1 SMD = −0.5; 95% CI, −0.79 to −0.21); however, this was consistently decreased only in bipolar disorder when individual diagnoses were examined. There was a small decrease in phylogenetic diversity (SMD = −0.24; 95% CI, −0.47 to −0.001) and no significant differences in Shannon and Simpson indices. Differences in beta diversity were consistently observed only for major depressive disorder and psychosis and schizophrenia. Regarding relative abundance, little evidence of disorder specificity was found. Instead, a transdiagnostic pattern of microbiota signatures was found. Depleted levels of Faecalibacterium and Coprococcus and enriched levels of Eggerthella were consistently shared between major depressive disorder, bipolar disorder, psychosis and schizophrenia, and anxiety, suggesting these disorders are characterized by a reduction of anti-inflammatory butyrate-producing bacteria, while pro-inflammatory genera are enriched. The confounding associations of region and medication were also evaluated. CONCLUSIONS AND RELEVANCE This systematic review and meta-analysis found that gut microbiota perturbations were associated with a transdiagnostic pattern with a depletion of certain anti-inflammatory butyrate-producing bacteria and an enrichment of pro-inflammatory bacteria in patients with depression, bipolar disorder, schizophrenia, and Author Affiliations: Author anxiety. affiliations are listed at the end of this article. Corresponding Author: Viktoriya L. Nikolova, MRes, Centre for Affective Disorders, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, De Crespigny Park, JAMA Psychiatry. 2021;78(12):1343-1354. doi:10.1001/jamapsychiatry.2021.2573 London SE5 8AF, United Kingdom Published online September 15, 2021. Last corrected on December 1, 2021. (viktoriya.nikolova@kcl.ac.uk). (Reprinted) 1343 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders espite evidence that probiotic formulations can im- prove mental health dating back to the early 20th Key Points 1,2 D century, it was only following advances in DNA/ Question Do psychiatric disorders present with distinct or shared RNA sequencing technologies that the involvement of the gut gut microbial alterations? microbiota in the pathophysiology of psychiatric disorders was Findings This review and meta-analysis of 59 case-control studies recognized. Preclinical studies have consistently demon- found that gut microbiota perturbations were associated with a strated that fecal microbiota transplants from patients with a transdiagnostic pattern with a depletion of certain wide range of psychiatric conditions result in the develop- anti-inflammatory butyrate-producing bacteria and an enrichment ment of the behavioral and physiological profile of the condi- of pro-inflammatory bacteria in depression, bipolar disorder, 3-7 tion in germ-free mice. This suggests that psychiatric dis- schizophrenia, and anxiety. orders may be associated with a distinct pattern of microbial Meaning These findings are in line with genetic and inflammatory perturbations, which may serve as a biomarker. marker studies and support the transdiagnostic dimensional Attempts to characterize the composition of the micro- model of psychiatric disorders by highlighting the gut microbiota biota in psychiatric populations have yielded plentiful yet con- as an additional dimensional component. tradictory results. Nevertheless, systematic reviews in indi- vidual disorders have been able to identify patterns that may 8-10 be promising biomarker targets. Indeed, the addition of such Selection Criteria biomarkers can improve diagnostic accuracy, guide treat- Systematic reviews and meta-analyses were considered eli- ment, and assist the monitoring of treatment response. For the gible if they followed established guidelines and included at definition of a biomarker to be met, ie, “substance, structure or least 1 eligible original study. Original studies were eligible if process that can be measured in the body and influence or pre- they (1) applied an observational case-control design, (2) per- dict the incidence of outcome or disease,” the specificity and formed gut microbiota analysis and reported diversity or abun- reproducibility of the alteration needs to be demonstrated. dance measures, and (3) sampled a general adult population Therefore,itiscrucialtocomparemicrobialperturbationsacross (age 18-65 years) with a psychiatric diagnosis of interest. In- the wider range of psychiatric conditions. terventional or longitudinal comparisons in the absence of a We performed an umbrella and updated review and meta- control group were excluded. Records were screened by 2 au- analysis of gut microbiota studies in adults with major depres- thors (V.L.N. and M.R.B.S) and discrepancies resolved via dis- sive disorder (MDD), bipolar disorder, psychosis and schizo- cussion and consultation with a third author (A.H.Y.). phrenia, anxiety disorders, obsessive compulsive disorder (OCD), eating disorders (anorexia nervosa and bulimia Data Extraction nervosa), autism spectrum disorder, attention-deficit/ Information was extracted using a predesigned template by 2 hyperactivity disorder (ADHD), and posttraumatic stress dis- authors (V.L.N. and M.R.B.S) and cross-checked. From sys- order (PTSD) to evaluate the specificity and reproducibility of tematic reviews and original studies, we extracted publica- gut microbiota alterations and delineate those with potential tion details, participant demographic and clinical character- to become biomarkers. istics, and methodological information. As primary outcomes of interest, we extracted community-level measures of gut mi- crobiota composition (alpha and beta diversity) and taxo- nomic findings at the phylum, family, and genus levels (rela- Methods tive abundance). Alpha diversity provides a summary of the The protocol for this review was preregistered with microbial community in individual samples and can be com- PROSPERO (CRD42021224342). We followed Preferred pared across groups to evaluate the role of a particular factor Reporting Items for Systematic Reviews and Meta-analyses (in this case psychiatric diagnosis) on the richness (number of (PRISMA) reporting guideline as well as Cochrane guidance species) and evenness (how well each species is represented) 14,15 10,16 for umbrella and updated reviews. in the sample. Beta diversity is a measure of interindi- vidual (between samples) diversity that assesses similarity of communities compared with the other samples analyzed. Search Details We searched Cochrane Library, PubMed, Embase, and This analysis allows us to see whether patient samples clus- PsycINFOonJanuary27,2021.Thesearchstringsusedareavail- ter significantly differently (ie, with little or no overlap) com- able in eAppendix 1 in the Supplement. This search was lim- pared with control participant samples or whether they over- ited to systematic reviews and meta-analyses in English, in- lap, thus suggesting the 2 groups are not distinct. Control cluding human studies, published since 2005. After reviewing samples were defined as individuals without the relevant the results, we realized that a large body of recent literature condition. was missed, as numerous studies have become available fol- lowing the publication of the latest reviews. To ensure thor- Quality Assessment ough coverage, we performed an updated search for each dis- We performed quality assessment of the systematic reviews order on February 2, 2021, from the search date recorded in using the ROBIS tool and of the original studies not covered the latest available high-quality review for that disorder (eAp- in any review with the Joanna Briggs Institute Critical pendix 1 in the Supplement). Appraisal Checklist for Case-Control Studies. No studies were 1344 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research Table. Summary Characteristics of the Identified Reviews and Original Studies by Psychiatric Disorder No. Mean patient Female, a c Disorder Reviews Studies Total patients Region of studies age, y mean % MDD 8 21 930 East: n = 14; west: n = 7 35 60 Schizophrenia and psychosis 5 11 699 East: n = 9; west: n = 2 36 45 Bipolar disorder 3 9 465 East: n = 5; west: n = 4 38 55 Anorexia nervosa 3 10 211 East: n = 2; west: n = 8 26 99 Anxiety 2 3 84 East: n = 2; west: n = 1 40 77 OCD 0 2 59 West: n = 2 36 54 PTSD 0 1 18 Africa: n = 1 42 14 ADHD 1 1 19 West: n = 1 20 32 MDD + anxiety NA 2 60 West: n = 2 39 82 MDD + bipolar disorder NA 2 98 East: n = 1; west: n = 1 37 69 Total 16 59 2643 East: n = 32; west: n = 24; NA NA Africa: n = 1 Abbreviations: ADHD, attention-deficit/hyperactivity disorder; MDD, major Some include >1 disorder. depressive disorder; NA, not applicable; OCD, obsessive compulsive disorder; c West region includes US, Canada, Europe, Australia, and New Zealand. East PTSD, posttraumatic stress disorder. region includes China, Japan, and Taiwan. Africa includes South Africa. a 21,22 Studies that examined combined cohorts (MDD + bipolar disorder or d Adult populations only. 23,24 MDD + anxiety ) are presented separately. excluded owing to quality concerns. The detailed assessment of psychiatric medication. All analyses were completed in R is available as eAppendix 2 in the Supplement. version 4.17-0 (meta package; R Foundation). Two-sided P values were statistically significant at less than .05. Qualitative Synthesis For the relative abundance of microbial taxa, we performed a qualitative synthesis owing to the large number and limited Results overlap of findings. Owing to the significant likelihood of false positives noted in previous meta-analyses, results reported Search Results only by a single study were excluded. Further, results re- We identified 16 systematic reviews (eAppendices 4 and 5 in ported only by 1 research group were also excluded because the Supplement for PRISMA flowcharts and details of the sys- these were considered potentially methodology or popula- tematic reviews) containing 39 eligible studies. There were no tion specific. To identify disease-specific and shared altera- reviews capturing OCD, PTSD, or autism spectrum disorder in tions, we performed a within- and between-diagnostic com- adults. In the second search, a further 20 studies were iden- parison.First,wesummarizedwithin-disorderfindingsforeach tified, resulting in 59 studies across 8 disorders. The most re- taxon reported in at least 2 studies and labeled those in- searched disorder was MDD, followed by psychosis and schizo- creased, decreased, or not consistent. Not consistent was any phrenia, bipolar disorder, and anorexia nervosa (Table). finding with less than 75% agreement between studies report- ing this taxon. A consistent finding by 2 studies was consid- Characteristics of Included Studies ered worth noting for future validation, whereas a finding by The 59 studies provided 64 case-control comparisons captur- 3 or more studies (from ≥2 research groups) was considered ing 2643 patients and 2336 controls (eAppendix 6 in the potentially associated with the disorder. A taxon was consid- Supplement provides a detailed summary of study character- ered a candidate for disease-specific response if it was istics). Most studies (32 [54.2%]) were conducted in East Asia altered (in a consistent direction) in a single disorder only. (China, Japan, and Taiwan), 24 (40.7%) in westernized popu- Alternatively, if a shift was replicated in several disorders with lations (US, Canada, Europe, Australia, and New Zealand; known symptomatic and pathophysiological overlap, this was grouped according to typical diet and lifestyle), and 1 (1.7%) in considered a transdiagnostic alteration. Taxa similarly al- Africa(SouthAfrica).Moststudieshadsmalltomoderatesample tered across all/multiple unrelated diagnostic categories were sizes (median, 62), ranging between 4 and 156 per group (eAp- interpreted as general disease response. pendix 6 in the Supplement). Studies were similar in exclusion criteria; however, few attempted to minimize dietary changes Quantitative Synthesis or control dietary intake (12 of 59 [20.3%]) or smoking status (8 Meta-analysis was performed on differences in alpha diver- of 59 [13.6%]). Use of psychiatric medication also varied sub- sity between patients and controls for indices with data re- stantially,with11of59studies(18.6%)conductedinmedication- ported in 10 or more studies. Detailed methods of data trans- free or drug-naive groups, 5 of 59 (8.5%) in groups undergoing formation and interpretation thresholds are available in treatment and the remainder not controlling this, resulting in eAppendix 3 in the Supplement. Publication bias was evalu- anywhere between 20% and 96% of patients taking medica- ated with funnel plots and Egger test. Preplanned subgroup tion. Methodology of stool processing (eAppendix 7 in the analyses were disorder, region of study (east/west), and use Supplement) and composition analysis (eAppendix 6 in the jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1345 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders Supplement) also varied widely, with 16S ribosomal RNA se- be noted that shotgun metagenomics showed increased quencing being most common (44 of 59 studies [74.6%]) fol- Shannon diversity in patients (4 studies) in comparison with lowed by 9 studies (15.2%) using quantitative polymerase chain 16SrRNA V3-V4 sequencing, which showed an overall de- reactionorreal-timequantitativepolymerasechainreactionand crease (12 studies). This could be because the shotgun ap- 7 (11.9%) using shotgun metagenomics. proach quantifies all genomic DNA (including mycobiome and virome) rather than just specific regions of bacterial DNA. Fur- Alpha Diversity ther studies using shotgun metagenomics or comparing the 2 Of 44 studies reporting alpha diversity, 34 provided data and methodologies on the same population are needed. were included in meta-analyses (1519 patients and 1429 con- trols). Eleven indices were used to assess alpha diversity, in- Beta Diversity cluding estimates of richness (observed species, Chao1, abun- Beta diversity comparison between patients and controls was dance coverage estimator, and incidence coverage estimator), reported in 43 studies, with 1 study reporting on 3 separate ), using a variety evenness, richness/evenness (Shannon, Simpson, inverse groups (MDD, anxiety, and MDD + anxiety Simpson, Fisher), biodiversity (Faith phylogenetic diversity), of measures (eAppendix 9 in the Supplement). Consistent non- and 1 newly developed index (eAppendix 6 in the Supple- significant differences were reported by 16 studies, and a fur- ment). The most widely used were observed species, Chao1, ther 3 reported conflicting results between the measures used. Shannon, Simpson, and phylogenetic diversity. There was no Patients’ samples clustered differently from controls in 12 of evidence of publication bias in any of the analyses (eAppen- 15 studies in MDD, 7 of 9 in psychosis and schizophrenia, 3 of dix 8 in the Supplement). 6 in bipolar disorder, 3 of 6 in anorexia nervosa, 2 of 3 in anxi- Regarding richness, 20 studies provided data on ob- ety, 0 of 2 in OCD, and 0 of 1 in PTSD (eAppendix 9 in the served species in patients (n = 897) vs controls (n = 789). The Supplement). One of 2 combined MDD + bipolar disorder co- pooled estimate showed a significant decrease in patients hort was also significantly different from controls, whereas the with a small effect size (standardized mean difference MDD + anxiety cohort was not. Although, while Mason et al [SMD] = −0.26; 95% CI, −0.47 to −0.06; P = .01) and high found no differences when looking at diagnostic categories, 2 3,4,22,26-42 heterogeneity (I =75%) (Figure 1A). Within diag- they found a significant difference when clustering partici- nostic categories, there was a significant decrease only in bi- pantsaccordingtoself-reportedsymptoms.Thesefindingssug- polar disorder (SMD = −0.61; 95% CI, −1.19 to −0.03; P = .04; gest there is reliable evidence for differences in the shared phy- I = 80%). Twenty-six studies provided data on Chao1 in pa- logenetic structure in MDD and psychosis and schizophrenia tients (n = 956) vs controls (n = 961). The pooled estimate compared with controls; however, method of measurement showed a significant decrease in patients with a medium ef- and method of patient classification (symptom vs diagnosis fect size (SMD = −0.5; 95% CI, −0.79 to −0.21; P = .001; based) may affect findings. I = 88%). Regarding individual diagnoses, there was a signifi- cant decrease only in bipolar disorder and anorexia nervosa Differentially Abundant Microbial Taxa (SMD = −0.53; 95% CI, −1.01 to −0.05; P = .03; I = 62% and All studies assessed the relative abundance of gut microbes and SMD = −0.86; 95% CI, −1.52 to −0.21; P = .01; I = 80%, respec- 57 of 59 (96.6%) identified significant differences between pa- 4,21,25,27,29,31-33,35-49 tively) (Figure 1B). tients and controls at phylum, family, or genus levels. Overall, Regarding diversity, 29 studies reported the Shannon in- in MDD (21 comparisons), 94 taxa were differentially abundant; dex in patients (n = 1176) vs controls (n = 1172). The pooled es- in psychosis and schizophrenia (11 comparisons), 136; in bipo- timate demonstrated a nonsignificant difference between lar disorder (9 comparisons), 60; in anxiety (2 comparisons), 36; groups (SMD = −0.12; 95% CI, −0.27 to 0.03; P = .11) inanorexianervosa(10comparisons),32;inOCD(2comparisons), 3,4,21,22,25,27,28,31-35,38-46,48,50-53 (Figure 2A). Simpson index data 15; and in ADHD and PTSD (1 study each), 9 and 3, respectively. were provided by 11 studies (n = 418 patients; n = 377 con- Afterremovalofnonreplicatedfindings,thedifferencesspanned trols). There was a nonsignificant difference between groups 7 phyla, 28 families, and 67 genera. Study-level findings are pre- (SMD = 0.04; 95% CI, −0.13 to 0.21; P = .66), with nonsignifi- sented in eAppendix 10 in the Supplement. 21,26,27,31-33,40,43,52 cant heterogeneity (Figure 2B). Finally, 10 Figure 3providesthesummaryofthewithin-andbetween- studies provided phylogenetic diversity data in patients disorder comparison for the disorders with sufficient studies (n = 412) vs controls (n = 454). The pooled estimate showed a (anorexia nervosa, MDD, bipolar disorder, anxiety, and psy- significant decrease in patients with a small effect size chosis and schizophrenia). There was high within-disorder in- (SMD = −0.24; 95% CI, −0.47 to −0.0012; P = .049; 64%) consistency and the majority of consistent within-disorder 3,4,28,32-34,39,40,42,44 (Figure 2C). changes were replicated by only 2 studies and thus require fur- To explore sources of interstudy heterogeneity, sub- ther investigation. Considerably fewer were replicated by more group analyses and meta-regressions were performed for the than 2 studies from different research groups. analyses with sufficient studies (observed species, Chao1, Shannon). Body mass index, age, sex, smoking, region (east/ Limited Evidence of Disorder Specificity west), psychiatric medication use, subgrouping of psychosis Disorder specificity was observed for the enrichment of gen- and schizophrenia into first episode, and chronic and sequenc- era Holdemania and Olsenella and the depletion of genera ing method (including hypervariable region sequenced) did not Fusicatenibacter, Dialister, and Sutterella in MDD (Figure 3C). have a significant association with findings. However, it should However, these findings were weakly reproduced (3 to 4 of 21 1346 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research Figure 1. Forest Plots of Alpha Diversity Richness Estimators in the Gut Microbiota of Patients With Psychiatric Disorders Compared With Healthy Controls A Observed species B Chao1 Source SMD (95% CI) Decreased Increased Source SMD (95% CI) Decreased Increased MDD MDD 26 43 Naseribafrouei et al, 2014 0.44 (–0.13 to 1.01) Jiang et al, 2015 1.32 (0.75 to 1.89) 4 4 Kelly et al, 2016 –0.90 (–1.40 to –0.39) Kelly et al, 2016 –0.82 (–1.32 to –0.32) 3 44 Zheng et al, 2016 0.32 (–0.03 to 0.68) Huang et al, 2018 –1.15 (–1.73 to –0.57) 27 25 Chen et al, 2021 0.14 (–0.24 to 0.53) Rong et al, 2019 –1.05 (–1.60 to –0.51) 28 27 Liu et al, 2020 –0.17 (–0.59 to 0.24) Chen et al, 2021 0.21 (–0.17 to 0.60) 22 21 Vinberg et al, 2019 –0.80 (–1.27 to –0.33) Jiang et al, 2020 –0.57 (–1.30 to 0.16) Total –0.16 (–0.58 to 0.27) Total –0.34 (–1.08 to 0.40) 2 2 2 2 Heterogeneity: χ = 28.71; P <.001; I = 83% Heterogeneity: χ = 58.52; P <.001; I = 91 % 5 5 Bipolar disorder Bipolar disorder 29 21 Painold et al, 2019 –0.59 (–1.31 to 0.13) Jiang et al, 2020 0.25 (–0.54 to 1.05) 30 29 Coello et al, 2019 –0.22 (–0.51 to 0.07) Painold et al, 2019 –0.33 (–1.04 to 0.39) 31 25 Hu et al, 2019 –1.05 (–1.47 to –0.62) Rong et al, 2019 –0.77 (–1.30 to –0.25) Total Hu et al, 2019 –0.61 (–1.19 to –0.03) –0.94 (–1.36 to –0.52) 2 2 Heterogeneity: χ = 9.9; P =.007; I = 80% Total –0.53 (–1.01 to –0.05) 2 2 Schizophrenia and psychosis Heterogeneity: χ = 7.8; P =.05; I = 62% Shen et al, 2018 –0.27 (–0.64 to 0.09) Schizophrenia and psychosis 33 32 Pan et al, 2020 0.41 (–0.11 to 0.93) Shen et al, 2018 –0.26 (–0.63 to 0.10) 34 45 Li et al, 2020 –0.03 (–0.34 to 0.27) Zheng et al, 2019 –0.18 (–0.52 to 0.16) 35 33 Zhang et al, 2020 –0.21 (–1.00 to 0.58) Pan et al, 2020 0.13 (–0.39 to 0.65) Total Ma et al, 2020 (FEP) –0.04 (–0.31 to 0.24) –0.04 (–0.43 to 0.35) 2 2 46 Heterogeneity: χ = 4.61; P =.20; I = 35% Ma et al, 2020 (schizophrenia) –0.63 (–0.96 to –0.31) Anxiety Zhang et al, 2020 –0.36 (–1.16 to 0.44) 36 47 Jiang et al, 2018 –0.55 (–1.01 to –0.09) Xu et al, 2020 –2.71 (–3.13 to –2.29) Total Total –0.55 (–1.01 to –0.09) –0.58 (–1.29 to 0.12) 2 2 Heterogeneity: NA Heterogeneity: χ = 121.69; P <.001; I = 95% Anorexia nervosa Anxiety 37 36 Kleiman et al, 2015 –1.97 (–2.90 to –1.03) Jiang et al, 2018 –0.56 (–1.02 to –0.10) Mack et al, 2016 Total –0.11 (–0.48 to 0.26) –0.56 (–1.02 to –0.10) Total –0.99 (–2.80 to 0.83) Heterogeneity NA 2 2 Heterogeneity: χ = 13.13; P <.001; I = 92% Anorexia nervosa PTSD Kleiman et al, 2015 –1.90 (–2.82 to –0.98) 39 38 Hemmings et al, 2017 –0.30 (–1.04 to 0.43) Mack et al, 2016 –0.15 (–0.53 to 0.22) –0.30 (–1.04 to 0.43) –0.92 (–1.49 to –0.35) Total Hanachi et al, 2019 Heterogeneity: NA Monteleone et al, 2021 –0.83 (–1.47 to –0.19) OCD Total –0.86 (–1.52 to –0.21) 40 2 2 Domènech et al, 2020 –0.53 (–1.00 to –0.05) Heterogeneity: χ = 14.81; P =.002; I = 80% Turna et al, 2020 0.09 (–0.51 to 0.68) PTSD Total –0.25 (–0.85 to 0.35) Hemmings et al, 2017 –0.33 (–1.07 to 0.40) 2 2 Heterogeneity: χ = 2.48; P =.12; I = 60% Total –0.33 (–1.07 to 0.40) ADHD Heterogeneity NA Aarts et al, 2017 0.25 (–0.25 to 0.76) OCD Total 0.25 (–0.25 to 0.76) Domènech et al, 2020 –0.53 (–1.00 to –0.05) Heterogeneity NA Turna et al, 2020 –0.15 (–0.75 to 0.45) Total –0.26 (–0.47 to –0.06) Total –0.38 (–0.75 to –0.01) 2 2 95% PI (–1.12 to 0.59) Heterogeneity: χ = .96; P =.33; I = 0% 2 2 Heterogeneity: χ = 74.75; P <.001; I = 75% ADHD –3 –2 –1 0 1 2 3 Aarts et al, 2017 0.10 (–0.40 to 0.60) SMD (95% CI) Total 0.10 (–0.40 to 0.60) Heterogeneity NA Total –0.50 (–0.79 to –0.21) 95% PI (–1.96 to 0.96) 2 2 Heterogeneity: χ = 217.38; P <.001; I = 88% –3 –2 –1 0 1 2 3 SMD (95% CI) ADHD indicates attention-deficit/hyperactivity disorder; FEP, first episode psychosis; MDD, major depressive disorder; NA, not applicable; OCD, obsessive compulsive disorder; PI, prediction interval; PTSD, posttraumatic stress disorder; SMD, standardized mean difference. studies). The archaeon Methanobrevibacter and genus estingly, an alteration in the same direction was also reported Anaerotruncus may also be candidates for disorder specific- in 2 studies from the other disorder, which could not be ex- ity because they were consistently associated with anorexia plained by apparent demographic, clinical, or methodologi- nervosa and psychosis and schizophrenia, respectively. Inter- cal factors. Nevertheless, specificity in anorexia nervosa jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1347 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders Figure 2. Forest Plots of Alpha Diversity in the Gut Microbiota of Patients With Psychiatric Disorders Compared With Healthy Controls A Shannon index B Simpson index Source SMD (95% CI) Decreased Increased Source SMD (95% CI) Decreased Increased MDD MDD 43 26 Jiang et al, 2015 Naseribafrouei et al, 2014 0.55 (0.03 to 1.08) 0.29 (–0.27 to 0.86) 4 43 Kelly et al, 2016 –1.03 (–1.54 to –0.52) Jiang et al, 2015 –0.35 (–0.86 to 0.16) 50 3 Liu et al, 2016 –1.56 (–2.34 to –0.79) Zheng et al, 2016 0.00 (–0.35 to 0.36) 3 27 Zheng et al, 2016 –0.17 (–0.53 to 0.19) Chen et al, 2021 0.35 (–0.03 to 0.74) 44 21 Huang et al, 2018 –0.78 (–1.34 to –0.23) Jiang et al, 2020 0.60 (–0.14 to 1.33) Lai et al, 2019 0.45 (–0.08 to 0.99) Total 0.14 (–0.14 to 0.43) 25 2 2 Rong et al, 2019 0.53 (0.01 to 1.04) Heterogeneity: χ = 6.95; P =.14; I = 42% Chen et al, 2021 –0.21 (–0.60 to 0.17) Bipolar disorder 21 21 Jiang et al, 2020 –0.59 (–1.32 to 0.15) Jiang et al, 2020 0.00 (–0.79 to 0.79) 28 31 Liu et al, 2020 –0.11 (–0.53 to 0.30) Hu et al, 2019 –0.22 (–0.62 to 0.18) 22 52 Vinberg et al, 2019 –0.50 (–0.96 to –0.04) Lai et al, 2021 0.28 (–0.26 to 0.82) Total –0.28 (–0.62 to 0.06) Total –0.03 (–0.34 to 0.28) 2 2 2 2 Heterogeneity: χ = 50.55; P <.001; I = 80% Heterogeneity: χ = 2.14; P =.34; I = 6% 10 2 Bipolar disorder Schizophrenia and psychosis 21 32 Jiang et al, 2020 –0.17 (–0.96 to 0.63) Shen et al, 2018 0.20 (–0.17 to 0.56) 25 33 Rong et al, 2019 Pan et al, 2020 –0.41 (–0.93 to 0.11) 0.46 (–0.05 to 0.97) Hu et al, 2019 –0.44 (–0.84 to –0.03) Total –0.08 (–0.68 to 0.52) 52 2 2 Lai et al, 2021 0.53 (–0.02 to 1.08) Heterogeneity: χ = 3.56; P =.06; I = 72% Total 0.09 (–0.43 to 0.62) OCD 2 2 40 Heterogeneity: χ = 11.19; P =.01; I = 73% Domènech et al, 2020 –0.12 (–0.58 to 0.35) –0.12 (–0.58 to 0.35) Schizophrenia and psychosis Total Shen et al, 2018 –0.16 (–0.52 to 0.21) Heterogeneity NA Zheng et al, 2019 –0.22 (–0.57 to 0.12) Total 0.04 (–0.13 to 0.21) Pan et al, 2020 0.36 (–0.16 to 0.88) 95% PI (–0.36 to 0.45) 53 2 2 Zhu et al, 2020 0.30 (–0.01 to 0.60) Heterogeneity: χ = 14.19; P =.16; I = 30% Li et al, 2020 –0.12 (–0.43 to 0.19) –2 –1 0 1 2 Ma et al, 2020 (FEP) 0.11 (–0.28 to 0.50) SMD (95% CI) Ma et al, 2020 (schizophrenia) –0.33 (–0.65 to –0.01) Zhang et al, 2020 0.29 (–0.51 to 1.08) Phylogenetic diversity Total –0.02 (–0.20 to 0.17) Source SMD (95% CI) Decreased Increased 2 2 Heterogeneity: χ = 13.23; P =.07; I = 47% MDD Anorexia nervosa Kelly et al, 2016 –1.16 (–1.68 to –0.64) Mack et al, 2016 0.13 (–0.24 to 0.51) Zheng et al, 2016 0.15 (–0.20 to 0.51) Hanachi et al, 2019 –0.40 (–0.94 to 0.15) Huang et al, 2018 –0.54 (–1.09 to 0.00) Total –0.09 (–0.61 to 0.42) 2 2 Liu et al, 2020 –0.22 (–0.63 to 0.20) Heterogeneity: χ = 2.48; P =.12; I = 60% Total –0.42 (–0.96 to 0.13) PTSD 2 2 Heterogeneity: χ = 17.6; P <.001; I = 83% Hemmings et al, 2017 0.15 (–0.58 to 0.88) Schizophrenia and psychosis Total 0.15 (–0.58 to 0.88) Shen et al, 2018 –0.07 (–0.43 to 0.29) Heterogeneity NA Pan et al, 2020 0.29 (–0.23 to 0.81) OCD Li et al, 2020 –0.07 (–0.38 to 0.23) Domènech et al, 2020 –0.31 (–0.78 to 0.15) 41 Total –0.01 (–0.22 to 0.20) Turna et al, 2020 –0.53 (–1.14 to 0.08) 2 2 Heterogeneity: χ = 1.54; P =.46; I = 0% Total –0.40 (–0.77 to –0.02) 2 PTSD 2 2 Heterogeneity: χ = 0.31; P =.58; I = 0% Hemmings et al, 2017 –0.28 (–1.01 to 0.46) ADHD Total –0.28 (–1.01 to 0.46) Aarts et al, 2017 –0.10 (–0.60 to 0.40) Heterogeneity NA Total –0.10 (–0.60 to 0.40) OCD Heterogeneity NA Domènech et al, 2020 –0.51 (–0.99 to –0.04) Total –0.12 (–0.27 to 0.03) Total –0.51 (–0.99 to –0.04) 95% PI (–0.81 to 0.58) Heterogeneity NA 2 2 Heterogeneity: χ = 85.63; P <.001; I = 67% ADHD –3 –2 –1 0 1 2 3 Aarts et al, 2017 –0.20 (–0.70 to 0.31) SMD (95% CI) Total –0.20 (–0.70 to 0.31) Heterogeneity NA Total –0.24 (–0.47 to 0.00) 95% PI (–0.97 to 0.50) 2 2 Heterogeneity: χ = 24.66; P =.003; I = 64% –2 –1 0 1 2 SMD (95% CI) ADHD indicates attention-deficit/hyperactivity disorder; FEP, first episode psychosis; MDD, major depressive disorder; NA, not applicable; OCD, obsessive compulsive disorder; PI, prediction interval; PTSD, posttraumatic stress disorder; SMD, standardized mean difference. 1348 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research Figure 3. Changes in Relative Abundance of Microbial Taxa Reported by at Least 2 Studies From a Diagnostic Category A Level: phylum B Level: family Increased Decreased Not consistent AN BD MDD ANX SCZ C Level: genus: phylum firmicutes AN BD MDD ANX SCZ Level: genus: all other phyla AN BD MDD ANX SCZ Gray cells indicate not examined, not reported, or not replicated. bipolar disorder (BD), 9; major depressive disorder (MDD), 21; anxiety (ANX), 2; psychosis and schizophrenia (SCZ), 11. Most replicated findings are indicated here, all of which have been reported by more than 1 research group. Number of studies: anorexia nervosa (AN), 10; cannot be assessed here because no studies in other eating dis- of 10 studies). Further, Atopobium was enriched in bipolar dis- orders were identified, and conditions such as obesity were be- order and MDD (5 of 5 studies), while Veillonella was en- yond the scope. No distinct disorder-specific alterations were riched in psychosis and schizophrenia and MDD (5 of 6 stud- observed for the remaining taxa. ies). There was also evidence for the increase of the pathogen Escherichia-Shigella in bipolar disorder, anxiety, and psycho- sis and schizophrenia (6 of 7 studies) but not MDD. The Transdiagnostic Alterations Our findings indicate an overlap between certain disorders: bi- Bifidobacterium and Bacteroides genera were reported fre- polar disorder, psychosis and schizophrenia, and anxiety were quently but inconsistently across these disorders (14 and 16 associated with MDD. The most consistent changes were deple- studies, respectively). tion of Faecalibacterium (in 15 of 17 studies reporting this ge- nus) and Coprococcus (10 of 10 studies) and the enrichment of Exploring Confounders: Region and Psychiatric Medication Eggerthella (in 10 of 11 studies) (eAppendix 10 in the Supple- We explored the association of study region (east/west) with ment). These were followed by enriched Lactobacillus (10 of microbial alterations. Owing to the limited overlap in find- 13 studies), Enterococcus (8 of 9 studies), and Streptococcus (8 ings and the imbalanced availability of studies by region (eg, jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1349 Actinomyces Anaerostipes Actinobacteria Bifidobacterium Blautia Bacteroidetes Atopobium Coprococcus Firmicutes Collinsella Dorea Proteobacteria Eggerthella Roseburia Fusobacteria Olsenella Lachnoclostridium Alistipes Fusicatenibacter Actinomycetaceae Bacteroides Anaerotruncus Bifidobacteriaceae Odoribacter Faecalibacterium Coriobacteriaceae Parabacteroides Ruminococcus Bacteroidaceae Paraprevotella Subdoligranulum Prevotellaceae Prevotella Oscillibacter Rikenellaceae Bilophila Gemmiger Alcaligenaceae Desulfovibrio Ruminiclostridium 9 Desulfovibrionaceae Escherichia-Shigella Butyricicoccus Enterobacteriaceae Citrobacter Clostridium Pasteurellaceae Enterobacter Clostridium cl. IV Succinivibrionaceae Klebsiella Clostridium cl.XI Sutterellaceae Parasutterella Clostridium cl. XIVa Acidaminococcaceae Sutterella Coprobacillus Clostridiaceae Haemophilus Erysipelotrichaceae is. Enterococcaceae Succinivibrio Holdemania Eubacteriaceae Methanobrevibacter Turicibacter Lachnospiraceae Eubacterium Lactobacillaceae Eubacterium ventriosum Oscillospiraceae Dialister Peptostreptococcaceae Megasphaera Ruminococcaceae Veillonella Streptococcaceae Turicibacteraceae Lactobacillus Acidaminococcus Veillonellaceae Enterococcus Parvimonas Flavonifractor Phascolarctobacterium Streptococcus Megamonas Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders MDD and psychosis and schizophrenia were largely investi- altered taxa, suggesting these likely harbor transdiagnostic gated in the east, while anorexia nervosa and OCD were alterations associated with overlapping pathophysiology as investigated in the west), this analysis should be considered has previously been seen in analyses of inflammatory mark- preliminary. Clustering according to region identified several ers, neutrophil-lymphocyte ratios, and genome-wide asso- 12,56,57 taxa that were altered only in studies from Eastern coun- ciation studies. tries: Acidaminococcus (increased), Blautia (not consistent), Most consistently, the genus Eggerthella was enriched in Megamonas (decreased), Megasphaera (increased), MDD, bipolar disorder, and psychosis and schizophrenia, while Atopobium (increased), and Bacteroides (not consistent). the genera Faecalibacterium and Coprococcus were decreased These differences were driven entirely by studies from in all. Eggerthella is associated with gastrointestinal 10,58 China, highlighting the need to distinguish the Chinese inflammation, while Faecalibacterium has known anti- microbiome from other East Asian nations as more evidence inflammatory properties and is depleted in immune- 58,60 becomes available. mediated inflammatory diseases. These associations are There is evidence that psychiatric medication can affect likely mediated by short-chain fatty acid butyrate, as Faecali- 16,54 61 microbiota composition. To investigate this, we com- bacterium and Coprococcus are involved in its production, pared results from medication-free studies (n = 11) with while Eggerthella has been associated with its depletion. those in which 80% or more of patients were taking medica- Butyrate has a key role in maintaining mucosal integrity and tion (n = 21). We found that increases in the family Lacto- reducing inflammation via macrophage function and de- bacillaceae (although not member genus Lactobacillus) and crease in proinflammatory cytokines, while increasing anti- 61-63 the genera Clostridium, Klebsiella, and Megasphaera were inflammatory mediators. Further, Faecalibacterium was in- only reported in medicated groups, while Dialister was versely associated with depression severity in 2 MDD studies, 28,43,64,65 decreased in medicated and increased in medication-free 1 bipolar disorder study, and 1 anorexia nervosa study, groups. Further, 6 of 8 studies in treated patients reported suggesting depletion of this genus may be characteristic of the increases in Streptococcus, which was not reported in drug- depressive state, irrespective of diagnosis. Therefore, clinical free studies. features and underlying pathophysiology that manifest across diagnoses may be better suited to explain the observed mi- crobial alterations than distinct diagnostic categories. The mer- its of incorporating the gut microbiota as a dimensional com- Discussion ponent to the Research Domain Criteria have previously been To our knowledge, this is the first review to assess gut micro- discussed and our results reinforce this by demonstrating that biota perturbations across a spectrum of psychiatric disor- while gut microbiota abnormalities were ubiquitously ob- ders with the aim of evaluating the reproducibility and speci- served, these do not seem to congregate according to distinct ficity of potential gut microbial biomarkers. The pattern of diagnoses but instead exhibit a transdiagnostic pattern. alterations observed suggests an increased magnitude and Interestingly, the family Lactobacillaceae and member ge- complexity of microbial disorganization for some disorders nus Lactobacillus, strains from which are components of pro- compared with others. For example, the highest number of dif- biotic supplements and linked to positive health outcomes, ferentially abundant taxa was in psychosis and schizophre- were enriched in MDD, psychosis and schizophrenia, and bi- nia (136 taxa; 11 studies), despite almost twice as many stud- polar disorder. A possible explanation could be that species ies in MDD (94 taxa; 21 studies). Conversely, anorexia nervosa from this genus have differential effect. For example, 1 study was associated with fewer differences (32 taxa; 10 studies), de- identified the increase in psychosis and schizophrenia to be spite the larger number of studies compared with anxiety (36 in subspecies not typically present in the healthy gut. Alter- taxa; 2 studies) and bipolar disorder (60 taxa; 9 studies). This natively, increased Lactobacillus has previously been associ- is reminiscent of genome-wide association studies’ findings, ated with antipsychotic use. This finding was somewhat cor- in which the highest number of loci have been associated with roborated here, as 4 psychosis and schizophrenia studies psychosis and schizophrenia followed by MDD and bipolar dis- reporting increased Lactobacillus were conducted in medi- 32,34,46,70 order, and fewer have been associated with anorexia ner- cated groups, while the one that reported decreased 55 71 vosa, PTSD, and ADHD. This increased complexity, also re- Lactobacillus was in a treatment-naive group. In our explor- flected in the microbiota, is consistent with the wider spectrum atory analyses, the family Lactobacillaceae was significantly of clinical presentations associated with the former com- increased only in medicated groups. This suggests that psy- pared with the latter set of disorders. chotropic medication may be exacerbating the presence of Overall, we did not find evidence for disorder specificity: illness-associated Lactobacilli species. whenever microbial alterations merited specificity, these Measures of alpha diversity (within sample) were widely were weakly reproduced, suggesting they may instead used, following the general assumption that higher diversity reflect specific population characteristics (eg, depression is more beneficial to the host and thus expected to be subtype) and thus need further verification. Instead, our decreased in psychiatric patients, as has previously been findings indicated that certain disorders share similar pat- observed for various diseases. However, our meta-analysis terns of microbial changes. Specifically, we observed an demonstrated a nonsignificant association with diversity overlap between psychosis and schizophrenia, bipolar disor- indices and small to medium decrease in richness, suggest- der, anxiety, and MDD in consistently and inconsistently ing that while richness is somewhat compromised (although 1350 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research the clinical significance of this decrease is unclear), diver- Limitations sity is overall preserved. The high residual heterogeneity Although there were insufficient studies to perform in-depth following subgrouping according to disorder type suggests analyses of OCD, PTSD, and anxiety, we believe the inclusion that diagnosis is not a good discriminator of alpha diversity. of these disorders provides a comprehensive overview of cur- Regarding beta diversity (between samples), patients with rentevidence.Therewerenostudiesinadultswithautismspec- MDD and psychosis and schizophrenia consistently clus- trumdisorderandonly1studyinADHD,thusprecludingusfrom tered differently from controls. However, it is yet unknown comparing the association of neurodevelopmental disorders whether psychiatric disorders cluster differently from one with the microbiota in adulthood. The decision to exclude stud- another, thus questioning the suitability of diversity mea- ies in children and elderly individuals was dictated by an ap- sures as biomarkers. From the studies summarized, only 2 preciation of the specialist nature of these populations and the studies compared beta diversity cross-diagnostically and substantialage-relateddifferencesinthemicrobiota. Next,we 21,25 neither found a significant difference. acknowledge that the division into Eastern and Western coun- Among the numerous clinical and demographic factors tries is a crude approach to controlling for geographical differ- that may have contributed to the widespread inconsisten- ences in diet and genetics and does not allow detection of re- cies between studies, current evidence allowed us to gional variations in the microbiome, which might also explain explore 2 key characteristics: geographical region and psy- why we found no alterations specific to Western populations. chiatric medication. Geographical region and the associated As more studies become available, more nuanced analyses will factor of diet can profoundly affect the composition of the bepossible.Additionally,moststudieshadmodestsamplesizes, 74,75 microbiota. Our analysis suggested that some of the suggesting our analyses may still be underpowered and pre- observed perturbations may be specific to Chinese popula- liminary.Similarly,asmoststudiesincludedbothmedicatedand tions (eg, increased Acidaminococcus), others may be owing unmedicated patients, our analyses of the confounding ef- to the effect of psychiatric medication (eg, increased Klebsi- fects of medication require further verification in larger strati- ella and decreased Dialister), while third may be influenced fied populations. Our summary may also suffer from the use of by a combination of both, such as the genus Megasphaera, different reference databases between studies, as inconsisten- which was enriched only in Chinese populations undergo- cies in assigning taxonomy have been described. Finally, the ing treatment. Future studies should be encouraged to aim of this review was to evaluate gut microbial composition, report findings (even nonsignificant) on all dominant taxa rather than function. Early evidence has suggested that func- to help delineate the effect of confounders from true dis- tionalpotentialsassociatedwithpsychiatricillnessincludeshort- ease effects. Additionally, more studies will be needed in chain fatty acid synthesis, tryptophan metabolism, and neu- 52,53,83,84 currently underrepresented populations from low- to rotransmitter synthesis/degradation. Given the noted middle-income countries, as mental health problems functional redundancy, functional analysis will be key in un- become an increasing concern. derstanding the role of host-microbiome interactions in neu- For brevity, we have not discussed methodological differ- ropsychiatric disorders. ences that may have contributed to inconsistent findings such as processing, sequencing, or analysis pipelines because these 9,10,74,77,78 have been extensively reviewed by others. Further, Conclusions some have suggested that the current approaches to microbi- ome analyses may be unreliable owing to inappropriate han- This review suggests a transdiagnostic commonality of micro- dling of inherently compositional data. The lack of power cal- bial disturbances in MDD, bipolar disorder, anxiety, and culationsisasignificantdeterrentinthefield.Tomoveforward, psychosis and schizophrenia, characterized by depleted anti- the reporting of quantitative effect sizes of abundance find- inflammatory butyrate-producing bacteria and enriched pro- ings in addition to P values is needed to enable meta- inflammatory bacteria. The effect of key confounders such as analyses and the evaluation of potentially relevant biological psychiatric medication and diet should be carefully considered. effects. Even then, technical and clinical variation between Researchersshouldinterprettheirfindingswithinthelargercon- studies may make it difficult to compare effect sizes, which re- text of psychiatric disorders to prevent unmerited claims of dis- inforces the need of harmonizing methodologies and encour- orderspecificityofgutmicrobialbiomarkers.Theevidencesum- aging data sharing with sufficient metadata. marized here is a good starting point for such comparisons. ARTICLE INFORMATION Author Affiliations: Centre for Affective Disorders, Life Sciences, ZIEL–Institute for Food & Health, Institute of Psychiatry, Psychology & Neuroscience, Technical University of Munich, Freising, Germany Accepted for Publication: July 21, 2021. King’s College London, London, United Kingdom (L. J. Hall); National Institute for Health Research Published Online: September 15, 2021. (Nikolova, Cleare, Stone, Young); Department of Biomedical Research Centre at South London and doi:10.1001/jamapsychiatry.2021.2573 Psychosis Studies, Institute of Psychiatry, Maudsley NHS Foundation Trust, King’s College Correction: This article was corrected on December Psychology and Neuroscience, King’s College of London, London, United Kingdom (Cleare, Young); 1, 2021, to add indication of the open access license. London, London, United Kingdom (M. R. B. Hall); South London and Maudsley NHS Foundation Trust, Quadram Institute Bioscience, Norwich Research Bethlem Royal Hospital, Beckenham, United Open Access: This is an open access article Park, Norwich, United Kingdom (L. J. Hall); Norwich Kingdom (Cleare, Young); Brighton and Sussex distributed under the terms of the CC-BY License. Medical School, University of East Anglia, Norwich Medical School, Brighton, United Kingdom (Stone). © 2021 Nikolova VL et al. JAMA Psychiatry. Research Park, Norwich, United Kingdom (L. J. Hall); Chair of Intestinal Microbiome, School of jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1351 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders Author Contributions: Ms Nikolova had full access induces neurobehavioural changes in the rat. systematic reviews was developed. J Clin Epidemiol. to all of the data in the study and takes J Psychiatr Res. 2016;82:109-118. doi:10.1016/ 2016;69:225-234. doi:10.1016/j.jclinepi.2015.06.005 responsibility for the integrity of the data and the j.jpsychires.2016.07.019 18. Moola S, Munn Z, Tufanaru C, et al Systematic accuracy of the data analysis. 5. Zhu F, Guo R, Wang W, et al. Transplantation of reviews of etiology and risk. In: Aromataris E, Munn Concept and design: Nikolova, Hall, Cleare, Stone, microbiota from drug-free patients with Z, eds. Joanna Briggs Institute Reviewer’s Manual. The Young. schizophrenia causes schizophrenia-like abnormal Joanna Briggs Institute; 2017. Acquisition, analysis, or interpretation of data: behaviors and dysregulated kynurenine 19. Duvallet C, Gibbons SM, Gurry T, Irizarry RA, All authors. metabolism in mice. Mol Psychiatry. 2020;25(11): Alm EJ. Meta-analysis of gut microbiome studies Drafting of the manuscript: Nikolova, Stone, Young. 2905-2918. doi:10.1038/s41380-019-0475-4 identifies disease-specific and shared responses. Critical revision of the manuscript for important 6. Li N, Wang Q, Wang Y, et al. Fecal microbiota Nat Commun. 2017;8(1):1784. doi:10.1038/s41467- intellectual content: All authors. transplantation from chronic unpredictable mild 017-01973-8 Statistical analysis: Nikolova. stress mice donors affects anxiety-like and Obtained funding: Hall, Cleare, Young. 20. Balduzzi S, Rücker G, Schwarzer G. How to depression-like behavior in recipient mice via the Administrative, technical, or material support: perform a meta-analysis with R: a practical tutorial. gut microbiota-inflammation-brain axis. Stress. Smith, Young. Evid Based Ment Health. 2019;22(4):153-160. 2019;22(5):592-602. doi:10.1080/10253890.2019. Supervision: Hall, Cleare, Stone, Young. doi:10.1136/ebmental-2019-300117 Conflict of Interest Disclosures: Dr Cleare 21. Jiang HY, Pan LY, Zhang X, Zhang Z, Zhou YY, 7. Sharon G, Cruz NJ, Kang D-W, et al. Human gut reported grants from Protexin Probiotics Ruan B. Altered gut bacterial-fungal interkingdom microbiota from autism spectrum disorder promote International (industrial partner of the Medical networks in patients with current depressive behavioral symptoms in mice. Cell. 2019;177(6): Research Council studentship that Ms Nikolova is episode. Brain Behav. 2020;10(8):e01677. 1600-1618.e17. doi:10.1016/j.cell.2019.05.004 funded by) outside the submitted work; received doi:10.1002/brb3.1677 honoraria for educational activities from Lundbeck 8. Sanada K, Nakajima S, Kurokawa S, et al. Gut 22. Vinberg M, Ottesen NM, Meluken I, et al. and Janssen in the last 3 years; honoraria for microbiota and major depressive disorder: Remitted affective disorders and high familial risk of consulting from Allergan, Livanova, and Janssen; a systematic review and meta-analysis. J Affect affective disorders associate with aberrant and sponsorship for conference attendance from Disord. 2020;266:1-13. doi:10.1016/j.jad.2020.01.102 intestinal microbiota. Acta Psychiatr Scand. 2019; Janssen. Ms Nikolova reported personal fees from 9. Di Lodovico L, Mondot S, Doré J, Mack I, Hanachi 139(2):174-184. doi:10.1111/acps.12976 Janssen. Dr Stone reported grants from Protexin M, Gorwood P. Anorexia nervosa and gut 23. Mason BL, Li Q, Minhajuddin A, et al. Reduced Probiotics International, personal fees from microbiota: a systematic review and quantitative anti-inflammatory gut microbiota are associated Janssen, and grants from Takeda outside the synthesis of pooled microbiological data. Prog with depression and anhedonia. J Affect Disord. submitted work. Dr Young has received honoraria Neuropsychopharmacol Biol Psychiatry. 2021;106: 2020;266:394-401. doi:10.1016/j.jad.2020.01.137 for speaking from AstraZeneca, Lundbeck, Eli Lilly 110114. doi:10.1016/j.pnpbp.2020.110114 and Company, and Sunovion; honoraria for 24. Stevens BR, Goel R, Seungbum K, et al. 10. Simpson CA, Diaz-Arteche C, Eliby D, Schwartz consulting from Allergan, Livanova, Lundbeck, Increased human intestinal barrier permeability OS, Simmons JG, Cowan CSM. The gut microbiota in Sunovion, and Janssen; and research grants from plasma biomarkers zonulin and FABP2 correlated anxiety and depression: a systematic review. Clin Janssen, Compass, and Protexin Probiotics with plasma LPS and altered gut microbiome in Psychol Rev. 2021;83:101943. doi:10.1016/j.cpr. International in the last 3 years. No other anxiety or depression. Gut. 2018;67(8):1555-1557. 2020.101943 disclosures were reported. doi:10.1136/gutjnl-2017-314759 11. IPCS INCHEM. Environmental health criteria Funding/Support: Ms Nikolova is funded by a 25. Rong H, Xie XH, Zhao J, et al. Similarly in 222: Biomarkers in risk assessment: validity and Medical Research Council PhD Studentship. This depression, nuances of gut microbiota: evidences validation. Accessed February 8, 2021. http://www. article represents independent research partly from a shotgun metagenomics sequencing study on inchem.org/documents/ehc/ehc/ehc222.htm funded by the National Institute for Health major depressive disorder versus bipolar disorder Research Biomedical Research Centre at South 12. Yuan N, Chen Y, Xia Y, Dai J, Liu C. with current major depressive episode patients. London and Maudsley National Health Service Inflammation-related biomarkers in major J Psychiatr Res. 2019;113:90-99. doi:10.1016/ Foundation Trust and King’s College London. psychiatric disorders: a cross-disorder assessment j.jpsychires.2019.03.017 of reproducibility and specificity in 43 Role of the Funder/Sponsor: The funders had 26. Naseribafrouei A, Hestad K, Avershina E, et al. meta-analyses. Transl Psychiatry. 2019;9(1):233. no role in the design and conduct of the study; Correlation between the human fecal microbiota doi:10.1038/s41398-019-0570-y collection, management, analysis, and and depression. Neurogastroenterol Motil. 2014;26 interpretation of the data; preparation, review, or 13. Hutton B, Salanti G, Caldwell DM, et al. The (8):1155-1162. doi:10.1111/nmo.12378 approval of the manuscript; and decision to submit PRISMA extension statement for reporting of 27. Chen YH, Xue F, Yu SF, et al. Gut microbiota the manuscript for publication. systematic reviews incorporating network dysbiosis in depressed women: the association of meta-analyses of health care interventions: Disclaimer: The views expressed are those of the symptom severity and microbiota function. J Affect checklist and explanations. Ann Intern Med. 2015; authors and not necessarily those of the UK Disord. 2021;282:391-400. doi:10.1016/j.jad.2020. 162(11):777-784. doi:10.7326/M14-2385 National Health Service, the National Institute for 12.143 Health Research, or Department of Health. 14. Pollock M, Fernandes RM, Becker LA, Pieper D, 28. Liu RT, Rowan-Nash AD, Sheehan AE, et al. Hartling L. Chapter V: overviews of reviews. In: Reductions in anti-inflammatory gut bacteria are REFERENCES Higgins JPT, Thomas J, Chandler J, Cumpston M, Li associated with depression in a sample of young T, Page MJ, Welch VA, eds. Cochrane Handbook for 1. Phillips JGP. The treatment of melancholia by the adults. Brain Behav Immun. 2020;88:308-324. Systematic Reviews of Interventions, version 6.1. lactic acid bacillus. J Mental Sci. 1910;56(234):422- doi:10.1016/j.bbi.2020.03.026 Cochrane; 2020. 430. doi:10.1192/bjp.56.234.422 29. Painold A, Mörkl S, Kashofer K, et al. A step 15. Cumpston M, Changler J. Chapter IV: updating a 2. Bested AC, Logan AC, Selhub EM. Intestinal ahead: exploring the gut microbiota in inpatients review. In: Higgins JPT, Thomas J, Chandler J, microbiota, probiotics and mental health: from with bipolar disorder during a depressive episode. Cumpston M, Li T, Page MJ, Welch VA, eds. Metchnikoff to modern advances: part I: Bipolar Disord. 2019;21(1):40-49. doi:10.1111/bdi.12682 Cochrane Handbook for Systematic Reviews of autointoxication revisited. Gut Pathog. 2013;5(1):5. 30. Coello K, Hansen TH, Sørensen N, et al. Gut Interventions, version 6.1. Cochrane; 2020. doi:10.1186/1757-4749-5-5 microbiota composition in patients with newly 16. Ait Chait Y, Mottawea W, Tompkins TA, 3. Zheng P, Zeng B, Zhou C, et al. Gut microbiome diagnosed bipolar disorder and their unaffected Hammami R. Unravelling the antimicrobial action of remodeling induces depressive-like behaviors first-degree relatives. Brain Behav Immun. 2019;75: antidepressants on gut commensal microbes. Sci Rep. through a pathway mediated by the host’s 112-118. doi:10.1016/j.bbi.2018.09.026 2020;10(1):17878. doi:10.1038/s41598-020- metabolism. Mol Psychiatry. 2016;21(6):786-796. 31. Hu S, Li A, Huang T, et al. Gut microbiota 74934-9 doi:10.1038/mp.2016.44 changes in patients with bipolar depression. Adv Sci 17. Whiting P, Savović J, Higgins JPT, et al; ROBIS 4. Kelly JR, Borre Y, O’ Brien C, et al. Transferring (Weinh). 2019;6(14):1900752. doi:10.1002/advs. group. ROBIS: a new tool to assess risk of bias in the blues: depression-associated gut microbiota 201900752 1352 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Perturbations in Gut Microbiota Composition in Psychiatric Disorders Original Investigation Research 32. Shen Y, Xu J, Li Z, et al. Analysis of gut 46. Ma X, Asif H, Dai L, et al. Alteration of the gut common dysbiosis exist? Microbiome. 2018;6(1):221. microbiota diversity and auxiliary diagnosis as a microbiome in first-episode drug-naïve and chronic doi:10.1186/s40168-018-0603-4 biomarker in patients with schizophrenia: medicated schizophrenia correlate with regional 59. Qiu X, Zhang M, Yang X, Hong N, Yu C. a cross-sectional study. Schizophr Res. 2018;197: brain volumes. J Psychiatr Res. 2020;123:136-144. Faecalibacterium prausnitzii upregulates regulatory 470-477. doi:10.1016/j.schres.2018.01.002 doi:10.1016/j.jpsychires.2020.02.005 T cells and anti-inflammatory cytokines in treating 33. Pan R, Zhang X, Gao J, Yi W, Wei Q, Su H. 47. Xu R, Wu B, Liang J, et al. Altered gut TNBS-induced colitis. J Crohns Colitis. 2013;7(11): Analysis of the diversity of intestinal microbiome microbiota and mucosal immunity in patients with e558-e568. doi:10.1016/j.crohns.2013.04.002 and its potential value as a biomarker in patients schizophrenia. Brain Behav Immun. 2020;85:120-127. 60. Sokol H, Seksik P, Furet JP, et al. Low counts of with schizophrenia: a cohort study. Psychiatry Res. doi:10.1016/j.bbi.2019.06.039 Faecalibacterium prausnitzii in colitis microbiota. 2020;291:113260. doi:10.1016/j.psychres.2020. 48. Hanachi M, Manichanh C, Schoenenberger A, Inflamm Bowel Dis. 2009;15(8):1183-1189. et al. Altered host-gut microbes symbiosis in doi:10.1002/ibd.20903 34. Li S, Zhuo M, Huang X, et al. Altered gut severely malnourished anorexia nervosa (AN) 61. Miquel S, Martín R, Bridonneau C, et al. Ecology microbiota associated with symptom severity in patients undergoing enteral nutrition: an and metabolism of the beneficial intestinal schizophrenia. PeerJ. 2020;8:e9574. doi:10.7717/ explicative factor of functional intestinal disorders? commensal bacterium Faecalibacterium prausnitzii. peerj.9574 Clin Nutr. 2019;38(5):2304-2310. doi:10.1016/ Gut Microbes. 2014;5(2):146-151. doi:10.4161/gmic. j.clnu.2018.10.004 35. Zhang X, Pan L-Y, Zhang Z, Zhou Y-Y, Jiang H-Y, 27651 Ruan B. Analysis of gut mycobiota in first-episode, 49. Monteleone AM, Troisi J, Fasano A, et al. 62. Chang PV, Hao L, Offermanns S, Medzhitov R. drug-naïve Chinese patients with schizophrenia: Multi-omics data integration in anorexia nervosa The microbial metabolite butyrate regulates a pilot study. Behav Brain Res. 2020;379:112374. patients before and after weight regain: intestinal macrophage function via histone doi:10.1016/j.bbr.2019.112374 A microbiome-metabolomics investigation. Clin Nutr. deacetylase inhibition. Proc Natl Acad SciUSA. 2021;40(3):1137-1146. doi:10.1016/j.clnu.2020.07. 36. Jiang H-Y, Zhang X, Yu Z-H, et al. Altered gut 2014;111(6):2247-2252. doi:10.1073/pnas.1322269111 microbiota profile in patients with generalized 63. Sublette ME, Cheung S, Lieberman E, et al. anxiety disorder. J Psychiatr Res. 2018;104:130-136. 50. Liu Y, Zhang L, Wang X, et al. Similar fecal Bipolar disorder and the gut microbiome: doi:10.1016/j.jpsychires.2018.07.007 microbiota signatures in patients with a systematic review. Bipolar Disord. Published online diarrhea-predominant irritable bowel syndrome 37. Kleiman SC, Watson HJ, Bulik-Sullivan EC, et al. January 29, 2021. doi:10.1111/bdi.13049 and patients with depression. Clin Gastroenterol The intestinal microbiota in acute anorexia nervosa 64. Evans SJ, Bassis CM, Hein R, et al. The gut Hepatol. 2016;14(11):1602-1611.e5. doi:10.1016/j.cgh. and during renourishment: relationship to microbiome composition associates with bipolar 2016.05.033 depression, anxiety, and eating disorder disorder and illness severity. J Psychiatr Res.2017; psychopathology. Psychosom Med. 2015;77(9): 51. Lai W-T, Deng W-F, Xu S-X, et al. Shotgun 87:23-29. doi:10.1016/j.jpsychires.2016.12.007 969-981. doi:10.1097/PSY.0000000000000247 metagenomics reveals both taxonomic and 65. Chen Y-H, Bai J, Wu D, et al. Association tryptophan pathway differences of gut microbiota 38. Mack I, Cuntz U, Grämer C, et al. Weight gain in between fecal microbiota and generalized anxiety in major depressive disorder patients. Psychol Med. anorexia nervosa does not ameliorate the faecal disorder: severity and early treatment response. 2021:51(1):90-101. doi:10.1017/S0033291719003027 microbiota, branched chain fatty acid profiles, and J Affect Disord. 2019;259:56-66. doi:10.1016/j.jad. gastrointestinal complaints. Sci Rep. 2016;6:26752. 52. Lai WT, Zhao J, Xu SX, et al. Shotgun 2019.08.014 doi:10.1038/srep26752 metagenomics reveals both taxonomic and 66. Insel T, Cuthbert B, Garvey M, et al. Research tryptophan pathway differences of gut microbiota 39. Hemmings SMJ, Malan-Müller S, van den domain criteria (RDoC): toward a new classification in bipolar disorder with current major depressive Heuvel LL, et al. The microbiome in posttraumatic framework for research on mental disorders. Am J episode patients. J Affect Disord. 2021;278:311-319. stress disorder and trauma-exposed controls: an Psychiatry. 2010;167(7):748-751. doi:10.1176/ doi:10.1016/j.jad.2020.09.010 exploratory study. Psychosom Med. 2017;79(8): appi.ajp.2010.09091379 936-946. doi:10.1097/PSY.0000000000000512 53. Zhu F, Ju Y, Wang W, et al. Metagenome-wide 67. Kelly JR, Clarke G, Cryan JF, Dinan TG. association of gut microbiome features for 40. Domènech L, Willis J, Alemany M, et al. Dimensional thinking in psychiatry in the era of the schizophrenia. Nat Commun. 2020;11(1):1612. Changes in the stool and oropharyngeal Research Domain Criteria (RDoC). Ir J Psychol Med. doi:10.1038/s41467-020-15457-9 microbiome in obsessive-compulsive disorder. 2018;35(2):89-94. doi:10.1017/ipm.2017.7 medRxiv. Preprint posted online May 28, 2020. 54. Macedo D, Filho AJMC, Soares de Sousa CN, 68. Yong SJ, Tong T, Chew J, Lim WL. doi:10.1101/2020.05.26.20113779 et al. Antidepressants, antimicrobials or both? gut Antidepressive mechanisms of probiotics and their microbiota dysbiosis in depression and possible 41. Turna J, Grosman Kaplan K, Anglin R, et al. The therapeutic potential. Front Neurosci. 2020;13:1361. implications of the antimicrobial effects of gut microbiome and inflammation in doi:10.3389/fnins.2019.01361 antidepressant drugs for antidepressant obsessive-compulsive disorder patients compared effectiveness. J Affect Disord. 2017;208:22-32. 69. Bahr SM, Tyler BC, Wooldridge N, et al. Use of to age- and sex-matched controls: a pilot study. doi:10.1016/j.jad.2016.09.012 the second-generation antipsychotic, risperidone, Acta Psychiatr Scand. 2020;142(4):337-347. and secondary weight gain are associated with an doi:10.1111/acps.13175 55. Ikeda M, Saito T, Kanazawa T, Iwata N. altered gut microbiota in children. Transl Psychiatry. Polygenic risk score as clinical utility in psychiatry: 42. Aarts E, Ederveen THA, Naaijen J, et al. Gut 2015;5(10):e652-e652. doi:10.1038/tp.2015.135 a clinical viewpoint. J Hum Genet. 2021;66(1):53-60. microbiome in ADHD and its relation to neural doi:10.1038/s10038-020-0814-y 70. Schwarz E, Maukonen J, Hyytiäinen T, et al. reward anticipation. PLoS One. 2017;12(9):e0183509. Analysis of microbiota in first episode psychosis doi:10.1371/journal.pone.0183509 56. Lee SH, Ripke S, Neale BM, et al; identifies preliminary associations with symptom Cross-Disorder Group of the Psychiatric Genomics 43. Jiang H, Ling Z, Zhang Y, et al. Altered fecal severity and treatment response. Schizophr Res. Consortium; International Inflammatory Bowel microbiota composition in patients with major 2018;192:398-403. doi:10.1016/j.schres.2017.04.017 Disease Genetics Consortium (IIBDGC). Genetic depressive disorder. Brain Behav Immun. 2015;48: relationship between five psychiatric disorders 71. Yuan X, Zhang P, Wang Y, et al. Changes in 186-194. doi:10.1016/j.bbi.2015.03.016 estimated from genome-wide SNPs. Nat Genet. metabolism and microbiota after 24-week 44. Huang Y, Shi X, Li Z, et al. Possible association 2013;45(9):984-994. doi:10.1038/ng.2711 risperidone treatment in drug naïve, normal weight of Firmicutes in the gut microbiota of patients with patients with first episode schizophrenia. Schizophr 57. Brinn A, Stone J. Neutrophil-lymphocyte ratio major depressive disorder. Neuropsychiatr Dis Treat. Res. 2018;201:299-306. doi:10.1016/j.schres.2018. across psychiatric diagnoses: a cross-sectional 2018;14:3329-3337. doi:10.2147/NDT.S188340 05.017 study using electronic health records. BMJ Open. 45. Zheng P, Zeng B, Liu M, et al. The gut 2020;10(7):e036859. doi:10.1136/bmjopen-2020- 72. Shade A. Diversity is the question, not the microbiome from patients with schizophrenia 036859 answer. ISME J. 2017;11(1):1-6. doi:10.1038/ismej. modulates the glutamate-glutamine-GABA cycle 2016.118 58. Forbes JD, Chen C-Y, Knox NC, et al. and schizophrenia-relevant behaviors in mice. Sci Adv. A comparative study of the gut microbiota in 73. Gerritsen J, Smidt H, Rijkers GT, de Vos WM. 2019;5(2):eaau8317. doi:10.1126/sciadv.aau8317 immune-mediated inflammatory diseases-does a Intestinal microbiota in human health and disease: jamapsychiatry.com (Reprinted) JAMA Psychiatry December 2021 Volume 78, Number 12 1353 Research Original Investigation Perturbations in Gut Microbiota Composition in Psychiatric Disorders the impact of probiotics. Genes Nutr. 2011;6(3): 78. Bastiaanssen TFS, Cowan CSM, Claesson MJ, 82. Balvočiūtė M, Huson DH. SILVA, RDP, 209-240. doi:10.1007/s12263-011-0229-7 Dinan TG, Cryan JF. Making sense of…the Greengenes, NCBI and OTT: how do these microbiome in psychiatry. Int J taxonomies compare? BMC Genomics. 2017;18(2) 74. Cheung SG, Goldenthal AR, Uhlemann A-C, Neuropsychopharmacol. 2019;22(1):37-52. (suppl 2):114. doi:10.1186/s12864-017-3501-4 Mann JJ, Miller JM, Sublette ME. Systematic review doi:10.1093/ijnp/pyy067 of gut microbiota and major depression. Front 83. Valles-Colomer M, Falony G, Darzi Y, et al. The Psychiatry. 2019;10:34. doi:10.3389/fpsyt.2019. 79. Gloor GB, Macklaim JM, Pawlowsky-Glahn V, neuroactive potential of the human gut microbiota 00034 Egozcue JJ. Microbiome datasets are in quality of life and depression. Nat Microbiol. compositional: and this is not optional. Front 2019;4(4):623-632. doi:10.1038/s41564-018- 75. Simpson HL, Campbell BJ. Review article: Microbiol. 2017;8:2224. doi:10.3389/fmicb.2017. 0337-x dietary fibre-microbiota interactions. Aliment Pharmacol Ther. 2015;42(2):158-179. doi:10.1111/apt. 84. Yang J, Zheng P, Li Y, et al. Landscapes of 13248 80. Debelius J, Song SJ, Vazquez-Baeza Y, Xu ZZ, bacterial and metabolic signatures and their Gonzalez A, Knight R. Tiny microbes, enormous interaction in major depressive disorders. Sci Adv. 76. Sankoh O, Sevalie S, Weston M. Mental health impacts: what matters in gut microbiome studies? 2020;6(49):eaba8555. doi:10.1126/sciadv.aba8555 in Africa. Lancet Glob Health. 2018;6(9):e954-e955. Genome Biol. 2016;17(1):217. doi:10.1186/s13059- doi:10.1016/S2214-109X(18)30303-6 85. Heintz-Buschart A, Wilmes P. Human gut 016-1086-x microbiome: function matters. Trends Microbiol. 77. Vindegaard N, Speyer H, Nordentoft M, 81. Odamaki T, Kato K, Sugahara H, et al. 2018;26(7):563-574. doi:10.1016/j.tim.2017.11.002 Rasmussen S, Benros ME. Gut microbial changes of Age-related changes in gut microbiota composition patients with psychotic and affective disorders: from newborn to centenarian: a cross-sectional a systematic review. Schizophr Res. 2020;S0920- study. BMC Microbiol. 2016;16(1):90. doi:10.1186/ 9964(19)30584-5. doi:10.1016/j.schres.2019.12.014 s12866-016-0708-5 1354 JAMA Psychiatry December 2021 Volume 78, Number 12 (Reprinted) jamapsychiatry.com Supplemental Online Content Nikolova VL, Smith MRB, Hall LJ, Cleare AJ, Stone JM, Young AH. Perturbations in gut microbiota composition in psychiatric disorders: a systematic review and meta-analysis. JAMA Psychiatry. Published online September 15, 2021. doi:10.1001/jamapsychiatry.2021.2573 eAppendix 1. Systematic search details eAppendix 2. Quality assessment eAppendix 3. Detailed methods of the meta-analysis performed eAppendix 4. PRISMA flowcharts for the umbrella review search and the updated review searches eAppendix 5. Details of the identified systematic reviews eAppendix 6. Detailed characteristics of the included studies eAppendix 7. Stool sample processing methods in the included studies eAppendix 8. Publication bias assessment for the alpha diversity meta-analyses eAppendix 9. Beta diversity eAppendix 10. Figures for study-level findings of relative abundance of microbial taxa This supplemental material has been provided by the authors to give readers additional information about their work. © 2021 American Medical Association. All rights reserved. eAppendix 1. Systematic search details 1. Systematic reviews and meta-analysis search – performed on 27 Jan 2021 1.1. Cochrane Library Search Hits Cochrane Reviews 8498 Topic Mental Health 637 AND gut OR gastrointestinal OR intestinal OR feacal OR fecal OR stool 56 AND microbiome OR microbiota OR ecosystem OR bacteria OR flora OR microflora OR dysbiosis 0 1.2. PubMed* Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 3699 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((depression) OR (depress*) OR (mdd) OR (trd) OR (bipolar) OR (mania) OR (bipolar depression) OR (anxiety) OR (psychosis) OR (schizophrenia) OR (obsessive compulsive disorder) OR (ocd) OR (ptsd) OR (post-traumatic stress disorder) OR (adhd) OR (attention deficit hyperactivity disorder) OR (autism) OR (autism spectrum disorder) OR (ASD) OR (eating disorder) OR (anorexia) OR (bulimia)) AND (systematicreview[Filter]) 80 Limits: 2005-current 79 Limits: human 47 Limits: English 45 1.3. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138475 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612345 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370239 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87064 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60712 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42743 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68233 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111701 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464415 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56856 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14943 © 2021 American Medical Association. All rights reserved. 14. depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1041348 15. depress*.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1193818 16. MDD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 32383 17. TRD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 3628 18. bipolar.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 159279 19. mania.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42988 20. bipolar depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 10336 21. anxiety.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 626469 22. psychosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 186882 23. schizophrenia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 341812 24. obsessive compulsive disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 50116 mh] 25. OCD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 26104 26. PTSD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 72988 27. post-traumatic stress disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 28176 mh] 28. autism.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 134153 29. Autism spectrum disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42153 30. pervasive developmental disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, 4354 tm, mh] 31. attention deficit hyperactivity disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, 61349 id, tm, mh] 32. ADHD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 69132 33. eating disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 49763 34. anorexia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108326 35. bulimia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 28823 36. systematic review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 386059 37. meta-analysis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 338776 38. 1 or 2 or 3 or 4 or 5 or 6 1066368 39. 7 or 8 or 9 or 10 or 11 or 12 or 13 693756 40. 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 2248227 30 or 31 or 32 or 33 or 34 or 35 41. 36 or 37 557086 42. 38 AND 39 AND 40 AND 41 252 43. limit 42 to english language 241 44. limit 43 to human 236 45. limit 44 to yr="2005 -Current" 234 © 2021 American Medical Association. All rights reserved. 2. Supplementary searches – all performed on 02 Feb 2021 2.1. PTSD & OCD – no date restriction as no systematic reviews were available 2.1.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 45 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((obsessive compulsive disorder) OR (ocd) OR (ptsd) OR (post-traumatic stress disorder) Limits: human, English 20 2.1.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. obsessive compulsive disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 50157 mh] 15. OCD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 26128 16. PTSD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 73062 17. post-traumatic stress disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, 28199 mh] 18. 1 or 2 or 3 or 4 or 5 or 6 1067443 19. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 20. 14 or 15 or 16 or 17 133639 21. 18 AND 19 AND 20 98 22. limit 42 to english language 95 23. limit 43 to human 66 2.2. Anxiety & Depression - last review search of both disorders done in March 2020 (Simpson et al., 2021) 2.2.1. Pubmed © 2021 American Medical Association. All rights reserved. Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 1714 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((depression) OR (depress*) OR (mdd) OR (trd) OR (anxiety)) AND NOT (review) Limits: human, English, start date: February 2020 42 2.2.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1047721 15. depress*.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 1200820 16. MDD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 32605 17. TRD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 3663 18. anxiety.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 630521 19. 1 or 2 or 3 or 4 or 5 or 6 1067443 20. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 21. 14 or 15 or 16 or 17 or 18 1511690 22. 19 AND 20 AND 21 2774 23. limit 22 to english language 2687 24. limit 23 to human 1894 25. limit 24 to yr="2020 -Current" 509 26. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 27. 25 NOT 26 278 © 2021 American Medical Association. All rights reserved. 2.3. Bipolar Disorder & Psychosis/Schizophrenia - last review search of both disorders done on 17 Jan 2019 (Vindegaard et al., 2020) 2.3.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 199 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((bipolar) OR (mania) OR (bipolar depression) OR (psychosis) OR (schizophrenia)) AND NOT (review) Limits: human, English, start date: January 2019 84 2.3.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. bipolar.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 160089 15. mania.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 43186 16. bipolar depression.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 10387 17. psychosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 187899 18. schizophrenia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 343421 19. 1 or 2 or 3 or 4 or 5 or 6 1067443 20. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 21. 14 or 15 or 16 or 17 or 18 577348 22. 19 AND 20 AND 21 661 23. limit 22 to 6English language 631 24. limit 23 to human 551 25. limit 24 to yr="2019 -Current" 291 26. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 © 2021 American Medical Association. All rights reserved. 27. 25 NOT 26 148 2.4. ADHD & ASD: last review search of both disorders done on 31Aug2018 (Jurek et al., 2020). 2.4.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 415 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((adhd) OR (attention deficit hyperactivity disorder) OR (autism) OR (autism spectrum disorder) OR (ASD)) AND NOT (review) Limits: human, English, start date: August 2018 20 2.4.2. Embase & PsychINFO (via Ovid) Search Hits 1. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 2. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 3. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 4. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 5. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 6. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 7. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 8. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 9. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 10. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 11. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 12. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 13. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 14. attention deficit hyperactivity disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, 61699 id, tm, mh] 15. adhd.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 69524 16. autism.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 135286 17. autism spectrum disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42650 18. ASD.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56686 19. 1 or 2 or 3 or 4 or 5 or 6 1067443 20. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 21. 14 or 15 or 16 or 17 or 18 221434 22. 19 AND 20 AND 21 1209 23. limit 22 to english language 1159 © 2021 American Medical Association. All rights reserved. 24. limit 23 to human 962 25. limit 24 to yr="2018 -Current" 531 26. adult. mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 8881500 27. 25 AND 26 63 28. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 29. 27 NOT 28 46 2.5. Eating disorders: last review search done in June 2020 (Di Ludovico et al., 2021). 2.5.1. PubMed Search Hits ((gut) OR (gastrointestinal) OR (intestinal) OR (feacal) OR (fecal) OR (stool)) 489 AND ((microbiome) OR (microbiota) OR (ecosystem) OR (bacteria) OR (flora) OR (microflora) OR (dysbiosis)) AND ((eating disorder) OR (anorexia) OR (bulimia)) AND NOT (review) Limits: human, English, start date: May 2020 13 2.5.2. Embase & PsychINFO (via Ovid) Search Hits 30. gut.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 138693 31. gastrointestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 612886 32. intestinal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 370633 33. feacal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 167 34. fecal.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 87197 35. stool.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 60767 36. microbiome.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 42848 37. microbiota.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 68365 38. ecosystem.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 111818 39. bacteria.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 464960 40. flora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108206 41. microflora.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 56939 42. dysbiosis.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 14988 43. eating disorder.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 50029 44. anorexia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 108993 45. bulimia.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 28918 46. 1 or 2 or 3 or 4 or 5 or 6 1067443 47. 7 or 8 or 9 or 10 or 11 or 12 or 13 694634 48. 14 or 15 or 16 152111 © 2021 American Medical Association. All rights reserved. 49. 17 AND 18 AND 19 550 50. limit 20 to english language 512 51. limit 21 to human 392 52. limit 22 to yr="2020 -Current" 83 53. review.mp. [mp=ti, ab, hw, tn, ot, dm, mf, dv, kw, fx, dq, tc, id, tm, mh] 4370959 54. 23 NOT 24 37 © 2021 American Medical Association. All rights reserved. eAppendix 2. Quality assessment Two authors (VLN and MRBS) performed the rating independently and resolved discrepancies via discussion. Summary: The overall risk of bias in the systematic reviews was low (12/16) (Table S5.1), however, only 3/16 had a pre-registered protocol (Table S3.1). The most frequent concerns were the lack of clarity how evidence was collected and evaluated for quality and the thoroughness of the search strategy (Figure S5.1). However, due to the significant overlap, a recent high-quality review was available for each disorder. Regarding original studies published after the reviews, quality was high as assessed with the JBI tool for case-control studies. The primary concern was incomplete consideration of confounders with 7/20 either not identifying these and/or not accounting for them in the analyses. We did not penalize studies that considered only some factors (e.g. age, gender, psychiatric medication) as other, particularly lifestyle factors such as diet, are difficult to fully control. Table e2.1. Quality assessment of the included systematic reviews using the ROBIS tool. Review Phase 1 Phase 2 Disorder First Author Year 1. Study 2. 3. Data 4. RISK OF eligibility Identificatio collection Synthesis BIAS criteria n and and study and IN THE selection of appraisal findings REVIEW* studies low low low low low AN Di Lodovico 2021 low high unclear high low AN Schalla 2019 unclear high low low low AN Schwensen 2018 low low low low low MDD, ANX Simpson 2021 high high unclear unclear high MDD, SCZ Fond 2020 unclear high unclear high high MDD Li 2020 low low low low low MDD, ANX Simpson 2020 low low low low low MDD Sanada 2020 MDD, BD, low low low low low Vindegaard 2020 SCZ low high high unclear unclear MDD Cheung 2019 low low low high low MDD Barandouzi 2020 low low unclear low low BD, SCZ Nguyen 2019 high unclear unclear low low BD, SCZ Nguyen 2018 high high high high high SCZ Cuomo 2018 low low low low low SCZ Kraeuter 2020 low low low low low ASD, ADHD Jurek 2020 Low = low risk of bias, high = high risk of bias, unclear = insufficient information to assess risk of bias; *this criterion is scored by answering the following questions: A. Did the interpretation of findings address all of the concerns identified the Phase 2 assessment?; B. Was the relevance of identified studies to the review's research question appropriately considered?; and C. Did the reviewers avoid emphasizing results on the basis of their statistical significance?. © 2021 American Medical Association. All rights reserved. Table e2.2. Quality assessment of the included original studies published after the systematic reviews using the Joanna Briggs Institute Critical Appraisal Checklist for Case Control Studies. © 2021 American Medical Association. All rights reserved. RISK OF BIAS IN THE REVIEW High Low 4. Synthesis and findings Unclear 3. Data collection and study appraisal 2. Identification and selection of studies 1. Study eligibility criteria 0% 20% 40% 60% 80% 100% Figure e5.1. Quality assessment of the included systematic reviews using the ROBIS tool. © 2021 American Medical Association. All rights reserved. eAppendix 3: Detailed methods of the meta-analysis performed Medians and inter-quartile ranges were transformed to means (M) and standard deviations (SD) using a web-based tool (http://www.math.hkbu.edu.hk/~tongt/papers/median2mean.html). For significantly skewed data, an alternative validated procedure was followed . Where necessary, numerical data were extracted from graphs using WebPlotDigitizer (v.4.4 ) and Adobe Acrobat's inbuilt measuring tool (Adobe Systems, California, USA), as previously done by others . A random-effects meta-analysis on Hedge’s g standardised mean difference (SMD) was performed applying the inverse-variance method. Effect size was categorized as small (SMD 0.2), moderate (SMD=0.5), or large (SMD=0.8). Inter-study heterogeneity was quantified using the DerSimonian–Laird estimator, reported with the I2 statistic and interpreted according to convention (25% - low, 50% - moderate, and 75% - high) . Publication bias was evaluated with funnel plots and Egger's regression test. Pre-planned subgroup analyses were disorder, region of study (east/west) and use of psychiatric medication. All analyses were completed with the meta package (v.4.17-0 ) in R. References: 1. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol. 2014;14. doi:10.1186/1471-2288-14-135 2. Rohatgi A. WebPlotDigitizer.; 2020. https://automeris.io/WebPlotDigitizer 3. Safadi JM, Quinton AMG, Lennox BR, Burnet PWJ, Minichino A. Gut dysbiosis in severe mental illness and chronic fatigue: a novel trans-diagnostic construct? A systematic review and meta-analysis. Molecular Psychiatry. Published online February 8, 2021:1- 13. doi:10.1038/s41380-021-01032-1 4. Higgins DJPT. Cochrane Handbook for Systematic Reviews of Interventions. John Wiley & Sons; 2008. Accessed December 17, 2020. 5. Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-analysis with R: a practical tutorial. Evid Based Ment Health. 2019;22(4):153-160. doi:10.1136/ebmental-2019- © 2021 American Medical Association. All rights reserved. eAppendix 4: PRISMA flowcharts for the umbrella review search and the updated review searches Records identified through Additional records identified through database searching other sources (n = 279) (n = 0) Records after duplicates removed (n = 228) Records excluded (n = 196), reasons: Clearly irrelevant (n = 136) Records screened Review of intervention (diet, (n = 228) probiotics, etc.) (n = 21) Not systematic review (n = 13) Gut microbiota not assessed (n = 8) Not disorder of interest (n = 7) Poster/Abstract (n = 8) Not adult (n = 2) Full-text articles assessed for eligibility (n = 32) Full-text articles excluded (n = 16), reasons: Not adult (n = 8) gut microbiota not assessed (n = 5) Eligible reviews and meta- Not disorder of interest (n = 2) analyses (n = 16) Not systematic review (n = 1) Eligible original cross- sectional studies (n = 39) Figure e4.1. PRISMA flowchart of the umbrella review search © 2021 American Medical Association. All rights reserved. Included Eligibility Screening Identification Table e4.1. PRISMA charts for the updated systematic review searches. OCD & MDD & ANX BD & SCZ ASD & ADHD Eating disorders PTSD Records 86 293 209 66 50 identified Reviewed 86 293 209 66 50 Excluded at 83 282 193 62 47 Title/Abstract Excluded at full 0 5 (n=1 gut 6 (n=2 gut 4 (n=2 non- 2 (n=1 secondary text (with microbiota not microbiota not adult, n=1 non- analysis from reasons) assessed; n=2 assessed, n=1 human, n=1 gut included study, outcome of interest outcome of microbiota not n=1 gut not reported; n=2 interest not assessed) microbiota not no control group) assessed; n=3 analysed) conference abstract) Included 3 (2 OCD, 6 (5 MDD, 1 MDD + 10 (6 SCZ, 4 BD) 0 1 (AN) 1 PTSD) BD) © 2021 American Medical Association. All rights reserved. eAppendix 5. Details of the identified systematic reviews Table e5.1. Details of the identified systematic reviews. Pre- First Author Year Ref. Disorder Studies included (eligible only)* registered? Di Lodovico 2021 [1] anorexia N = 9 no Armougoum 2009; Million 2013; Kleiman 2015; Morita 2015; Mack 2016; Mörkl 2017; Borgo 2017; Hanachi 2018; Hata 2019 Schalla 2019 [2] anorexia N = 5 no Morita et al., 2015; Mörkl et al., 2017; Kleiman et al., 2015; Borgo et al., 2017; Mack et al., 2016 Schwensen 2018 [3] anorexia N = 7 Yes Mörkl et al., 2017; Borgo et al 2017; Kleiman et al 2015; Morita et al 2015; Amougom et al 2009; Million et al 2013; Mack et al 2016 Simpson 2021 [4] depression N = 18 no Aizawa et al (2016); Chahwan et al. (2019); Chen, Li et al. (2018); Chen, Zheng et al. (2018); Chen et al. (2019); Chung et al. (2019); Huang et al. (2018); Jiang et al. (2015); Kelly et al. (2016); Lai et al., (2019); Lin et al. (2017); Liu et al. (2016); Mason et al., (2020); Naseribafrouei et al. (2014); Rong et al. (2019); Valles- Colomer et al. (2019); Vinberg et al. (2019); Zheng et al. (2016) anxiety N = 3 Mason et al., (2020); Jiang et al. (2018); Chen et al. (2019) Fond 2020 [5] depression N = 7 no Chen et al. 2018; Kelly et al. 2016; Liu et al. 2016; Jiang et al. 2015; Madan et al. 2020; Mason et al. 2020; Naseribafrouei et al., 2014 schizophrenia N = 0 Li 2020 [6] depression N= 9 no Naseribafrouei et al., 2014; Jiang et al., 2015; Kelly et al., 2016; Zheng et al., 2016; Huang et al., 2018; Chen et al., 2018b; Chung et al., 2019; Chen et al., 2018a; Rong et al., 2019 Simpson 2020 [7] depression N = 2 no Liu et al. (2016), Aizawa et al. (2016) Sanada 2020 [8] depression N = 10 no Chen 2018a; Chen 2018b; Huang 2018; Lin 2017; Aizawa 2016; Kelly 2016; Liu 2016; Zheng 2016; Jiang 2015; Naseribafrouei 2014 © 2021 American Medical Association. All rights reserved. Vindegaard 2020 [9] depression N = 9 yes Chen et al., 2018; Huang et al., 2018; Stevens et al., 2018; Lin et al., 2017; Aizawa et al., 2016; Kelly et al., 2016; Zheng et al., 2016; Jiang et al., 2015; Naseribafrouei et al., 2014 bipolar N = 4 disorder Coello et al., 2019; Aizawa et al., 2018; Painold et al., 2018; Evans et al., 2017 psychosis N = 4 Nguyen et al., 2018; Schwarz et al., 2018; Shen et al., 2018; Yuan et al., 2018 Cheung 2019 [10] depression N = 6 no Naseribafrouei et al 2014; Jiang et al 2015; Aizawa et al 2016; Zheng et al 2016; Lin et al 2017; Chen et al 2018 Barandouzi 2020 [11] depression N = 9 no Chen et al 2018; Zheng et al 2016; Liu et al 2016; Chen et al 2018; Jiang et al 2015; Naserbafrouei et al 2014; Lin et al 2017; Aizawa et al 2016; Kelly et al 2016; Nguyen 2019 [12] bipolar N = 4 no disorder Painold et al., 2019; Evans et al 2017; Coello et al., 2019; Aizawa et al., 2018 psychosis N = 5 Schwarz et al., 2018; Yuan et al., 2018; Nagamine et al 2018; Nguyen et al 2019; Shen et al 2018 Nguyen 2018 [13] bipolar N = 1 no disorder Evans et al 2017 psychosis N = 1 Schwarz et al. (2017) Cuomo 2018 [14] psychosis N = 3 no Schwarz et al. (2018); Shen et al (2018); Yuan et al (2018) Kraeuter 2020 [15] psychosis N = 5 no Schwarz et al 2018; Shen et al 2018; Yuan et al 2018; Nguyen et al 2018; Zheng 2019 Jurek 2020 [16] autism, N = 1 yes ADHD Aarts et al. 2017 *after duplicate studies from these systematic reviews were removed, the total number of studies added through this route was 39. eReferences: 1. Di Lodovico L, Mondot S, Doré J, Mack I, Hanachi M, Gorwood P. Anorexia nervosa and gut microbiota: A systematic review and quantitative synthesis of pooled microbiological data. Prog Neuropsychopharmacol Biol Psychiatry. 2020 Sep 22;110114. 2. Schalla MA, Stengel A. Gastrointestinal alterations in anorexia nervosa - A systematic review. Eur Eat Disord Rev. 2019;27(5):447–61. 3. Schwensen HF, Kan C, Treasure J, Høiby N, Sjögren M. A systematic review of studies on the faecal microbiota in anorexia nervosa: future research may need to include microbiota from the small intestine. Eat Weight Disord. 2018 Aug;23(4):399–418. © 2021 American Medical Association. All rights reserved. 4. Simpson CA, Mu A, Haslam N, Schwartz OS, Simmons JG. Feeling down? A systematic review of the gut microbiota in anxiety/depression and irritable bowel syndrome. Journal of Affective Disorders [Internet]. 2020 Apr [cited 2020 Jun 2];266:429–46. Available from: https://linkinghub.elsevier.com/retrieve/pii/S0165032719325777 5. Fond GB, Lagier JC, Honore S, Lancon C, Korchia T, De Verville PLS, Llorca PM, Auquier P, Guedj E, Boyer L. Microbiota-orientated treatments for major depression and schizophrenia. Nutrients [Internet]. 2020;12(4). 6. Li S, Hua D, Wang Q, Yang L, Wang X, Luo A, Yang C. The Role of Bacteria and Its Derived Metabolites in Chronic Pain and Depression: Recent Findings and Research Progress. Int J Neuropsychopharmacol. 2020 10;23(1):26–41. 7. Simpson CA, Diaz-Arteche C, Eliby D, Schwartz OS, Simmons JG, Cowan CSM. The gut microbiota in anxiety and depression - A systematic review. Clin Psychol Rev. 2020 Oct 29;83:101943. 8. Sanada K, Nakajima S, Kurokawa S, Barceló-Soler A, Ikuse D, Hirata A, Yoshizawa A, Tomizawa Y, Salas-Valero M, Noda Y, Mimura M, Iwanami A, Kishimoto T. Gut microbiota and major depressive disorder: A systematic review and meta-analysis. J Affect Disord. 2020 01;266:1–13. 9. Vindegaard N, Speyer H, Nordentoft M, Rasmussen S, Benros ME. Gut microbial changes of patients with psychotic and affective disorders: A systematic review. Schizophr Res. 2020 Jan 14; 10. Cheung SG, Goldenthal AR, Uhlemann A-C, Mann JJ, Miller JM, Sublette ME. Systematic Review of Gut Microbiota and Major Depression. Front Psychiatry [Internet]. 2019 Feb 11 [cited 2020 Jun 2];10:34. Available from: https://www.frontiersin.org/article/10.3389/fpsyt.2019.00034/full 11. Barandouzi ZA, Starkweather AR, Henderson WA, Gyamfi A, Cong XS. Altered Composition of Gut Microbiota in Depression: A Systematic Review. Front Psychiatry. 2020;11:541. 12. Nguyen TT, Hathaway H, Kosciolek T, Knight R, Jeste DV. Gut microbiome in serious mental illnesses: A systematic review and critical evaluation. Schizophr Res. 2019 Sep 5; 13. Nguyen TT, Kosciolek T, Eyler LT, Knight R, Jeste DV. Overview and systematic review of studies of microbiome in schizophrenia and bipolar disorder. J Psychiatr Res. 2018;99:50–61. 14. Cuomo A, Maina G, Rosso G, Beccarini Crescenzi B, Bolognesi S, Di Muro A, Giordano N, Goracci A, Neal SM, Nitti M, Pieraccini F, Fagiolini A. The Microbiome: A New Target for Research and Treatment of Schizophrenia and its Resistant Presentations? A Systematic Literature Search and Review. Front Pharmacol [Internet]. 2018 Oct 15 [cited 2021 Mar 17];9. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6196757/ 15. Kraeuter A-K, Phillips R, Sarnyai Z. The Gut Microbiome in Psychosis From Mice to Men: A Systematic Review of Preclinical and Clinical Studies. Front Psychiatry. 2020;11:799. 16. Jurek L, Sevil M, Jay A, Schröder C, Baghdadli A, Héry-Arnaud G, Geoffray M-M. Is there a dysbiosis in individuals with a neurodevelopmental disorder compared to controls over the course of development? A systematic review. Eur Child Adolesc Psychiatry. 2020 May 8; © 2021 American Medical Association. All rights reserved. eAppendix 6. Detailed characteristics of the included studies The 59 studies provided 64 case-control comparisons capturing 2643 patients and 2336 controls (Table 6.1). Most studies (n=32) were conducted in East Asia (China, Taiwan or Japan), 24 in westernised populations (grouped on the basis of typical diet and lifestyle: USA, Canada, Europe, Australia, New Zealand) and one in Africa. All but one used formal diagnostic criteria to define their population. Studies were similar in exclusion criteria such as major medical and gastrointestinal conditions, pregnancy and recent consumption of antibiotics (except four studies in which 17,25,28,33 antibiotics weren’t mentioned ). Recent probiotic consumption was excluded in 35/59 studies, was not mentioned in 21, and three studies 3,27,50 included a small number of participants taking probiotics . Few studies imposed restrictions on diet such as no major changes in the months 10,19,35,36 14,15,18,20,31 39,44 preceding enrolment or no weight loss, vegetarian or vegan diets . Two studies matched groups according to diet and one 55 60 controlled dietary intake . Despite its known impact on microbial communities , smoking was generally not controlled: only three studies 18,20,38 19,23,28,45,51 excluded smokers and five controlled for it during analyses . Amplicon 16S rRNA sequencing was used in 44 studies, although choice of hypervariable region (V1-V9) varied, seven studies used shotgun metagenomics to sample all microbial genes, nine studies used either qPCR or RT-qPCR to target a pre-specified range of microbial taxa, and one study employed metaproteomic analysis (Table 6.1). Supplementary Table 6.1. Key sample and methodology characteristics of case-control comparisons of the gut microbiome by disorder. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Smokers size n Age Female BMI Stage medication MDD Naseribafrou Norway ICD-10 P: 37 P: 49.2 56% P: 25.9 nr Nr 16S rRNA : Observed sp., Simpsons ei et al. 2014 HC: 18 HC: 46.1 HC: 24.7 nr : not measured MDD Jiang et al. China DSM-IV P: 29 P: 25.3 38% P: 20.3 P: 10% most, total 16S rRNA : Chao1, ACE, Shannon, Simpson, 2015 HC: 30 HC: 26.8 50% HC: 19.6 HC: 7% % nr V1-V3 evenness; : UniFrac (unweighted) MDD Aizawa et al. Japan DSM-IV P: 43 P: 39.4 41% P:23.2 nr 65% RT-qPCR : not measured 2016 HC: 57 HC: 42.8 61% HC: 22.3 16S rRNA : not measured MDD Kelly et al. Ireland DSM-IV P: 34 P: 45.8 38% P: 26.2 P: 13% 96% 16S rRNA : Observed sp., Chao1, Shannon, PD 2016 HC: 33 HC: 45.8 HC: 24.6 HC: 3% nr : UniFrac (weighted & unweighted), Bray-Curtis MDD Liu et al. China DSM-IV P: 15 P: 44.8 69% P: 22.0 nr 0% 16S rRNA : Observed sp., Chao1, Shannon, PD 2016 HC: 20 HC: 43.9 HC: 22.0 V1-V3 : measured, nr © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication MDD Zheng et al. China DSM-IV P: 58 P: 40.6 63% P: 22.0 P: 16% 67% 16S rRNA : Observed sp., Shannon, Simpson, PD 2016 HC: 63 HC: 41.8 HC: 22.6 H: 21% V3-V5 : Bray-Curtis MDD Lin et al. China DSM-IV P: 10 P: 36.2 60% P: 23.8 P:40% 100% 16S rRNA : measured, nr 2017 HC: 10 HC: 38.1 HC: 24.2 HC:30% V3-V4 : UniFrac (weighted) MDD Chen et al. China HAMD P: 44 P: 40.9 55% P: 22.1 nr 0% 16S rRNA : Observed sp., Shannon, Simpson, PD 2018a HC: 44 HC: 43.4 HC: 22.6 (drug- seq : UniFrac (nr), PLS-DA naïve) V3-V5 MDD Chen et al. China DSM-IV P: 10 P: 43.9 50% P: 23.5 nr 20% Meta- : not measured 2018b HC: 10 HC:39.6 HC: 22.6 proteomics : not measured MDD Huang et al. China ICD-10 P: 27 P: 48.7 74% P: 23.8 nr Nr 16S rRNA : Chao1, ACE, Shannon, PD 2018 HC: 27 HC: 42.3 HC: 23.4 V3-V4 : UniFrac (weighted & unweighted) MDD Chahwan et Australi M.I.N.I P: 68 P: 36.1 70% nr P: 25% 0% 16S rRNA : Observed sp., Chao1, Shannon al. 2019 a HC: 20 HC: 40.0 HC: V3-V4 : UniFrac (weighted) 11% MDD Valles- Belgium GP & P: 80 50.9 55% 24.9 nr 50% 16S rRNA : not measured Colomer / NL self-report HC: 70 nr : not measured MDD Chung et al. Taiwan DSM-5 P: 36 P: 45.8 70% P: 22.8 P: 19% 86% 16S rRNA : Shannon 2019 HC: 37 HC: 41.2 HC: 24 HC: 3% V3-V4 : UniFrac (weighted) MDD Lai et al. China DSM-5 P: 26 P: 43.7 P: P: 27.2 nr 81% Shotgun : Shannon, Fisher 2019 HC: 29 HC: 39.4 69% HC: 21.1 Metagenomic : Bray–Curtis HC: s 55% MDD Rong et al. China DSM-5 P: 31 P: 41.6 P: P: 21.5 nr 74% Shotgun : Chao 1, Shannon, Inverse Simpson, 2019 HC: 30 HC: 39.5 71% HC: 22.0 Metagenomic Gm coefficient; : Bray-Curtis HC:53 s MDD Mason et al. USA DSM-IV P: 14 P: 41.9 P:79% P: 31.0 nr 64% 16S rRNA : Shannon 2020 HC: 10 HC: 33.0 HC: HC: 25.6 V4 : UniFrac (weighted) 60% MDD Chen et al. China DSM-IV Young Young 72% Young P: nr Young: 16SrRNA : Chao1, ACE 2020 P: 25 P: 24.0 22.1 28% V3-V5 : OPLS-DA HC: 27 HC: 25.0 HC: 21.5 Mid-age: Mid-age Mid-age Mid-age 31% P: 45 P: 45.0 P:22.6 © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication HC: 44 HC: 47.2 HC: 23.2 MDD Chen et al. China DSM-5 P: 62 P: 39.6 100% P: 22.0 0% 0% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2021 HC: 46 HC: 34.0 HC: 22.2 V3-V4 Simpson; : UniFrac (weighted & unweighted) MDD Yang et al. China DSM-IV P: 156 P: 29.6 P: 36 P: 22.3 nr 24% Shotgun : Chao1, Shannon, Inv. Shannon 2020 HC: 155 HC: 29.1 % HC: 22.4 Metagenomic : Bray-Curtis HC: s 54% MDD Liu et al. USA DSM-5 P: 43 P: 21.9 P: nr 0% 65% 16SrRNA : Observed sp., Shannon, PD 2020 HC: 47 HC: 22.1 88% V4 : UniFrac (weighted & unweighted), HC: Bray-Curtis 72% MDD Stevens et USA DSM-IV P: 20 P & HC: P: nr nr 75% nr : Chao1, Shannon al. 2020 HC: 20 34 50% : Bray-Curtis HC: 70% MDD + Stevens et USA DSM-5 P: 22 nr nr nr nr Nr nr : not measured ANX al. 2018 HC: 28 : not measured MDD & Mason et al. USA DSM-IV P: 38 P:39.2 P: P: 20.4 nr 42% 16S rRNA : Shannon ANX 2020 HC: 10 HC:33.0 82% HC: 25.6 V4 : UniFrac (weighted) HC: 60% MDD & Vinberg et al. Denmar ICD-8 & P: 74 P: 37.7 77% P: 26.5 P > HC, 61% 16SrRNA : Observed sp., Shannon BD 2019 k ICD-10 HC: 25 HC: 37.0 HC: 24.5 % nr V3-V4 : generalized UniFrac remission MDD & Jiang et al. China DSM-IV MDD:14 P: 37.2 45% P: 23.6 P: 8% most, total 16SrRNA : Chao1, ACE, Shannon, Simpson BD 2020 BD: 10 HC: 35.8 HC: 22.3 HC: 6% % nr V1-V3 : UniFrac (weighted & unweighted), HC: 16 Bray-Curtis BD Evans et al. USA DSM-IV P:115 P: 50.2 73% P: 29.3 nr most, total 16SrRNA : not measured 2017 HC: 64 HC: 48.6 HC: 26.0 % nr V4 : Yue & Clayton BD Painold et al. Austria DSM-IV, P: 32 P: 41.3 44% P: 28.4 nr 100% 16SrRNA : Observed sp., Chao1, Shannon, 2018 current HC: 10 HC: 31.4 HC: 24.3 V1-V2 Simpson : UniFrac (weighted & unweighted) © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication depression or euthymia BD Aizawa et al. Japan DSM-IV, P: 39 P: 40.3 56% P: 23.9 nr 92% RT qPCR for : not measured 2019 any HC: 58 HC: 43.1 HC: 22.4 16S/23S : not measured episode rRNA BD Rong et al. China DSM-5, P: 30 P: 38.4 52% P: 21.9 nr 83% Shotgun : Chao 1, Shannon, Inv. Simpson, Gm 2019 current HC: 30 HC: 39.5 HC: 22.0 Metagenomic coefficient depression s : Bray-Curtis BD Coello et al. Denmar ICD-10, P: 113 P: 31 62% P: 24.8 P: 36% 88% 16SrRNA : Observed sp., Shannon 2019 k HC: 77 HC: 29 HC:24.2 HC: V3-V4 : UniFrac (weighted & unweighted) any 11% episode BD McIntyre et Canada DSM-5, P: 23 P: 45 70% P: 30 HC: P: 21% nr 16SrRNA : Observed sp., Shannon, Inv. Simpson al. 2019 current HC: 23 HC: 43.8 26 HC: 9% V3 : Bray-Curtis depression BD Hu et al. China DSM-IV- P: 52 P: 24.2 48% P: 21.6 nr 0% 16SrRNA : Observed sp., Chao1, Shannon, 2019 TR, current HC: 45 HC:36.3 HC: 22.4 V3-V4 Simpson, Inv. Simpson, ICE depression : UniFrac (weighted & unweighted) BD Lai et al. China DSM-5, P: 25 P: 36.9 48% P: 22.1 nr 80% Shotgun : Shannon, Simpson, Fisher 2021 current HC: 28 HC: 39.2 HC:21.1 Metagenomic : Bray-Curtis depression s BD Lu et al. China DSM-IV- P: 36 P: 32.6 43% P: 22.2 0% 0% qPCR : not measured 2019 TR, current HC: 27 HC: 28.9 HC: 21.8 : not measured depression SCZ Nguyen et al. USA DSM-IV-TR P: 25 P: 52.9 44% P: 31.8 P: 56% 100% 16SrRNA : Observed sp., Shannon, Faith's PD 2019 HC: 25 HC: 54.7 HC:28.9 HC: 4% V4 : UniFrac (unweighted), Bray-Curtis SCZ Schwarz et Finland DSM-IV; P: 28 P: 25.9 43% P: 23.8 nr 93% qPCR for : not measured al. 2018 FEP HC: 16 HC: 27.1 HC:23.9 16S rRNA : not measured primers, Metagenomic SCZ Shen et al. China ICD-10 P: 64 P: 42 44% P:23.5 P: 19% 100% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2018 HC: 53 HC: 39 HC:23.1 HC: V3-V4 Simpson, Faith's PD 23% : UniFrac (unweighted) © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication SCZ Yuan et al. China DSM-IV; P: 41 P: 23.1 44% P:20.5 P: 5% 0% qPCR for : not measured 2018 FEP HC: 41 HC: 24.7 HC: 20.8 HC: 6% 16S rRNA : not measured primers SCZ Zheng et al. China DSM-IV P: 63 P: 43.5 P:33% P: 22.9 nr 92% 16SrRNA : Chao 1, Shannon 2019 HC: 69 HC: 40.0 HC: HC: 23.2 V3-V4 : PLS-DA 48% SCZ Pan et al. China DSM-IV P: 29 P: 34.9 34% P: 23.7 0% 90% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2020 HC: 29 HC: 34.8 HC: 23.5 V3-V4 Simpson, Faith's PD : UniFrac (unweighted) SCZ Zhu et al. China DSM-IV P: 90 P: 28.6 49% P: 20.6 P: 30% 0% Shotgun : Shannon 2020 HC: 81 HC: 32.3 HC: 21.7 HC: Metagenomic : Bray-Curtis 25% s SCZ Li et al. China DSM-IV-TR P: 82 P: 42.2 47% P:24.5 P:21% 91% 16SrRNA : Observed Sp., Evenness, Shannon, 2020 HC: 80 HC: 41.0 HC: 23.0 HC: 5% V4 Faith’s PD; : Bray-Curtis SCZ Ma et al. China DSM-IV; FEP: 40 P: 24.2 46% nr nr FEP: 0% 16SrRNA : Chao1, Shannon 2020 FEP & SCZ SCZ: 85 HC:23.1 (drug- V4 : UniFrac (weighted & unweighted) HC: 69 naïve) SCZ: 100% SCZ Zhang et al. China DSM-IV; P:10 P: 37.6 42% P: 23.3 P: 10% 0% 16SrRNA : Observed sp., Chao1, Shannon, 2020 FEP HC:16 HC: 35.8 HC:22.3 HC: (drug nr Simpson; : UniFrac (weighted & 6.3% naïve) unweighted), Bray-Curtis SCZ Xu et al., China DSM-5 P: 84 P: 35.0 43% P:22 nr 98% 16SrRNA : Chao1 2020 HC: 84 HC: 35.0 HC:23.1 V4 & : non-metric multidimensional Shotgun scaling Metagenomic ANX Jiang et al. China DSM-IV P: 40 P: 33.4 P: P:21.7 P: 2.5% 70% 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2018 HC: 36 HC: 35.6 75% HC: 21.4 HC: 3% V3-V4 Simpson; : UniFrac unweighted HC: 64% ANX Chen et al. China DSM-5 P: 36 P: 46.1 57% P: 23.1 P: nr 16SrRNA : Observed sp., Chao1, ACE, Shannon, 2019 HC: 24 HC: 41.8 HC: 22.5 19.4% V3-V4 Simpson HC: : UniFrac (weighted & unweighted) 16.7% © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication ANX Mason et al. USA DSM-IV P: 8 P: 40.0 P: P: 33.3 nr 62% 16SrRNA : Shannon 2020 HC: 10 HC: 33.0 100% HC: 25.6 V4 : UniFrac weighted HC: 60% AN Armougom et France DSM-IV P: 9 P: 19-36 nr P: 12.7 nr nr RT qPCR : not measured al. 2009 HC: 20 HC: 13-68 HC: 20.7 : not measured AN Million et al. France DSM-IV P: 15 P: 27.3 P: P: 13.5 nr nr qPCR : not measured 2013 HC: 76 HC: 49.5 93% HC: 22.4 : not measured HC:43 AN Kleiman et USA DSM-IV-TR P: 16 P: 28.0 100% P: 16.2 nr nr 16SrRNA : Observed sp., Chao1 al. 2015 HC: 12 HC: 29.8 HC: 21.5 V1-V3 : UniFrac (weighted & unweighted) AN Morita et al. Japan DSM-IV-TR P: 25 P: 30.0 100% P: 12.8 nr nr qPCR for : not measured 2015 HC: 21 HC: 31.5 HC: 20.5 16S/23S : not measured rRNA AN Mack et al. German ‘Diagnosis’ P: 55 P: 23.8 100% P: 15.3 nr nr 16SrRNA : Observed sp., Chao 1, Shannon 2016 y (not HC: 55 HC: 23.7 HC: 21.6 V4 : UniFrac (weighted & unweighted), specified) Bray-Curtis AN Mörkl et al. Austria ICD-10 P: 18 P: 22.4 100% P: 15. 25% nr 16SrRNA : Observed sp., Chao 1, Shannon 2017 HC: 26 HC: 24.9 HC: 21.9 V1-V2 : UniFrac (weighted & unweighted) AN Borgo et al. Italy DSM-5 P: 15 P: 25.6 100% P: 13.9 nr nr 16SrRNA : measured, nr 2017 HC: 15 HC: 24.4 HC: 22.1 V3-V4 : measured, nr AN Hanachi et France DSM-IV-TR P: 33 P: 32 100% P: 11.7 nr nr 16SrRNA : Chao 1, Shannon al. 2018 HC: 22 HC: 36 HC: 21.0 V3-V4 : UniFrac (weighted & unweighted) AN Hata et al. Japan DSM-IV-TR P: 4 P: 23.0 100% P:13.7 nr nr 16SrRNA : Observed sp., Chao 1, Shannon 2019 restrictive HC: 4 HC: 25.3 HC:21.6 V3-V4 : UniFrac (weighted & unweighted) only AN Monteleone Italy DSM-5 P:21 P: 21.7 100% P: 14.6 nr nr 16SrRNA : Chao1, Fisher et al. 2021 HC: 20 HC:23.0 HC: 20.3 V4 : non-metric multidimensional scaling OCD Domenech et Spain DSM-IV P: 38 P: 40.2 53% nr nr nr 16SrRNA : Observed sp., Chao 1, Shannon, al. pre-print HC: 33 HC: 36.0 V3-V4 Simpson, Inv. Simpson, Faith's PD : UniFrac (weighted & unweighted), Bray-Curtis, Jensen-Shannon, Canberra © 2021 American Medical Association. All rights reserved. Definition % Patients Sample Mean % Mean % Disorder Study Country of disorder on Sequencing Diversity assessments Female Smokers size n Age BMI Stage medication OCD Turna et al. Canada DSM-5 P: 21 P: 31.0 54% P: 24.6 nr 0% 16SrRNA : Observed sp., Chao 1, Inv. Simpson, 2020 HC: 22 HC:29.3 HC: 23.2 V3 Shannon; : UniFrac (weighted & unweighted), Bray-Curtis, Jaccard PTSD Hemmings et South DSM-5 P: 18 P: 42.0 P: P:28.5 P: 50% 33% 16SrRNA : Observed sp., Chao1, Shannon, al. 2017 Africa HC: 22 HC: 38.7 14% HC:28.6 HC:42% V3-V4 Faith's PD HC: : UniFrac (weighted & unweighted), 7% Bray Curtis ADHD Aarts et al. NL DSM-IV P: 19 P: 19.5 P:32% P: 23.8 nr nr 16SrRNA : Observed sp., Chao1, Shannon, 2017 HC:77 HC: 27.1 HC:47 HC: 23.0 V3-V4 Faith's PD % : not measured ADHD attention deficit hyperactivity disorder, AN anorexia nervosa, ANX anxiety, BD bipolar disorder, OCD obsessive compulsive disorder, MDD major depressive disorder, PTSD post-traumatic stress disorder, SCZ schizophrenia and psychosis, BMI body mass index, P patient, HC healthy control, NL Netehrlands, ICD International Classification of Diseases, DSM Diagnostic and Statistical Manual of Mental Disorders, MINI Mini-International Neuropsychiatric Interview, FEP first episode psychosis, seq sequencing, (RT) qPCR (real time) quantitative polymerase chain reaction, ACE abundance-based coverage estimator, ICE incidence-based estimator, Faith’s PD Faith’s phylogenetic diversity, PLS-DA partial least squares discriminant analysis, OPLS-DA orthogonal projections to latent structures discriminant analysis, n number, nr not reported eReferences: 1. Naseribafrouei A, Hestad K, Avershina E, et al. Correlation between the human fecal microbiota and depression. Neurogastroenterol Motil Off J Eur Gastrointest Motil Soc. 2014;26(8):1155-1162. doi:10.1111/nmo.12378 2. Jiang H, Ling Z, Zhang Y, et al. Altered fecal microbiota composition in patients with major depressive disorder. Brain Behav Immun. 2015;48:186-194. doi:10.1016/j.bbi.2015.03.016 3. Aizawa E, Tsuji H, Asahara T, et al. Possible association of Bifidobacterium and Lactobacillus in the gut microbiota of patients with major depressive disorder. J Affect Disord. 2016;202:254-257. doi:10.1016/j.jad.2016.05.038 4. Kelly JR, Borre Y, O’ Brien C, et al. Transferring the blues: Depression-associated gut microbiota induces neurobehavioural changes in the rat. J Psychiatr Res. 2016;82:109-118. doi:10.1016/j.jpsychires.2016.07.019 © 2021 American Medical Association. All rights reserved. 5. Liu Y, Zhang L, Wang X, et al. Similar Fecal Microbiota Signatures in Patients With Diarrhea-Predominant Irritable Bowel Syndrome and Patients With Depression. Clin Gastroenterol Hepatol. 2016;14(11):1602-1611.e5. doi:10.1016/j.cgh.2016.05.033 6. Zheng P, Zeng B, Zhou C, et al. Gut microbiome remodeling induces depressive-like behaviors through a pathway mediated by the host’s metabolism. Mol Psychiatry. 2016;21(6):786-796. doi:10.1038/mp.2016.44 7. Lin P, Ding B, Feng C, et al. Prevotella and Klebsiella proportions in fecal microbial communities are potential characteristic parameters for patients with major depressive disorder. J Affect Disord. 2017;207:300-304. doi:10.1016/j.jad.2016.09.051 8. Chen J, Zheng P, Liu Y, et al. Sex differences in gut microbiota in patients with major depressive disorder. Neuropsychiatr Dis Treat. 2018;14:647-655. doi:10.2147/NDT.S159322 9. Chen Z, Li J, Gui S, et al. Comparative metaproteomics analysis shows altered fecal microbiota signatures in patients with major depressive disorder. Neuroreport. 2018;29(5):417-425. doi:10.1097/WNR.0000000000000985 10. Huang Y, Shi X, Li Z, et al. Possible association of Firmicutes in the gut microbiota of patients with major depressive disorder. Neuropsychiatr Dis Treat. 2018;14:3329-3337. doi:10.2147/NDT.S188340 11. Chahwan B, Kwan S, Isik A, van Hemert S, Burke C, Roberts L. Gut feelings: A randomised, triple-blind, placebo-controlled trial of probiotics for depressive symptoms. J Affect Disord. 2019;253:317-326. doi:10.1016/j.jad.2019.04.097 12. Valles-Colomer M, Falony G, Darzi Y, et al. The neuroactive potential of the human gut microbiota in quality of life and depression. Nat Microbiol. 2019;4(4):623-632. doi:10.1038/s41564-018-0337-x 13. Chung Y-CE, Chen H-C, Chou H-CL, et al. Exploration of microbiota targets for major depressive disorder and mood related traits. J Psychiatr Res. 2019;111:74-82. doi:10.1016/j.jpsychires.2019.01.016 14. Lai W-T, Deng W-F, Xu S-X, et al. Shotgun metagenomics reveals both taxonomic and tryptophan pathway differences of gut microbiota in major depressive disorder patients. Psychol Med. Published online November 5, 2019:1-12. doi:10.1017/S0033291719003027 15. Rong H, Xie X, Zhao J, et al. Similarly in depression, nuances of gut microbiota: Evidences from a shotgun metagenomics sequencing study on major depressive disorder versus bipolar disorder with current major depressive episode patients. J Psychiatr Res. 2019;113:90-99. doi:10.1016/j.jpsychires.2019.03.017 © 2021 American Medical Association. All rights reserved. 16. Mason BL, Li Q, Minhajuddin A, et al. Reduced anti-inflammatory gut microbiota are associated with depression and anhedonia. J Affect Disord. 2020;266:394-401. doi:10.1016/j.jad.2020.01.137 17. Chen J-J, He S, Fang L, et al. Age-specific differential changes on gut microbiota composition in patients with major depressive disorder. Aging. 2020;12(3):2764-2776. doi:10.18632/aging.102775 18. Chen Y, Xue F, Yu S, et al. Gut microbiota dysbiosis in depressed women: The association of symptom severity and microbiota function. J Affect Disord. 2021;282:391-400. doi:10.1016/j.jad.2020.12.143 19. Yang J, Zheng P, Li Y, et al. Landscapes of bacterial and metabolic signatures and their interaction in major depressive disorders. Sci Adv. 2020;6(49):eaba8555. doi:10.1126/sciadv.aba8555 20. Liu RT, Rowan-Nash AD, Sheehan AE, et al. Reductions in anti-inflammatory gut bacteria are associated with depression in a sample of young adults. Brain Behav Immun. 2020;88:308-324. doi:10.1016/j.bbi.2020.03.026 21. Stevens BR, Roesch L, Thiago P, et al. Depression phenotype identified by using single nucleotide exact amplicon sequence variants of the human gut microbiome. Mol Psychiatry. Published online January 27, 2020:1-11. doi:10.1038/s41380-020-0652-5 22. Stevens BR, Goel R, Seungbum K, et al. Increased human intestinal barrier permeability plasma biomarkers zonulin and FABP2 correlated with plasma LPS and altered gut microbiome in anxiety or depression. Gut. 2018;67(8):1555-1557. doi:10.1136/gutjnl-2017-314759 23. Vinberg M, Ottesen NM, Meluken I, et al. Remitted affective disorders and high familial risk of affective disorders associate with aberrant intestinal microbiota. Acta Psychiatr Scand. 2019;139(2):174-184. doi:https://doi.org/10.1111/acps.12976 24. Jiang H, Pan L, Zhang X, Zhang Z, Zhou Y, Ruan B. Altered gut bacterial–fungal interkingdom networks in patients with current depressive episode. Brain Behav. 2020;10(8):e01677. doi:https://doi.org/10.1002/brb3.1677 25. Evans SJ, Bassis CM, Hein R, et al. The gut microbiome composition associates with bipolar disorder and illness severity. J Psychiatr Res. 2017;87:23-29. doi:10.1016/j.jpsychires.2016.12.007 26. Painold A, Mörkl S, Kashofer K, et al. A step ahead: Exploring the gut microbiota in inpatients with bipolar disorder during a depressive episode. Bipolar Disord. 2019;21(1):40-49. doi:10.1111/bdi.12682 27. Aizawa E, Tsuji H, Asahara T, et al. Bifidobacterium and Lactobacillus Counts in the Gut Microbiota of Patients With Bipolar Disorder and Healthy Controls. Front Psychiatry. 2019;9:730. doi:10.3389/fpsyt.2018.00730 © 2021 American Medical Association. All rights reserved. 28. Coello K, Hansen TH, Sørensen N, et al. Gut microbiota composition in patients with newly diagnosed bipolar disorder and their unaffected first-degree relatives. Brain Behav Immun. 2019;75:112-118. doi:10.1016/j.bbi.2018.09.026 29. McIntyre RS, Subramaniapillai M, Shekotikhina M, et al. Characterizing the gut microbiota in adults with bipolar disorder: a pilot study. Nutr Neurosci. Published online May 28, 2019:1-8. doi:10.1080/1028415X.2019.1612555 30. Hu S, Li A, Huang T, et al. Gut Microbiota Changes in Patients with Bipolar Depression. Adv Sci. 2019;6(14):1900752. doi:https://doi.org/10.1002/advs.201900752 31. Lai W, Zhao J, Xu S, et al. Shotgun metagenomics reveals both taxonomic and tryptophan pathway differences of gut microbiota in bipolar disorder with current major depressive episode patients. J Affect Disord. 2021;278:311-319. doi:10.1016/j.jad.2020.09.010 32. Lu Q, Lai J, Lu H, et al. Gut Microbiota in Bipolar Depression and Its Relationship to Brain Function: An Advanced Exploration. Front Psychiatry. 2019;10. doi:10.3389/fpsyt.2019.00784 33. Nguyen TT, Hathaway H, Kosciolek T, Knight R, Jeste DV. Gut microbiome in serious mental illnesses: A systematic review and critical evaluation. Schizophr Res. Published online September 5, 2019. doi:10.1016/j.schres.2019.08.026 34. Schwarz E, Maukonen J, Hyytiäinen T, et al. Analysis of microbiota in first episode psychosis identifies preliminary associations with symptom severity and treatment response. Schizophr Res. 2018;192:398-403. doi:10.1016/j.schres.2017.04.017 35. Shen Y, Xu J, Li Z, et al. Analysis of gut microbiota diversity and auxiliary diagnosis as a biomarker in patients with schizophrenia: A cross-sectional study. Schizophr Res. 2018;197:470-477. doi:10.1016/j.schres.2018.01.002 36. Yuan X, Zhang P, Wang Y, et al. Changes in metabolism and microbiota after 24-week risperidone treatment in drug naïve, normal weight patients with first episode schizophrenia. Schizophr Res. 2018;201:299-306. doi:10.1016/j.schres.2018.05.017 37. Zheng P, Zeng B, Liu M, et al. The gut microbiome from patients with schizophrenia modulates the glutamate-glutamine-GABA cycle and schizophrenia-relevant behaviors in mice. Sci Adv. 2019;5(2):eaau8317. doi:10.1126/sciadv.aau8317 38. Pan R, Zhang X, Gao J, Yi W, Wei Q, Su H. Analysis of the diversity of intestinal microbiome and its potential value as a biomarker in patients with schizophrenia: A cohort study. Psychiatry Res. 2020;291:113260. doi:10.1016/j.psychres.2020.113260 39. Zhu F, Ju Y, Wang W, et al. Metagenome-wide association of gut microbiome features for schizophrenia. Nat Commun. 2020;11(1):1612. doi:10.1038/s41467-020-15457-9 © 2021 American Medical Association. All rights reserved. 40. Li S, Zhuo M, Huang X, et al. Altered gut microbiota associated with symptom severity in schizophrenia. PeerJ. 2020;8. doi:10.7717/peerj.9574 41. Ma X, Asif H, Dai L, et al. Alteration of the gut microbiome in first-episode drug-naïve and chronic medicated schizophrenia correlate with regional brain volumes. J Psychiatr Res. 2020;123:136-144. doi:10.1016/j.jpsychires.2020.02.005 42. Zhang X, Pan L-Y, Zhang Z, Zhou Y-Y, Jiang H-Y, Ruan B. Analysis of gut mycobiota in first-episode, drug-naïve Chinese patients with schizophrenia: A pilot study. Behav Brain Res. 2020;379:112374. doi:10.1016/j.bbr.2019.112374 43. Xu R, Wu B, Liang J, et al. Altered gut microbiota and mucosal immunity in patients with schizophrenia. Brain Behav Immun. 2020;85:120-127. doi:10.1016/j.bbi.2019.06.039 44. Jiang H-Y, Zhang X, Yu Z-H, et al. Altered gut microbiota profile in patients with generalized anxiety disorder. J Psychiatr Res. 2018;104:130-136. doi:10.1016/j.jpsychires.2018.07.007 45. Chen Y-H, Bai J, Wu D, et al. Association between fecal microbiota and generalized anxiety disorder: Severity and early treatment response. J Affect Disord. 2019;259:56-66. doi:10.1016/j.jad.2019.08.014 46. Armougom F, Henry M, Vialettes B, Raccah D, Raoult D. Monitoring Bacterial Community of Human Gut Microbiota Reveals an Increase in Lactobacillus in Obese Patients and Methanogens in Anorexic Patients. PLOS ONE. 2009;4(9):e7125. doi:10.1371/journal.pone.0007125 47. Million M, Angelakis E, Maraninchi M, et al. Correlation between body mass index and gut concentrations of Lactobacillus reuteri , Bifidobacterium animalis , Methanobrevibacter smithii and Escherichia coli. Int J Obes. 2013;37(11):1460-1466. doi:10.1038/ijo.2013.20 48. Kleiman SC, Watson HJ, Bulik-Sullivan EC, et al. The Intestinal Microbiota in Acute Anorexia Nervosa and During Renourishment: Relationship to Depression, Anxiety, and Eating Disorder Psychopathology. Psychosom Med. 2015;77(9):969–981. doi:10.1097/PSY.0000000000000247 49. Morita C, Tsuji H, Hata T, et al. Gut Dysbiosis in Patients with Anorexia Nervosa. PLOS ONE. 2015;10(12):e0145274. doi:10.1371/journal.pone.0145274 50. Mack I, Cuntz U, Grämer C, et al. Weight gain in anorexia nervosa does not ameliorate the faecal microbiota, branched chain fatty acid profiles, and gastrointestinal complaints. Sci Rep. 2016;6:26752. doi:10.1038/srep26752 © 2021 American Medical Association. All rights reserved. 51. Mörkl S, Lackner S, Müller W, et al. Gut microbiota and body composition in anorexia nervosa inpatients in comparison to athletes, overweight, obese, and normal weight controls. Int J Eat Disord. 2017;50(12):1421-1431. doi:10.1002/eat.22801 52. Borgo F, Riva A, Benetti A, et al. Microbiota in anorexia nervosa: The triangle between bacterial species, metabolites and psychological tests. Sanz Y, ed. PLOS ONE. 2017;12(6):e0179739. doi:10.1371/journal.pone.0179739 53. Hanachi M, Manichanh C, Schoenenberger A, et al. Altered host-gut microbes symbiosis in severely malnourished anorexia nervosa (AN) patients undergoing enteral nutrition: An explicative factor of functional intestinal disorders? Clin Nutr Edinb Scotl. 2019;38(5):2304- 2310. doi:10.1016/j.clnu.2018.10.004 54. Hata T, Miyata N, Takakura S, et al. The Gut Microbiome Derived From Anorexia Nervosa Patients Impairs Weight Gain and Behavioral Performance in Female Mice. Endocrinology. 2019;160(10):2441-2452. doi:10.1210/en.2019-00408 55. Monteleone AM, Troisi J, Fasano A, et al. Multi-omics data integration in anorexia nervosa patients before and after weight regain: A microbiome-metabolomics investigation. Clin Nutr Edinb Scotl. 2021;40(3):1137-1146. doi:10.1016/j.clnu.2020.07.021 56. Domènech L, Willis J, Alemany M, et al. Changes in the stool and oropharyngeal microbiome in obsessive-compulsive disorder. medRxiv. Published online May 28, 2020:2020.05.26.20113779. doi:10.1101/2020.05.26.20113779 57. Turna J, Grosman Kaplan K, Anglin R, et al. The gut microbiome and inflammation in obsessive-compulsive disorder patients compared to age- and sex-matched controls: a pilot study. Acta Psychiatr Scand. 2020;142(4):337-347. doi:10.1111/acps.13175 58. Hemmings SMJ, Malan-Müller S, van den Heuvel LL, et al. The Microbiome in Posttraumatic Stress Disorder and Trauma-Exposed Controls: An Exploratory Study. Psychosom Med. 2017;79(8):936-946. doi:10.1097/PSY.0000000000000512 59. Aarts E, Ederveen THA, Naaijen J, et al. Gut microbiome in ADHD and its relation to neural reward anticipation. PloS One. 2017;12(9):e0183509. doi:10.1371/journal.pone.0183509 60. Savin Z, Kivity S, Yonath H, Yehuda S. Smoking and the intestinal microbiome. Arch Microbiol. 2018;200(5):677-684. doi:10.1007/s00203-018-1506-2 © 2021 American Medical Association. All rights reserved. eAppendix 7. Stool sample processing methods in the included studies Table e7.1. Stool sample collection, storage and DNA extraction procedures of the included studies. Study Collection & handling by participant Long-term storage DNA extraction method Naseribafrouei et al. Outpatient samples frozen at 20 °C in home at 70 °C until use MagTM mini kit (LGC), 2014 freezer upon collection and transported at below following the manufacturer recommendations zero. Inpatients stored directly at -70 °C. Jiang et al. Sterile plastic cups were used to collect samples at 80 °C within 15 min of QIAamp® DNA Stool Mini Kit (QIAGEN), following 2015 and were kept in an icebox collection, until use manufacturer instructions, with additional glass-bead beating steps on a Mini-beadbeater (FastPrep; Thermo Electron Corp) Aizawa et al. 2016 collected with RNA stabilizer an stored at room - Total RNA fractions were extracted (Yakult Central temperature or at 4°C until sent to lab Institute), method not reported Kelly et al. Collected in plastic containers containing an Homogenized, aliquoted and QIAamp DNA Stool Mini Kit (QIAGEN) 2016 anaerobic generator AnaeroGen Compact Oxoid stored at 80 °C until further use sachet Liu et al. - Immediately stored at 80 °C until PowerSoil DNA Isolation Kit (MoBio), following Human 2016 use Microbiome Project recommendations Zheng et al. 2016 - Immediately stored at 80 °C until PowerSoil DNA Isolation Kit (MoBio), following use standard protocols Lin et al. - Immediately stored at 70 °C until Tiagen DNA Stool Mini Kit (Tiagen Biotech), following 2017 use manufacturer protocols Chen et al. 2018a - at 80 °C until use PowerSoil DNA kit following standard protocol Chen et al. 2018b Sterile plastic cups were used to collect samples at 80 °C until use Tandem mass spectra were extracted and analyzed using Mascot (Matrix Science) against a combined Swiss prot-human (20151226) and TrEMBL bacteria database Huang et al. 2018 sterile containers were used to collect samples Immediately stored at 80 °C until PowerSoil DNA Kit (Missouri Biotechnology use Association) Chahwan et al. kept on ice or refrigerated before delivery to study Samples placed at 4 °C and PowerFecal DNA Isolation Kit (MoBio), following 2019 staff aliquots stored at 80 °C within manufacturer instuctions several days © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method Valles-Colomer Frozen at -18°C at home and cool transported to stored at -18°C until transport on PowerMicrobiome RNA Isolation kit (MoBio 2019 collection point dry ice to the research facility for - Laboratories) 80°C storage Chung et al. 2019 delivered in 4 °C to staff and 80 °C until use QIAamp DNA Stool Mini Kit (QIAGEN) or phenol– chloroform extraction method Lai et al. 2019 - immediately stored at 80 °C until StoolGen DNA kit (CWBiotech Co) use Rong et al. 2019 - immediately stored at 80 °C until StoolGen DNA kit (CWBiotech Co) shipped to lab Chen et al. 2020 - - standard PowerSoil kit protocol Chen et al. 2021 Sterile plastic cups were used to collect samples stored at 80 °C within 30 min of Qiagen QIAamp DNA Stool Mini Kit (Qiagen) according collection until use to manufacturer’s instructions Yang et al. 2020 - - E.Z.N.A. Soil DNA Kit (Omega Bio-tek, Norcross, GA, USA) according to manufacturer’s instructions Liu et al. 2020 OMNIgene•GUT stool collection kits at 80 °C until use ZymoBIOMICS 96 DNA Kit (Zymo Research) according to manufacturer’s instructions Stevens et al. 2020 collected with OMNIgeneGUT fecal at 80 °C until use Stool Extraction Kit following the manufacturer’s collection kits (DNAGenotek, Ontario, Canada) instructions (Omega Bio-tek, Doraville, CA). Stevens et al. 2018 [ - at 80 °C until use - Vinberg et al. 2019 stool collection kits (Sarsted Kit sterile) at 80 °C within 24-72 hours, until NucleoSpin 96 Soil kit (Macherey-Nagel) use Jiang et al. 2020 collected in a sterile plastic cup at the stored at 80 °C within 30 min of FastDNATM SPIN Kit for Feces (MP Biomedicals) hospital and refrigerated collection until use according to manufacturer's instructions Evans et al. 2017 Home stool collection kits (DNA Genotek, Ontario at 80 °C until use PowerMag soil isolation kit (MoBio) CA) Painold et al. 2018 - PowerLyzer PowerSoil DNA Isolation Kit (MoBio) - 20 C until use Aizawa et al. 2019 collected with RNA stabilizer and stored at 4°C at at 4°C until use Total RNA fractions were extracted (Yakult Central home Institute), method not reported Coello et al. 2019 collected with OMNIgeneGUT fecal at 80 °C until use NucleoSpin ® 96 Soil kit collection kits (DNAGenotek, Ontario, Canada) © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method McIntyre et al. 2019 Collected in sterile screw-capped sample jar and Processed upon receipt; back-ups Using in-house protocol on a MagMax™ robot frozen in home freezer stored at 80 °C (Thermo Fischer Scientific,WalthamMass) Hu et al. 2019 at -80°C within 30 min of PSP Spin Stool DNA Plus Kit (Stratec, Germany) collection, until use according to the manufacturer’s instructions. Lai et al. 2021 - At -80 until use StoolGen DNA kit (CWBiotech Co., Beijing, China) Lu et al. 2019 - at -80°C within 30 min of Qiagen Stool Kit (Qiagen, Hilden, Germany), according collection, until use to a modified protocol for cell lysis Nguyen et al. 2019 home stool collection kits (BD SWUBE Dual Swab at 80 °C until use Earth Microbiome Project (EMP), modified from Collection System; BD Worldwide MagAttract® PowerSoil® DNA KF Kit Schwarz et al. 2018 Collected in a larger sampling bowl, from which at 80 °C until use FastDNA Spin Kit for Soil (QBIOgene duplicate samples were transferred to smaller tubes 80 °C until use Shen et al. 2018 - at PowerSoil DNA kit (MoBio) Yuan et al. 2018 - - QIAamp Fast DNA Stool Mini Kit (QIAGEN) Zheng et al. 2019 - immediately stored at 80 °C until QIAamp DNA Stool Mini Kit (QIAGEN, Hilden, use Germany). Pan et al. 2020 Collected and immediately transported to lab with at 80 °C until use E.Z.N.A. soil kit (Omega Bio-tek, Norcross, GA, U.S.) ice packs according to manufacturer's instructions Zhu et al. 2020 - - - Li et al. 2020 - at 80 °C until use MOBIO PowerSoil DNA Isolation Kit 12,888–100 protocol Ma et al. 2020 collected in sterile plastic containers immediately stored at 80 °C until QIAamp DNA Mini Kit (QIAGEN, Hilden, Germany) use according to the manufacturer's instructions Zhang et al. 2020 Sterile plastic cups were used to collect samples at 80 °C within 30 min of FastDNA™ SPIN Kit for Feces (MP Biomedical Inc., and were kept in an icebox collection, until use Santa Ana, CA, USA) according to the manufacturer’s instructions with additional glass-bead beating steps Xu et al., 2020 Collected in disposable sterile potty and Same or next day moved to 80 StoolGen fecal DNA extraction kit (CWBiotech, Beijing, transferred to tubes then frozen at -20°C °C until use China) Jiang et al. 2018 Sterile plastic cups were used to collect samples at 80 °C within 30 min of QIAamp DNA Stool Mini Kit (QIAGEN), following and were kept in an icebox collection, until use manufacturer instructions, with the addition of a glass- © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method bead beating step on a Mini-beadbeater (FastPrep; Thermo Electron Corp) Chen et al. 2019 Sterile plastic cups were used to collect samples at 80 °C within 15 min of QIAamp DNA Stool Mini Kit (QIAGEN), following collection, until use manufacturer instructions Mason et al. 2020 Frozen in home freezer after collection at 80 °C until use Crude DNA extracts were treated with RNAseA (QIAGEN) and column-purified (PCR Purification Kit, QIAGEN) Armougom et al. - - NucleoSpinH Tissue Mini Kit according to 2009 manufacturer’s instructions Million et al. 2013 collected using sterile plastic containers and at -80°C until use NucleoSpin® Tissue Mini Kit (information taken from transported “as soon as possible” to the lab cross-referenced source [1] Kleiman et al. 2015 collected by nurses trained in collection protocols at -80°C until use Qiagen DNeasy® Blood and Tissue extraction kit (Qiagen, Valencia, CA, USA) Morita et al. 2015 Collected in tubes containing the RNA stabilizer at 4°C until use Total RNA fractions were extracted (Yakult Central Institute), method not reported Mack et al. 2016 Collected with a stool-collecting kit (Süsse immediately stored at 80 °C until PSP Spin Stool Kit (Stratec Molecular, Berlin, Labortechnik, Gudensberg, Germany) from eight use Germany) according different sites of the stool to the manufacturers’ instructions Mörkl et al. 2017 Collected with the PSP spin stool DNA stool immediately stored at 20 °C until PowerLyzer PowerSoil DNA Isolation Kit (MO BIO collection use Laboratories, CA) according to the manufacturer’s kit (Stratec, Birkenfeld, Germany) instructions. Borgo et al. 2017 - at -80°C until use Spin stool DNA kit (Stratec Molecular, Berlin, Germany), according to the manufacturer's instructions Hanachi et al. 2018 - - Standard Operating Procedure 07 of the IHMS Hata et al. 2019 Collected and sealed in a plastic bag containing a - - disposable oxygen-absorbing and carbon dioxide– generating agent and transported on ice to the laboratory within several hours. Monteleone et al. - at -80°C until use PowerSoil DNA isolation kit (Qiagen, Germantown, 2021 MD, USA) Domenech et al. collected in Stool Collection Tubes (Stratec At -20°C until use PSP Spin Stool 15 DNA Basic Kit (Stratec Molecular) pre-print 14 Molecular) © 2021 American Medical Association. All rights reserved. Study Collection & handling by participant Long-term storage DNA extraction method Turna et al. 2020 collected and transferred into a sterile screw - - capped sample jar and placed in a household freezer (for up to 1 week) Hemmings et al. - - PSP® Spin Stool DNA Plus Kit (STRATEC Molecular, 2017 Birkenfeld, Germany) according to the manufacturer’s protocol 2 (“Isolation of total DNA from 1.4 ml stabilized stool homogenate with enrichment of bacterial DNA”). Aarts et al. 2017 pea-sized amount stored it in a 50ml Falcon tube, At -80°C within 24hrs, until use DNeasy1Blood and Tissue Kit (Qiagen, Venlo, The then Netherlands) stored at 4°C until delivery to site – information not available eReferences not included elsewhere: 1. Dridi B, Henry M, El Khéchine A, et al. High Prevalence of Methanobrevibacter smithii and Methanosphaera stadtmanae Detected in the Human Gut Using an Improved DNA Detection Protocol. PLoS One; 4. Epub ahead of print 17 September 2009. DOI: 10.1371/journal.pone.0007063. © 2021 American Medical Association. All rights reserved. eAppendix 8. Publication bias assessment for the alpha diversity meta-analyses Figure e8.1. Funnel plots assessing publication bias in the meta-analyses of alpha diversity. A. Chao1, B. Observed species, C. Phylogenetic diversity, D. Shannon, E. Simpson index. © 2021 American Medical Association. All rights reserved. eAppendix 9. Beta diversity Table e9.1. Methodology and findings of the included studies assessing beta diversity for the patient vs. control group comparison. Disorder Study Year Metric Analysis Finding MDD Jiang 2015 Unweighted Unifrac - no sig. difference MDD Kelly 2016 Bray-Curtis PCoA, Adonis sig. different Unweighted Unifrac PERMANOVA sig. different Weighted Unifrac sig. different MDD Zheng 2016 Bray-Curtis PCoA sig. different MDD Lin 2017 Weighted Unifrac PCoA sig. different MDD Chen 2018a Unifrac (nr) PCoA, PLS-DA sig. different MDD Huang 2018 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different MDD Chawan 2019 Weighted Unifrac PCoA, no sig. difference PERMANOVA MDD Chung 2019 Weighted Unifrac PERMANOVA sig. different MDD Lai 2019 Bray-Curtis PCoA, sig. different PERMANOVA MDD Chen 2020 - OPLS-DA sig. different MDD Chen 2021 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different MDD Yang 2020 Bray-Curtis PERMANOVA sig. different MDD Liu 2020 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different Bray - Curtis sig. different MDD Stevens 2020 Bray-Curtis (unfiltered PERMANOVA no sig. difference data) sig. different Bray-Curtis (filtered data) MDD, ANX, Mason 2020 Weighted Unifrac PCoA no sig. difference MDD +ANX MDD, Jiang 2020 Unweighted Unifrac PCoA no sig. difference BD Weighted Unifrac no sig. difference Bray - Curtis sig. different MDD, BD Rong 2019 Bray-Curtis PCoA no sig. difference MDD+BD Vinberg 2019 Generalized UniFrac PCoA, no sig. difference PERMANOVA BD Evans 2017 Yue & Clayton distance PCoA, AMOVA sig. different BD Painold 2018 Unweighted Unifrac PCoA no sig. difference Weighted Unifrac BD Coello 2019 Weighted Unifrac - no sig. difference Unweighted Unifrac sig. different BD Mcintyre 2019 Bray–Curtis PCoA no sig. difference BD Hu 2019 Unweighted Unifrac PCoA sig. different Weighted Unifrac BD Lai 2021 Bray-Curtis PERMANOVA sig. different SCZ Nguyen 2019 Unweighted Unifrac PCoA sig. different Bray-Curtis sig. different SCZ Shen 2018 Unweighted Unifrac PCoA, ANOSIM sig. different SCZ Zheng 2019 - PLS-DA sig. different SCZ Pan 2020 Unweighted Unifrac PCoA, ANOSIM no sig. difference © 2021 American Medical Association. All rights reserved. SCZ Zhu 2020 Bray-Curtis - sig. different SCZ Li 2020 Bray-Curtis PCoA, sig. different PERMANOVA SCZ Ma 2020 Unweighted Unifrac PCoA, sig. different Weighted Unifrac PERMANOVA no difference SCZ Zhang 2020 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different Bray-Curtis sig. different SCZ Xu 2020 - NMDS sig. different ANX Jiang 2018 UniFrac unweighted PCoA, sig. different PERMANOVA ANX Chen 2019 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different AN Kleiman 2015 Unweighted Unifrac PCoA no sig. difference Weighted Unifrac no sig. difference AN Mack 2016 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different Bray Curtis sig. different AN Mörkl 2017 Unweighted Unifrac PCoA, ANOSIM, sig. different Weighted Unifrac Adonis sig. different AN Borgo 2017 - RDA, ANOSIM no sig. difference AN Hanachi 2018 Unweighted Unifrac PCoA sig. different Weighted Unifrac sig. different AN Montelione 2020 - NMDS, no sig. difference PERMANOVA PTSD Hemmings 2017 Unweighted Unifrac PCoA, ANOSIM no sig. difference Weighted Unifrac no sig. difference Bray-Curtis no sig. difference OCD Domenech pre- Unweighted Unifrac PCoA, no sig. difference print Weighted Unifrac PERMANOVA no sig. difference Bray-Curtis no sig. difference Jensen-Shannon no sig. difference Canberra no sig. difference OCD Turna 2020 Unweighted Unifrac PCoA, no sig. difference Weighted Unifrac PERMANOVA no sig. difference Bray-Curtis no sig. difference Jaccard no sig. difference PCoA = principal coordinates analysis; PERMANOVA = permutational analysis of variance; PLS-DA = principal least squares discriminant analysis; OPLS-DA = orthogonal principal least squares discriminant analysis; NMDS = non-metric multidimensional scaling © 2021 American Medical Association. All rights reserved. eAppendix 10. Figures for study-level findings of relative abundance of microbial taxa Figure e10.1. Study-level findings of relative abundance of microbial taxa in patients with psychiatric disorders compared to healthy controls at the: A. Phylum level, B. Family level, C1-2. Genus level. ADHD= attention deficit hyperactivity disorder, AN= anorexia nevrosa, ANX= anxiety, BD= bipolar disorder, MDD= depression, OCD= obsessive compulsive disorder, PTSD= post-traumatic stress disorder, SCZ= psychosis & schizophrenia © 2021 American Medical Association. All rights reserved. Figure e10.1. Continued © 2021 American Medical Association. All rights reserved. Figure e10.1. Continued © 2021 American Medical Association. All rights reserved. Figure e10.1. Continued © 2021 American Medical Association. All rights reserved.

Journal

JAMA PsychiatryAmerican Medical Association

Published: Dec 15, 2021

References