View in EDS HTML Full Text PDF Full Text

Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis

Saved in:
Bibliographic Details
Title: Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis
Language: English
Authors: Wanyu Xie (ORCID 0009-0004-5361-9971), Jie Yu, Ping Wang (ORCID 0009-0002-3406-6713)
Source: Journal of Attention Disorders. 2026 30(5):615-628.
Availability: SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: https://sagepub.com
Peer Reviewed: Y
Page Count: 14
Publication Date: 2026
Document Type: Journal Articles
Reports - Research
Descriptors: Attention Deficit Hyperactivity Disorder, Socioeconomic Status, Correlation, Brain Hemisphere Functions, Incidence, Family Income, Foreign Countries
Geographic Terms: Europe
DOI: 10.1177/10870547251385350
ISSN: 1087-0547
1557-1246
Abstract: Background: Research has consistently demonstrated a negative correlation between socioeconomic status (SES) and the prevalence of ADHD, with SES exerting a significant influence on brain development. ADHD, closely intertwined with neurological development, often manifests as impairments within brain regions associated with memory, executive function, and emotion regulation. Nevertheless, the specific brain structural mediators linking SES to ADHD remain unclear. Method: We explored whether the brain surface area (SA) and thickness (TH) mediated the relationship between SES indicators (Townsend deprivation index at recruitment, average total household income before tax, and job involves heavy manual or physical work) and ADHD utilizing two-step Mendelian Randomization (MR) and multivariate MR method. Results: The MR analysis indicated that higher SES corresponds to a lower prevalence of ADHD. Genetically predicted household income was positively correlated with the SA of insula (β = 0.31, p = 1.02 × 10⁴), and physical work was positively correlated with the TH of entorhinal cortex (β = 0.74, p = 3.73 × 10⁵). Mediation analysis showed that the SA of insula was identified as a partial mediator in the protective effect of household income against ADHD prevalence, with a mediation ratio of 5.6%. Concerning potential causal relationships between IDPs and ADHD, reduced total brain SA increased ADHD risk (OR = 0.77, p = 5.60 × 10⁹), while reduced the TH of lateral occipital was protective (OR = 1.54, p = 2.02 × 10⁴). Conclusions: SES influences ADHD through brain structural changes, offering insights for prevention and intervention strategies.
Abstractor: As Provided
Entry Date: 2026
Accession Number: EJ1501799
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwF3Fz-UYA6Bl1EFsSHe4YMmAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDLo0wL3zGM_xGhEo3wIBEICBmt_GC4saAs15gWocVoUT_tBTOmqEJt4MnLoX8dMdsjtv7RJ4ozFinNYL_zmTzhrSOB9Q89qW1qKvO-2ec4_Xiw2l3L4QZkGbbfSfG-VQ4y9S0V2a5l5XU8UILGZHsQ4M7gJLsNKHOyr06Kfx8ykc0AeH_ZOUxcgtGzgH7KiQ3rWC-hfHWz9v_eplFGRinc275PicfH8BYgyXgX4=
Text:
  Availability: 1
  Value: <anid>AN0192584782;gs001may.26;2026Mar31.02:51;v2.2.500</anid> <title id="AN0192584782-1">Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis </title> <p>Background: Research has consistently demonstrated a negative correlation between socioeconomic status (SES) and the prevalence of ADHD, with SES exerting a significant influence on brain development. ADHD, closely intertwined with neurological development, often manifests as impairments within brain regions associated with memory, executive function, and emotion regulation. Nevertheless, the specific brain structural mediators linking SES to ADHD remain unclear. Method: We explored whether the brain surface area (SA) and thickness (TH) mediated the relationship between SES indicators (Townsend deprivation index at recruitment, average total household income before tax, and job involves heavy manual or physical work) and ADHD utilizing two-step Mendelian Randomization (MR) and multivariate MR method. Results: The MR analysis indicated that higher SES corresponds to a lower prevalence of ADHD. Genetically predicted household income was positively correlated with the SA of insula (β =.31, p = 1.02 × 10<sup>−4</sup>), and physical work was positively correlated with the TH of entorhinal cortex (β =.74, p = 3.73 × 10<sup>−5</sup>). Mediation analysis showed that the SA of insula was identified as a partial mediator in the protective effect of household income against ADHD prevalence, with a mediation ratio of 5.6%. Concerning potential causal relationships between IDPs and ADHD, reduced total brain SA increased ADHD risk (OR = 0.77, p = 5.60 × 10<sup>−9</sup>), while reduced the TH of lateral occipital was protective (OR = 1.54, p = 2.02 × 10<sup>−4</sup>). Conclusions: SES influences ADHD through brain structural changes, offering insights for prevention and intervention strategies.</p> <p>Keywords: brain image-derived phenotypes; ADHD; socioeconomic status; Mendelian randomization; mediation analysis</p> <hd id="AN0192584782-2">Introduction</hd> <p>ADHD is a prevalent neurodevelopmental disorder, affecting approximately 5% of children and adolescents worldwide, along with 2.5% of adults ([<reflink idref="bib19" id="ref1">19</reflink>]). It is characterized by difficulties in concentration, hyperactivity, and impulsive behaviors, with symptoms of dysfunction persisting into adulthood in approximately two-thirds of affected adolescents ([<reflink idref="bib20" id="ref2">20</reflink>]), causing great disturbance to the patient's daily study and life. As a neurodevelopmental disorder, the nexus between ADHD and the brain is profound, as risk factors for the disorder intricately influence the structure and function of brain networks, thereby eliciting ADHD symptoms, neurocognitive deficits, and pervasive functional impairments ([<reflink idref="bib19" id="ref3">19</reflink>]).</p> <p>The etiology of ADHD is multifaceted, arising from a blend of genetic and environmental factors. Twin studies have indicated that the heritability of ADHD is up to 70% ([<reflink idref="bib21" id="ref4">21</reflink>]), underscoring the genetic component. Nonetheless, environmental factors also played a significant role and should not be overlooked. It is noted that, in recent decades, there has been some observational research examining the influence of socioeconomic status (SES) as a significant risk factor for ADHD ([<reflink idref="bib49" id="ref5">49</reflink>]; [<reflink idref="bib61" id="ref6">61</reflink>]; [<reflink idref="bib62" id="ref7">62</reflink>]). A meta-analysis revealed that children from low SES backgrounds were notably more susceptible to developing ADHD compared to their counterparts from high SES families, with a risk ratio ranging from 1.85 to 2.21 ([<reflink idref="bib63" id="ref8">63</reflink>]). SES serves as a comprehensive measure encompassing both economic and social standing within a society. It delineates individuals from those who are economically and socially advantaged to those who are disadvantaged. Factors such as income, education, occupational prestige, and neighborhood quality collectively contribute to determining one's SES ([<reflink idref="bib18" id="ref9">18</reflink>]). Although the explanation of the association between SES and health was usually focused on environmental factors as being more plausible, it's noteworthy that genetics had also been linked to SES, with common single nucleotide polymorphisms (SNPs) explaining 21% of the variance in social deprivation and 11% of the variance in household income ([<reflink idref="bib30" id="ref10">30</reflink>]). Taking advantage of this feature, [<reflink idref="bib48" id="ref11">48</reflink>] employed Mendelian randomization (MR) to corroborate a causal link between SES and ADHD.</p> <p>SES is not only related to ADHD, but also to various aspects of brain-related abilities as well as brain structure. Studies demonstrated that individuals with higher SES exhibited superior performance across domains such as cognition ([<reflink idref="bib70" id="ref12">70</reflink>]), language skills ([<reflink idref="bib22" id="ref13">22</reflink>]), and execution function ([<reflink idref="bib40" id="ref14">40</reflink>]; [<reflink idref="bib59" id="ref15">59</reflink>]). Moreover, a positive correlation existed between high SES and academic achievement, importantly, disparities in these abilities have been linked to differences in brain structure development ([<reflink idref="bib26" id="ref16">26</reflink>]; [<reflink idref="bib43" id="ref17">43</reflink>]). Hence, differences in SES are, to some extent, mirrored in alterations in brain structure. In an observation study, it was found that the SES indicators were positively related to the surface area (SA) of total brain. Specifically, family income particularly influenced the SA of specific brain regions associated with language processing and executive function, including the bilateral inferior temporal gyrus, insula, inferior frontal gyrus, and the right occipital and medial prefrontal lobes ([<reflink idref="bib54" id="ref18">54</reflink>]). Previous study has reported reduced gray matter volumes in certain brain regions among children from low-income families, including the bilateral hippocampi, middle temporal gyri, left fusiform gyrus, and right inferior occipito-temporal gyrus ([<reflink idref="bib34" id="ref19">34</reflink>]).</p> <p>Notably, a striking similarity existed between the structural brain alterations observed in individuals with ADHD and those associated with differences in SES ([<reflink idref="bib33" id="ref20">33</reflink>]; [<reflink idref="bib56" id="ref21">56</reflink>]). Both conditions are related to brain regions crucial for cognition, executive function, emotion regulation, and memory. Observational studies have shown an association between ADHD, SES, and changes in brain structure ([<reflink idref="bib33" id="ref22">33</reflink>]; [<reflink idref="bib54" id="ref23">54</reflink>]; [<reflink idref="bib63" id="ref24">63</reflink>]). However, it remains unclear whether and to what extent brain structural changes mediate the association between SES and ADHD.</p> <p>MR represents a widely utilized approach leveraging genetic variation as an instrumental variable (IV) to infer causal relationships between exposure and outcome, effectively circumventing confounding factors and reverse causation ([<reflink idref="bib39" id="ref25">39</reflink>]). Multivariate MR (MVMR) serves as an extension of MR, facilitating the exploration of independent effects stemming from multiple exposures on outcome. Compared to conventional mediation analysis, MR-based mediation analysis offers advantages in terms of mitigating result biases arising from unmeasured confounders and measurement errors ([<reflink idref="bib12" id="ref26">12</reflink>]). The role of mediators can be elucidated through two-step MR ([<reflink idref="bib11" id="ref27">11</reflink>]) and MVMR ([<reflink idref="bib64" id="ref28">64</reflink>]). In this study, we chose the Townsend deprivation index at recruitment, average pre-tax household income, and whether the job involves manual labor as indicators of SES to probe the mediating role of brain, characterized as imaging-derived phenotypes (IDPs) including SA and thickness (TH), on the association between SES and ADHD, as brain surface area and thickness have different effects on brain volume reduction in ADHD patients ([<reflink idref="bib67" id="ref29">67</reflink>]). Given that the MR study ([<reflink idref="bib48" id="ref30">48</reflink>]) demonstrated that the estimated causal effects of ADHD on SES outcomes were minimal, with effect sizes in the range of 0.05 to 0.06 standard deviations. We posited that genetically predicted changes in SES would result in modifications in brain structure, which in turn contribute to ADHD susceptibility. Through research on this issue, we aim to gain a deeper understanding of the pathogenesis of ADHD and identify the specific brain regions that may mediate SES-induced ADHD.</p> <hd id="AN0192584782-3">Materials and Methods</hd> <p></p> <hd id="AN0192584782-4">Study Design</hd> <p>The primary objective of this study is to investigate whether individual differences in brain cortical structures mediated the causality between SES and ADHD as well as to quantify the extent of this mediation. Initially, we confirmed the causal effects of all three SES indicators on ADHD through two-sample MR, utilizing SES-related genetic variants as IVs. Subsequently, we identified IDPs influenced by SES that exhibited a causal effect on ADHD through a two-step MR, suggesting their potential role as mediators. Finally, the reliability of potential mediators was assessed by MVMR and the mediated proportion was determined using the product of coefficients method ([<reflink idref="bib64" id="ref31">64</reflink>]). Additionally, as a complementary analysis, we performed MR analyses of residual brain phenotypes on ADHD to find the specific brain regions that influence the prevalence of ADHD. The flowchart of the study design is shown in Figure 1. This study is reported according to the Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization guideline (STROBE-MR; [<reflink idref="bib68" id="ref32">68</reflink>]).</p> <p>Graph: Figure 1. The study design diagram. βtotal is the MR total effects of SES on ADHD; α represents the MR effects of SES on IDPs; β1 represents the MR effects of IDPs on ADHD; βdirect represents the MVMR direct effect of SES on ADHD; β2 represents the MVMR direct effects of IDPs on ADHD. MR, mendelian randomization; MVMR, multivariate mendelian randomization; SES, socioeconomic status; TDI, Townsend deprivation index at recruitment; SA, surface area; TH, thickness; ADHD, attention-deficit/hyperactivity disorder.</p> <hd id="AN0192584782-5">Data Sources</hd> <p>Information about the data sources and sample sizes used in this study is summarized in Table 1, which relied on published abstract-level data; all original studies received ethical approval.</p> <p>Table 1. Summary of the GWAS Data Used in the MR Study.</p> <p>Graph</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" colspan="2">Phenotype</th><th align="center">Sample size</th><th align="center">Ancestry</th><th align="center">Consortium/cohort</th><th align="center">Year</th><th align="center">PMID</th></tr></thead><tbody><tr><td rowspan="3">SES</td><td>TDI</td><td>462,464</td><td rowspan="3">European</td><td rowspan="3">MRC-IEU</td><td rowspan="3">2018</td><td rowspan="3">30305743</td></tr><tr><td>Household income</td><td>397,751</td></tr><tr><td>Physical work</td><td>263,615</td></tr><tr><td rowspan="2">IDP</td><td>SA</td><td rowspan="2">23,909</td><td rowspan="2">European</td><td rowspan="2">ENIGMA</td><td rowspan="2">2020</td><td rowspan="2">32193296</td></tr><tr><td>TH</td></tr><tr><td colspan="2">ADHD</td><td>225,534 (38,691 cases and 186,843 controls)</td><td>European</td><td>PGC</td><td>2022</td><td>36702997</td></tr></tbody></table> </ephtml> </p> <p>1 SES = socioeconomic status; TDI = Townsend deprivation index at recruitment; Household income = average total household income before tax; Physical work = job involves heavy manual or physical work; MRC-IEU = Medical Research Council Integrative Epidemiology Unit; ENIGMA = Enhancing NeuroImaging Genetics through Meta-Analysis; PGC = Psychiatric Genomic Consortium; IDP = Image-derived phenotype; SA = surface area; TH = thickness.</p> <hd id="AN0192584782-6">SES Data</hd> <p>We selected three characteristics related to SES, namely Townsend deprivation index at recruitment (TDI), average total household income before tax (household income), and job involves heavy manual or physical work (physical work). The genome-wide association studies (GWAS) summary data for the traits were all obtained from the Integrative Epidemiology Unit GWAS database (MRC-IEU; [<reflink idref="bib17" id="ref33">17</reflink>]), which is a large publicly available GWAS summary database. Household income represents the average pre-tax household economic status of 397,751 Europeans in the UK Biobank database between 2006 and 2010, divided into five categories: less than £18,000, £18,000 to £30,999, £31,000 to £51,999, £52,000 to £100,000, and greater than £100,000, as the original study. The GWAS summary data of TDI comprised 462,464 Europeans and was computed at recruitment, considering factors such as unemployment, car ownership, home ownership, cramped living space, and household overcrowding. Notably, TDI is linked to the postal codes of UK Biobank participants rather than to specific individuals ([<reflink idref="bib46" id="ref34">46</reflink>]), reflecting the overall socioeconomic level of the community in which the participant resides, with higher TDI corresponding to poorer economic status, and the GWAS summary data of physical work included 263,615 Europeans.</p> <hd id="AN0192584782-7">IDPs Data</hd> <p>The GWAS summary data for IDPs (Supplemental Table S1) came from the Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) consortium study ([<reflink idref="bib3" id="ref35">3</reflink>]). We utilized data from a meta-analysis of SA and TH in 34 brain regions using the Desikan-Killiany brain atlas delineation as well as the total SA and average TH of the brain by Grasby et al, and the genomic control was applied, the global measure was treated as a covariate for 34 brain regions ([<reflink idref="bib24" id="ref36">24</reflink>]). To avoid bias in the MR results caused by sample overlap, we used GWAS summary from 23,909 from 49 cohorts participating in the ENIGMA consortium and excluded 10,083 UK Biobank samples. Given that the raw data were not normalized for beta-coefficient and standard error, like the previous study ([<reflink idref="bib74" id="ref37">74</reflink>]), we standardized the units for each brain phenotype to standard deviation (SD).</p> <hd id="AN0192584782-8">ADHD Data</hd> <p>For ADHD, we utilized information from a GWAS meta-analysis in the Psychiatric Genomics Consortium (PGC), a study that encompassed 38,691 ADHD patients and 186,843 normal individuals, including three European cohorts ([<reflink idref="bib16" id="ref38">16</reflink>]). The first is the iPSYCH consortium, comprised of 38,899 participants in the iPSYCH1 sample (17,019 cases, 21,880 controls) and 24,144 participants in the iPSYCH2 sample (8,876 cases and 15,268 controls) after quality control procedures. ADHD cases were diagnosed according to ICD10 criteria. The second is the deCODE cohort, consisting of 8,281 patients with ADHD and 137,993 controls. These patients were either individuals with a clinical diagnosis of ADHD according to ICD10 criteria (<emph>n</emph> = 5,583) or were taking medications that target ADHD symptoms (<emph>n</emph> = 2,698). The third is the PGC cohort, using pooled statistics from 10 with European ancestry (4,515 cases and 11,702 controls). Two cohorts were present in both IDP and ADHD GWAS datasets: the CARDIFF sample (<emph>n</emph> = 270) and IMAGEN-I cohort (<emph>n</emph> = 1,358). As it's uncertain that which subjects were used from GWAS summary, these potential overlapping participants represented at most only 0.7% (1,628/225,534) of the total ADHD GWAS samples (38,691 cases and 186,843 controls) and 6.8% of IDP GWAS samples.</p> <hd id="AN0192584782-9">Selection of Instrumental Variable</hd> <p>To ensure the reliability of the results, screening of eligible IVs is pivotal in the MR research. MR relies on three fundamental assumptions: Relevance, wherein the SNP must strongly correlate with the exposure; Independence, SNP shouldn't be associated with confounding factors; and Exclusivity, ensuring the SNP only influences the outcome through the exposure, without any other pathways. To fulfill these assumptions, we selected genetic variants that were robustly correlated (<emph>p</emph> < 5 × 10<sups>−8</sups>) and independent (<emph>r</emph><sups>2</sups> =.01, window size = 10,000 kb) with SES as IVs. Given the small effect sizes of genetic variants, which may only explain a fraction of the exposure, and applying less than 10 independent SNPs as IV may reduce MR statistical efficiency ([<reflink idref="bib77" id="ref39">77</reflink>]), we applied a more lenient threshold (<emph>p</emph> < 5 × 10<sups>−6</sups>) to select IVs related with IDPs, akin to previous research ([<reflink idref="bib35" id="ref40">35</reflink>]). Our screening process involved several measures: (<reflink idref="bib1" id="ref41">1</reflink>) Removal of SNPs associated with the outcome at genome-wide significance. (Supplemental Table S2) (<reflink idref="bib2" id="ref42">2</reflink>) Considering that confounders are also an important piece of the puzzle for the precision of the result, we used the PhenoScanner database ([<reflink idref="bib36" id="ref43">36</reflink>]) to search for SNPs as IVs. Based on known literature ([<reflink idref="bib13" id="ref44">13</reflink>]; [<reflink idref="bib25" id="ref45">25</reflink>]; [<reflink idref="bib31" id="ref46">31</reflink>]) and prior knowledge, we eliminated SNPs related to smoking, drinking, and education years (Supplemental Table S3). (<reflink idref="bib3" id="ref47">3</reflink>) Implementation of Steiger filtering to exclude potentially invalid genetic variants (Supplemental Table S4), ensuring that the SNP-exposure association is stronger than the SNP-outcome association ([<reflink idref="bib28" id="ref48">28</reflink>]). (<reflink idref="bib4" id="ref49">4</reflink>) Harmonization of exposure and outcome data to ensure that genetic associations are expressed per additional copy of the same allele and eliminate palindromic SNPs. (<reflink idref="bib5" id="ref50">5</reflink>) Employment of Radial MR ([<reflink idref="bib8" id="ref51">8</reflink>]) to detect genetic variants with potential pleiotropic effects through heterogeneity detection (<emph>p</emph> <.05), See Supplemental Table S5 for outlier SNPs. Moreover, to avoid the bias caused by weak IVs, we utilized the <emph>F</emph>-statistic to gauge the power of IVs, and in general, the <emph>F</emph>-statistic greater than 10 indicates an effective IV ([<reflink idref="bib58" id="ref52">58</reflink>]). In MVMR, we also considered the conditional <emph>F</emph>-statistic which is typically weaker than standard <emph>F</emph>-statistics, with a value exceeding 10 suggesting the absence of weak IVs ([<reflink idref="bib65" id="ref53">65</reflink>]). IVs are available in Supplemental Table S7 and Table S10.</p> <hd id="AN0192584782-10">MR Analysis</hd> <p>In our study, we utilized the inverse variance weighted method (IVW; [<reflink idref="bib10" id="ref54">10</reflink>]) with multiplicative random effects as the main analytical approach. In two-sample MR, the MR estimate of an IV is equal to the effect value of SNP-outcome divided by the effect value of SNP-exposure. Then the IVW method aggregates the MR estimate of each IV through weighted meta-analysis, with the inverse variance of the SNP-outcome association serving as the weight. While the IVW method is the most effective analysis method and takes into account the heterogeneity of results obtained from each variant, it assumes that all IVs adhere to the three fundamental assumptions without exhibiting heterogeneity or pleiotropic effects, ensuring accurate results. Based on this situation, we used four other methods to provide valid causal inferences under weaker assumptions than the standard IVW method, as a complement and validation. The Weight-Median method can provide reliable results when 50% or more of the IVs are valid IVs ([<reflink idref="bib6" id="ref55">6</reflink>]). The Weight-Mode method clusters IVs into groups based on the similarity of causal effects and returns estimates based on the cluster with the largest number of SNPs ([<reflink idref="bib27" id="ref56">27</reflink>]). It yields unbiased causal effects if the SNPs in the largest cluster are valid instruments. The difference between the MR-Egger method and IVW is that it allows the existence of intercepts, which means that it allows for horizontal or directional pleiotropy. This method can also return unbiased causal effects, but it assumes that horizontal pleiotropic isn't related to the effect of SNP-exposure (InSIDE hypothesis; [<reflink idref="bib5" id="ref57">5</reflink>]). The MR RAPS method performs parameter estimation based on the likelihood model, providing robust results in the presence of a large number of weak IVs and increasing statistical power ([<reflink idref="bib76" id="ref58">76</reflink>]).</p> <p>To explore the relationship between the three, we conducted pairwise univariate MR (UVMR) analyses: (a) the causal estimate of SES on ADHD, which yields the total effect (β<subs>total</subs>); (b) the MR analysis between SES and IDPs, identifying IDPs causally affected by SES for subsequent mediation analysis; (c) the MR analysis between IDPs and ADHD. In all UVMR analyses, MR results were considered significant after Bonferroni correction.</p> <hd id="AN0192584782-11">Mediation Analysis</hd> <p></p> <hd id="AN0192584782-12">Two-Step MR</hd> <p>In the first step, UVMR analysis was performed to filter out IDPs causally associated with SES and derive the effect value (α) of SES on IDPs. Subsequently, reverse MR analysis was executed to mitigate bias stemming from the bidirectional causal relationship between exposure and mediator, thus refining the effect in the subsequent mediation analysis. In the second step, we conducted UVMR analysis with the IDPs selected in the first step as exposure and ADHD as the outcome, to ascertain whether the brain phenotype was a potential mediator.</p> <hd id="AN0192584782-13">MVMR</hd> <p>IDPs identified as significant in the two-step MR analysis alongside the corresponding SES-related indicator were performed MVMR analysis to estimate the independent impact of IDPs on ADHD while adjusting for SES. MVMR employed two methods, MVMR-IVW and MVMR-Egger. The effect size (β<subs>2</subs>) estimated by MVMR-IVW represents the direct effect of IDP on ADHD.</p> <p>By discerning the disparity between the direct effect (β<subs>direct</subs>) and the total effect of SES on ADHD, we delineated whether IDPs function as mediators. The indirect effect was computed by multiplying α and β<subs>2</subs>, with the standard error estimated via the delta method ([<reflink idref="bib44" id="ref59">44</reflink>]). Subsequently, the mediation ratio was derived by dividing the indirect effect by the total effect.</p> <hd id="AN0192584782-14">Sensitivity Analyses</hd> <p>We conducted a series of sensitivity analyses to assess the robustness of the results. The first is heterogeneity detection. Cochran's <emph>Q</emph> statistic and Rücker's <emph>Q</emph> statistic are respectively utilized in the IVW method and MR-Egger method to detect the heterogeneity of each IV ([<reflink idref="bib7" id="ref60">7</reflink>]). The second is pleiotropy detection. A significant deviation of the intercept from zero in the MR-Egger method indicates the presence of directional pleiotropy. Third, we examined whether the five methods used in the MR yielded consistent directional effects. The fourth is the leave-one-out analysis, which tests whether the significant result is caused by a single IV. For all sensitivity analyses, a significance level of.05 was used for the <emph>p</emph>-values of the <emph>Q</emph>-statistic, intercept, and leave-one-out test.</p> <hd id="AN0192584782-15">Statistical Analyses</hd> <p>Statistical analysis was performed by the Two-sample MR package (version 0.5.11; [<reflink idref="bib29" id="ref61">29</reflink>]), MVMR package (version 0.4; [<reflink idref="bib66" id="ref62">66</reflink>]), and Radial MR (version 1.0; [<reflink idref="bib8" id="ref63">8</reflink>]) package of the R program. The "ggplot2" and "ggseg ([<reflink idref="bib51" id="ref64">51</reflink>])" packages in R were used to create plots.</p> <hd id="AN0192584782-16">Results</hd> <p></p> <hd id="AN0192584782-17">Total Effect of SES on ADHD</hd> <p>By two-sample MR, the three SES indicators: TDI, household income, and physical work, showed a significant causal association with ADHD in the IVW estimates (Table 2, Supplemental Figure S3), which was consistent with previous reports ([<reflink idref="bib48" id="ref65">48</reflink>]). For every SD increase in either genetically predicted TDI or physical work, there is a corresponding increase in the prevalence of ADHD (OR = 3.44, <emph>p</emph> = 1.60 × 10<sups>−6</sups>; OR = 2.26, <emph>p</emph> = 5.78 × 10<sups>−5</sups>, respectively). For every SD increase in genetically predicted household income, the odds of ADHD decrease by 48% (OR = 0.52, <emph>p</emph> = 2.69 × 10<sups>−8</sups>; Table 2). Importantly, there was no evidence of heterogeneity or directional pleiotropy (Table 2), and no distortion in the leave-one-out plot (Supplemental Figure S4), which showed that causality was not driven by a single SNP. All four other MR methods, except for the causal effect of household income estimated by MR-Egger, produce consistent estimates in the direction with those predicted by IVW (Supplemental Figure S4 and Table S6).</p> <p>Table 2. The MR Result of SES on ADHD.</p> <p>Graph</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="center">Exposure</th><th align="center">Outcome</th><th align="center">Method</th><th align="center">NSNP</th><th align="center">OR [95% CI]</th><th align="center"><italic>p</italic></th><th align="center"><italic>p</italic><sub>he</sub></th><th align="center"><italic>p</italic><sub>intercept</sub></th></tr></thead><tbody><tr><td>TDI</td><td>ADHD</td><td>IVW</td><td>13</td><td>3.44 [2.08, 5.69]</td><td>1.60 × 10−6</td><td>.21</td><td>.58</td></tr><tr><td>Household income</td><td>ADHD</td><td>IVW</td><td>24</td><td>0.52 [0.41, 0.65]</td><td>2.69 × 10−8</td><td>.60</td><td>.13</td></tr><tr><td>Physical work</td><td>ADHD</td><td>IVW</td><td>11</td><td>2.26 [1.52, 3.37]</td><td>5.78 × 10−5</td><td>.78</td><td>.61</td></tr></tbody></table> </ephtml> </p> <p>2 <emph>p</emph> is the <emph>p</emph>-value of the MR result by IVW method, and <emph>p</emph><subs>he</subs> is the <emph>p</emph>-value of the heterogeneity test, <emph>p</emph><subs>intercept</subs> is the <emph>p</emph>-value of the directional pleiotropy test (MR Egger intercept). All statistical tests were two-sided. <emph>p</emph> < 0.05 was considered significant in the sensitivity analyses. IVW = inverse variance weighted; OR = odds ratio; CI = confidence interval; NSNP = number of single nucleotide polymorphism; TDI = Townsend deprivation index at recruitment.</p> <hd id="AN0192584782-18">Mediation Analysis</hd> <p></p> <hd id="AN0192584782-19">Two-Step MR</hd> <p></p> <hd id="AN0192584782-20">Causal Effects of SES on IDPs</hd> <p>By the IVW method, we conducted two-sample MR analyses of SES on 70 IDPs, and after Bonferroni correction, a significance threshold of <emph>p</emph> < 2.38 ×10 <sups>−4</sups> (0.05/3/70) was applied (Figure 2). Two IDPs were significantly influenced by SES, including the SA of insula associated with household income (β =.31; 95% CI [0.16, 0.47]; <emph>p</emph> = 1.02 ×10 <sups>−4</sups>), and the TH of entorhinal cortex associated with physical work (β =.74; 95% CI [0.39, 1.09]; <emph>p</emph> = 3.73 × 10<sups>−5</sups>). The MR-RAPS, the weighted median, the weighted mode, and the MR-Egger yielded similar patterns of effects (Supplemental Figure S5 and Table S9), and the estimated values of MR RAPS were also significant suggesting robustness. The forest plots illustrating the results for these two pairs are depicted in Figure 3a. There were no significant values for the heterogeneity detection and the directional pleiotropy detection (Supplemental Table S9), and no distortion in the leave-one-out plots (Supplemental Figure S5).</p> <p>Graph: Figure 2. Casual effect of SES on IDPs. (a) The volcano plot of IVW results for SES on IDPs, the grey dashed line represents p-value = 0.05, and the red one represents p-value = 2.38×10-4. (b) The anatomic location of the insula and the entorhinal. SA, surface area; TH, thickness; TDI, Townsend deprivation index at recruitment; IVW, inverse variance weighted.</p> <p>Graph: Figure 3. The results of two-step MR analysis. (a) Forest plots for MR estimates of SES on significant IDPs. The significant p-value for the estimate of SES on IDPs is 2.38×10-4 (0.05/210). (b) Forest plots for reverse MR estimates of significant IDPs on SES, the significant p-value for the reverse MR is 0.025 (0.05/2). (c) Forest plots for the relationship between IDPs with non-significant p-value in the reverse MR and ADHD. The significant p-value is 0.025. NSNP, number of single nucleotide polymorphism; CI, confidence interval; OR, odds ratio; SA, surface area; TH, thickness; TDI, Townsend deprivation index at recruitment; IVW, inverse variance weighted.</p> <hd id="AN0192584782-21">Causal Effects of IDPs on ADHD</hd> <p>Building upon the findings from the previous step, we proceeded with further investigation utilizing the two identified brain phenotypes. To mitigate the bias caused by bidirectional causation to the mediating results, we performed reverse MR of two significant IDPs on SES. The results showed that neither the SA of insula nor the TH of entorhinal cortex exhibited a significant reverse causal impact on SES (Figure 3b; Supplemental Table S9). In the subsequent analysis of the two IDPs and ADHD, we found that every SD increase in SA of insula correlated with a 15% reduction in the odds of developing ADHD (OR = 0.85; 95% CI [0.76, 0.97]; <emph>p</emph> =.01) by IVW (Figure 3c), conversely, the TH of entorhinal cortex did not exhibit a causal relationship with ADHD (<emph>p</emph> =.57, Figure 3c). Thus, only the SA of insula showed a significant causal relationship with ADHD after Bonferroni correction. The other four MR methods also exhibited consistent directional alignment with the IVW method, and the estimated values were comparable (Supplemental Figure S6). There was no evidence to prove the existence of directional pleiotropy and heterogeneity (Supplemental Table S9), and no distortion in the leave-one-out plots (Supplemental Figure S6). Thus, the above two-step MR analysis revealed that the SA of insula may serve as the mediator between household income and ADHD.</p> <p>Further analysis among the remaining 68 brain SA or TH phenotypes unveiled two additional significant causal effects on ADHD (significant <emph>p</emph>-value threshold by Bonferroni correction is 0.05/68 = 7.35 × 10<sups>−4</sups>). These included a protective effect of total brain SA on ADHD, with a 23% reduction in the risk of ADHD for every SD increase in total brain SA (OR = 0.77; 95% CI [0.70, 0.84]; <emph>p</emph> = 5.60 × 10<sups>−9</sups>), and a risk factor association between the TH of lateral occipital and ADHD, with the risk increasing by 54% for every SD increase in the TH of lateral occipital (OR = 1.54; 95% CI [1.23, 1.94]; <emph>p</emph> = 2.02 × 10<sups>−4</sups>). The other four methods are consistent with the direction of IVW (Supplemental Figures S7 and S8), and the MR RAPS method corroborated this association. No significant values were detected in heterogeneity or directional pleiotropy tests (Supplemental Table S11), and no distortion in the leave-one-out plots (Supplemental Figure S7)</p> <hd id="AN0192584782-22">Mediating Effects of IDP in the Association Between SES and ADHD</hd> <p>To understand the potential mediator role of IDP in the causal association between SES and ADHD, we need to control SES and IDP respectively, and estimate the direct effect between the remaining one and ADHD. Therefore, the MVMR analysis of household income and the SA of insula to ADHD was conducted (Figure 1). The conditional <emph>F</emph>-statistic, exceeding 10 for both household income and the SA of insula (20 and 11, respectively), indicates the validity of the IVs. The findings by the IVW method revealed that even after controlling for the genetic influence of the SA of insula, the impact of household income on ADHD remains significant (β<subs>direct =</subs> −.61, <emph>p</emph> = 1.02 × 10<sups>−8</sups>), albeit with reduced effect size (Table 3), indicating that the SA of insula is a partial mediator (β<subs>2 =</subs> −.12, <emph>p</emph> =.04). We also confirmed these findings by MR-Egger (Table 3). Next, based on IVW results, we calculated the indirect effect to be −0.037 via the product of coefficients method (using α and β<subs>2</subs>), resulting in a mediation ratio of 5.6% (Table 4).</p> <p>Table 3. The Result of MVMR Estimates Between Household Income, the SA of Insula, and ADHD.</p> <p>Graph</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="center">Adjustment</th><th align="center">Exposure</th><th align="center">Outcome</th><th align="center">Method</th><th align="center">NSNP</th><th align="center">Beta [95%CI]</th><th align="center"><italic>p</italic></th></tr></thead><tbody><tr><td>Insula_SA</td><td>Household income</td><td rowspan="2">ADHD</td><td rowspan="2">MV_IVW</td><td>24</td><td>−.61 [−0.82, −0.40]</td><td>1.02 × 10−8</td></tr><tr><td>Household income</td><td>Insula_SA</td><td>21</td><td>−.12 [−0.23, −0.01]</td><td>.04</td></tr><tr><td>Insula_SA</td><td>Household income</td><td rowspan="2">ADHD</td><td rowspan="2">MVMR_Egger</td><td>24</td><td>−.61 [−0.81, −0.41]</td><td>3.07 × 10−9</td></tr><tr><td>Household income</td><td>Insula_SA</td><td>21</td><td>−.13 [−0.25, −0.03]</td><td>.01</td></tr></tbody></table> </ephtml> </p> <p>Table 4. The Mediation Effect of Household Income on ADHD via the SA of Insula.</p> <p>Graph</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="center">Exposure</th><th align="center">Mediator</th><th align="center">Outcome</th><th align="center">β<sub>total</sub> [95% CI]</th><th align="center">α [95% CI]</th><th align="center">β<sub>2</sub> [95% CI]</th><th align="center">β<sub>indirect</sub> [95%CI]</th><th align="center">Mediated proportion (%) [95% CI]</th></tr></thead><tbody><tr><td>Household income</td><td>Insula_SA</td><td>ADHD</td><td>−.66 [−0.90, −0.43]</td><td>.31 [0.16, 0.47]</td><td>−.12 [−0.23, −0.01]</td><td>−.037 [−0.082, −0.001]</td><td>5.6% [0.15%, 12.4%]</td></tr></tbody></table> </ephtml> </p> <p>3 β<subs>total</subs> represents the MR effect of household income on ADHD, α represents the MR effect of household income on insula_SA, and β<subs>2</subs> represents the MR effect of insula_SA on ADHD adjusted for household income, β<subs>indirect</subs> is obtained by multiplying α by β<subs>2</subs>. All effects are obtained using the IVW method. <emph>SE</emph> = standard error; CI = confidence interval; IVW = inverse variance weighted.</p> <hd id="AN0192584782-23">Discussion</hd> <p>In this study, we endeavored to investigate whether brain structure is a mediator between SES and ADHD, considering a genetic perspective. We initially confirmed a protective causal effect of SES indicators on ADHD, with effect sizes aligning closely with the previous study. Then, our results demonstrated two specific brain regions exhibited significant susceptibility to SES, including insula and entorhinal cortex, with the SA of insula identified as a partial mediator in the causal pathway between household income and ADHD using two-step MR and MVMR methods, albeit with a proportion of mediation that wasn't notably large. Additionally, the UVMR results provided compelling evidence of a causal link between the SA of total brain and the TH of the lateral occipital with ADHD.</p> <p>The most noteworthy point of this study lay in finding the mediating role of brain structure between SES and ADHD. We suggested that the SA of insula was a mediator between household income and ADHD. The previous study concluded that increases in local gyrification index and surface area expansion are more effective in enhancing brain connectivity and functional development compared to increases in thickness ([<reflink idref="bib32" id="ref66">32</reflink>]). The insula, situated deep within the lateral sulcus of the brain, plays a pivotal role in a myriad of brain functions, encompassing social-emotional processing, language, and complex higher cognitive activities. Numerous existing studies have also highlighted the relationship between the insula and SES. Observational investigations have explored the relationship between insula structure and SES, such as cortical thickness ([<reflink idref="bib60" id="ref67">60</reflink>]), cortical surface area ([<reflink idref="bib54" id="ref68">54</reflink>]), and gray matter volume ([<reflink idref="bib34" id="ref69">34</reflink>]) of the insula, all of which have shown a significant positive correlation with SES. Similarly, one study delved into how the income-needs ratio (INR) relates to the topology of the brain's structural network in childhood, revealing that lower INR was associated with network inefficiency in insula ([<reflink idref="bib37" id="ref70">37</reflink>]). Our findings are consistent with a recent work ([<reflink idref="bib55" id="ref71">55</reflink>]), who reported a positive association between parental SES and insula SA in a large pediatric sample from the ABCD cohort. Our results further revealed a causal effect of household income on the SA of insula, indicating that the insula was particularly sensitive to SES disparities. On the other hand, one study pinpointed the ventral-lateral cortex, limbic regions, and insula volumes as key indicators for identifying ADHD patients ([<reflink idref="bib41" id="ref72">41</reflink>]). Existing study using MRI and artificial intelligence to identify ADHD have found that the trained model achieved optimal accuracy when weighing the reduced volumes of the inferior frontal cortex, bilateral sensorimotor cortex, and insula ([<reflink idref="bib75" id="ref73">75</reflink>]). Extending these structural insights, a study ([<reflink idref="bib57" id="ref74">57</reflink>]) demonstrated altered insula and subcortical functional connectivity in youths with ADHD, supporting the role of the insula not only as a structural but also functional node implicated in ADHD pathophysiology. The convergence of these findings with our study underscored the specificity of insula alterations in identifying individuals with ADHD.</p> <p>It has been suggested that neural responses to threat and stress, emotion regulation may be mechanisms through which SES affects health ([<reflink idref="bib52" id="ref75">52</reflink>]), meanwhile, the insula was reported to have related functions. Initially, individuals from low SES might experience heightened exposure to stressful events, rendering them more susceptible to various stress-related effects. During stress-inducing tasks, threat-related brain regions like the insula were activated ([<reflink idref="bib53" id="ref76">53</reflink>]), and the insula was a transmitter of information for stress-triggered neural circuits ([<reflink idref="bib72" id="ref77">72</reflink>]), with prolonged activation of these circuits potentially exerting detrimental effects on health through physiological pathways associated with the autonomic nervous system (ANS), the hypothalamic-pituitary-adrenal (HPA) axis, and inflammatory processes. In addition, emotion dysregulation is a common symptom that accompanies ADHD ([<reflink idref="bib15" id="ref78">15</reflink>]). The anterior insula, recognized for its crucial role in emotional experience and subjective feelings, exhibited activation in response to various emotional stimuli ([<reflink idref="bib71" id="ref79">71</reflink>]). One study also found that lower childhood SES was associated with reduced insula activation during the emotion regulation task in adults ([<reflink idref="bib38" id="ref80">38</reflink>]). Furthermore, executive function (EF) was closely linked to ADHD, and [<reflink idref="bib2" id="ref81">2</reflink>] argued that EF deficits were a central cause of ADHD because impaired EF led to dysfunctional self-control and goal behaviors in ADHD patients. Meanwhile, existing research has found deficits in EF among the poor throughout their lives ([<reflink idref="bib26" id="ref82">26</reflink>]). Response inhibition, which refers to the ability to inhibit or restrain one's behavior, was an integral component of EF ([<reflink idref="bib42" id="ref83">42</reflink>]). Impulsive behavior in ADHD patients was a manifestation of impaired response inhibition ([<reflink idref="bib23" id="ref84">23</reflink>]), with studies revealing bilateral under-activation of the insula in these individuals during response inhibition tasks utilizing function MRI ([<reflink idref="bib56" id="ref85">56</reflink>]). The insula, a constituent part of the "salience network" ([<reflink idref="bib47" id="ref86">47</reflink>]), serves as a key hub for information integration and coordination of the brain network state, and the previous study considered insula as the gatekeeper of executive control ([<reflink idref="bib50" id="ref87">50</reflink>]). Thus, alterations in the structure or function of the insula will harm human health, we believe that household income may partially impact ADHD through insula in three areas: stress, emotion regulation, and executive function. Cognitive training enhances interoceptive accuracy by strengthening resting-state functional connectivity (rsFC) between the anterior insular cortex (AIC) – a key interoceptive hub – and the dorsolateral prefrontal cortex (DLPFC), supramarginal gyrus (SMG), and anterior cingulate cortex (ACC). The AIC-DLPFC-SMG circuit, involved in top-down interoceptive modulation and cognitive control of emotion, highlights the therapeutic potential of targeting insula processing for ADHD intervention or prevention ([<reflink idref="bib69" id="ref88">69</reflink>]).</p> <p>Although the mediation ratio is only 5.6%, our study revealed the mediating role of insula for household income and ADHD from a genetic perspective, and provided a new understanding of the pathogenesis of ADHD and the effect of SES on health. However, much of the effect of SES on ADHD remained unclear mechanistically, warranting further exploration in future studies. To our surprise, several brain regions, mainly the frontal lobes ([<reflink idref="bib4" id="ref89">4</reflink>]; [<reflink idref="bib54" id="ref90">54</reflink>]), exhibited significant associations with both SES and ADHD in the observational study but did not act as mediators in the progression from SES to ADHD. The reason may be that we used the conservative Bonferroni correction for multiple testing or insufficient statistical power. Consequently, some previously reported significant relationships may not have remained in our study or would have required a larger sample size.</p> <p>In addition, our results suggested that abnormalities in brain structure may be influencing the prevalence of ADHD. We identified that the SA of total brain and the TH of lateral occipital have inverse and positive causal effects on the risk of ADHD prevalence, respectively. The generalized reduction in total brain volume in ADHD patients has been reported in many articles ([<reflink idref="bib14" id="ref91">14</reflink>]; [<reflink idref="bib73" id="ref92">73</reflink>]), and the total brain surface area correlated inversely with the Child Behavior Checklist (CBCL) syndrome scale attention problems score ([<reflink idref="bib33" id="ref93">33</reflink>]), which was consistent with our finding. The change in cortical thickness in patients with ADHD is a controversial topic. Previous studies have consistently observed reduced cortical thickness in most brain regions among ADHD patients ([<reflink idref="bib45" id="ref94">45</reflink>]). Another study reported that patients with ADHD decreased cortical thickness primarily in the frontoparietal region, while increased cortical thickness was observed mainly in the occipital lobe ([<reflink idref="bib1" id="ref95">1</reflink>]). Our results provided new evidence that every SD increase in cortical thickness of the lateral occipital caused the increased risk of ADHD prevalence by 54% from a genetic perspective.</p> <p>To the best of our knowledge, this study represented the first attempt to elucidate the role of brain surface area and thickness in mediating the relationship between SES and ADHD using MR methods. This work had several strengths. First, compared with traditional mediation analysis, MR offers the advantage of overcoming some of the stringent assumptions necessary for causal inference in mediation analysis, such as unmeasured confounders and measurement error. Moreover, the causal effect between all three can be examined. Second, the study utilized a large sample size with minimal sample overlap (approximately 0.7%) among the three phenotype datasets, thereby increasing statistical power and reducing bias associated with overlapping samples. Third, we screened for mediating brain phenotypes to minimize the reverse causality of brain on SES, thereby ensuring the rationality of the mediator effect explanatory model.</p> <p>Admittedly, there were some limitations in our study. Firstly, SES were self-reported indicators, which may be susceptible to selection and reporting bias due to questionnaire quality. Secondly, interaction and nonlinearity were not taken into account because both two-step MR and MVMR assumed that there was no interaction between the exposure and the mediator. Future research should explore the role of mediation in the presence of interaction effects between exposure and mediator. Thirdly, despite conducting several sensitivity analyses to detect pleiotropy, including heterogeneity analysis, directional horizontal pleiotropy test, leave-one-out analysis, and examining the directional consistency across the five methods (while the weighted median, weighted mode, and MR-Egger methods did not all reach statistical significance, the direction of the effect were generally consistent with the IVW methods, suggesting nonsignificance may reflect limited power), IV pleiotropy cannot be completely excluded, which may affect the robustness of the results. Fourthly, our study used European data, considering that populations of different races have different linkage disequilibrium structures and allele frequencies. Thus, additional validation was required to extend the findings to other races. Fifthly, considering the limited sample size of brain image GWAS from European children or adolescents, we utilized GWAS data on SES and brain images from adults, which may differ from GWAS data on SES and brain images from children ([<reflink idref="bib9" id="ref96">9</reflink>]) and may cause collider bias. Recent observational study ([<reflink idref="bib55" id="ref97">55</reflink>]) with childhood data revealed that parental SES positively correlates with insula surface area. The finding suggested that SES-related genetic liability may have influenced brain development from early in life. To strengthen causal inference, future validation using longitudinal cohorts such as ABCD or Generation R could help clarify how early-life SES shapes brain development and ADHD onset, addressing limitations of adult GWAS and MR-based designs. Lastly, the present study only considered limited brain cortical structure factors as a mediator between SES and ADHD, and future studies should explore the roles of other brain and non-brain factors to fully explain the mediating effect between SES and ADHD.</p> <hd id="AN0192584782-24">Conclusions</hd> <p>In summary, this study, from a genetic perspective, described the causal relationship between three SES-related indicators and ADHD, outlined the SA of insula as a causal mediator in the effect of household income on ADHD, and shed light on the relationship between the SA of total brain, the TH of lateral occipital and the risk of ADHD. This study added causal evidence to the etiology of ADHD and provided direction for mediating ADHD prevention and intervention goals through SES-related indicators.</p> <hd id="AN0192584782-25">Supplemental Material</hd> <p>Graph: Supplemental material, sj-docx-1-jad-10.1177_10870547251385350 for Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis by Wanyu Xie, Jie Yu and Ping Wang in Journal of Attention Disorders</p> <hd id="AN0192584782-26">Supplemental Material</hd> <p>Graph: Supplemental material, sj-xlsx-2-jad-10.1177_10870547251385350 for Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis by Wanyu Xie, Jie Yu and Ping Wang in Journal of Attention Disorders</p> <p>We gratefully acknowledged the authors and participants of all GWASs from which we used summary statistics data. We were extremely grateful to Prof. Grasby, Prof. Medland, and Prof. Stein for their assistance with the data of brain phenotypes.</p> <ref id="AN0192584782-27"> <title> References </title> <blist> <bibl id="bib1" idref="ref41" type="bt">1</bibl> <bibtext> Almeida Montes L. G., Prado Alcántara H., Martínez García R. B., De La Torre L. B., Ávila Acosta D., Duarte M. G. (2013). Brain cortical thickness in ADHD: Age, sex, and clinical correlations. Journal of Attention Disorders, 17, 641–654. https://doi.org/10.1177/1087054711434351</bibtext> </blist> <blist> <bibl id="bib2" idref="ref42" type="bt">2</bibl> <bibtext> Barkley R. A. (1997). Behavioral inhibition, sustained attention, and executive functions: Constructing a unifying theory of ADHD. Psychological Bulletin, 121(1), 65–94.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref35" type="bt">3</bibl> <bibtext> Bearden C. E., Thompson P. M. (2017). Emerging global initiatives in neurogenetics: The Enhancing Neuroimaging Genetics through Meta-analysis (ENIGMA) Consortium. Neuron, 94, 232–236. https://doi.org/10.1016/j.neuron.2017.03.033</bibtext> </blist> <blist> <bibl id="bib4" idref="ref49" type="bt">4</bibl> <bibtext> Biederman J. (2005). Attention-deficit/hyperactivity disorder: A selective overview. Biological Psychiatry, 57, 1215–1220. https://doi.org/10.1016/j.biopsych.2004.10.020</bibtext> </blist> <blist> <bibl id="bib5" idref="ref50" type="bt">5</bibl> <bibtext> Bowden J., Davey Smith G., Burgess S. (2015). Mendelian randomization with invalid instruments: Effect estimation and bias detection through Egger regression. International Journal of Epidemiology, 44, 512–525. https://doi.org/10.1093/ije/dyv080</bibtext> </blist> <blist> <bibl id="bib6" idref="ref55" type="bt">6</bibl> <bibtext> Bowden J., Davey Smith G., Haycock P. C., Burgess S. (2016). Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genetic Epidemiology, 40, 304–314. https://doi.org/10.1002/gepi.21965</bibtext> </blist> <blist> <bibl id="bib7" idref="ref60" type="bt">7</bibl> <bibtext> Bowden J., Del Greco M F., Minelli C., Davey Smith G., Sheehan N., Thompson J. (2017). A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization. Statistics in Medicine, 36, 1783–1802. https://doi.org/10.1002/sim.7221</bibtext> </blist> <blist> <bibl id="bib8" idref="ref51" type="bt">8</bibl> <bibtext> Bowden J., Spiller W., Del Greco M F., Sheehan N., Thompson J., Minelli C., Davey Smith G. (2018). Improving the visualization, interpretation and analysis of two-sample summary data Mendelian randomization via the Radial plot and Radial regression. International Journal of Epidemiology, 47, 1264–1278. https://doi.org/10.1093/ije/dyy101</bibtext> </blist> <blist> <bibl id="bib9" idref="ref96" type="bt">9</bibl> <bibtext> Brouwer R. M., Klein M., Grasby K. L., Schnack H. G., Jahanshad N., Teeuw J., Thomopoulos S. I., Sprooten E., Franz C. E., Gogtay N., Kremen W. S., Panizzon M. S., Loohuis L. M. O., Whelan C. D., Aghajani M., Alloza C., Alnæs D., Artiges E., Ayesa-Arriola R... Hulshoff Pol H. E. (2022). Genetic variants associated with longitudinal changes in brain structure across the lifespan. Nat Neurosci, 25, 421–432. https://doi.org/10.1038/s41593-022-01042-4</bibtext> </blist> <blist> <bibtext> Burgess S., Butterworth A., Thompson S. G. (2013). Mendelian randomization analysis with multiple genetic variants using summarized data. Genetic Epidemiology, 37, 658–665. https://doi.org/10.1002/gepi.21758</bibtext> </blist> <blist> <bibtext> Burgess S., Daniel R. M., Butterworth A. S., Thompson S. G., & the EPIC-InterAct Consortium. (2015). Network Mendelian randomization: Using genetic variants as instrumental variables to investigate mediation in causal pathways. International Journal of Epidemiology, 44(2), 484–495. https://doi.org/10.1093/ije/dyu176</bibtext> </blist> <blist> <bibtext> Carter A. R., Sanderson E., Hammerton G., Richmond R. C., Davey Smith G., Heron J., Taylor A. E., Davies N. M., Howe L. D. (2021). Mendelian randomisation for mediation analysis: current methods and challenges for implementation. European Journal of Epidemiology, 36, 465–478. https://doi.org/10.1007/s10654-021-00757-1</bibtext> </blist> <blist> <bibtext> Casswell S., Pledger M., Hooper R. (2003). Socioeconomic status and drinking patterns in young adults. Addiction, 98, 601–610. https://doi.org/10.1046/j.1360-0443.2003.00331.x</bibtext> </blist> <blist> <bibtext> Castellanos F. X. (2002). Developmental trajectories of brain volume abnormalities in children and adolescents with attention-deficit/hyperactivity disorder. JAMA, 288(14), 1740–1748. https://doi.org/10.1001/jama.288.14.1740</bibtext> </blist> <blist> <bibtext> Christiansen H., Hirsch O., Albrecht B., Chavanon M. -L. (2019). Attention-deficit/hyperactivity disorder (ADHD) and emotion regulation over the life span. Current Psychiatry Reports, 21, 17. https://doi.org/10.1007/s11920-019-1003-6</bibtext> </blist> <blist> <bibtext> Demontis D., Walters G. B., Athanasiadis G., Walters R., Therrien K., Nielsen T. T., Farajzadeh L., Voloudakis G., Bendl J., Zeng B., Zhang W., Grove J., Als T. D., Duan J., Satterstrom F. K., Bybjerg-Grauholm J., Bækved-Hansen M., Gudmundsson O. O., Magnusson S. H... Børglum A. D. (2023). Genome-wide analyses of ADHD identify 27 risk loci, refine the genetic architecture and implicate several cognitive domains. Nature Genetics, 55, 198–208. https://doi.org/10.1038/s41588-022-01285-8</bibtext> </blist> <blist> <bibtext> Elsworth B., Lyon M., Alexander T., Liu Y., Matthews P., Hallett J., Bates P., Palmer T., Haberland V., Smith G. D., Zheng J., Haycock P., Gaunt T. R., Hemani G. (2020). The MRC IEU OpenGWAS data infrastructure [preprint]. Genetics. https://doi.org/10.1101/2020.08.10.244293</bibtext> </blist> <blist> <bibtext> Farah M. J. (2017). The neuroscience of socioeconomic status: Correlates, causes, and consequences. Neuron, 96, 56–71. https://doi.org/10.1016/j.neuron.2017.08.034</bibtext> </blist> <blist> <bibtext> Faraone S. V., Asherson P., Banaschewski T., Biederman J., Buitelaar J. K., Ramos-Quiroga J. A., Rohde L. A., Sonuga-Barke E. J. S., Tannock R., Franke B. (2015). Attention-deficit/hyperactivity disorder. Nature Reviews Disease Primers, 1, Article 15020. https://doi.org/10.1038/nrdp.2015.20</bibtext> </blist> <blist> <bibtext> Faraone S. V., Biederman J., Mick E. (2006). The age-dependent decline of attention deficit hyperactivity disorder: A meta-analysis of follow-up studies. Psychological Medicine, 36, 159–165. https://doi.org/10.1017/S003329170500471X</bibtext> </blist> <blist> <bibtext> Faraone S. V., Perlis R. H., Doyle A. E., Smoller J. W., Goralnick J. J., Holmgren M. A., Sklar P. (2005). Molecular genetics of attention-deficit/hyperactivity disorder. Biological Psychiatry, 57, 1313–1323. https://doi.org/10.1016/j.biopsych.2004.11.024</bibtext> </blist> <blist> <bibtext> Fernald A., Marchman V. A., Weisleder A. (2013). SES differences in language processing skill and vocabulary are evident at 18 months. Developmental Science, 16, 234–248. https://doi.org/10.1111/desc.12019</bibtext> </blist> <blist> <bibtext> Fineberg N. A., Chamberlain S. R., Goudriaan A. E., Stein D. J., Vanderschuren L. J. M. J., Gillan C. M., Shekar S., Gorwood P. A. P. M., Voon V., Morein-Zamir S., Denys D., Sahakian B. J., Moeller F. G., Robbins T. W., Potenza M. N. (2014). New developments in human neurocognition: Clinical, genetic, and brain imaging correlates of impulsivity and compulsivity. CNS Spectrums, 19, 69–89. https://doi.org/10.1017/S1092852913000801</bibtext> </blist> <blist> <bibtext> Grasby K. L., Jahanshad N., Painter J. N., Colodro-Conde L., Bralten J., Hibar D. P., Lind P. A., Pizzagalli F., Ching C. R. K., McMahon M. A. B., Shatokhina N., Zsembik L. C. P., Thomopoulos S. I., Zhu A. H., Strike L. T., Agartz I., Alhusaini S., Almeida M. A. A., Alnæs D... Medland S. E., & Enhancing NeuroImaging Genetics through Meta-Analysis Consortium (ENIGMA)—Genetics working group. (2020). The genetic architecture of the human cerebral cortex. Science, 367, eaay6690. https://doi.org/10.1126/science.aay6690</bibtext> </blist> <blist> <bibtext> Gudjonsson G. H., Sigurdsson J. F., Sigfusdottir I. D., Young S. (2012). An epidemiological study of ADHD symptoms among young persons and the relationship with cigarette smoking, alcohol consumption and illicit drug use. Child Psychology Psychiatry, 53, 304–312. https://doi.org/10.1111/j.1469-7610.2011.02489.x</bibtext> </blist> <blist> <bibtext> Hair N. L., Hanson J. L., Wolfe B. L., Pollak S. D. (2015). Association of child poverty, brain development, and academic achievement. JAMA Pediatrics, 169, 822. https://doi.org/10.1001/jamapediatrics.2015.1475</bibtext> </blist> <blist> <bibtext> Hartwig F. P., Davey Smith G., Bowden J. (2017). Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. International Journal of Epidemiology, 46, 1985–1998. https://doi.org/10.1093/ije/dyx102</bibtext> </blist> <blist> <bibtext> Hemani G., Tilling K., Davey Smith G. (2017). Orienting the causal relationship between imprecisely measured traits using GWAS summary data. PLoS Genet, 13, Article e1007081. https://doi.org/10.1371/journal.pgen.1007081</bibtext> </blist> <blist> <bibtext> Hemani G., Zheng J., Elsworth B., Wade K. H., Haberland V., Baird D., Laurin C., Burgess S., Bowden J., Langdon R., Tan V. Y., Yarmolinsky J., Shihab H. A., Timpson N. J., Evans D. M., Relton C., Martin R. M., Davey Smith G., Gaunt T. R... Haycock P. C. (2018). The MR-Base platform supports systematic causal inference across the human phenome. eLife, 7, Article e34408. https://doi.org/10.7554/eLife.34408</bibtext> </blist> <blist> <bibtext> Hill W. D., Hagenaars S. P., Marioni R. E., Harris S. E., Liewald D. C. M., Davies G., Okbay A., McIntosh A. M., Gale C. R., Deary I. J. (2016). Molecular genetic contributions to social deprivation and household income in UK Biobank. Current Biology, 26, 3083–3089. https://doi.org/10.1016/j.cub.2016.09.035</bibtext> </blist> <blist> <bibtext> Hiscock R., Bauld L., Amos A., Fidler J. A., Munafò M. (2012). Socioeconomic status and smoking: A review. Annals of the New York Academy of Sciences, 1248, 107–123. https://doi.org/10.1111/j.1749-6632.2011.06202.x</bibtext> </blist> <blist> <bibtext> Hogstrom L. J., Westlye L. T., Walhovd K. B., Fjell A. M. (2013). The structure of the cerebral cortex across adult life: Age-related patterns of surface area, thickness, and gyrification. Cerebral Cortex, 23, 2521–2530. https://doi.org/10.1093/cercor/bhs231</bibtext> </blist> <blist> <bibtext> Hoogman M., Muetzel R., Guimaraes J. P., Shumskaya E., Mennes M., Zwiers M. P., Jahanshad N., Sudre G., Wolfers T., Earl E. A., Soliva Vila J. C., Vives-Gilabert Y., Khadka S., Novotny S. E., Hartman C. A., Heslenfeld D. J., Schweren L. J. S., Ambrosino S., Oranje B... Franke B. (2019). Brain imaging of the cortex in ADHD: A coordinated analysis of large-scale clinical and population-based samples. American Journal of Psychiatry, 176, 531–542. https://doi.org/10.1176/appi.ajp.2019.18091033</bibtext> </blist> <blist> <bibtext> Jednoróg K., Altarelli I., Monzalvo K., Fluss J., Dubois J., Billard C., Dehaene-Lambertz G., Ramus F. (2012). The influence of socioeconomic status on children's brain structure. PLoS ONE, 7, Article e42486. https://doi.org/10.1371/journal.pone.0042486</bibtext> </blist> <blist> <bibtext> Jia H., Li Z., Guo F., Hua Z., Zhou X., Li X., Li R., Liu Q., Liu Y., Dong H. (2024). Cortical structure and the risk of amyotrophic lateral sclerosis: A bidirectional Mendelian randomization study. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 129, 110872. https://doi.org/10.1016/j.pnpbp.2023.110872</bibtext> </blist> <blist> <bibtext> Kamat M. A., Blackshaw J. A., Young R., Surendran P., Burgess S., Danesh J., Butterworth A. S., Staley J. R. (2019). PhenoScanner V2: An expanded tool for searching human genotype–phenotype associations. Bioinformatics, 35, 4851–4853. https://doi.org/10.1093/bioinformatics/btz469</bibtext> </blist> <blist> <bibtext> Kim D. -J., Davis E. P., Sandman C. A., Glynn L., Sporns O., O'Donnell B. F., Hetrick W. P. (2019). Childhood poverty and the organization of structural brain connectome. NeuroImage, 184, 409–416. https://doi.org/10.1016/j.neuroimage.2018.09.041</bibtext> </blist> <blist> <bibtext> Kim P., Evans G. W., Angstadt M., Ho S. S., Sripada C. S., Swain J. E., Liberzon I., Phan K. L. (2013). Effects of childhood poverty and chronic stress on emotion regulatory brain function in adulthood. Proceedings of the National Academy of Sciences of the United States of America, 110, 18442–18447. https://doi.org/10.1073/pnas.1308240110</bibtext> </blist> <blist> <bibtext> Lawlor D. A., Harbord R. M., Sterne J. A. C., Timpson N., Davey Smith G. (2008). Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology. Statistics in Medicine, 27, 1133–1163. https://doi.org/10.1002/sim.3034</bibtext> </blist> <blist> <bibtext> Lawson G. M., Hook C. J., Farah M. J. (2018). A meta-analysis of the relationship between socioeconomic status and executive function performance among children. Developmental Science, 21, Article e12529. https://doi.org/10.1111/desc.12529</bibtext> </blist> <blist> <bibtext> Lim L., Marquand A., Cubillo A. A., Smith A. B., Chantiluke K., Simmons A., Mehta M., Rubia K. (2013). Disorder-specific predictive classification of adolescents with attention deficit hyperactivity disorder (ADHD) relative to autism using structural magnetic resonance imaging. PLoS ONE, 8, Article e63660. https://doi.org/10.1371/journal.pone.0063660</bibtext> </blist> <blist> <bibtext> Logue S. F., Gould T. J. (2014). The neural and genetic basis of executive function: Attention, cognitive flexibility, and response inhibition. Pharmacology Biochemistry and Behavior, 123, 45–54. https://doi.org/10.1016/j.pbb.2013.08.007</bibtext> </blist> <blist> <bibtext> Mackey A. P., Finn A. S., Leonard J. A., Jacoby-Senghor D. S., West M. R., Gabrieli C. F. O., Gabrieli J. D. E. (2015). Neuroanatomical correlates of the income-achievement gap. Psychological Science, 26, 925–933. https://doi.org/10.1177/0956797615572233</bibtext> </blist> <blist> <bibtext> MacKinnon D. P., Fairchild A. J., Fritz M. S. (2007). Mediation analysis. Annual Review of Psychology, 58, 593–614. https://doi.org/10.1146/annurev.psych.58.110405.085542</bibtext> </blist> <blist> <bibtext> Makris N., Biederman J., Valera E. M., Bush G., Kaiser J., Kennedy D. N., Caviness V. S., Faraone S. V., Seidman L. J. (2007). Cortical thinning of the attention and executive function networks in adults with attention-deficit/hyperactivity disorder. Cerebral Cortex, 17, 1364–1375. https://doi.org/10.1093/cercor/bhl047</bibtext> </blist> <blist> <bibtext> Martens L. G., Van Hamersveld D., Le Cessie S., Willems Van Dijk K., Van Heemst D., Noordam R. (2023). The impact of sociodemographic status on the association of classical cardiovascular risk factors with coronary artery disease: A stratified Mendelian randomization study. Journal of Clinical Epidemiology, 162, 56–62. https://doi.org/10.1016/j.jclinepi.2023.07.009</bibtext> </blist> <blist> <bibtext> Menon V., Uddin L. Q. (2010). Saliency, switching, attention and control: A network model of insula function. Brain Structure and Function, 214, 655–667. https://doi.org/10.1007/s00429-010-0262-0</bibtext> </blist> <blist> <bibtext> Michaëlsson M., Yuan S., Melhus H., Baron J. A., Byberg L., Larsson S. C., Michaëlsson K. (2022). The impact and causal directions for the associations between diagnosis of ADHD, socioeconomic status, and intelligence by use of a bi-directional two-sample Mendelian randomization design. BMC Medicine, 20, 106. https://doi.org/10.1186/s12916-022-02314-3</bibtext> </blist> <blist> <bibtext> Miller L. L., Gustafsson H. C., Tipsord J., Song M., Nousen E., Dieckmann N., Nigg J. T. (2018). Is the association of ADHD with socio-economic disadvantage explained by child comorbid externalizing problems or parent ADHD? Journal of Abnormal Child Psychology, 46, 951–963. https://doi.org/10.1007/s10802-017-0356-8</bibtext> </blist> <blist> <bibtext> Molnar-Szakacs I., Uddin L. Q. (2022). Anterior insula as a gatekeeper of executive control. Neuroscience & Biobehavioral Reviews, 139, Article 104736. https://doi.org/10.1016/j.neubiorev.2022.104736</bibtext> </blist> <blist> <bibtext> Mowinckel A. M., Vidal-Piñeiro D. (2020). Visualization of brain statistics with R packages ggseg and ggseg3d. Advances in Methods and Practices in Psychological Science, 3, 466–483. https://doi.org/10.1177/2515245920928009</bibtext> </blist> <blist> <bibtext> Muscatell K. A. (2018). Socioeconomic influences on brain function: Implications for health. Annals of the New York Academy of Sciences, 1428, 14–32. https://doi.org/10.1111/nyas.13862</bibtext> </blist> <blist> <bibtext> Muscatell K. A., Eisenberger N. I. (2012). A social neuroscience perspective on stress and health. Social and Personality Psychology Compass, 6, 890–904. https://doi.org/10.1111/j.1751-9004.2012.00467.x</bibtext> </blist> <blist> <bibtext> Noble K. G., Houston S. M., Brito N. H., Bartsch H., Kan E., Kuperman J. M., Akshoomoff N., Amaral D. G., Bloss C. S., Libiger O., Schork N. J., Murray S. S., Casey B. J., Chang L., Ernst T. M., Frazier J. A., Gruen J. R., Kennedy D. N., Van Zijl P... Sowell E. R. (2015). Family income, parental education and brain structure in children and adolescents. Nature Neuroscience, 18, 773–778. https://doi.org/10.1038/nn.3983</bibtext> </blist> <blist> <bibtext> Norbom L. B., Rokicki J., Eilertsen E. M., Wiker T., Hanson J., Dahl A., Alnæs D., Fernández-Cabello S., Beck D., Agartz I., Andreassen O. A., Westlye L. T., Tamnes C. K. (2024). Parental education and income are linked to offspring cortical brain structure and psychopathology at 9–11 years. JCPP Advances, 4, Article e12220. https://doi.org/10.1002/jcv2.12220</bibtext> </blist> <blist> <bibtext> Norman L. J., Carlisi C., Lukito S., Hart H., Mataix-Cols D., Radua J., Rubia K. (2016). Structural and functional brain abnormalities in attention-deficit/hyperactivity disorder and obsessive-compulsive disorder: A comparative meta-analysis. JAMA Psychiatry, 73, 815–825. https://doi.org/10.1001/jamapsychiatry.2016.0700</bibtext> </blist> <blist> <bibtext> Norman L. J., Sudre G., Price J., Shaw P. (2024). Subcortico-cortical dysconnectivity in ADHD: A voxel-wise mega-analysis across multiple cohorts. American Journal of Psychiatry, 181, 553–562. https://doi.org/10.1176/appi.ajp.20230026</bibtext> </blist> <blist> <bibtext> Palmer T. M., Lawlor D. A., Harbord R. M., Sheehan N. A., Tobias J. H., Timpson N. J., Smith G. D., Sterne J. A. (2012). Using multiple genetic variants as instrumental variables for modifiable risk factors. Statistical Methods in Medical Research, 21, 223–242. https://doi.org/10.1177/0962280210394459</bibtext> </blist> <blist> <bibtext> Raver C. C., Blair C., Willoughby M. (2013). Poverty as a predictor of 4-year-olds' executive function: New perspectives on models of differential susceptibility. Developmental Psychology, 49, 292–304. https://doi.org/10.1037/a0028343</bibtext> </blist> <blist> <bibtext> Romeo R. R., Christodoulou J. A., Halverson K. K., Murtagh J., Cyr A. B., Schimmel C., Chang P., Hook P. E., Gabrieli J. D. E. (2018). Socioeconomic status and reading disability: Neuroanatomy and plasticity in response to intervention. Cerebral Cortex, 28, 2297–2312. https://doi.org/10.1093/cercor/bhx131</bibtext> </blist> <blist> <bibtext> Rowland A. S., Skipper B. J., Rabiner D. L., Qeadan F., Campbell R. A., Naftel A. J., Umbach D. M. (2018). Attention-deficit/hyperactivity disorder (ADHD): Interaction between socioeconomic status and parental history of ADHD determines prevalence. Journal of Child Psychology and Psychiatry, 59, 213–222. https://doi.org/10.1111/jcpp.12775</bibtext> </blist> <blist> <bibtext> Russell A. E., Ford T., Russell G. (2015). Socioeconomic associations with ADHD: Findings from a Mediation analysis. PLoS ONE, 10, Article e0128248. https://doi.org/10.1371/journal.pone.0128248</bibtext> </blist> <blist> <bibtext> Russell A. E., Ford T., Williams R., Russell G. (2016). The association between socioeconomic disadvantage and attention deficit/hyperactivity disorder (ADHD): A systematic review. Child Psychiatry & Human Development, 47, 440–458. https://doi.org/10.1007/s10578-015-0578-3</bibtext> </blist> <blist> <bibtext> Sanderson E. (2021). Multivariable Mendelian randomization and Mediation. Cold Spring Harbor Perspectives in Medicine, 11, a038984. https://doi.org/10.1101/cshperspect.a038984</bibtext> </blist> <blist> <bibtext> Sanderson E., Davey Smith G., Windmeijer F., Bowden J. (2019). An examination of multivariable Mendelian randomization in the single-sample and two-sample summary data settings. International Journal of Epidemiology, 48, 713–727. https://doi.org/10.1093/ije/dyy262</bibtext> </blist> <blist> <bibtext> Sanderson E., Spiller W., Bowden J. (2021). Testing and correcting for weak and pleiotropic instruments in two-sample multivariable Mendelian randomization. Statistics in Medicine, 40, 5434–5452. https://doi.org/10.1002/sim.9133</bibtext> </blist> <blist> <bibtext> Silk T. J., Beare R., Malpas C., Adamson C., Vilgis V., Vance A., Bellgrove M. A. (2016). Cortical morphometry in attention-deficit hyperactivity disorder: Contribution of thickness and surface area to volume. Cortex, 82, 1–10. https://doi.org/10.1016/j.cortex.2016.05.012</bibtext> </blist> <blist> <bibtext> Skrivankova V. W., Richmond R. C., Woolf B. A. R., Yarmolinsky J., Davies N. M., Swanson S. A., VanderWeele T. J., Higgins J. P. T., Timpson N. J., Dimou N., Langenberg C., Golub R. M., Loder E. W., Gallo V., Tybjaerg-Hansen A., Davey Smith G., Egger M., Richards J. B. (2021). Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: The STROBE-MR statement. JAMA, 326, 1614. https://doi.org/10.1001/jama.2021.18236</bibtext> </blist> <blist> <bibtext> Sugawara A., Terasawa Y., Katsunuma R., Sekiguchi A. (2020). Effects of interoceptive training on decision making, anxiety, and somatic symptoms. BioPsychoSocial Medicine, 14, 7. https://doi.org/10.1186/s13030-020-00179-7</bibtext> </blist> <blist> <bibtext> Turrell G., Lynch J. W., Kaplan G. A., Everson S. A., Helkala E. -L., Kauhanen J., Salonen J. T. (2002). Socioeconomic position across the lifecourse and cognitive function in late middle age. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 57, S43–S51. https://doi.org/10.1093/geronb/57.1.S43</bibtext> </blist> <blist> <bibtext> Uddin L. Q., Nomi J. S., Hebert-Seropian B., Ghaziri J., Boucher O. (2017). Structure and function of the human insula. Journal of Clinical Neurophysiology, 34, 300–306. https://doi.org/10.1097/WNP.0000000000000377</bibtext> </blist> <blist> <bibtext> Ulrich-Lai Y. M., Herman J. P. (2009). Neural regulation of endocrine and autonomic stress responses. Nature Reviews Neuroscience, 10, 397–409. https://doi.org/10.1038/nrn2647</bibtext> </blist> <blist> <bibtext> Valera E. M., Faraone S. V., Murray K. E., Seidman L. J. (2007). Meta-analysis of structural imaging findings in attention-deficit/hyperactivity disorder. Biological Psychiatry, 61, 1361–1369. https://doi.org/10.1016/j.biopsych.2006.06.011</bibtext> </blist> <blist> <bibtext> Wu B. -S., Zhang Y. -R., Li H. -Q., Kuo K., Chen S. -D., Dong Q., Liu Y., Yu J.-T. (2021). Cortical structure and the risk for Alzheimer's disease: A bidirectional Mendelian randomization study. Translational Psychiatry, 11, 476. https://doi.org/10.1038/s41398-021-01599-x</bibtext> </blist> <blist> <bibtext> Yoo J. H., Kim J. I., Kim B.-N., Jeong B. (2020). Exploring characteristic features of attention-deficit/hyperactivity disorder: Findings from multi-modal MRI and candidate genetic data. Brain Imaging and Behavior, 14, 2132–2147. https://doi.org/10.1007/s11682-019-00164-x</bibtext> </blist> <blist> <bibtext> Zhao Q., Wang J., Hemani G., Bowden J., Small D. S. (2019). Statistical inference in two-sample summary-data Mendelian randomization using robust adjusted profile score. <ulink href="http://arxiv.org/abs/1801.09652">http://arxiv.org/abs/1801.09652</ulink></bibtext> </blist> <blist> <bibtext> Zhu Z., Zheng Z., Zhang F., Wu Y., Trzaskowski M., Maier R., Robinson M. R., McGrath J. J., Visscher P. M., Wray N. R., Yang J. (2018). Causal associations between risk factors and common diseases inferred from GWAS summary data. Nature Communications, 9, 224. https://doi.org/10.1038/s41467-017-02317-2</bibtext> </blist> </ref> <ref id="AN0192584782-28"> <title> Footnotes </title> <blist> <bibtext> Jie Yu is now affiliated to Department of Magnetic Resonance, Henan Provincial Hospital of Traditional Chinese Medicine (The Second Affiliated Hospital of Henan University of Chinese Medicine), Zhengzhou, Henan Province, China.</bibtext> </blist> <blist> <bibtext> Wanyu Xie</bibtext> </blist> <blist> <bibtext>Graph</bibtext> </blist> <blist> <bibtext>https://orcid.org/0009-0004-5361-9971 Ping Wang</bibtext> </blist> <blist> <bibtext>Graph https://orcid.org/0009-0002-3406-6713</bibtext> </blist> <blist> <bibtext> Not applicable. This MR study used the de-identified summary-level data that have been made publicly available. The informed consent and ethical approval had been obtained in all original GWAS studies, thus this study did not require additional ethical approval.</bibtext> </blist> <blist> <bibtext> Not applicable.</bibtext> </blist> <blist> <bibtext> Not applicable.</bibtext> </blist> <blist> <bibtext> Wanyu Xie: Writing – original draft, Investigation, Methodology, Data curation, Formal analysis, and Visualization. Jie Yu: Writing – review & editing, Methodology, and Validation. Ping Wang: Writing – review & editing, Supervision, and Funding acquisition.</bibtext> </blist> <blist> <bibtext> The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the scientific research project of Tianjin municipal education commission [grant number No. 2021KJ264].</bibtext> </blist> <blist> <bibtext> The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.</bibtext> </blist> <blist> <bibtext> The dataset of ADHD analyzed during the current study are obtained from https://pgc.unc.edu/. The summary-level datasets of the human brain structure used during the study are available from the ENIGMA Consortium website (<ulink href="http://enigma.ini.usc.edu/research/download-enigma-gwas-results/">http://enigma.ini.usc.edu/research/download-enigma-gwas-results/</ulink>). The datasets of the SES-related indictors analyzed in the current study are available from https://gwas.mrcieu.ac.uk/.</bibtext> </blist> <blist> <bibtext> Supplemental material for this article is available online.</bibtext> </blist> </ref> <aug> <p>By Wanyu Xie; Jie Yu and Ping Wang</p> <p>Reported by Author; Author; Author</p> <p></p> <p>Wanyu Xie is a researcher at the School of Medical Imaging, Division of Medical Technology, Tianjin Medical University, specializing in imaging genetics of neuropsychiatric diseases.</p> <p>Jie Yu is a researcher in the Department of Magnetic Resonance at Henan Provincial Hospital of Traditional Chinese Medicine, Henan, China, where she conducts research on brain MRI.</p> <p>Ping Wang is an Associate Professor in the School of Medical Imaging, Division of Medical Technology, Tianjin Medical University, Tianjin, China.</p> </aug> <nolink nlid="nl1" bibid="bib19" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib20" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib21" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib49" firstref="ref5"></nolink> <nolink nlid="nl5" bibid="bib61" firstref="ref6"></nolink> <nolink nlid="nl6" bibid="bib62" firstref="ref7"></nolink> <nolink nlid="nl7" bibid="bib63" firstref="ref8"></nolink> <nolink nlid="nl8" bibid="bib18" firstref="ref9"></nolink> <nolink nlid="nl9" bibid="bib30" firstref="ref10"></nolink> <nolink nlid="nl10" bibid="bib48" firstref="ref11"></nolink> <nolink nlid="nl11" bibid="bib70" firstref="ref12"></nolink> <nolink nlid="nl12" bibid="bib22" firstref="ref13"></nolink> <nolink nlid="nl13" bibid="bib40" firstref="ref14"></nolink> <nolink nlid="nl14" bibid="bib59" firstref="ref15"></nolink> <nolink nlid="nl15" bibid="bib26" firstref="ref16"></nolink> <nolink nlid="nl16" bibid="bib43" firstref="ref17"></nolink> <nolink nlid="nl17" bibid="bib54" firstref="ref18"></nolink> <nolink nlid="nl18" bibid="bib34" firstref="ref19"></nolink> <nolink nlid="nl19" bibid="bib33" firstref="ref20"></nolink> <nolink nlid="nl20" bibid="bib56" firstref="ref21"></nolink> <nolink nlid="nl21" bibid="bib39" firstref="ref25"></nolink> <nolink nlid="nl22" bibid="bib12" firstref="ref26"></nolink> <nolink nlid="nl23" bibid="bib11" firstref="ref27"></nolink> <nolink nlid="nl24" bibid="bib64" firstref="ref28"></nolink> <nolink nlid="nl25" bibid="bib67" firstref="ref29"></nolink> <nolink nlid="nl26" bibid="bib68" firstref="ref32"></nolink> <nolink nlid="nl27" bibid="bib17" firstref="ref33"></nolink> <nolink nlid="nl28" bibid="bib46" firstref="ref34"></nolink> <nolink nlid="nl29" bibid="bib24" firstref="ref36"></nolink> <nolink nlid="nl30" bibid="bib74" firstref="ref37"></nolink> <nolink nlid="nl31" bibid="bib16" firstref="ref38"></nolink> <nolink nlid="nl32" bibid="bib77" firstref="ref39"></nolink> <nolink nlid="nl33" bibid="bib35" firstref="ref40"></nolink> <nolink nlid="nl34" bibid="bib36" firstref="ref43"></nolink> <nolink nlid="nl35" bibid="bib13" firstref="ref44"></nolink> <nolink nlid="nl36" bibid="bib25" firstref="ref45"></nolink> <nolink nlid="nl37" bibid="bib31" firstref="ref46"></nolink> <nolink nlid="nl38" bibid="bib28" firstref="ref48"></nolink> <nolink nlid="nl39" bibid="bib58" firstref="ref52"></nolink> <nolink nlid="nl40" bibid="bib65" firstref="ref53"></nolink> <nolink nlid="nl41" bibid="bib10" firstref="ref54"></nolink> <nolink nlid="nl42" bibid="bib27" firstref="ref56"></nolink> <nolink nlid="nl43" bibid="bib76" firstref="ref58"></nolink> <nolink nlid="nl44" bibid="bib44" firstref="ref59"></nolink> <nolink nlid="nl45" bibid="bib29" firstref="ref61"></nolink> <nolink nlid="nl46" bibid="bib66" firstref="ref62"></nolink> <nolink nlid="nl47" bibid="bib51" firstref="ref64"></nolink> <nolink nlid="nl48" bibid="bib32" firstref="ref66"></nolink> <nolink nlid="nl49" bibid="bib60" firstref="ref67"></nolink> <nolink nlid="nl50" bibid="bib37" firstref="ref70"></nolink> <nolink nlid="nl51" bibid="bib55" firstref="ref71"></nolink> <nolink nlid="nl52" bibid="bib41" firstref="ref72"></nolink> <nolink nlid="nl53" bibid="bib75" firstref="ref73"></nolink> <nolink nlid="nl54" bibid="bib57" firstref="ref74"></nolink> <nolink nlid="nl55" bibid="bib52" firstref="ref75"></nolink> <nolink nlid="nl56" bibid="bib53" firstref="ref76"></nolink> <nolink nlid="nl57" bibid="bib72" firstref="ref77"></nolink> <nolink nlid="nl58" bibid="bib15" firstref="ref78"></nolink> <nolink nlid="nl59" bibid="bib71" firstref="ref79"></nolink> <nolink nlid="nl60" bibid="bib38" firstref="ref80"></nolink> <nolink nlid="nl61" bibid="bib42" firstref="ref83"></nolink> <nolink nlid="nl62" bibid="bib23" firstref="ref84"></nolink> <nolink nlid="nl63" bibid="bib47" firstref="ref86"></nolink> <nolink nlid="nl64" bibid="bib50" firstref="ref87"></nolink> <nolink nlid="nl65" bibid="bib69" firstref="ref88"></nolink> <nolink nlid="nl66" bibid="bib14" firstref="ref91"></nolink> <nolink nlid="nl67" bibid="bib73" firstref="ref92"></nolink> <nolink nlid="nl68" bibid="bib45" firstref="ref94"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1501799
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Wanyu+Xie%22">Wanyu Xie</searchLink> (ORCID <externalLink term="https://orcid.org/0009-0004-5361-9971">0009-0004-5361-9971</externalLink>)<br /><searchLink fieldCode="AR" term="%22Jie+Yu%22">Jie Yu</searchLink><br /><searchLink fieldCode="AR" term="%22Ping+Wang%22">Ping Wang</searchLink> (ORCID <externalLink term="https://orcid.org/0009-0002-3406-6713">0009-0002-3406-6713</externalLink>)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Journal+of+Attention+Disorders%22"><i>Journal of Attention Disorders</i></searchLink>. 2026 30(5):615-628.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: https://sagepub.com
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 14
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2026
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Attention+Deficit+Hyperactivity+Disorder%22">Attention Deficit Hyperactivity Disorder</searchLink><br /><searchLink fieldCode="DE" term="%22Socioeconomic+Status%22">Socioeconomic Status</searchLink><br /><searchLink fieldCode="DE" term="%22Correlation%22">Correlation</searchLink><br /><searchLink fieldCode="DE" term="%22Brain+Hemisphere+Functions%22">Brain Hemisphere Functions</searchLink><br /><searchLink fieldCode="DE" term="%22Incidence%22">Incidence</searchLink><br /><searchLink fieldCode="DE" term="%22Family+Income%22">Family Income</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Europe%22">Europe</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1177/10870547251385350
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1087-0547<br />1557-1246
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Background: Research has consistently demonstrated a negative correlation between socioeconomic status (SES) and the prevalence of ADHD, with SES exerting a significant influence on brain development. ADHD, closely intertwined with neurological development, often manifests as impairments within brain regions associated with memory, executive function, and emotion regulation. Nevertheless, the specific brain structural mediators linking SES to ADHD remain unclear. Method: We explored whether the brain surface area (SA) and thickness (TH) mediated the relationship between SES indicators (Townsend deprivation index at recruitment, average total household income before tax, and job involves heavy manual or physical work) and ADHD utilizing two-step Mendelian Randomization (MR) and multivariate MR method. Results: The MR analysis indicated that higher SES corresponds to a lower prevalence of ADHD. Genetically predicted household income was positively correlated with the SA of insula (β = 0.31, p = 1.02 × 10⁴), and physical work was positively correlated with the TH of entorhinal cortex (β = 0.74, p = 3.73 × 10⁵). Mediation analysis showed that the SA of insula was identified as a partial mediator in the protective effect of household income against ADHD prevalence, with a mediation ratio of 5.6%. Concerning potential causal relationships between IDPs and ADHD, reduced total brain SA increased ADHD risk (OR = 0.77, p = 5.60 × 10⁹), while reduced the TH of lateral occipital was protective (OR = 1.54, p = 2.02 × 10⁴). Conclusions: SES influences ADHD through brain structural changes, offering insights for prevention and intervention strategies.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2026
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1501799
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1501799
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1177/10870547251385350
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 615
    Subjects:
      – SubjectFull: Attention Deficit Hyperactivity Disorder
        Type: general
      – SubjectFull: Socioeconomic Status
        Type: general
      – SubjectFull: Correlation
        Type: general
      – SubjectFull: Brain Hemisphere Functions
        Type: general
      – SubjectFull: Incidence
        Type: general
      – SubjectFull: Family Income
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Europe
        Type: general
    Titles:
      – TitleFull: Dissecting the Mediating Role of Cortical Structures in the Pathogenesis of Socioeconomic Status to ADHD: A Mendelian Randomization Study and Mediation Analysis
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Wanyu Xie
      – PersonEntity:
          Name:
            NameFull: Jie Yu
      – PersonEntity:
          Name:
            NameFull: Ping Wang
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 1087-0547
            – Type: issn-electronic
              Value: 1557-1246
          Numbering:
            – Type: volume
              Value: 30
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: Journal of Attention Disorders
              Type: main
ResultId 1