Mendelian Randomization Studies of Myopia: Choosing the right Summary Statistics

Mendelian Randomization Studies of Myopia: Choosing the right Summary Statistics | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Mendelian Randomization Studies of Myopia: Choosing the right Summary Statistics View ORCID Profile Thu-Nga Nguyen , View ORCID Profile Louise Terry , View ORCID Profile Jeremy A. Guggenheim doi: https://doi.org/10.1101/2025.06.04.25328975 Thu-Nga Nguyen 1 School of Optometry and Vision Sciences, Cardiff University , Cardiff, United Kingdom Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Thu-Nga Nguyen Louise Terry 1 School of Optometry and Vision Sciences, Cardiff University , Cardiff, United Kingdom Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Louise Terry Jeremy A. Guggenheim 1 School of Optometry and Vision Sciences, Cardiff University , Cardiff, United Kingdom Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Jeremy A. Guggenheim For correspondence: guggenheimj1{at}cardiff.ac.uk Abstract Full Text Info/History Metrics Supplementary material Preview PDF Abstract Purpose To examine if the choice of genome-wide association study (GWAS) summary statistics can yield invalid or misleading conclusions in Mendelian randomization (MR) studies of myopia. Methods The relationships between (1) years of full-time education and myopia, and (2) myopia and primary open-angle glaucoma (POAG), were used as exemplar testcases. MR analyses were performed with nine different sets of summary statistics for myopia: seven from sources widely used in published MR studies, plus two newly derived sets (a GWAS in either 66,773 unrelated participants or 93,036 participants that included relatives). Results Using the two newly derived sets of summary statistics from GWAS for myopia in unrelated and related samples, MR analyses demonstrated the expected positive causal relationship between education and myopia: odds ratio (OR) for myopia = 1.18, 95% confidence interval (CI) = 1.10 to 1.26 and OR = 1.16, 95% CI = 1.09 to 1.23 per additional year of education, respectively, and the expected positive relationship between myopia and POAG: OR = 1.11, 95% CI = 1.03 to 1.19 and OR = 1.12, 95% CI = 1.03 to 1.21, respectively. MR analyses performed using existing published GWAS summary statistics yielded highly inconsistent results, including MR estimates that suggested education protected against myopia and that myopia reduced the risk of POAG. Conclusions Care is required when designing MR analyses. Our findings imply that the results of some past MR studies of myopia were invalid. Introduction Eye diseases have become increasingly prevalent, with detrimental effects on individuals, healthcare systems, and society. 1 , 2 For example, myopia now affects approximately one-third of the global population. 3 Myopia increases the risk of sight-threatening eye disorders and results in an estimated productivity loss of around US$250 billion per year, globally. 4 – 9 To address the burden of eye diseases, it is imperative to develop comprehensive preventive and management strategies by elucidating the risk factors and pathophysiological mechanisms underlying them. Mendelian randomization (MR) is a statistical method for investigating the causal relationship between a modifiable environmental exposure and an outcome, relying solely on observational data. 10 MR takes advantage of the random assortment of alleles during gametogenesis to define groups that, on average, differ in their level of the exposure-of-interest: Individuals who inherit “high-risk” alleles tend to experience a higher level of the exposure in comparison to those who inherit “low-risk” alleles. The random assortment of alleles during meiosis (Mendel’s second law) ensures that this genetically-conferred level of exposure is largely independent of socio- economic and other environmental or lifestyle risk factors (but this can only be guaranteed in the absence of population stratification, assortative mating, and genetic nurture 11 ). If strict, so-called “instrumental variable”, assumptions are met, then MR is robust to the presence of confounding factors and reverse causation. Numerous vision-related studies have utilized MR to assess the causal relationship between variables, such as years of education and myopia risk, 12 , 13 myopia and primary open-angle glaucoma, 14 , 15 and many more. 16 – 24 One of the fundamental steps in conducting a two-sample MR analysis is to select appropriate data sources. Summary statistics from published genome-wide association studies (GWAS) that include regression coefficients for single nucleotide polymorphism (SNP)-exposure or SNP- outcome relationships are readily available for download from public databases, repositories, or research consortia websites. These sets of GWAS summary statistics enable MR analyses to be conducted in a short time with minimal resource requirements, which has led to a rapid expansion of MR publications in the research literature. 10 , 25 Comprehensive guidelines on how to conduct or report MR studies have been published. 25 , 26 However, these guidelines have focused little attention on the importance of choosing appropriate data sources for an MR analysis. Given the availability of multiple data repositories, researchers typically face the question of which set of summary statistics for an outcome or exposure should they choose? The various GWAS analyses for a trait-of-interest may have used different participant cohorts, case definitions, or analysis methods. Each of these parameters will influence the estimated SNP-trait regression coefficients. This source of variability will feed through to influence the causal effect estimate of the MR analysis, potentially resulting in spurious findings that compromise the robustness and reliability of the research literature. The aim of the current work was to investigate if the choice of GWAS summary statistics can influence MR estimates. We focused on two exemplar testcases: (1) a two-sample MR analysis examining the role of educational attainment on myopia, and (2) a two-sample MR analysis examining the risk of primary open- angle glaucoma (POAG) in myopic vs. non-myopic individuals. Our findings advocate for a more scientifically rigorous approach in selecting summary statistics for MR and interpreting the results obtained from such studies. Methods Study cohorts We used publicly available summary statistics of GWAS from four sources: the UK Biobank, the FinnGen study, the Social Science Genetic Association Consortium (SSGAC), and an international consortium of glaucoma genetics researchers 27 , as well as performing two new GWAS analyses for myopia in UK Biobank participants. UK Biobank : Approximately 500,000 adults aged between 40 and 69 were recruited from 2006 to 2010. 28 Participants visited one of 22 assessment centers across Great Britain for baseline and follow-up evaluations, during which their sociodemographic and clinical information was collected, including ophthalmic assessments completed by approximately 23% of participants. Genotype data were obtained using either the Biobank Axiom array (Affymetrix, High Wycombe, UK) or the BiLEVE Axiom array (Affymetrix), followed by imputation. FinnGen : Approximately 500,000 individuals from Finland, averaging 53 years of age, consented to the use of their electronic health record (EHR) information and biological samples, as part of disease-based and population-based studies. The FinnGen cohort has been intentionally enriched with individuals suffering from various diseases. Phenotype data, which includes International Classification of Diseases (ICD)-10 codes, primarily derive from the Finnish national health registers, given Finland’s comprehensive population-wide registry coverage. Genotype data were generated using the FinnGen ThermoFisher Axiom custom chip array (versions 1 and 2). 29 SSGAC : A total of 293,723 adults of European ancestry from different cohorts with educational attainment information were included in a large GWAS meta-analysis conducted by Okbay et al. 30 The primary outcome variable, EduYears, was imputed based on the participants’ years of schooling, which was standardized according to the 1997 International Standard Classification of Education. Gharahkhani et al. 27 glaucoma genetics consortium . A GWAS meta-analysis of POAG was conducted in 216,257 adults of European ancestry. POAG diagnosis was based on ICD9/ICD10 criteria. Ethical approval for the study was obtained from each of the participating institutions. All participants provided informed consent. GWAS summary statistics for years of education Summary statistics for the SSGAC GWAS for EduYears (n=293,723) were reported by Okbay et al. 30 The SNP-EduYears regression coefficients were reported in units of standard deviation in the original paper; for the current analyses, these were converted to units of years of full-time education using a conversion factor of 1 standard deviation = 3.6 years in school. 30 GWAS summary statistics for primary open-angle glaucoma Summary statistics for a GWAS meta-analysis of POAG across 21 independent samples of European ancestry (n = 16,677 POAG cases and n = 199,580 controls) were reported by Gharahkhani et al. 27 UK Biobank participants (n = 1,448 cases and n = 22,107 controls) and FinnGen participants (n = 1,824 cases and n = 93,036 controls) were included in the Gharahkhani et al. GWAS meta-analysis of POAG. However, this small degree of sample overlap with UK Biobank and FinnGen was not expected to appreciably bias 2-sample MR analyses. 31 GWAS summary statistics for myopia Publicly available GWAS summary statistics for myopia in European-ancestry individuals were identified by searching the following public databases: GeneATLAS (Roslin Institute and Medical Research Council Human Genetics Unit, University of Edinburgh), GWASATLAS (VU University of Amsterdam), GWAS Catalog (National Human Genome Research Institute, European Molecular Biology Laboratory – European Bioinformatics Institute), IEU OpenGWAS project (UK Medical Research Council Integrative Epidemiology Unit [IEU], University of Bristol), and FinnGen. To facilitate the search, the terms “myopia,” “nearsightedness,” “shortsightedness,” and “refractive error” were employed in the database queries. The results are presented in Table 1 . The majority of published MR studies used GWAS summary statistics for analyses conducted in the UK Biobank cohort, making use of participants’ self-reported “Reason for glasses/contact lens: For short-sightedness (called ‘myopia’)” (UK Biobank data field 6147; response option 1). The remainder of the published studies used GWAS summary statistics from analyses conducted in FinnGen, which used the ICD code H52.1 or Phecode 367.1 to define myopia. Ultimately, seven existing sets of GWAS summary statistics were included in a series of MR analyses. Details of each dataset are provided in Table 1 and Supplementary Note S1. View this table: View inline View popup Download powerpoint Table 1. Publicly available and newly derived GWAS summary statistics for myopia. Newly performed GWAS for myopia in UK Biobank Given the limitations of the currently available GWAS summary statistics for myopia (see below), we performed two new GWAS analyses for myopia in UK Biobank participants of European ancestry who had non-cycloplegic autorefraction measurement data and no other eye disorders. Details of the GWAS analyses can be found in Supplementary Note S2. Briefly, spherical equivalent refractive error (SER) was calculated as the autorefraction sphere power plus half the cylinder power. Participants were classified as myopia cases if the SER ≤ -0.50 diopters (D) in at least one eye. Controls were classified as individuals with SER > -0.50 D in both eyes. A GWAS for myopia in unrelated individuals (n = 25,804 cases; n = 40,969 controls) was conducted using PLINK2, 32 while a GWAS for myopia that included related individuals (n = 35,531 cases; n = 57,505 controls) was conducted using SAIGE/GATE. 33 Summary statistics for the newly performed GWAS analyses are available at: doi: 10.17035/cardiff.29211071. Ethical approval for the UK Biobank study was obtained from the Northwest Multicentre Research Ethics Committee (Reference: 11/NW/0382). Participants provided informed consent and were free to withdraw from the study at any time. The research adhered to the tenets of the Declaration of Helsinki. Selection of instrumental variables for years of education We used summary statistics from the Okbay et al. 30 GWAS for EduYears discovery analysis to avoid overlap with UK Biobank. GWAS variants independently associated with EduYears at P < 5 x 10 - 8 were selected; subsequently, variants not included in the Haplotype Reference Consortium (HRC) were excluded, as were SNPs not present in any of the myopia GWAS datasets, to ensure consistency in the genetic instrumental variables utilized across all of the MR analyses. Ultimately, 62 genetic variants remained suitable for use as instrumental variables (Supplementary Table S1). Selection of instrumental variables for myopia We used a consistent approach to select IVs for myopia from each of the nine sets of GWAS summary statistics for myopia listed in Table 1 . We first excluded variants not available in the summary statistics for POAG 27 or not available in the clumping reference panel (N = 10,000 UK Biobank participants of European ancestry). Variants were clumped using PLINK v1.9 using a p- value threshold of P < 5.0e-08, a distance metric of ±1000 kb and a linkage disequilibrium threshold of r 2 < 0.05. 32 Mendelian randomization analysis Full details of the MR analyses can be found in Supplementary Note S3 and code to reproduce the analyses is provided in Supplementary Note S4. The inverse-variance weighted (IVW) MR method was chosen as the primary analysis method. 34 The following sensitivity analyses were performed to evaluate the robustness of the IVW-MR analysis against the assumption of no horizontal pleiotropy: MR-EGGER 35 , weighted median MR 36 , mode-based MR 37 , and MR PRESSO 38 . Results A search of online GWAS repositories yielded 16 sets of myopia summary statistics publicly available for download. Twelve of the 16 sets of summary statistics were from the FinnGen study. We included the myopia summary statistics from FinnGen release 5, 9 and 10 in the current work, since these 3 sets have been utilized in published MR studies of myopia ( Table 1 ). Two of the 16 sets of myopia summary statistics were from the IEU OpenGWAS database; we included these as they have been very widely used in published MR studies of myopia ( Table 1 ). Finally, 2 of the 16 sets of myopia summary statistics were from the GWAS Catalog. Although these have not been used in published MR studies to our knowledge, we included them in the current analyses because their convenient availability would make them potentially suitable for an MR study of myopia. To provide gold standard MR causal effect estimates to compare against those obtained using the publicly available myopia summary statistics, we performed two new GWAS analyses for myopia in samples of UK Biobank participants whose SER had been measured by autorefraction. These new sets of summary statistics have been made openly accessible. Mendelian randomization analysis of the relationship between years of education and myopia An IVW-MR analysis using summary statistics from our newly-performed GWAS for myopia in unrelated participants (25,804 cases and 40,969 controls) yielded an estimate for the causal effect of education on myopia of OR = 1.18 per year of education (95% CI 1.10 to 1.26; P = 3.1e-06). An IVW-MR analysis using summary statistics from our newly-performed GWAS for myopia that included related individuals (35,531 cases and 57,505 controls) yielded a similar causal effect estimate of OR = 1.16 per year of education (95% CI 1.09 to 1.23; P = 2.2e-06). The results of the IVW-MR analyses using the seven sets of publicly-available GWAS summary statistics are presented in Table 2 , Figure 1 and Supplementary Figure S1. Complete results using the full range of MR methods are presented in Supplementary Table S2. These MR analyses yielded varying causal effect estimates for the relationship between education and myopia. As well as variability in effect size, the level of statistical significance of the causal effect varied widely, too. For instance, the MR analysis using GWAS summary statistics ukb-a-419 from the Neale lab repository suggested a small but highly significant negative causal relationship between years spent in full-time education and myopia status (OR = 0.99 per year of education, 95% CI = 0.988 to 0.996, P = 1.7e-05), implying that additional education was protective against myopia. GWAS summary statistics ukb-b-6353 from the IEU OpenGWAS project suggested evidence of a very small, highly significant positive causal effect of EduYears on myopia (OR = 1.01 per year of education, 95% CI = 1.006 to 1.014, P = 1.7e-07). The other set of GWAS summary statistics provided evidence for a positive causal effects ranging from OR = 1.07 to OR = 1.24. Download figure Open in new tab Figure 1: Causal effect estimates from Mendelian Randomization analyses of education as a risk factor for myopia obtained using different sets of GWAS summary statistics for myopia . Each data point represents an analysis using a different set of GWAS summary statistics for myopia. Error bars indicate 95% confidence intervals. View this table: View inline View popup Download powerpoint Table 2. Causal effect of years of education on the risk of myopia using different sets of summary statistics. Results were obtained using the IVW-MR method. Mendelian randomization analysis of the relationship between myopia and POAG An IVW-MR analysis using summary statistics from the newly-performed GWAS for myopia in unrelated participants (25,804 myopia cases and 40,969 controls) yielded an estimate for the risk of POAG in myopic vs. non-myopic individuals of OR = 1.11 (95% CI = 1.03 to 1.19, P = 6.5e- 03). An IVW-MR analysis using summary statistics from the newly-performed GWAS for myopia that included relatives (35,531 myopia cases and 57,505 controls) yielded an estimate for the risk of POAG in myopic vs. non-myopic individuals of OR = 1.12 (95% CI = 1.03 to 1.21, P = 5.7e-03). The results of the IVW-MR analyses using the publicly-available GWAS summary statistics are presented in Table 3 , Figure 2 and Supplementary Figure S2. Complete results using the full range of MR methods are presented in Supplementary Table S3. Once again, MR analyses employing different sets of GWAS summary statistics for myopia yielded highly varied results. Strikingly, the MR analysis using myopia summary statistics ukb-a-419 from the Neale lab repository suggested myopia had a highly protective effect against POAG (OR = 0.07, 95% CI = 0.02 to 0.24, P = 4.1e-05). Equally striking was the estimated causal effect of myopia on POAG obtained with summary statistics ukb-b-6353 from the IEU OpenGWAS database, which suggested myopia significantly increased the risk of POAG several fold (OR = 8.98, 95% CI = 2.19 to 36.83, P = 2.3e-03). MR analyses using myopia summary statistics from Zhou et al. 33 or FinnGen R9 and R10 suggested there was no evidence of a causal effect of myopia on POAG (all P > 0.05). Only the myopia summary statistics of Jiang et al. 39 yielded results similar to those from the newly derived summary statistics: OR = 1.14, 95% CI = 1.04 to 1.25, P = 5.6e-03. The number of IVs for myopia in these MR analyses also varied widely ( Table 3 ). Indeed, for the FinnGen R5 myopia summary statistics, no GWAS variant met our p-value threshold of P < 5.0e-08, hence there were no IVs available for an MR analysis using the FinnGen R5 summary statistics. Download figure Open in new tab Figure 2: Causal effect estimates from Mendelian Randomization analyses of myopia as a risk factor for primary open-angle glaucoma obtained using different sets of GWAS summary statistics for myopia. Each data point represents an analysis using a different set of GWAS summary statistics for myopia. Error bars indicate 95% confidence intervals. View this table: View inline View popup Download powerpoint Table 3. Causal effect of myopia on the risk of primary open-angle glaucoma using different sets of summary statistics. Results were obtained using the IVW-MR method. Discussion This work revealed that MR analyses employing different sets of publicly available GWAS summary statistics for myopia can yield contradictory findings. Our first exemplar testcase examined the relationship between years of education and myopia. Prior research using a range of methods has suggested that education is a causal risk factor for myopia. 12 , 13 , 43 , 44 Here, we found that researchers performing an MR analysis of education and myopia would have obtained evidence of a positive causal relationship, a null relationship or even a negative causal association, depending on the choice of GWAS summary statistics. Our second exemplar testcase examined the risk of POAG in myopic vs. non-myopic individuals. Prior research generally suggests that myopia modestly increases the risk of POAG, although the findings have not always been conclusive. 14 , 45 – 47 Here, we found that researchers performing an MR analysis would have obtained evidence of a large protective effect of myopia on the risk of POAG (OR < 0.1), a large increased risk (OR ≍ 9.0), or a no significant relationship, depending on the choice of summary statistics for myopia. As discussed in detail below, the reason for all these disparate findings is that publicly available summary statistics for myopia are derived from poorly designed GWAS analyses in which myopia cases and controls were often misclassified. Myopia is a common refractive error. Among the middle and older-aged populations of Europe the prevalence of myopia is approximately 30.6%. 48 However, the benign nature of myopia in most individuals has led to myopia being under-reported in electronic health records (EHR). 49 , 50 For example, the myopia prevalence in the FinnGen sample based on EHR information (ICD code H52.1) is approximately 1% ( Table 1 ), yet the true prevalence of myopia in Finland is 22- 30%. 41 , 42 Work by Wittenborn and colleagues suggested that as well as myopia, several other eye conditions were markedly under-reported in an EHR system. 50 Thus, a GWAS for myopia in the FinnGen sample using EHR-based information would analyze a sample of participants with a case-control ratio of approximately 1:100; the cases would be bona fide myopia cases, but 20- 30% of the controls would be myopic individuals who were misclassified. When using MR to evaluate the relationship between education and myopia, these misclassification issues with the FinnGen myopia summary statistics led to loss of statistical power, such that the causal effect of education on myopia was difficult to distinguished from a null causal effect ( Table 2 ; Figure 1 ). Similarly, in the MR examining the relationship between myopia and POAG, the misclassification issues with the FinnGen myopia summary statistics led to few SNPs being available as instrumental variables, leading to limited statistical power to distinguish a causal effect from a null effect. By contrast to the FinnGen study, the UK Biobank study performed direct assessment of refractive error using autorefraction and specifically asked participants if they were nearsighted. However, since the ophthalmic assessment component was introduced late in the UK Biobank recruitment process, only 23% of participants underwent the eye examinations and completed the ophthalmic questionnaire. 40 The true prevalence of myopia in these UK Biobank participants was approximately 38%. 40 When researchers from the Jiang et al. 39 study, the Neale lab, and the IEU OpenGWAS project classified UK Biobank participants as myopia cases or controls, the lack of ophthalmic assessment information was not taken into account; instead, the 385,000 participants who were not asked about their myopia status were all classified as non-myopic controls. Specifically, the GWAS for myopia performed by the Neale lab and IEU OpenGWAS project used 27,000 and 37,000 correctly-classified cases, respectively, but 309,000 and 423,000 controls, respectively, of whom about 38% were misclassified. As discussed below, we found that the misclassification of controls had a greater impact on downstream MR studies for the Neale lab and OpenGWAS myopia summary statistics compared to the Jiang et al. and Zhou et al. summary statistics. An additional concern regarding existing GWAS summary statistics for myopia is the use of a highly imbalanced case-control ratio. As mentioned above, published studies have used case- control ratios as high as of 1:100 or even 1:300 ( Table 1 ). Logistic regression analysis using a highly imbalanced case-control ratio may result in an inflated type I error rate and inaccurate estimated effect sizes; 33 the latter phenomenon would directly impact MR causal effect estimates. Fortunately, methods for performing case-control GWAS analyses that account for case-control imbalance have been developed, such as SAIGE, 33 fastGWA-GLMM, 39 GMMAT, 51 and others. 52 Of the sets of myopia summary statistics from publicly available databases, most used an appropriate method to address the case-control imbalance: Jiang et al. 39 used fastGWA-GLMM, while Zhou et al. 33 and the FinnGen team 33 employed SAIGE. However, the Neale lab and IEU OpenGWAS project utilized approaches that did not account for case-control imbalance: Hail ( https://hail.is/docs/0.2/overview/index.html ) and BOLT-LMM 53 , respectively. Therefore, the published GWAS summary statistics for myopia from these two sources may include inaccurate regression coefficients for some SNPs and even include false-positive SNPs that reached the genome-wide significance threshold for spurious reasons. The combination of misclassification of controls and insufficient account of case-control imbalance is a potential reason for the worse performance of the Neale lab and IEU OpenGWAS project myopia summary statistics in our exemplar MR scenarios, compared to the summary statistics released by Jiang et al. and Zhou et al. Thus, whereas the Jiang et al. and Zhou et al. myopia summary statistics yielded MR causal effect estimates comparable to those of our gold standard summary statistics, MR causal effect estimates obtained using the Neale lab and IEU OpenGWAS project were grossly misleading in suggesting highly significant causal effects of very different magnitude from our gold standard causal effect estimates ( Tables 2 - 3 , Figures 1 - 2 ). Finally, the GWAS summary statistics from the Neale lab used in the current study had wrongly labeled risk and reference alleles, which led to MR causal effect estimates in the opposite direction to that expected, for example suggesting that additional education reduced the risk of myopia. We downloaded the Neale lab myopia summary statistics from the IEU OpenGWAS repository, since the original Neale lab GWAS repository no longer exists and because the IEU OpenGWAS repository is the most widely used source of myopia summary statistics for use in MR studies ( Table 1 ). We were able to confirm that a more recent release of myopia summary statistics by the Neale lab has the risk and effect alleles correctly labelled (however, the misclassification of controls and the lack of account for case-control imbalance remain as potential issues). Conclusions Mendelian randomization analyses performed using different sets of publicly available GWAS summary statistics for myopia can yield conflicting results. The results of previously published MR studies that have used these resources may not be valid ( Table 1 ). Although we investigated just two exemplar scenarios – the impact of education on myopia and the risk of POAG conferred by myopia – we suspect this issue of inappropriate GWAS summary statistics may extend to other ophthalmic diseases. We urge researchers to exercise caution when selecting GWAS summary statistics for an MR study: The adage “choose the GWAS study with the largest sample size” may not always hold. 10 Specifically, researchers should consider the underlying population, case definition, and method of association analyses when selecting summary statistics for MR. We have made our two newly derived sets of myopia summary statistics openly available and encourage researchers to utilize these for future MR studies of myopia. Data Availability UK Biobank data are available via an application to the study’s access team ( https://www.ukbiobank.ac.uk/enable-your-research/about-our-data ). GWAS summary statistics for POAG were downloaded from the GWAS catalog ( https://www.ebi.ac.uk/gwas/studies/GCST90011766 ). Existing GWAS summary statistics for myopia were downloaded from the GWAS catalog (Jiang et al. http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90044001-GCST90045000/GCST90044326 ; Zhou et al. http://ftp.ebi.ac.uk/pub/databases/gwas/summary_statistics/GCST90435001-GCST90436000/GCST90435990 ), the IEU OpenGWAS database (Neale lab: https://gwas.mrcieu.ac.uk/files/ukb-a-419/ukb-a-419.vcf.gz ; MRC IEU: https://gwas.mrcieu.ac.uk/files/ukb-b-6353/ukb-b-6353.vcf.gz ) and the FinnGen project (R5: https://storage.googleapis.com/finngen-public-data-r5/summary_stats/finngen_R5_H7_MYOPIA.gz ; R9: https://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_H7_MYOPIA.gz ; R10: https://storage.googleapis.com/finngen-public-data-r10/summary_stats/finngen_R10_H7_MYOPIA.gz ). New summary statistics from a GWAS of myopia in related and unrelated UK Biobank participants are openly available (doi: 10.17035/cardiff.29211071). Acknowledgements This research was conducted using the UK Biobank Resource under Application Number 83325. Data analysis was performed on the HAWK computing cluster, managed by Supercomputing Wales and Cardiff University ARCCA. We would like to thank the UK Biobank, FinnGen, and all participants who contributed to the two cohorts. Additionally, we extend our sincere gratitude to the authors of all GWAS who made their summary statistics available to facilitate this work. Footnotes Funding : This work was funded by UK Research and Innovation (UKRI) under the UK government’s Horizon Europe funding guarantee [grant number EP/Y032292/1]. Funded by the European Union (Project 101119501 — MyoTreat — HORIZON-MSCA-2022-DN-01). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or UKRI. Neither the European Union nor the granting authority can be held responsible for them. Commercial Relationships Disclosure : T.N. Nguyen, None; L. Terry, None; J.A. Guggenheim, Associate Editor for IOVS (S: non-remunerative) References 1. ↵ World Health Organization. World report on vision ; 2019. 2. ↵ Burton MJ , Ramke J , Marques AP , et al. The Lancet Global Health Commission on Global Eye Health: vision beyond 2020 . The Lancet Global Health 2021 ; 9 : e489 – e551 . OpenUrl 3. ↵ Holden BA , Fricke TR , Wilson DA , et al. Global Prevalence of Myopia and High Myopia and Temporal Trends from 2000 through 2050 . Ophthalmol 2016 ; 123 : 1036 – 1042 . OpenUrl 4. ↵ Haarman AEG , Enthoven CA , Tideman JWL , Tedja MS , Verhoeven VJM , Klaver CCW . The Complications of Myopia: A Review and Meta-Analysis . Invest Ophthalmol Vis Sci 2020 ; 61 : 49 . 5. Ohno-Matsui K , Kawasaki R , Jonas JB , et al. International photographic classification and grading system for myopic maculopathy . Am J Ophthalmol 2015 ; 159 : 877 – 883 . OpenUrl CrossRef PubMed 6. Zou M , Wang S , Chen A , et al. Prevalence of myopic macular degeneration worldwide: a systematic review and meta-analysis . Br J Ophthalmol 2020 ; 104 : 1748 – 1754 . OpenUrl Abstract / FREE Full Text 7. Marcus MW , de Vries MM , Junoy Montolio FG , Jansonius NM. Myopia as a risk factor for open-angle glaucoma: a systematic review and meta-analysis . Ophthalmol 2011 ;118:1989-1994 e1982. 8. ↵ Younan C , Mitchell P , Cumming RG , Rochtchina E , Wang JJ . Myopia and incident cataract and cataract surgery: the blue mountains eye study . Invest Ophthalmol Vis Sci 2002 ; 43 : 3625 – 3632 . OpenUrl Abstract / FREE Full Text 9. ↵ Naidoo KS , Fricke TR , Frick KD , et al. Potential Lost Productivity Resulting from the Global Burden of Myopia: Systematic Review , Meta-analysis, and Modeling. Ophthalmol 2019 ; 126 : 338 – 346 . OpenUrl 10. ↵ Sanderson E , Glymour MM , Holmes MV , et al. Mendelian randomization . Nature Reviews Methods Primers 2022 ; 2 : 6 . 11. ↵ Howe LJ , Nivard MG , Morris TT , et al. Within-sibship genome-wide association analyses decrease bias in estimates of direct genetic effects . Nat Genet 2022 ; 54 : 581 – 592 . OpenUrl CrossRef PubMed 12. ↵ Mountjoy E , Davies NM , Plotnikov D , et al. Education and myopia: assessing the direction of causality by mendelian randomisation . BMJ 2018 ; 361 : k2022 . 13. ↵ Clark R , Kneepkens SCM , Plotnikov D , et al. Time Spent Outdoors Partly Accounts for the Effect of Education on Myopia . Invest Ophthalmol Vis Sci 2023 ; 64 : 38 . 14. ↵ Choquet H , Khawaja AP , Jiang C , et al. Association Between Myopic Refractive Error and Primary Open-Angle Glaucoma: A 2-Sample Mendelian Randomization Study . JAMA Ophthalmol 2022 ; 140 : 864 – 871 . OpenUrl CrossRef PubMed 15. ↵ Chong RS , Li H , Cheong AJ , et al. Mendelian Randomization Implicates Bidirectional Association between Myopia and Primary Open-Angle Glaucoma or Intraocular Pressure . Ophthalmol 2023 ; 130 : 394 – 403 . OpenUrl 16. ↵ Han X , Ong J-S , An J , et al. Association of Myopia and Intraocular Pressure With Retinal Detachment in European Descent Participants of the UK Biobank Cohort: A Mendelian Randomization Study . JAMA Ophthalmol 2020 ; 138 : 671 – 678 . OpenUrl PubMed 17. Zhang X , Yuan W , Xu J , Zhao F . Application of mendelian randomization in ocular diseases: a review . Hum Genomics 2024 ; 18 : 66 . 18. Jiang C , Melles RB , Sangani P , et al. Association of Behavioral and Clinical Risk Factors With Cataract: A Two-Sample Mendelian Randomization Study . Invest Ophthalmol Vis Sci 2023 ; 64 : 19 – 19 . OpenUrl CrossRef 19. Burgess S , Davey Smith G . Mendelian Randomization Implicates High-Density Lipoprotein Cholesterol-Associated Mechanisms in Etiology of Age-Related Macular Degeneration . Ophthalmol 2017 ; 124 : 1165 – 1174 . OpenUrl 20. Nusinovici S , Li H , Thakur S , et al. High-Density Lipoprotein 3 Cholesterol and Primary Open-Angle Glaucoma: Metabolomics and Mendelian Randomization Analyses . Ophthalmol 2022 ; 129 : 285 – 294 . OpenUrl 21. Kuan V , Warwick A , Hingorani A , et al. Association of Smoking, Alcohol Consumption, Blood Pressure, Body Mass Index, and Glycemic Risk Factors With Age-Related Macular Degeneration: A Mendelian Randomization Study . JAMA Ophthalmol 2021 ; 139 : 1299 – 1306 . OpenUrl PubMed 22. Han X , Lains I , Li J , et al. Integrating genetics and metabolomics from multi-ethnic and multi-fluid data reveals putative mechanisms for age-related macular degeneration . Cell Rep Med 2023 ; 4 : 101085 . 23. Plotnikov D , Huang Y , Khawaja AP , et al. High Blood Pressure and Intraocular Pressure: A Mendelian Randomization Study . Invest Ophthalmol Vis Sci 2022 ; 63 : 29 . 24. ↵ Plotnikov D , Sheehan NA , Williams C , Atan D , Guggenheim JA , UK Biobank Eye and Vision Consortium . Hyperopia Is Not Causally Associated With a Major Deficit in Educational Attainment . Translational Vision Science & Technology 2021 ; 10 : 34 . 25. ↵ Skrivankova VW , Richmond RC , Woolf BAR , et al. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE- MR Statement . JAMA 2021 ; 326 : 1614 – 1621 . OpenUrl CrossRef PubMed 26. ↵ Burgess S , Davey Smith G , Davies NM , et al. Guidelines for performing Mendelian randomization investigations . Wellcome Open Res 2019 ; 4 : 186 . 27. ↵ Gharahkhani P , Jorgenson E , Hysi P , et al. Genome-wide meta-analysis identifies 127 open-angle glaucoma loci with consistent effect across ancestries . Nat Commun 2021 ; 12 : 1258 . OpenUrl CrossRef PubMed 28. ↵ Bycroft C , Freeman C , Petkova D , et al. The UK Biobank resource with deep phenotyping and genomic data . Nature 2018 ; 562 : 203 – 209 . OpenUrl CrossRef PubMed 29. ↵ Kurki MI , Karjalainen J , Palta P , et al. FinnGen provides genetic insights from a well- phenotyped isolated population . Nature 2023 ; 613 : 508 – 518 . OpenUrl CrossRef PubMed 30. ↵ Okbay A , Beauchamp JP , Fontana MA , et al. Genome-wide association study identifies 74 loci associated with educational attainment . Nature 2016 ; 533 : 539 – 542 . OpenUrl CrossRef PubMed 31. ↵ Burgess S , Davies NM , Thompson SG . Bias due to participant overlap in two-sample Mendelian randomization . Genet Epidemiol 2016 ; 40 : 597 – 608 . OpenUrl CrossRef PubMed 32. ↵ Chang CC , Chow CC , Tellier LC , Vattikuti S , Purcell SM , Lee JJ . Second-generation PLINK: rising to the challenge of larger and richer datasets . GigaScience 2015 ; 4 : 7 . 33. ↵ Zhou W , Nielsen JB , Fritsche LG , et al. Efficiently controlling for case-control imbalance and sample relatedness in large-scale genetic association studies . Nat Genet 2018 ; 50 : 1335 – 1341 . OpenUrl CrossRef PubMed 34. ↵ Burgess S , Butterworth A , Thompson SG . Mendelian Randomization Analysis With Multiple Genetic Variants Using Summarized Data . Genet Epidemiol 2013 ; 37 : 658 – 665 . OpenUrl CrossRef PubMed 35. ↵ Bowden J , Del Greco M F , Minelli C , Davey Smith G , Sheehan N , Thompson J. A framework for the investigation of pleiotropy in two-sample summary data Mendelian randomization . Statistics in Medicine 2017 ; 36 : 1783 – 1802 . OpenUrl CrossRef PubMed 36. ↵ Bowden J , Davey Smith G , Haycock PC , Burgess S . Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator . Genet Epidemiol 2016 ; 40 : 304 – 314 . OpenUrl CrossRef PubMed 37. ↵ Hartwig FP , Davey Smith G , Bowden J . Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption . Int J Epidemiol 2017 ; 46 : 1985 – 1998 . OpenUrl CrossRef PubMed 38. ↵ Verbanck M , Chen C-Y , Neale B , Do R . Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases . Nat Genet 2018 ; 50 : 693 – 698 . OpenUrl CrossRef PubMed 39. ↵ Jiang L , Zheng Z , Fang H , Yang J . A generalized linear mixed model association tool for biobank-scale data . Nat Genet 2021 ; 53 : 1616 – 1621 . OpenUrl CrossRef PubMed 40. ↵ Cumberland PM , Bao Y , Hysi PG , et al. Frequency and Distribution of Refractive Error in Adult Life: Methodology and Findings of the UK Biobank Study . PLoS ONE 2015 ; 10 : e0139780 . OpenUrl CrossRef PubMed 41. ↵ Pärssinen O . The increased prevalence of myopia in Finland . Acta Ophthalmologica 2012 ; 90 : 497 – 502 . OpenUrl CrossRef PubMed 42. ↵ Vannas AE , Ying GS , Stone RA , Maguire MG , Jormanainen V , Tervo T . Myopia and natural lighting extremes: risk factors in Finnish army conscripts . Acta Ophthalmologica Scandinavica 2003 ; 81 : 588 – 595 . OpenUrl CrossRef PubMed Web of Science 43. ↵ Cuellar-Partida G , Lu Y , Kho PF , et al. Assessing the Genetic Predisposition of Education on Myopia: A Mendelian Randomization Study . Genet Epidemiol 2016 ; 40 : 66 – 72 . OpenUrl CrossRef PubMed 44. ↵ Plotnikov D , Williams C , Atan D , et al. Effect of Education on Myopia: Evidence from the United Kingdom ROSLA 1972 Reform . Invest Ophthalmol Vis Sci 2020 ; 61 : 7 . 45. ↵ Marcus MW , de Vries MM , Montolio FGJ , Jansonius NM . Myopia as a Risk Factor for Open-Angle Glaucoma: A Systematic Review and Meta-Analysis . Ophthalmol 2011 ; 118 : 1989 – 1994 .e1982. OpenUrl 46. Hsu CH , Chen RI , Lin SC . Myopia and glaucoma: sorting out the difference . Curr Opin Ophthalmol 2015 ; 26 : 90 – 95 . OpenUrl PubMed 47. ↵ Iglesias AI , Ong JS , Khawaja AP , et al. Determining Possible Shared Genetic Architecture Between Myopia and Primary Open-Angle Glaucoma . Invest Ophthalmol Vis Sci 2019 ; 60 : 3142 – 3149 . OpenUrl CrossRef PubMed 48. ↵ Williams KM , Verhoeven VJ , Cumberland P , et al. Prevalence of refractive error in Europe: the European Eye Epidemiology (E3) Consortium . Eur J Epidemiol 2015 ; 30 : 305 – 315 . OpenUrl CrossRef PubMed 49. ↵ Pignot M , Kossack N , Shi-van Wielink K . RWD133 Cross Sectional Claims Data Analysis on Myopia Epidemiology and Treatment Options in Germany . Value Health 2023 ; 26 : S529 . 50. ↵ Wittenborn JS , Lee AY , Lundeen EA , et al. Validity of Administrative Claims and Electronic Health Registry Data From a Single Practice for Eye Health Surveillance . JAMA Ophthalmol 2023 ; 141 : 534 – 541 . OpenUrl PubMed 51. ↵ Chen H , Wang C , Conomos Matthew P , et al. Control for Population Structure and Relatedness for Binary Traits in Genetic Association Studies via Logistic Mixed Models . Am J Hum Genet 2016 ; 98 : 653 – 666 . OpenUrl CrossRef PubMed 52. ↵ Mbatchou J , Barnard L , Backman J , et al. Computationally efficient whole-genome regression for quantitative and binary traits . Nat Genet 2021 ; 53 : 1097 – 1103 . OpenUrl CrossRef PubMed 53. ↵ Loh P-R , Tucker G , Bulik-Sullivan BK , et al. Efficient Bayesian mixed-model analysis increases association power in large cohorts . Nat Genet 2015 ; 47 : 284 – 290 . OpenUrl CrossRef PubMed 54. Xu X , Liu N , Yu W . No Evidence of an Association between Genetic Factors Affecting Response to Vitamin A Supplementation and Myopia: A Mendelian Randomization Study and Meta-Analysis . Nutrients 2024 ; 16 : doi: 10.3390/nu16121933 . OpenUrl CrossRef 55. Deng B , Zhou M , Kong X , et al. The lack of causal link between myopia and intraocular pressure: Insights from cross-sectional analysis and Mendelian randomization study . Photodiagnosis Photodyn Ther 2024 ; 49 : 104334 . 56. Jiang X , Xu B , Li Q , Zhao YE . Association between Plasma Metabolite Levels and Myopia: A 2-Sample Mendelian Randomization Study . Ophthalmol Sci 2025 ; 5 : 100699 . 57. Li F-F , Zhu M-C , Shao Y-L , Lu F , Yi Q-Y , Huang X-F . Causal Relationships Between Glycemic Traits and Myopia . Invest Ophthalmol Vis Sci 2023 ; 64 : 7 . 58. Liang R , Li T , Gao H , et al. Causal relationships between inflammatory cytokines and myopia: an analysis of genetic and observational studies . Ann Med Surg 2024 ; 86 : 5179 – 5190 . OpenUrl 59. Lv H , Wang Z , Huang C , Yu X , Li X , Song X . Causal Links between Gut Microbiota, Blood Metabolites, Immune Cells , Inflammatory Proteins, and Myopia: A Mendelian Randomization Study. Ophthalmol Sci 2025 ; 5 : 100684 . 60. Mo Q , Liu X , Gong W , et al. Pinpointing Novel Plasma and Brain Proteins for Common Ocular Diseases: A Comprehensive Cross-Omics Integration Analysis . Int J Mol Sci ; 2024 : doi: 10.3390/ijms251910236 . OpenUrl CrossRef 61. Wei P , Han G , Su Q , Jia L , Xue C , Wang Y . Corneal biomechanics as a causal factor in myopia and astigmatism: Evidence from Mendelian randomization . Ophthalmology Science 2025 ; 100738 . 62. Xu J , Mo Y . Mendelian randomization study confirms causal relationship between myopia and vitreous disorders . BMC Med Genomics 2023 ; 16 : 238 . 63. Zhang XB , Jiang HH , Zhang LL , et al. Potential causal associations between leisure sedentary behaviors, physical activity, sleep traits, and myopia: a Mendelian randomization study . BMC Ophthalmol 2024 ; 24 : 104 . 64. Zhu G , Tian R , Zhou D , Qin X . Genetic correlation and causal relationship between sleep and myopia: a mendelian randomization study . Front Genet 2024 ; 15 : 1378802 . 65. Li H , Du Y , Cheng K , et al. Gut microbiota-derived indole-3-acetic acid suppresses high myopia progression by promoting type I collagen synthesis . Cell Discovery 2024 ; 10 : 89 . 66. Wei D , Wang H , Huang L , et al. A Mendelian randomization study on the causal relationship between smoking, alcohol consumption, and the development of myopia and astigmatism . Sci Rep 2024 ; 14 : 1868 . OpenUrl PubMed 67. Wei X , Li W , Liu R . Causal effects of allergic diseases on the risk of myopia: a two- sample Mendelian randomization study . Eye 2025 ; doi: 10.1038/s41433-41025-03749-41437 . OpenUrl CrossRef 68. Hui J , Tang K , Zhou Y , Cui X , Han Q . The causal impact of gut microbiota and metabolites on myopia and pathological myopia: a mediation Mendelian randomization study . Sci Rep 2025 ; 15 : 12928 . 69. Huang Z , Zhou J , Liu S , et al. The interplay between systemic inflammation and myopia: A bidirectional Mendelian randomization and experimental validation study . International Immunopharmacology 2025 ; 157 : 114803 . View the discussion thread. Back to top Previous Next Posted June 05, 2025. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Mendelian Randomization Studies of Myopia: Choosing the right Summary Statistics Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Mendelian Randomization Studies of Myopia: Choosing the right Summary Statistics Thu-Nga Nguyen , Louise Terry , Jeremy A. Guggenheim medRxiv 2025.06.04.25328975; doi: https://doi.org/10.1101/2025.06.04.25328975 Share This Article: Copy Citation Tools Mendelian Randomization Studies of Myopia: Choosing the right Summary Statistics Thu-Nga Nguyen , Louise Terry , Jeremy A. Guggenheim medRxiv 2025.06.04.25328975; doi: https://doi.org/10.1101/2025.06.04.25328975 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Genetic and Genomic Medicine Subject Areas All Articles Addiction Medicine (567) Allergy and Immunology (863) Anesthesia (297) Cardiovascular Medicine (4411) Dentistry and Oral Medicine (443) Dermatology (380) Emergency Medicine (606) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1505) Epidemiology (15205) Forensic Medicine (30) Gastroenterology (1119) Genetic and Genomic Medicine (6574) Geriatric Medicine (666) Health Economics (994) Health Informatics (4511) Health Policy (1365) Health Systems and Quality Improvement (1608) Hematology (537) HIV/AIDS (1263) Infectious Diseases (except HIV/AIDS) (15903) Intensive Care and Critical Care Medicine (1103) Medical Education (620) Medical Ethics (144) Nephrology (666) Neurology (6573) Nursing (345) Nutrition (998) Obstetrics and Gynecology (1139) Occupational and Environmental Health (954) Oncology (3319) Ophthalmology (968) Orthopedics (369) Otolaryngology (420) Pain Medicine (435) Palliative Medicine (129) Pathology (662) Pediatrics (1689) Pharmacology and Therapeutics (691) Primary Care Research (710) Psychiatry and Clinical Psychology (5422) Public and Global Health (9205) Radiology and Imaging (2191) Rehabilitation Medicine and Physical Therapy (1367) Respiratory Medicine (1191) Rheumatology (593) Sexual and Reproductive Health (709) Sports Medicine (529) Surgery (709) Toxicology (99) Transplantation (288) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9feb9e7ffa584807',t:'MTc3OTI4MzAyMA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();