Avoiding dynastic, assortative mating, and population stratification biases in Mendelian randomization through within-family analyses

Causal Links Between Systemic Disorders and Keratoconus in European Population

PURPOSE

To establish the presence of a causal linkage between prevalent systemic diseases and keratoconus (KC).

DESIGN

Mendelian randomization (MR) analysis.

METHODS

After an exhaustive screening process, genetic variants linked to various systemic diseases were identified as instrumental variables at the genome-wide significance level. Subsequently, MR analyses were conducted to elucidate their potential causal connection with KC (N = 26,742). The encompassed systemic ailments comprise diabetes, hay fever/allergic rhinitis/eczema, obstructive sleep apnea, thyroid dysfunction, aortic aneurysm, major depressive disorder, inflammatory bowel disease (including Crohn's disease and ulcerative colitis), and mitral valve prolapse. Our study adheres to the principles of Strengthening the Reporting of Observational Studies in Epidemiology Using MR guidelines.

RESULTS

Using inverse variance weighting as the primary MR analysis method, our findings revealed that hay fever/allergic rhinitis/eczema (odds ratio, 10.144; 95% CI, 2.441-42.149; P = .001) and ulcerative colitis (odds ratio, 1.147; 95% CI, 1.054-1.248; P = .002) were associated with an increased risk of KC within the largest population under scrutiny. Conversely, the prolonged hyperglycemic state did not exhibit a potentially protective effect in delaying the pathogenesis of KC, and no correlation was observed between the two (odds ratio, 0.320; 95% CI, 0.029-3.549; P = .353). Also, obstructive sleep apnea, thyroid function, aortic aneurysm, major depressive disorder, Crohn's disease, and mitral valve prolapse did not exhibit a causal association with KC (P > .05 for all comparisons).

CONCLUSIONS

This study indicates an increased risk of KC related to hay fever/allergic rhinitis/eczema and ulcerative colitis, with diabetes not providing a protective effect. These findings may potentially contribute some insights to inform clinical interventions.

Child maltreatment as a transdiagnostic risk factor for the externalizing dimension: a Mendelian randomization study

Observational studies suggest that child maltreatment increases the risk of externalizing spectrum disorders such as attention deficit hyperactivity disorder (ADHD), conduct disorder (CD), antisocial personality disorder (ASPD), and substance use disorder (SUD). Yet, only few of such associations have been investigated by approaches that provide strong evidence for causation, such as Mendelian Randomization (MR). Establishing causal inference is essential given the growing recognition of gene-environment correlations, which can confound observational research in the context of childhood maltreatment. Evaluating causality between child maltreatment and the externalizing phenotypes, we used genome-wide association study (GWAS) summary data for child maltreatment (143,473 participants), ADHD (20,183 cases; 35,191 controls), CD (451 cases; 256,859 controls), ASPD (381 cases; 252,877 controls), alcohol use disorder (AUD; 13,422 cases; 244,533 controls), opioid use disorder (OUD; 775 cases; 255,921 controls), and cannabinoid use disorder (CUD; 14,080 cases; 343,726 controls). We also generated a latent variable 'common externalizing factor' (EXT) using genomic structural equation modeling. Genetically predicted childhood maltreatment was consistently associated with ADHD (odds ratio [OR], 10.09; 95%-CI, 4.76-21.40; P = 1.63 × 10-09), AUD (OR, 3.72; 95%-CI, 1.85-7.52; P = 2.42 × 10-04), and the EXT (OR, 2.64; 95%-CI, 1.52-4.60; P = 5.80 × 10-04) across the different analyses and pleiotropy-robust methods. A subsequent GWAS on childhood maltreatment and the externalizing dimension from Externalizing Consortium (EXT-CON) confirmed these results. Two of the top five genes with the strongest associations in EXT GWAS, CADM2 and SEMA6D, are also ranked among the top 10 in the EXT-CON. The present results confirm the existence of a common externalizing factor and an increasing vulnerability caused by child maltreatment, with crucial implications for prevention. However, the partly diverging results also indicate that specific influences impact individual phenotypes separately.

Mapping associations of polygenic scores with autistic and ADHD traits in a single city region

BACKGROUND

The genetic and environmental aetiology of autistic and Attention Deficit Hyperactivity Disorder (ADHD) traits is known to vary spatially, but does this translate into variation in the association of specific common genetic variants?

METHODS

We mapped associations between polygenic scores for autism and ADHD and their respective traits in the Avon Longitudinal Study of Parents and Children (N = 4,255-6,165) across the area surrounding Bristol, UK, and compared them to maps of environments associated with the prevalence of autism and ADHD.

RESULTS

Our results suggest genetic associations vary spatially, with consistent patterns for autistic traits across polygenic scores constructed at different p-value thresholds. Patterns for ADHD traits were more variable across thresholds. We found that the spatial distributions often correlated with known environmental influences.

CONCLUSIONS

These findings shed light on the factors that contribute to the complex interplay between the environment and genetic influences in autistic and ADHD traits.

Educational attainment and mental health conditions: a within-sibship Mendelian randomization study

Importance

Observational studies have demonstrated consistent protective effects of higher educational attainment (EA) on the risk of suffering mental health conditions (MHC). Determining whether these beneficial effects are causal is challenging given the potential role of dynastic effects and demographic factors (assortative mating and population structure) in this association.

Objective

To evaluate to what extent the relationship between EA and various MHC is independent from dynastic effects and demographic factors.

Design

Within-sibship Mendelian randomization (MR) study.

Setting

One-sample MR analyses included participants' data from the Trøndelag Health Study (HUNT, Norway) and UK Biobank (United Kingdom). For two-sample MR analyses we used summary statistics from publicly available genome-wide-association-studies.

Participants

61 880 siblings (27 507 sibships).

Exposure

Years of education.

Main outcomes

Scores for symptoms of anxiety, depression and neuroticism using the Hospital Anxiety Depression Scale (HADS), the 7-item Generalized Anxiety Disorder Scale (GAD-7), the 9-item Patient Health Questionnaire (PHQ-9), and the Eysenck Personality Questionnaire, as well as self-reported consumption of psychotropic medication.

Results

One standard deviation (SD) unit increase in years of education was associated with a lower symptom score of anxiety (-0.20 SD [95%CI: -0.26, -0.14]), depression (-0.11 SD [-0.43, 0.22]), neuroticism (-0.30 SD [-0.53, -0.06]), and lower odds of psychotropic medication consumption (OR: 0.60 [0.52, 0.69]). Estimates from the within-sibship MR analyses showed some attenuation, which however were suggestive of a causal association (anxiety: -0.17 SD [-0.33, -0.00]; depression: -0.18 SD [-1.26, 0.89]; neuroticism: -0.29 SD [-0.43, -0.15]); psychotropic medication consumption: OR, 0.52 [0.34, 0.82]).

Conclusions and Relevance

Associations between EA and MHC in adulthood, although to some extend explained by dynastic effects and demographic factors, overall remain robust, indicative of a causal effect. However, larger studies are warranted to improve statistical power and further validate our conclusions.

Benchmarking Mendelian randomization methods for causal inference using genome-wide association study summary statistics

Mendelian randomization (MR), which utilizes genetic variants as instrumental variables (IVs), has gained popularity as a method for causal inference between phenotypes using genetic data. While efforts have been made to relax IV assumptions and develop new methods for causal inference in the presence of invalid IVs due to confounding, the reliability of MR methods in real-world applications remains uncertain. Instead of using simulated datasets, we conducted a benchmark study evaluating 16 two-sample summary-level MR methods using real-world genetic datasets to provide guidelines for the best practices. Our study focused on the following crucial aspects: type I error control in the presence of various confounding scenarios (e.g., population stratification, pleiotropy, and family-level confounders like assortative mating), the accuracy of causal effect estimates, replicability, and power. By comprehensively evaluating the performance of compared methods over one thousand exposure-outcome trait pairs, our study not only provides valuable insights into the performance and limitations of the compared methods but also offers practical guidance for researchers to choose appropriate MR methods for causal inference.

Cross-sectional and Mendelian randomization study of fibroblast growth factor 19 reveals causal associations with metabolic diseases

BACKGROUND AND AIM

Fibroblast growth factor 19 (FGF19) is an intestinal-derived factor that plays a role in metabolic diseases. We performed a differential study of circulating FGF19 levels and investigated the causal effects of FGF19 on metabolic diseases using Mendelian randomization (MR).

METHODS

Firstly, 958 subjects were included in the physical examination center of affiliated hospital from January 2019 to January 2021. Dividing the subjects into different subgroups to compare FGF19 levels. We conducted a two-sample MR analysis of genetically predicted circulating FGF19 in relation to alcohol, cardiovascular and metabolic biomarkers and diseases, and liver function biomarkers using publicly available genome-wide association study summary statistics data.

RESULTS

The circulating FGF19 levels in nonalcoholic fatty liver disease (NAFLD) patients were lower than those without NAFLD (P < 0.001). The FGF19 levels in participants with obese were lower than those without obese (P < 0.001). In two-sample MR analyses, genetically predicted higher circulating FGF19 levels was significantly associated with lower aspartate aminotransferase, γ-glutamyltransferase, triglycerides, total cholesterol, low-density lipoprotein, and C-reactive protein concentrations (P < 0.05) and a negative correlation with cardiovascular disease and cirrhosis whereas a positive association with type 2 diabetes mellitus (P < 0.05).

CONCLUSIONS

Our study found that circulating FGF19 levels were lower in NAFLD and obese populations. Additionally, our MR research results support the causal effects of FGF19 on improved liver function, lipids, and reduced the occurrence of inflammation, cardiovascular disease, and cirrhosis. We found a positive correlation with diabetes, which may indicate a compensatory increase in regulating above FGF19 resistance states in humans.

Genetic Insights Into Perinatal Outcomes of Maternal Antihypertensive Therapy During Pregnancy

Importance

Limited information exists regarding the impact of pharmacotherapy in pregnancy due to ethical concerns of unintended fetal harm. Yet, maternal prescriptive drug use for chronic conditions such as hypertension is common.

Objective

To investigate potential causal relationships between perturbing maternal genetic variants influencing antihypertensive drug targets and perinatal outcomes among offspring using mendelian randomization (MR).

Design, Setting, and Participants

This 2-sample MR study used individual-level single-nucleotide variation (SNV) outcome data from mother-father-offspring trios with complete genetic and phenotypic information from the Norwegian Mother, Father and Child Cohort Study (MoBa) and summary-level SNV exposure data from UK Biobank participants sourced from the Integrative Epidemiology Unit OpenGWAS project. Pregnant individuals were recruited across Norway during their routine ultrasonography examination at 18 weeks' gestation between June 1999 and December 2008, and mothers, fathers, and offspring were followed up after birth. Novel genetic instruments for maternal antihypertensive drug targets that act via systolic blood pressure (SBP) were derived from individual-level data analyzed in January 2018. Two-sample multivariable MR analysis of these maternal drug targets and offspring outcomes were performed between January 2023 and April 2024.

Exposures

Maternal genetic variants associated with drug targets for treatments of hypertension, as specified in the National Health Service dictionary of medicines and devices.

Main Outcomes and Measures

Offspring outcomes were Apgar score at 1 minute and 5 minutes, offspring developmental score at 6 months, birth length, birth weight z score, gestational age, head circumference, and congenital malformation. Maternal hypertensive disorders of pregnancy were a positive control.

Results

The MoBa sample contained 29 849 family trios, with a mean (SD) maternal age of 30.2 (18.6) years and a mean (SD) paternal age of 32.8 (13.1) years; 51.1% of offspring were male. Seven independent SNVs were identified as influencing maternal SBP via the antihypertensive drug target instruments. For higher levels of maternal SBP acting through the CACNB2 calcium channel blocker target, the estimated change in gestational age was 3.99 days (95% CI, 0.02-7.96 days) per 10-mm Hg decrease in SBP. There was no evidence of differential risk for measured perinatal outcomes from maternal SBP acting through drug targets for multiple hypertensive subclasses, such as between the ADRB1 β-adrenoceptor-blocking target and risk of congenital malformation (estimated odds ratio, 0.28 [95% CI, 0.02-4.71] per 10-mm Hg decrease in SBP). Maternal and paternal SBP acting through the EDNRA vasodilator antihypertensive target did not have a potential causal effect on birth weight z score, with respective β estimates of 0.71 (95% CI, -0.09 to 1.51) and 0.72 (95% CI, -0.08 to 1.53) per 10-mm Hg decrease in SBP.

Conclusions and Relevance

The findings provided little evidence to indicate that perturbation of maternal genetic variants for SBP that influence antihypertensive drug targets had potential causal relationships with measures of perinatal development and health within this study. These findings may be triangulated with existing literature to guide physicians and mothers in decisions about antihypertensive use during pregnancy.

Polygenic Indices (a.k.a. Polygenic Scores) in Social Science: A Guide for Interpretation and Evaluation

Polygenic indices (PGI)-the new recommended label for polygenic scores (PGS) in social science-are genetic summary scales often used to represent an individual's liability for a disease, trait, or behavior based on the additive effects of measured genetic variants. Enthusiasm for linking genetic data with social outcomes and the inclusion of premade PGIs in social science datasets have facilitated increased uptake of PGIs in social science research-a trend that will likely continue. Yet, most social scientists lack the expertise to interpret and evaluate PGIs in social science research. Here, we provide a primer on PGIs for social scientists focusing on key concepts, unique statistical genetic considerations, and best practices in calculation, estimation, reporting, and interpretation. We summarize our recommended best practices as a checklist to aid social scientists in evaluating and interpreting studies with PGIs. We conclude by discussing the similarities between PGIs and standard social science scales and unique interpretative considerations.

Mendelian randomization studies of periodontitis: Understanding benefits and natural limitations in an applied context

Mendelian randomization (MR) is a flexible analytical tool that has been widely applied to strengthen causal inference in observational epidemiology and is now gaining attention in many areas including periodontal research. The interpretation of results drawn from MR is based on a series of assumptions, which can be unrealistic or difficult to meet faithfully in some settings. However, we argue that with care, this does not necessarily prevent valuable deployment of the approach. We argue that clarity of presentation as well as careful assessment of specific analytical conditions is a fundamental part of all MR analyses. To that end, awareness of its limitations should also guide the design of MR investigations and the presentation of results rather than rule out its use altogether. Notably, considerations similar to those known to be important in conventional epidemiological settings apply to MR. While MR studies are valuable in their contrast to other study limitations, the application of this technique must be carefully cross-examined. Specific considerations include possible confounders, recruitment strategy and phenotypic measurement and differential analysis properties across studies. In the case of periodontal research, current MR applications are limited by the available evidence base for genetic contributions to periodontitis; however, this sets a specific scene for the strategic use of MR and shines light on a need for greater research emphasis on the genetics of the condition and intermediaries. This article provides a perspective on the uses and inherent limitations of MR studies and the importance of adhering to basic epidemiological principles when designing them.

Association between serum total cholesterol and the development of gastric cancer: A two-way two-sample Mendelian randomization study

Previous epidemiologic studies have suggested a potential negative correlation between total cholesterol (TC) and Gastric cancer (GC); however, several observational studies have shown conflicting results and have failed to provide definitive evidence for a causal relationship between TC and GC. Therefore, we conducted a 2-sample bidirectional Mendelian randomization (MR) study to explore the genetic correlation and potential causal relationship between the 2 variables. We screened for single nucleotide polymorphisms (SNPs) associated with TC and GC utilizing a large-scale genome-wide association study (GWAS) public database. The causal relationship was analyzed using 5 MR analysis methods: inverse variance weighting (IVW), weighted median, MR-Egger regression, weighted mode, and simple mode. Additionally, reverse MR analysis was performed to evaluate the possibility of reverse causality. Sensitivity analyses were conducted, including heterogeneity tests, horizontal multiple validity tests, and leave-one-out tests. After meticulous screening, 79 SNPs were identified as instrumental variables (IVs). The IVW method revealed a causal relationship between TC and GC (OR = 0.844; 95% CI: 0.741-0.961; P = .01). Sensitivity analyses did not detect significant horizontal pleiotropy. Though heterogeneity was observed in the forward MR analysis (IVW, Qp = 0.0006), the results remained reliable as we utilized the IVW random-effects model as the primary analytical method. Furthermore, inverse MR analysis found no evidence of reverse causality between TC and GC, effectively ruling out the influence of GC on the reverse causality of TC. Our MR study provided evidence of a causal association between TC and GC, suggesting that TC acts as a protective factor against GC due to its negative association with the disease.

Investigating the impact of poverty on mental illness in the UK Biobank using Mendelian randomization

It is unclear whether poverty and mental illness are causally related. Using UK Biobank and Psychiatric Genomic Consortium data, we examined evidence of causal links between poverty and nine mental illnesses (attention deficit and hyperactivity disorder (ADHD), anorexia nervosa, anxiety disorder, autism spectrum disorder, bipolar disorder, major depressive disorder, obsessive-compulsive disorder, post-traumatic stress disorder and schizophrenia). We applied genomic structural equation modelling to derive a poverty common factor from household income, occupational income and social deprivation. Then, using Mendelian randomization, we found evidence that schizophrenia and ADHD causally contribute to poverty, while poverty contributes to major depressive disorder and schizophrenia but decreases the risk of anorexia nervosa. Poverty may also contribute to ADHD, albeit with uncertainty due to unbalanced pleiotropy. The effects of poverty were reduced by approximately 30% when we adjusted for cognitive ability. Further investigations of the bidirectional relationships between poverty and mental illness are warranted, as they may inform efforts to improve mental health for all.

Mendelian randomization analyses clarify the effects of height on cardiovascular diseases

An inverse correlation between stature and risk of coronary artery disease (CAD) has been observed in several epidemiologic studies, and recent Mendelian randomization (MR) experiments have suggested causal association. However, the extent to which the effect estimated by MR can be explained by cardiovascular, anthropometric, lung function, and lifestyle-related risk factors is unclear, with a recent report suggesting that lung function traits could fully explain the height-CAD effect. To clarify this relationship, we utilized a well-powered set of genetic instruments for human stature, comprising >1,800 genetic variants for height and CAD. In univariable analysis, we confirmed that a one standard deviation decrease in height (~6.5 cm) was associated with a 12.0% increase in the risk of CAD, consistent with previous reports. In multivariable analysis accounting for effects from up to 12 established risk factors, we observed a >3-fold attenuation in the causal effect of height on CAD susceptibility (3.7%, p = 0.02). However, multivariable analyses demonstrated independent effects of height on other cardiovascular traits beyond CAD, consistent with epidemiologic associations and univariable MR experiments. In contrast with published reports, we observed minimal effects of lung function traits on CAD risk in our analyses, indicating that these traits are unlikely to explain the residual association between height and CAD risk. In sum, these results suggest the impact of height on CAD risk beyond previously established cardiovascular risk factors is minimal and not explained by lung function measures.

Association of Crohn's disease and ulcerative colitis with the risk of neurological diseases: a large-scale Mendelian randomization study

Observational studies suggested increased risks of Alzheimer's disease (AD), Parkinson's disease (PD), and multiple sclerosis (MS) in patients with Crohn's disease (CD) and ulcerative colitis (UC). We aimed to assess the causality for the associations of CD and UC with the risks of AD, PD, and MS through a two-sample Mendelian randomization (MR) study. Independent single nucleotide polymorphisms associated with CD (17,897 cases and 33,977 controls) and UC (13,768 cases and 33,977 controls) were identified as genetic instruments based on a European-descent genome-wide association study (GWAS) released by the International Inflammatory Bowel Disease Genetics Consortium. Summary statistics for AD (combined: 25,881 cases and 256,837 controls), PD (combined: 35,836 cases and 665,686 controls), and MS (combined: 48,477 cases and 285,515 controls) were obtained from the largest GWASs and FinnGen study of European ancestry, respectively. MR estimates were generated using the inverse-variance weighted method in the main analysis with a series of sensitivity analyses. MR analyses were conducted per outcome database and were subsequently meta-analyzed to generate combined estimates. Genetically predicted UC was significantly associated with increased risks of AD (combined: OR, 1.03; 95% CI, 1.01-1.05; P = 1.80 × 10-3) and MS (combined: OR, 1.37; 95% CI, 1.23-1.53; P = 1.18 × 10-8), while there was no association between genetically predicted UC and the risk of PD. In contrast, no significant associations were observed for genetically predicted CD with AD, PD, and MS. MR-Egger regression showed no directional pleiotropy for the identified associations, and sensitivity analyses with different MR methods further confirmed these findings. This study suggested significant adverse effects of UC on AD and MS, highlighting that UC patients should receive early intervention with optimal adjunctive medical therapy to reduce the risks of AD and MS.

Use of Mendelian randomization to assess the causal status of modifiable exposures for rheumatic diseases

The explosion in Mendelian randomization (MR) publications is hard to ignore and shows no signs of slowing. Clinician readers, who may not be familiar with jargon-ridden methods, are expected to discern the good from the many low-quality studies that make overconfident claims of causality or stretch the plausibility of what MR can investigate. We aim to equip readers with foundational concepts, contextualized using examples in rheumatology, to appraise the many MR papers that are or will appear in their journals. We highlight the importance of assessing whether exposures are under plausibly specific genetic influence, whether the hypothesized causal pathways make biological sense, and whether results stand up to replication and use of control outcomes. Quality of research can vary substantially using MR as with any design, and all methods have inherent limitations. MR studies have provided and can still contribute valuable insights in the context of evidence triangulation.

Mendelian randomization: causal inference leveraging genetic data

Mendelian randomization (MR) leverages genetic information to examine the causal relationship between phenotypes allowing for the presence of unmeasured confounders. MR has been widely applied to unresolved questions in epidemiology, making use of summary statistics from genome-wide association studies on an increasing number of human traits. However, an understanding of essential concepts is necessary for the appropriate application and interpretation of MR. This review aims to provide a non-technical overview of MR and demonstrate its relevance to psychiatric research. We begin with the origins of MR and the reasons for its recent expansion, followed by an overview of its statistical methodology. We then describe the limitations of MR, and how these are being addressed by recent methodological advances. We showcase the practical use of MR in psychiatry through three illustrative examples - the connection between cannabis use and psychosis, the link between intelligence and schizophrenia, and the search for modifiable risk factors for depression. The review concludes with a discussion of the prospects of MR, focusing on the integration of multi-omics data and its extension to delineating complex causal networks.

The genetic landscape of substance use disorders

Substance use disorders represent a significant public health concern with considerable socioeconomic implications worldwide. Twin and family-based studies have long established a heritable component underlying these disorders. In recent years, genome-wide association studies of large, broadly phenotyped samples have identified regions of the genome that harbour genetic risk variants associated with substance use disorders. These regions have enabled the discovery of putative causal genes and improved our understanding of genetic relationships among substance use disorders and other traits. Furthermore, the integration of these data with clinical information has yielded promising insights into how individuals respond to medications, allowing for the development of personalized treatment approaches based on an individual's genetic profile. This review article provides an overview of recent advances in the genetics of substance use disorders and demonstrates how genetic data may be used to reduce the burden of disease and improve public health outcomes.

Cardiovascular disease risk factors and infertility: multivariable analyses and one-sample Mendelian randomization analyses in the Trøndelag Health Study

STUDY QUESTION

Are cardiovascular disease (CVD) risk factors causally associated with higher risk of infertility among women and men?

SUMMARY ANSWER

We found evidence to support a causal relationship between smoking initiation and history of infertility in women.

WHAT IS KNOWN ALREADY

Several CVD risk factors are associated with history of infertility. Previous studies using Mendelian randomization (MR) further support a causal relationship between BMI and infertility in women.

STUDY DESIGN SIZE DURATION

We used data from the Trøndelag Health Study (HUNT) in Norway, a prospective population-based cohort study, including 26 811 women and 15 598 men participating in three survey collections in 1995-1997 (HUNT2), 2006-2008 (HUNT3), and 2017-2019 (HUNT4).

PARTICIPANTS/MATERIALS SETTING METHODS

Our outcome was women's self-reported history of infertility, defined as ever having tried to conceive for 12 months or more or having used ART. We assigned the history of infertility reported by women to their male partners; therefore, the measure of infertility was on the couple level. We used both conventional multivariable analyses and one-sample MR analyses to evaluate the association between female and male CVD risk factors (including BMI, blood pressure, lipid profile measurements, and smoking behaviours) and history of infertility in women and men, separately.

MAIN RESULTS AND THE ROLE OF CHANCE

A total of 4702 women (18%) and 2508 men (16%) were classified with a history of infertility. We found a higher risk of infertility among female smokers compared to non-smokers in both multivariable and MR analyses (odds ratio (OR) in multivariable analysis, 1.20; 95% CI, 1.12-1.28; OR in MR analysis, 1.13; CI, 1.02-1.26), and potentially for higher BMI (OR in multivariable analysis, 1.13; CI, 1.09-1.18; OR in MR analysis, 1.11, CI, 0.92-1.34). In multivariable analysis in women, we also found evidence of associations between triglyceride levels, high-density lipoprotein cholesterol, lifetime smoking index, and smoking intensity with higher risk of infertility. However, these results were not consistent in MR analyses. We found no robust or consistent associations between male CVD risk factors and infertility.

LIMITATIONS REASONS FOR CAUTION

Our main limitation was that the CVD risk factors measured might not adequately capture the relevant time periods for when couples were trying to conceive. Additionally, we did not have information on causes of infertility in either women or men.

WIDER IMPLICATIONS OF THE FINDINGS

Women with infertility could have a worse CVD risk factor profile and thus public health interventions aimed at reducing the impact of some CVD risk factors, such as smoking and BMI, could reduce the burden of infertility. However, additional MR studies of the relationship between CVD risk factors and infertility with a larger sample size would be of value.

STUDY FUNDING/COMPETING INTERESTS

The study was supported by a grant from the European Research Council under the European Union's Horizon 2020 research and innovation program (grant agreements no. 947684). This research was also supported by the Research Council of Norway through its Centres of Excellence funding scheme (project no. 262700) and partly funded by the Research Council of Norway, project: Women's fertility-an essential component of health and well-being (project no. 320656). D.A.L. and A.F. work in a unit that is supported by the University of Bristol and the UK Medical Research Council (MC_UU_00011/6). D.A.L.'s contribution to the article is supported by the European Research Council (101021566), the British Heart Foundation (CH/F/20/90003 and AA/18/7/34219). S.B.'s contribution to the article is supported by the Wellcome Trust (225790/Z/22/Z). B.M.B. is funded by The Liaison Committee for education, research and innovation in Central Norway; and the Joint Research Committee between St. Olavs Hospital and the Faculty of Medicine and Health Sciences, NTNU. The genotyping in HUNT was financed by the National Institute of Health (NIH); University of Michigan; The Research Council of Norway; The Liaison Committee for education, research and innovation in Central Norway; and the Joint Research Committee between St. Olavs Hospital and the Faculty of Medicine and Health Sciences, NTNU. None of the funding organizations influenced the study design, reporting, or interpretation of results. The views expressed in the present article are those of the authors and not necessarily any acknowledged funding organization. D.A.L. reports grants from Medtronic Ltd and Roche Diagnostics outside the submitted work. The other authors have no conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.

Analysis of admixed Greenlandic siblings shows that the mean genotypic values for metabolic phenotypes differ between Inuit and Europeans

BACKGROUND

Disease prevalence and mean phenotype values differ between many populations, including Inuit and Europeans. Whether these differences are partly explained by genetic differences or solely due to differences in environmental exposures is still unknown, because estimates of the genetic contribution to these means, which we will here refer to as mean genotypic values, are easily confounded, and because studies across genetically diverse populations are lacking.

METHODS

Leveraging the unique genetic properties of the small, admixed and historically isolated Greenlandic population, we estimated the differences in mean genotypic value between Inuit and European genetic ancestry using an admixed sibling design. Analyses were performed across 26 metabolic phenotypes, in 1474 admixed sibling pairs present in a cohort of 5996 Greenlanders.

RESULTS

After FDR correction for multiple testing, we found significantly lower mean genotypic values in Inuit genetic ancestry compared to European genetic ancestry for body weight (effect size per percentage of Inuit genetic ancestry (se), -0.51 (0.16) kg/%), body mass index (-0.20 (0.06) kg/m2/%), fat percentage (-0.38 (0.13) %/%), waist circumference (-0.42 (0.16) cm/%), hip circumference (-0.38 (0.11) cm/%) and fasting serum insulin levels (-1.07 (0.51) pmol/l/%). The direction of the effects was consistent with the observed mean phenotype differences between Inuit and European genetic ancestry. No difference in mean genotypic value was observed for height, markers of glucose homeostasis, or circulating lipid levels.

CONCLUSIONS

We show that mean genotypic values for some metabolic phenotypes differ between two human populations using a method not easily confounded by possible differences in environmental exposures. Our study illustrates the importance of performing genetic studies in diverse populations.

Is low-level alcohol consumption really health-protective? A critical review of approaches to promote causal inference and recent applications

Heavy and disordered alcohol consumption is a known risk factor for several health conditions and is associated with considerable disease burden. However, at low-to-moderate levels, evidence suggests that drinking is associated with reduced risk for certain health outcomes. Whether these findings represent genuine protective effects or mere methodological artifacts remains unclear, but has substantial consequences for policy and practice. This critical review introduces methodological advances capable of enhancing causal inference from observational research, focusing on the 'G-methods' and Mendelian Randomization. We also present and evaluate recent research applying these methods and compare findings to the existing evidence base. Future directions are proposed for improving our causal understanding of the relationships between alcohol and long-term health outcomes.

Disentangling potential causal effects of educational duration on well-being, and mental and physical health outcomes

BACKGROUND

Extensive research has focused on the potential benefits of education on various mental and physical health outcomes. However, whether the associations reflect a causal effect is harder to establish.

METHODS

To examine associations between educational duration and specific aspects of well-being, anxiety and mood disorders, and cardiovascular health in a sample of European Ancestry UK Biobank participants born in England and Wales, we apply four different causal inference methods (a natural policy experiment leveraging the minimum school-leaving age, a sibling-control design, Mendelian randomization [MR], and within-family MR), and assess if the methods converge on the same conclusion.

RESULTS

A comparison of results across the four methods reveals that associations between educational duration and these outcomes appears predominantly to be the result of confounding or bias rather than a true causal effect of education on well-being and health outcomes. Although we do consistently find no associations between educational duration and happiness, family satisfaction, work satisfaction, meaning in life, anxiety, and bipolar disorder, we do not find consistent significant associations across all methods for the other phenotypes (health satisfaction, depression, financial satisfaction, friendship satisfaction, neuroticism, and cardiovascular outcomes).

CONCLUSIONS

We discuss inconsistencies in results across methods considering their respective limitations and biases, and additionally discuss the generalizability of our findings in light of the sample and phenotype limitations. Overall, this study strengthens the idea that triangulation across different methods is necessary to enhance our understanding of the causal consequences of educational duration.

Graves' disease and the risk of five autoimmune diseases: A Mendelian randomization and colocalization study

BACKGROUND

Epidemiological studies have consistently demonstrated a high prevalence of concurrent autoimmune diseases in individuals with Graves' disease (GD).

OBJECTIVE

The objective of this study is to establish a causal association between GD and autoimmune diseases.

METHODS

We employed a two-sample Mendelian randomization (MR) to infer a causal association between GD and five autoimmune diseases, namely rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), Crohn's disease (CD), ulcerative colitis (UC), and amyotrophic lateral sclerosis (ALS), in the East Asian and European population. Genetic correlations were explored through linkage disequilibrium score regression analysis (LDSC). Finally, colocalization analyses were performed to investigate possible genetic foundations.

RESULTS

Bidirectional MR analysis indicated that genetically predicted GD increased the risk of RA (Odds Ratio (OR): 1.34, 95 % Confidence Interval (CI): 1.21 to 1.47, P < 0.001) and SLE (OR: 1.21, 95%CI: 1.08 to 1.35, P < 0.001) in the East Asian population. In contrast, we found that genetically predicted RA (OR: 1.14, 95%CI: 1.05 to 1.24, P = 0.002) and SLE (OR: 1.10, 95%CI: 1.03 to 1.17, P = 0.003) were associated with a higher risk of GD. The results have been partially validated in European cohorts. Colocalization analysis suggested the potential existence of shared causal variants between GD and other autoimmune diseases. In particular, gene ARID5B may play an important role in the incidence of autoimmune diseases.

CONCLUSION

This study has confirmed that GD was associated with RA and SLE and found a possible key gene ARID5B. It may be necessary to strengthen detection to prevent the occurrence of comorbidities in clinical practice.

Identifying the causal relationship between immune factors and osteonecrosis: a two-sample Mendelian randomization study

A wealth of evidence intimates a profound connection between the immune system and osteonecrosis, albeit the specific immune factors underlying this connection remain largely veiled. A bidirectional Mendelian randomization (MR) study was conducted based on genome-wide association study summary data to identify causal links between 731 immune factors and osteonecrosis including drug-induced osteonecrosis. Preliminary MR analysis was accomplished utilizing the inverse-variance weighted method under a multiplicative random effects model, and heterogeneity and potential horizontal pleiotropy were evaluated through Cochrane's Q-test, MR-Egger intercept test, MR-PRESSO global test, and leave-one-out analysis. Upon false discovery rate correction, the gene-predicted level of one immune factor (CD62L - monocyte %monocyte) exhibited a significant positive correlation with osteonecrosis, while eight immune traits associated with monocytes, dendritic cells, and NK cells demonstrated significant causal effects with drug-induced osteonecrosis. Reverse MR revealed no significant correlations. This MR research provides genetic evidence for the causal associations between a broad spectrum of immune factors and osteonecrosis. Such a study aids in unraveling the intricate interaction patterns between the immune and skeletal systems, elucidating the pathogenesis of osteonecrosis, and identifying potential novel therapeutic approaches.

Tissue-specific genetic variation suggests distinct molecular pathways between body shape phenotypes and colorectal cancer

It remains unknown whether adiposity subtypes are differentially associated with colorectal cancer (CRC). To move beyond single-trait anthropometric indicators, we derived four multi-trait body shape phenotypes reflecting adiposity subtypes from principal components analysis on body mass index, height, weight, waist-to-hip ratio, and waist and hip circumference. A generally obese (PC1) and a tall, centrally obese (PC3) body shape were both positively associated with CRC risk in observational analyses in 329,828 UK Biobank participants (3728 cases). In genome-wide association studies in 460,198 UK Biobank participants, we identified 3414 genetic variants across four body shapes and Mendelian randomization analyses confirmed positive associations of PC1 and PC3 with CRC risk (52,775 cases/45,940 controls from GECCO/CORECT/CCFR). Brain tissue-specific genetic instruments, mapped to PC1 through enrichment analysis, were responsible for the relationship between PC1 and CRC, while the relationship between PC3 and CRC was predominantly driven by adipose tissue-specific genetic instruments. This study suggests distinct putative causal pathways between adiposity subtypes and CRC.

Assessing causal links between age at menarche and adolescent mental health: a Mendelian randomisation study

BACKGROUND

The timing of puberty may have an important impact on adolescent mental health. In particular, earlier age at menarche has been associated with elevated rates of depression in adolescents. Previous research suggests that this relationship may be causal, but replication and an investigation of whether this effect extends to other mental health domains is warranted.

METHODS

In this Registered Report, we triangulated evidence from different causal inference methods using a new wave of data (N = 13,398) from the Norwegian Mother, Father, and Child Cohort Study. We combined multiple regression, one- and two-sample Mendelian randomisation (MR), and negative control analyses (using pre-pubertal symptoms as outcomes) to assess the causal links between age at menarche and different domains of adolescent mental health.

RESULTS

Our results supported the hypothesis that earlier age at menarche is associated with elevated depressive symptoms in early adolescence based on multiple regression (β =  - 0.11, 95% CI [- 0.12, - 0.09], pone-tailed < 0.01). One-sample MR analyses suggested that this relationship may be causal (β =  - 0.07, 95% CI [- 0.13, 0.00], pone-tailed = 0.03), but the effect was small, corresponding to just a 0.06 standard deviation increase in depressive symptoms with each earlier year of menarche. There was also some evidence of a causal relationship with depression diagnoses during adolescence based on one-sample MR (OR = 0.74, 95% CI [0.54, 1.01], pone-tailed = 0.03), corresponding to a 29% increase in the odds of receiving a depression diagnosis with each earlier year of menarche. Negative control and two-sample MR sensitivity analyses were broadly consistent with this pattern of results. Multivariable MR analyses accounting for the genetic overlap between age at menarche and childhood body size provided some evidence of confounding. Meanwhile, we found little consistent evidence of effects on other domains of mental health after accounting for co-occurring depression and other confounding.

CONCLUSIONS

We found evidence that age at menarche affected diagnoses of adolescent depression, but not other domains of mental health. Our findings suggest that earlier age at menarche is linked to problems in specific domains rather than adolescent mental health in general.

Genetic associations of risk behaviours and educational achievement

Risk behaviours are common in adolescent and persist into adulthood, people who engage in more risk behaviours are more likely to have lower educational attainment. We applied genetic causal inference methods to explore the causal relationship between adolescent risk behaviours and educational achievement. Risk behaviours were phenotypically associated with educational achievement at age 16 after adjusting for confounders (-0.11, 95%CI: -0.11, -0.09). Genomic-based restricted maximum likelihood (GREML) results indicated that both traits were heritable and have a shared genetic architecture (Riskh 2  = 0.18, 95% CI: -0.11,0.47; educationh 2  = 0.60, 95%CI: 0.50,0.70). Consistent with the phenotypic results, genetic variation associated with risk behaviour was negatively associated with education (r g  = -0.51, 95%CI: -1.04,0.02). Lastly, the bidirectional MR results indicate that educational achievement or a closely related trait is likely to affect risk behaviours PGI (β=-1.04, 95% CI: -1.41, -0.67), but we found little evidence that the genetic variation associated with risk behaviours affected educational achievement (β=0.00, 95% CI: -0.24,0.24). The results suggest engagement in risk behaviour may be partly driven by educational achievement or a closely related trait.

Interpreting population- and family-based genome-wide association studies in the presence of confounding

A central aim of genome-wide association studies (GWASs) is to estimate direct genetic effects: the causal effects on an individual's phenotype of the alleles that they carry. However, estimates of direct effects can be subject to genetic and environmental confounding and can also absorb the "indirect" genetic effects of relatives' genotypes. Recently, an important development in controlling for these confounds has been the use of within-family GWASs, which, because of the randomness of mendelian segregation within pedigrees, are often interpreted as producing unbiased estimates of direct effects. Here, we present a general theoretical analysis of the influence of confounding in standard population-based and within-family GWASs. We show that, contrary to common interpretation, family-based estimates of direct effects can be biased by genetic confounding. In humans, such biases will often be small per-locus, but can be compounded when effect-size estimates are used in polygenic scores (PGSs). We illustrate the influence of genetic confounding on population- and family-based estimates of direct effects using models of assortative mating, population stratification, and stabilizing selection on GWAS traits. We further show how family-based estimates of indirect genetic effects, based on comparisons of parentally transmitted and untransmitted alleles, can suffer substantial genetic confounding. We conclude that, while family-based studies have placed GWAS estimation on a more rigorous footing, they carry subtle issues of interpretation that arise from confounding.

Genome-wide association studies have problems due to confounding: Are family-based designs the answer?

Genome-wide association studies (GWASs) can be affected by confounding. Family-based GWAS uses random, within-family genetic variation to avoid this. A study in PLOS Biology details how different sources of confounding affect GWAS and whether family-based designs offer a solution.

Breast and bowel cancers diagnosed in people 'too young to have cancer': A blueprint for research using family and twin studies

Young breast and bowel cancers (e.g., those diagnosed before age 40 or 50 years) have far greater morbidity and mortality in terms of years of life lost, and are increasing in incidence, but have been less studied. For breast and bowel cancers, the familial relative risks, and therefore the familial variances in age-specific log(incidence), are much greater at younger ages, but little of these familial variances has been explained. Studies of families and twins can address questions not easily answered by studies of unrelated individuals alone. We describe existing and emerging family and twin data that can provide special opportunities for discovery. We present designs and statistical analyses, including novel ideas such as the VALID (Variance in Age-specific Log Incidence Decomposition) model for causes of variation in risk, the DEPTH (DEPendency of association on the number of Top Hits) and other approaches to analyse genome-wide association study data, and the within-pair, ICE FALCON (Inference about Causation from Examining FAmiliaL CONfounding) and ICE CRISTAL (Inference about Causation from Examining Changes in Regression coefficients and Innovative STatistical AnaLysis) approaches to causation and familial confounding. Example applications to breast and colorectal cancer are presented. Motivated by the availability of the resources of the Breast and Colon Cancer Family Registries, we also present some ideas for future studies that could be applied to, and compared with, cancers diagnosed at older ages and address the challenges posed by young breast and bowel cancers.

Causation and familial confounding as explanations for the associations of polygenic risk scores with breast cancer: Evidence from innovative ICE FALCON and ICE CRISTAL analyses

A polygenic risk score (PRS) combines the associations of multiple genetic variants that could be due to direct causal effects, indirect genetic effects, or other sources of familial confounding. We have developed new approaches to assess evidence for and against causation by using family data for pairs of relatives (Inference about Causation from Examination of FAmiliaL CONfounding [ICE FALCON]) or measures of family history (Inference about Causation from Examining Changes in Regression coefficients and Innovative STatistical AnaLyses [ICE CRISTAL]). Inference is made from the changes in regression coefficients of relatives' PRSs or PRS and family history before and after adjusting for each other. We applied these approaches to two breast cancer PRSs and multiple studies and found that (a) for breast cancer diagnosed at a young age, for example, <50 years, there was no evidence that the PRSs were causal, while (b) for breast cancer diagnosed at later ages, there was consistent evidence for causation explaining increasing amounts of the PRS-disease association. The genetic variants in the PRS might be in linkage disequilibrium with truly causal variants and not causal themselves. These PRSs cause minimal heritability of breast cancer at younger ages. There is also evidence for nongenetic factors shared by first-degree relatives that explain breast cancer familial aggregation. Familial associations are not necessarily due to genes, and genetic associations are not necessarily causal.

Educational attainment, income, and attention deficit hyperactivity disorder: A mediation analysis based on two-step Mendelian randomization

Previous studies have reported the relationship between educational attainment and attention deficit hyperactivity disorder (ADHD). However, the mechanism of this relationship remains unknown. It is well known that educational attainment correlates with income. Therefore, based on summary data from a genome-wide association study we used two-step Mendelian randomization (MR) to explore the role of income between education and ADHD. The inverse variance weighted (IVW) method was used in our analysis. The IVW results suggested that educational attainment and income were protective factors against ADHD. Educational attainment affects ADHD through income [ADHD: Beta = -0.68, 95% confidence interval (CI) = -0.87, -0.49; female: Beta = -0.87, 95% CI = -1.28, -0.47; male: Beta = -1.01, 95% CI = -1.34, -0.68; childhood: Beta = -0.52, 95% CI = -0.74, -0.30; late-diagnosed: Beta = -0.78, 95% CI = -1.11, -0.47; persistent: Beta = -0.82, 95% CI = -1.33, -0.31]. Income also affected ADHD through educational attainment [female: Beta = -1.08, 95% CI = -1.35, -0.83; male: Beta = -1.16, 95% CI = -1.57, -0.77; persistent: Beta = -1.48, 95% CI = -2.09, -0.94]. In the final analysis, data with heterogeneity were analyzed using IVW random effects results. The mechanism is that income will mediate the relationship between educational attainment and ADHD.

What maternal educational mobility tells us about the mother's parenting routines, offspring school achievement and intelligence

BACKGROUND

Educational mobility at the macro-level is a common measure of social inequality. Nonetheless, the correlates of mobility of education at the individual level are less well studied. We evaluated whether educational mobility of the second generation (compared to the first generation level) predicts differences in parenting practices of the second generation and school achievement and intelligence in the third generation.

METHODS

Data from a population-based cohort of children in the Netherlands (N = 3547; 49.4% boys) were analyzed. Maternal, grandparental education and family routines, a parenting practice, were reported by the mother. Child school achievement at the end of primary school (∼12 years, with the national Dutch academic test score) and child intelligence (∼6 and 13 years) were measured in a standardized manner. Also, a child genome-wide polygenic score of academic attainment was calculated. To estimate the effect of educational mobility, inverse probability-weighted linear models and Diagonal Reference Models (DRM) were used.

RESULTS

Upward maternal educational mobility was associated with better offspring school achievement, higher intelligence, and more family routines if compared to offspring of mothers with no upward mobility. However, mothers did not implement the same level of family routines as similarly educated mothers and grandfathers who already had achieved this educational level. Likewise, children of mothers with upward educational mobility had lower school achievement and intelligence than children of similarly educated mothers with no mobility. Child's genetic potential for education followed a similar association pattern with higher potential in children of upward mobile mothers.

CONCLUSION

Policymakers might overlook social inequalities when focused on parental socioeconomic status. Grandparental socioeconomic status, which independently predicts child school achievement, intelligence, and parental family routines, should also be assessed. The child's genetic endowment reflects the propensity for education across generations that partly underlies mobility and some of its effect on the offspring.

Association of educational attainment with hypertension and type-2 diabetes: A Mendelian randomization study

BACKGROUNDDue to the long time interval between exposure and outcome, it is difficult to infer the causal relationship between educational attainment (EA) and common chronic diseases. Therefore, we utilized Mendelian randomization (MR) to predict the causal relationships of EA with hypertension and type-2 diabetes (T2DM). METHODSA two-sample MR analysis was conducted using genome-wide association studies (GWASs) combined with inferential measurements. A GWAS meta-analysis including 1,131,881 European individuals was used to identify instruments for EA. Hypertension and T2DM data were obtained from a Finnish database. MR analyses were performed using inverse-variance weighted meta-analysis (IVW), weighted median regression, MR‒Egger regression, simple mode regression, weighted mode regression and the MR-Pleiotropy RESidual Sum and Outlier test. Sensitivity analyses were further performed using the leave-one-out method to test the robustness of our findings. RESULTSUsing the MR approach, our results showed that EA was significantly associated with a reduced risk of hypertension (OR = 0.63; P = 2.94 × 10-47; [95% CI: 0.59, 0.67]) and type-2 diabetes (OR = 0.59; P = 1.25 × 10-16; [95% CI: 0.52, 0.67]). CONCLUSIONSThis study showed that EA is causally linked to the risk of chronic diseases, including high blood pressure and T2DM.

Associations between type 2 diabetes mellitus and chronic liver diseases: evidence from a Mendelian randomization study in Europeans and East Asians

Objective

Multiple observational studies have demonstrated an association between type 2 diabetes mellitus (T2DM) and chronic liver diseases (CLDs). However, the causality of T2DM on CLDs remained unknown in various ethnic groups.

Methods

We obtained instrumental variables for T2DM and conducted a two-sample mendelian randomization (MR) study to examine the causal effect on nonalcoholic fatty liver disease (NAFLD), hepatocellular carcinoma (HCC), viral hepatitis, hepatitis B virus (HBV) infection, and hepatitis C virus (HCV) infection risk in Europeans and East Asians. The primary analysis utilized the inverse variance weighting (IVW) technique to evaluate the causal relationship between T2DM and CLDs. In addition, we conducted a series of rigorous analyses to bolster the reliability of our MR results.

Results

In Europeans, we found that genetic liability to T2DM has been linked with increased risk of NAFLD (IVW : OR =1.3654, 95% confidence interval [CI], 1.2250-1.5219, p=1.85e-8), viral hepatitis (IVW : OR =1.1173, 95%CI, 1.0271-1.2154, p=0.0098), and a suggestive positive association between T2DM and HCC (IVW : OR=1.2671, 95%CI, 1.0471-1.5333, p=0.0150), HBV (IVW : OR=1.1908, 95% CI, 1.0368-1.3677, p=0.0134). No causal association between T2DM and HCV was discovered. Among East Asians, however, there was a significant inverse association between T2DM and the proxies of NAFLD (ALT: IVW OR=0.9752, 95%CI 0.9597-0.9909, p=0.0021; AST: IVW OR=0.9673, 95%CI, 0.9528-0.9821, p=1.67e-5), and HCV (IVW: OR=0.9289, 95%CI, 0.8852-0.9747, p=0.0027). Notably, no causal association was found between T2DM and HCC, viral hepatitis, or HBV.

Conclusion

Our MR analysis revealed varying causal associations between T2DM and CLDs in East Asians and Europeans. Further research is required to investigate the potential mechanisms in various ethnic groups, which could yield new insights into early screening and prevention strategies for CLDs in T2DM patients.

Obesity, Glycemic Traits, Lifestyle Factors, and Risk of Facial Aging: A Mendelian Randomization Study in 423,999 Participants

BACKGROUND

Several recent observational studies have associated obesity, lifestyle factors (smoking, sleep duration, and alcohol drinking), and glycemic traits with facial aging. However, whether this relationship is causal due to confounding and reverse causation is yet to be substantiated.

AIMS

We aimed to assess these relationships using Mendelian randomization (MR).

METHODS

For the instrumental variables, this paper selected independent single nucleotide polymorphisms (SNPs) linked to the exposures at a genome-wide state (P < 5 × 10-8) in equivalent genome-wide association studies (GWAS). Using the UK Biobank, we obtained summary-level data for facial aging on 423,999 individuals. The primary assessments were performed through the combination of complementing techniques (simple method approaches, weighted model, MR-Egger, and weighted median) and the inverse-variance-weighted method. Along with that, we examined the heterogeneity and horizontal pleiotropy through different types of sensitivity analyses.

RESULTS

The correlations were (a) facial aging for body mass index (BMI, OR = 1.054, 95% CI 1.044-1.64), (b) waist/hip ratio (OR = 1.056, 95% CI 1.023-1.091), and (c) smoking (OR = 1.023, 95% CI 1.007-1.039). Equally important, the correlations for waist/hip ratio remained robust after adjusting for the genetically predicted BMI (OR = 1.028, 95% CI 1.003-1.054). However, no causal effects of alcoholic drinking, glycemic traits, and sleep duration on facial aging were observed.

CONCLUSIONS

The outcomes shed light on the potential correlation of obesity and cigarette smoking with facial aging while putting forward a more comprehensive and credible foundation for the optimization of facial aging strategies.

NO LEVEL ASSIGNED

This journal requires that authors assign a level of evidence to each submission to which Evidence-Based Medicine rankings are applicable. This excludes Review Articles, Book Reviews, and manuscripts that concern Basic Science, Animal Studies, Cadaver Studies, and Experimental Studies. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Using the global randomization test as a Mendelian randomization falsification test for the exclusion restriction assumption

Mendelian randomization may give biased causal estimates if the instrument affects the outcome not solely via the exposure of interest (violating the exclusion restriction assumption). We demonstrate use of a global randomization test as a falsification test for the exclusion restriction assumption. Using simulations, we explored the statistical power of the randomization test to detect an association between a genetic instrument and a covariate set due to (a) selection bias or (b) horizontal pleiotropy, compared to three approaches examining associations with individual covariates: (i) Bonferroni correction for the number of covariates, (ii) correction for the effective number of independent covariates, and (iii) an r2 permutation-based approach. We conducted proof-of-principle analyses in UK Biobank, using CRP as the exposure and coronary heart disease (CHD) as the outcome. In simulations, power of the randomization test was higher than the other approaches for detecting selection bias when the correlation between the covariates was low (r2 < 0.1), and at least as powerful as the other approaches across all simulated horizontal pleiotropy scenarios. In our applied example, we found strong evidence of selection bias using all approaches (e.g., global randomization test p < 0.002). We identified 51 of the 58 CRP genetic variants as horizontally pleiotropic, and estimated effects of CRP on CHD attenuated somewhat to the null when excluding these from the genetic risk score (OR = 0.96 [95% CI: 0.92, 1.00] versus 0.97 [95% CI: 0.90, 1.05] per 1-unit higher log CRP levels). The global randomization test can be a useful addition to the MR researcher's toolkit.

Combining Mendelian randomization with the sibling comparison design

Mendelian randomization (MR) is a popular epidemiologic study design that uses genetic variants as instrumental variables (IVs) to estimate causal effects, while accounting for unmeasured confounding. The validity of the MR design hinges on certain IV assumptions, which may sometimes be violated due to dynastic effects, population stratification, or assortative mating. Since these mechanisms act through parental factors it was recently suggested that the bias resulting from violations of the IV assumptions can be reduced by combing the MR design with the sibling comparison design, which implicitly controls for all factors that are constant within families. In this article, we provide a formal discussion of this combined MR-sibling design. We derive conditions under which the MR-sibling design is unbiased, and we relate these to the corresponding conditions for the standard MR and sibling comparison designs. We proceed by considering scenarios where all three designs are biased to some extent, and discuss under which conditions the MR-sibling design can be expected to have less bias than the other two designs. We finally illustrate the theoretical results and conclusions with an application to real data, in a study of low-density lipoprotein and diastolic blood pressure using data from the Swedish Twin Registry.

Sex-stratified genome-wide association and transcriptome-wide Mendelian randomization studies reveal drug targets of heart failure

The prevalence of heart failure (HF) subtypes, which are classified by left ventricular ejection fraction (LVEF), demonstrate significant sex differences. Here, we perform sex-stratified genome-wide association studies (GWASs) on LVEF and transcriptome-wide Mendelian randomization (MR) on LVEF, all-cause HF, HF with reduced ejection fraction (HFrEF), and HF with preserved ejection fraction (HFpEF). The sex-stratified GWASs of LVEF identified three sex-specific loci that were exclusively detected in the sex-stratified GWASs. Three drug target genes show sex-differential effects on HF/HFrEF via influencing LVEF, with NPR2 as the target gene for the HF drug Cenderitide under phase 2 clinical trial. Our study highlights the importance of considering sex-differential genetic effects in sex-balanced diseases such as HF and emphasizes the value of sex-stratified GWASs and MR in identifying putative genetic variants, causal genes, and candidate drug targets for HF, which is not identifiable using a sex-combined strategy.

PheWAS-based clustering of Mendelian Randomisation instruments reveals distinct mechanism-specific causal effects between obesity and educational attainment

Mendelian Randomisation (MR) estimates causal effects between risk factors and complex outcomes using genetic instruments. Pleiotropy, heritable confounders, and heterogeneous causal effects violate MR assumptions and can lead to biases. To alleviate these, we propose an approach employing a Phenome-Wide association Clustering of the MR instruments (PWC-MR) and apply this method to revisit the surprisingly large apparent causal effect of body mass index (BMI) on educational attainment (EDU): [Formula: see text] = -0.19 [-0.22, -0.16]. First, we cluster 324 BMI-associated genetic instruments based on their association with 407 traits in the UK Biobank, which yields six distinct groups. Subsequent cluster-specific MR reveals heterogeneous causal effect estimates on EDU. A cluster enriched for socio-economic indicators yields the largest BMI-on-EDU causal effect estimate ([Formula: see text] = -0.49 [-0.56, -0.42]) whereas a cluster enriched for body-mass specific traits provides a more likely estimate ([Formula: see text] = -0.09 [-0.13, -0.05]). Follow-up analyses confirms these findings: within-sibling MR ([Formula: see text] = -0.05 [-0.09, -0.01]); MR for childhood BMI on EDU ([Formula: see text] = -0.03 [-0.06, -0.002]); step-wise multivariable MR ([Formula: see text] = -0.05 [-0.07, -0.02]) where socio-economic indicators are jointly modelled. Here we show how the in-depth examination of the BMI-EDU causal relationship demonstrates the utility of our PWC-MR approach in revealing distinct pleiotropic pathways and confounder mechanisms.

Pleiotropy and genetically inferred causality linking multisite chronic pain to substance use disorders

Individuals suffering from chronic pain develop substance use disorders (SUDs) more often than others. Understanding the shared genetic influences underlying the comorbidity between chronic pain and SUDs will lead to a greater understanding of their biology. Genome-wide association statistics were obtained from the UK Biobank for multisite chronic pain (MCP, Neffective = 387,649) and from the Million Veteran Program and the Psychiatric Genomics Consortium meta-analyses for alcohol use disorder (AUD, Neffective = 296,974), cannabis use disorder (CanUD, Neffective = 161,053), opioid use disorder (OUD, Neffective = 57,120), and problematic tobacco use (PTU, Neffective = 270,120). SNP-based heritability was estimated for each of the traits and genetic correlation (rg) analyses were performed to assess MCP-SUD pleiotropy. Bidirectional Mendelian Randomization analyses evaluated possible causal relationships. Finally, to identify and characterize individual loci, we performed a genome-wide pleiotropy analysis and a brain-wide analysis using imaging phenotypes available from the UK Biobank. MCP was positively genetically correlated with AUD (rg = 0.26, p = 7.55 × 10-18), CanUD (rg = 0.37, p = 8.21 × 10-37), OUD (rg = 0.20, p = 1.50 × 10-3), and PTU (rg = 0.29, p = 8.53 × 10-12). Although the MR analyses supported bi-directional relationships, MCP had larger effects on AUD (pain-exposure: beta = 0.18, p = 8.21 × 10-4; pain-outcome: beta = 0.07, p = 0.018), CanUD (pain-exposure: beta = 0.58, p = 2.70 × 10-6; pain-outcome: beta = 0.05, p = 0.014) and PTU (pain-exposure: beta = 0.43, p = 4.16 × 10-8; pain-outcome: beta = 0.09, p = 3.05 × 10-6) than the reverse. The genome-wide analysis identified two SNPs pleiotropic between MCP and all SUD investigated: IHO1 rs7652746 (ppleiotropy = 2.69 × 10-8), and CADM2 rs1248857 (ppleiotropy = 1.98 × 10-5). In the brain-wide analysis, rs7652746 was associated with multiple cerebellum and amygdala imaging phenotypes. When analyzing MCP pleiotropy with each SUD separately, we found 25, 22, and 4 pleiotropic variants for AUD, CanUD, and OUD, respectively. To our knowledge, this is the first large-scale study to provide evidence of potential causal relationships and shared genetic mechanisms underlying MCP-SUD comorbidity.

Beyond the factor indeterminacy problem using genome-wide association data

Latent factors, such as general intelligence, depression and risk tolerance, are invoked in nearly all social science research where a construct is measured via aggregation of symptoms, question responses or other measurements. Because latent factors cannot be directly observed, they are inferred by fitting a specific model to empirical patterns of correlations among measured variables. A long-standing critique of latent factor theories is that the correlations used to infer latent factors can be produced by alternative data-generating mechanisms that do not include latent factors. This is referred to as the factor indeterminacy problem. Researchers have recently begun to overcome this problem by using information on the associations between individual genetic variants and measured variables. We review historical work on the factor indeterminacy problem and describe recent efforts in genomics to rigorously test the validity of latent factors, advancing the understanding of behavioural science constructs.

More than nature and nurture, indirect genetic effects on children's academic achievement are consequences of dynastic social processes

Families transmit genes and environments across generations. When parents' genetics affect their children's environments, these two modes of inheritance can produce an 'indirect genetic effect'. Such indirect genetic effects may account for up to half of the estimated genetic variance in educational attainment. Here we tested if indirect genetic effects reflect within-nuclear-family transmission ('genetic nurture') or instead a multi-generational process of social stratification ('dynastic effects'). We analysed indirect genetic effects on children's academic achievement in their fifth to ninth years of schooling in N = 37,117 parent-offspring trios in the Norwegian Mother, Father, and Child Cohort Study (MoBa). We used pairs of genetically related families (parents were siblings, children were cousins; N = 10,913) to distinguish within-nuclear-family genetic-nurture effects from dynastic effects shared by cousins in different nuclear families. We found that indirect genetic effects on children's academic achievement cannot be explained by processes that operate exclusively within the nuclear family.

The causal effect of HbA1c on white matter brain aging by two-sample Mendelian randomization analysis

Background

Poor glycemic control with elevated levels of hemoglobin A1c (HbA1c) is associated with increased risk of cognitive impairment, with potentially varying effects between sexes. However, the causal impact of poor glycemic control on white matter brain aging in men and women is uncertain.

Methods

We used two nonoverlapping data sets from UK Biobank cohort: gene-outcome group (with neuroimaging data, (N = 15,193; males/females: 7,101/8,092)) and gene-exposure group (without neuroimaging data, (N = 279,011; males/females: 122,638/156,373)). HbA1c was considered the exposure and adjusted "brain age gap" (BAG) was calculated on fractional anisotropy (FA) obtained from brain imaging as the outcome, thereby representing the difference between predicted and chronological age. The causal effects of HbA1c on adjusted BAG were studied using the generalized inverse variance weighted (gen-IVW) and other sensitivity analysis methods, including Mendelian randomization (MR)-weighted median, MR-pleiotropy residual sum and outlier, MR-using mixture models, and leave-one-out analysis.

Results

We found that for every 6.75 mmol/mol increase in HbA1c, there was an increase of 0.49 (95% CI = 0.24, 0.74; p-value = 1.30 × 10-4) years in adjusted BAG. Subgroup analyses by sex and age revealed significant causal effects of HbA1c on adjusted BAG, specifically among men aged 60-73 (p-value = 2.37 × 10-8).

Conclusion

Poor glycemic control has a significant causal effect on brain aging, and is most pronounced among older men aged 60-73 years, which provides insights between glycemic control and the susceptibility to age-related neurodegenerative diseases.

Unraveling shared susceptibility loci and Mendelian genetic associations linking educational attainment with multiple neuropsychiatric disorders

Background

Empirical studies have demonstrated that educational attainment (EA) is associated with neuropsychiatric disorders (NPDs), suggesting a shared etiological basis between them. However, little is known about the shared genetic mechanisms and causality behind such associations.

Methods

This study explored the shared genetic basis and causal relationships between EA and NPDs using the high-definition likelihood (HDL) method, cross phenotype association study (CPASSOC), transcriptome-wide association study (TWAS), and bidirectional Mendelian randomization (MR) with summary-level data for EA (N = 293,723) and NPDs (N range = 9,725 to 455,258).

Results

Significant genetic correlations between EA and 12 NPDs (rg range - 0.49 to 0.35; all p < 3.85 × 10-3) were observed. CPASSOC identified 37 independent loci shared between EA and NPDs, one of which was novel (rs71351952, mapped gene: ARFGEF2). Functional analyses and TWAS found shared genes were enriched in brain tissue, especially in the cerebellum and highlighted the regulatory role of neuronal signaling, purine nucleotide metabolic process, and cAMP-mediated signaling pathways. CPASSOC and TWAS supported the role of three regions of 6q16.1, 3p21.31, and 17q21.31 might account for the shared causes between EA and NPDs. MR confirmed higher genetically predicted EA lower the risk of ADHD (ORIVW: 0.50; 95% CI: 0.39 to 0.63) and genetically predicted ADHD decreased the risk of EA (Causal effect: -2.8 months; 95% CI: -3.9 to -1.8).

Conclusion

These findings provided evidence of shared genetics and causation between EA and NPDs, advanced our understanding of EA, and implicated potential biological pathways that might underlie both EA and NPDs.

A Bayesian approach to Mendelian randomization using summary statistics in the univariable and multivariable settings with correlated pleiotropy

Mendelian randomization uses genetic variants as instrumental variables to make causal inferences on the effect of an exposure on an outcome. Due to the recent abundance of high-powered genome-wide association studies, many putative causal exposures of interest have large numbers of independent genetic variants with which they associate, each representing a potential instrument for use in a Mendelian randomization analysis. Such polygenic analyses increase the power of the study design to detect causal effects; however, they also increase the potential for bias due to instrument invalidity. Recent attention has been given to dealing with bias caused by correlated pleiotropy, which results from violation of the "instrument strength independent of direct effect" assumption. Although methods have been proposed that can account for this bias, a number of restrictive conditions remain in many commonly used techniques. In this paper, we propose a Bayesian framework for Mendelian randomization that provides valid causal inference under very general settings. We propose the methods MR-Horse and MVMR-Horse, which can be performed without access to individual-level data, using only summary statistics of the type commonly published by genome-wide association studies, and can account for both correlated and uncorrelated pleiotropy. In simulation studies, we show that the approach retains type I error rates below nominal levels even in high-pleiotropy scenarios. We demonstrate the proposed approaches in applied examples in both univariable and multivariable settings, some with very weak instruments.

The causal effects of education on adult health, mortality and income: evidence from Mendelian randomization and the raising of the school leaving age

BACKGROUND

On average, educated people are healthier, wealthier and have higher life expectancy than those with less education. Numerous studies have attempted to determine whether education causes differences in later health outcomes or whether another factor ultimately causes differences in education and subsequent outcomes. Previous studies have used a range of natural experiments to provide causal evidence. Here we compare two natural experiments: a policy reform, raising the school leaving age in the UK in 1972; and Mendelian randomization.

METHODS

We used data from 334 974 participants of the UK Biobank, sampled between 2006 and 2010. We estimated the effect of an additional year of education on 25 outcomes, including mortality, measures of morbidity and health, ageing and income, using multivariable adjustment, the policy reform and Mendelian randomization. We used a range of sensitivity analyses and specification tests to assess the plausibility of each method's assumptions.

RESULTS

The three different estimates of the effects of educational attainment were largely consistent in direction for diabetes, stroke and heart attack, mortality, smoking, income, grip strength, height, body mass index (BMI), intelligence, alcohol consumption and sedentary behaviour. However, there was evidence that education reduced rates of moderate exercise and increased alcohol consumption. Our sensitivity analyses suggest that confounding by genotypic or phenotypic confounders or specific forms of pleiotropy are unlikely to explain our results.

CONCLUSIONS

Previous studies have suggested that the differences in outcomes associated with education may be due to confounding. However, the two independent sources of exogenous variation we exploit largely imply consistent causal effects of education on outcomes later in life.

Lifestyle differences between co-twins are associated with decreased similarity in their internal and external exposome profiles

Whether differences in lifestyle between co-twins are reflected in differences in their internal or external exposome profiles remains largely underexplored. We therefore investigated whether within-pair differences in lifestyle were associated with within-pair differences in exposome profiles across four domains: the external exposome, proteome, metabolome and epigenetic age acceleration (EAA). For each domain, we assessed the similarity of co-twin profiles using Gaussian similarities in up to 257 young adult same-sex twin pairs (54% monozygotic). We additionally tested whether similarity in one domain translated into greater similarity in another. Results suggest that a lower degree of similarity in co-twins' exposome profiles was associated with greater differences in their behavior and substance use. The strongest association was identified between excessive drinking behavior and the external exposome. Overall, our study demonstrates how social behavior and especially substance use are connected to the internal and external exposomes, while controlling for familial confounders.

Large-scale genome-wide association study to identify causal relationships and potential mediators between education and autoimmune diseases

Objectives

Epidemiological studies suggested a potential connection between education and autoimmune disorders. This study investigated the possible cause-and-effect relationship using a Mendelian randomization approach.

Methods

We explored the causality between four education traits (n = 257,841~1,131,881) and 22 autoimmune diseases. The mediating role of smoking (632,802 individuals), BMI (681,275 individuals), alcohol (335,394 individuals), and income (397,751 individuals) was also investigated. Transcriptome-wide association study (TWAS) and enriched signaling pathways analysis were used to investigate the underlying biological mechanisms.

Results

Especially, higher cognitive performance was protective for psoriasis (odds ratio (OR) = 0.69, 95% confidence interval (CI) = 0.60-0.79, p = 6.12×10-8), rheumatoid arthritis (RA) (OR = 0.75, 95% CI = 0.67-0.83, p = 4.62×10-6), and hypothyroidism (OR = 0.83, 95% CI = 0.77-0.90, p = 9.82×10-6). Higher levels of educational attainment decreased risks of psoriasis (OR = 0.61, 95% CI = 0.52-0.72, p = 1.12×10-9), RA (OR = 0.68, 95% CI = 0.59-0.79, p = 1.56×10-7), and hypothyroidism (OR = 0.80, 95% CI = 0.72-0.88, p = 5.00×10-6). The completion of highest-level math class genetically downregulates the incidence of psoriasis (OR = 0.66, 95% CI = 0.58-0.76, p = 2.47×10-9), RA (OR = 0.71, 95% CI = 0.63-0.81, p = 5.28×10-8), and hypothyroidism (OR = 0.85, 95% CI = 0.79-0.92, p = 8.88×10-5). Higher self-reported math ability showed protective effects on Crohn's disease (CD) (OR = 0.67, 95% CI = 0.55-0.81, p = 4.96×10-5), RA (OR = 0.76, 95% CI = 0.67-0.87, p = 5.21×10-5), and psoriasis (OR = 0.76, 95% CI = 0.65-0.88, p = 4.08×10-4). Protein modification and localization, response to arsenic-containing substances may participate in the genetic association of cognitive performance on UC, RA, psoriasis, and hypothyroidism. According to mediation analyses, BMI, smoking, and income served as significant mediators in the causal connection between educational traits and autoimmune diseases.

Conclusion

Higher levels of education-related factors have a protective effect on the risk of several autoimmune disorders. Reducing smoking and BMI and promoting income equality can mitigate health risks associated with low education levels.

The causality between gut microbiota and ankylosing spondylitis: Insights from a bidirectional two-sample Mendelian randomization analysis

BACKGROUND

The association between gut microbiota and ankylosing spondylitis (AS) has been reported in the literature; however, whether the two are correlative is unclear.

METHODS

Single nucleotide polymorphisms associated with the gut microbiome composition and AS (968 AS cases and 336 191 controls) were obtained from published genome-wide association studies in this two-sample Mendelian randomization (MR) study. The causal relationship between gut microbiota and AS was estimated using the inverse-variance weighted method, and the robustness of our findings was confirmed through a comprehensive series of sensitivity analyses.

RESULTS

Anaerotruncus (OR = 0.9984, 95% CI, 0.9968-0.9999, p = .0405) and Ruminococcaceae UCG002 (OR = 0.9989, 95% CI, 0.9979-0.9999, p = .0375) were protective against AS. Defluviitaleaceae (OR = 1.0015, 95% CI, 1.0005-1.0025, p = .0048), Butyricicoccus (OR = 1.0016, 95% CI, 1.0001-1.0032, p = .0429), Coprococcus 3 (OR = 1.0016, 95% CI, 1.0000-1.0032, p = .0463), and Defluviitaleaceae UCG011 (OR = 1.0016, 95% CI, 1.0005-1.0027, p = .0041) exhibited significant positive correlations with heightened susceptibility to AS. Reverse MR revealed that AS does not affect the gut microbial composition.

CONCLUSION

Our study has established a genetically-based causal relationship between gut microbiota and AS. This finding suggests that we may be able to target and regulate specific bacterial groups in the gut to prevent and treat AS.