CoChat
Sign Up
Cited by 2,229Genetic Associations and EpidemiologyAdvanced Causal Inference TechniquesStatistical Methods in Epidemiology

Mendelian randomization

Open access
Sanderson, Eleanor, Glymour, M. Maria, Holmes, Michael V., Kang, Hyunseung, Morrison, Jean, Munafò, Marcus R., Palmer, Tom, Schooling, C. Mary +3 more·2022·Nature Reviews Methods Primers
DOI 10.1038/s43586-021-00092-5PMC PMC7614635
Open this paper in CoChat — free account

Abstract

Mendelian randomization (MR) is a term that applies to the use of genetic variation to address causal questions about how modifiable exposures influence different outcomes. The principles of MR are based on Mendel’s laws of inheritance and instrumental variable estimation methods, which enable the inference of causal effects in the presence of unobserved confounding. In this Primer, we outline the principles of MR, the instrumental variable conditions underlying MR estimation and some of the methods used for estimation. We go on to discuss how the assumptions underlying an MR study can be assessed and describe methods of estimation that are robust to certain violations of these assumptions. We give examples of a range of studies in which MR has been applied, the limitations of current methods of analysis and the outlook for MR in the future. The differences between the assumptions required for MR analysis and other forms of epidemiological studies means that MR can be used as part of a triangulation across multiple sources of evidence for causal inference. Mendelian randomization is a technique for using genetic variation to examine the causal effect of a modifiable exposure on an outcome such as disease status. This Primer by Sanderson et al. explains the concepts of and the conditions required for Mendelian randomization analysis, describes key examples of its application and looks towards applying the technique to growing genomic datasets.

AI snapshot

Written from the paper’s full text.

TL;DR

Mendelian randomization (MR) uses genetic variants as instrumental variables to address causal questions about modifiable exposures while avoiding bias from unobserved confounding. This Primer outlines MR principles, IV assumptions, estimation and sensitivity methods, applications, limitations, and future directions.

Key findings

  • MR relies on four conditions: the three instrumental variable conditions plus a point-estimate-identifying condition; a first-stage F-statistic >10 is described as a minimum criterion indicating weak instrument bias is small, though it does not prove instrument validity.
  • Summary-data MR commonly uses inverse-variance weighting of variant-specific Wald ratios; pleiotropy-robust methods include weighted median, weighted mode, MR Egger, MR PRESSO, MR RAPS, MR GRAPPLE, and multivariable MR, each imposing different assumptions and often reducing power.
  • In applications, MR using ALDH2 and ADH1B variants across average alcohol consumption from 4 g/week to 256 g/week found a dose-response relationship with stroke and no strong protective or detrimental effect on coronary heart disease; MR studies of HDL-C showed no protective effect on CHD, consistent with subsequent RCTs.
  • An MR study of childhood and adult body size detected 295 and 557 independent SNPs, respectively; multivariable MR suggested only adult body size had an independent causal effect on CHD and type 2 diabetes, while early-life body size was more strongly inversely related to breast cancer.

Methods

  • This is a Primer/review, not a new empirical study; it describes MR principles, IV conditions, individual-level two-stage least squares, summary-level inverse-variance weighting, sensitivity analyses, software packages, and extensions such as bidirectional, multivariable, mediation, and non-linear MR.
  • Data discussed include individual-level and summary-level genetic association data, often from GWAS and biobanks such as UK Biobank and China Kadoorie Biobank, used in published MR applications.
  • The main analysis is conceptual and methodological synthesis; no new quantitative meta-analysis or primary dataset analysis is reported.

Limitations

  • The paper states that MR cannot easily correct confounding of genetic variants and the outcome from population stratification, dynastic effects, and assortative mating; selection bias, survival bias, canalization, and misspecification of the primary phenotype as GWAS grow are also highlighted.
  • The paper notes that pleiotropy-robust methods are not truly robust to all types of pleiotropy, often have less statistical power and wider confidence intervals, and that consistent results across methods do not guarantee absence of bias.
  • Evident from the text: as a Primer, it does not present a systematic review or exhaustive list of all MR methods, and the described applications are illustrative rather than comprehensive.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s43586-021-00092-5

Cites

228 of 228 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 130 cited sources checked · 0 supported · 0 flagged

  • Premisereasoning

    The capacity to use two different samples for MR analyses has dramatically broadened the scope of MR studies.

    “The capacity to use two different samples for MR analyses has dramatically broadened the scope of MR studies because when either the desired exposure or outcome of a study is rare or expensive to measure, it can be difficult to identify a dataset with data on the genetic instrument, exposure and outcome.”
    Quote in the paperSubstantial claim
  • Interpretationreasoning

    Bias originating from confounding of the exposure and outcome should not influence the MR estimate.

    “Bias originating from confounding of the exposure and outcome should not influence the MR estimate.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: Genetic variants provide a source of exogenous variation in the exposure and can act as an instrumental variable.
  • Premisereasoning

    In MR using unrelated individuals, the similarity between the allele groups is not guaranteed as with a well-conducted RCT.

    “In MR using unrelated individuals, the similarity between the allele groups is not guaranteed as with a well-conducted RCT.”
    Quote in the paperSubstantial claim
  • Premisenone given

    When individual-level data are used for estimation, the statistical power of an MR analysis increases in proportion to the sample size and the variance in the exposure explained by the genetic instruments.

    “When individual-level data are used for estimation, the statistical power of an MR analysis (or equivalently, the precision of the estimate that can be derived) increases in proportion to the sample size and the variance in the exposure explained by the genetic instruments.”
    Quote in the paperNot a substantial claim
  • Premisenone given

    Associations between allele distribution and traits can exist at a population level owing to population stratification or assortative mating.

    “Further, associations between allele distribution and traits can exist at a population level owing to population stratification or assortative mating.”
    Quote in the paperSubstantial claim
  • May combine two claims. 54
  • Classified as probably not a substantial claim. 23

  • May combine two claims.

    Premisecitation

    MR can be applied using any genetic variation that satisfies the requirements of an IV, although it is most often implemented using single nucleotide polymorphisms (SNPs).

    “MR can be applied using any genetic variation that satisfies the requirements of an IV,5 although it is most often implemented using single nucleotide polymorphisms (SNPs).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Zuccolo, L. & Holmes, M. V. Commentary: Mendelian randomization-inspired causal inference in the absence of genetic data. Int. J. Epidemiol. 46, 962–965 (2017).

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Using genetic variants in this way, MR avoids bias from unobserved confounding of the exposure and outcome.

    “Using genetic variants in this way, MR avoids bias from unobserved confounding of the exposure and outcome.”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: Genetic variants provide a source of exogenous variation in the exposure and can act as an instrumental variable.

  • May combine two claims.

    Premisereasoning

    Genetic variants typically only explain a small proportion of the variation in the relevant phenotype; as a result, low statistical power and imprecise effect estimates are common in MR studies and well-powered studies usually require large datasets.

    “Genetic variants typically only explain a small proportion of the variation in the relevant phenotype; as a result, low statistical power and imprecise effect estimates are common in MR studies and well-powered studies usually require large datasets.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    MR was first proposed using family data where the difference in alleles between siblings is random, however data limitations mean most MR is conducted using data on unrelated individuals.

    “MR was first proposed using the family data where the difference in alleles between siblings is random, however data limitations mean most MR is conducted using data on unrelated individuals.22”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [22] Swanson, S. A., Tiemeier, H., Ikram, M. A. & Hernán, M. A. Nature as a trialist? Deconstructing the analogy between Mendelian randomization and randomized trials. Epidemiology 28, 653–659 (2017). source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisecitation

    IV analysis uses an instrument related to the exposure but not the outcome other than through its association with the exposure to make causal effect estimates in the presence of unobserved confounding of the exposure and the outcome.

    “IV analysis was first proposed a century ago and is an approach to causal inference that uses an IV, or “instrument” — which is related to the exposure but not the outcome of interest other than through its association with the exposure — to make causal effect estimates in the presence of unobserved confounding of the exposure and the outcome.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [2] Angrist, J. D., Imbens, G. W. & Rubin, D. B. Identification of causal effects using instrumental variables. J. Am. Stat. Assoc. 91, 444–455 (1996). source

      Not checked in the free run.

    • cites [3] Hernán, M. A. & Robins, J. M. Causal Inference: What If (Chapman & Hall/CRC, 2020).

      Not checked in the free run.

    • cites [4] Greenland, S. An introduction to instrumental variables for epidemiologists. Int. J. Epidemiol. 29, 722–729 (2000). source

      Not checked in the free run.

  • May combine two claims.

    Premisereasoning

    In RCTs, randomly assigned treatment evaluates the effect of treatment on the outcome, whereas in MR, a genetic variant is treated as a naturally occurring form of randomization.

    “In RCTs, randomly assigned treatment therefore evaluates the effect of treatment on the outcome, whereas in MR, a genetic variant is treated as a naturally occurring form of randomization.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    MR is especially appealing because it relies on different assumptions to conventional observational studies and therefore circumvents some of their common biases.

    “MR is especially appealing because it relies on different assumptions to conventional observational studies and therefore circumvents some of their common biases.8”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [8] Richmond, R. C. & Davey Smith, G. Mendelian randomization: concepts and scope. Cold Spring Harb. Perspect. Med. https://doi.org/10.1101/cshperspect.a040501 (2022). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Causal effect estimates from MR can be evaluated within a triangulation of evidence framework, which involves interpreting findings alongside results from complementary approaches that rely on different assumptions.

    “Causal effect estimates from MR can be evaluated within a triangulation of evidence framework, which involves interpreting findings alongside results from complementary approaches that rely on different assumptions.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [6] Munafò, M. R., Higgins, J. P. & Davey Smith, G. Triangulating evidence through the inclusion of genetically informed designs. Cold Spring Harb. Perspect. Med. 11, a040659 (2021). source

      Not checked in the free run.

    • cites [7] Lawlor, D. A., Tilling, K. & Davey Smith, G. Triangulation in aetiological epidemiology. Int. J. Epidemiol. 45, 1866–1886 (2017).

      Not checked in the free run.

    • cites [8] Richmond, R. C. & Davey Smith, G. Mendelian randomization: concepts and scope. Cold Spring Harb. Perspect. Med. https://doi.org/10.1101/cshperspect.a040501 (2022). source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Premisenone given

    MR studies - especially two-sample studies using previously published summary-level genetic association data - provide a rapid and affordable approach to evaluating causal questions.

    “MR studies — especially two-sample studies using previously published summary-level genetic association data — provide a rapid and affordable approach to evaluating causal questions.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim. May combine two claims.

    Premisenone given

    The first three of these conditions are commonly referred to as the conditions for a valid instrumental variable and are required for any IV analysis to test whether the exposure has a causal effect on the outcome.

    “The first three of these conditions are commonly referred to as the conditions for a valid instrumental variable and are required for any IV analysis to test whether the exposure has a causal effect on the outcome.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisereasoning

    The rationale of MR studies parallels that of RCTs in which randomization influences the treatment received by participants, is not associated with any confounders of the treatment and outcome and has no other plausible mechanism to influence health outcomes other than through treatment.

    “The rationale of MR studies parallels that of RCTs in which randomization influences the treatment received by participants, is not associated with any confounders of the treatment and outcome and has no other plausible mechanism to influence health outcomes other than through treatment (see Fig. 1).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    RCTs estimate the effect of a particular intervention or treatment over the timeframe of the study, whereas MR estimates the lifetime effects.

    “The above example highlights an important difference between RCTs and MR: RCTs estimate the effect of a particular intervention or treatment over the timeframe of the study, whereas MR estimates the lifetime effects.21”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [21] Morris, T. T., Heron, J., Sanderson, E., Davey Smith, G. & Tilling, K. Interpretation of Mendelian randomization using one measure of an exposure that varies over time. Preprint at medRxiv https://doi.org/10.1101/2021.11.18.21266515 (2021). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    MR GxE and MR GENIUS have extended and formalised concepts of using gene-environment interactions and enable the estimation of causal effects in more general settings.

    “Two methods, MR GxE and MR GENIUS, have extended and formalised these concepts and enable the estimation of causal effects in more general settings.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [76] Spiller, W., Slichter, D., Bowden, J. & Davey Smith, G. Detecting and correcting for bias in Mendelian randomization analyses using gene-by-environment interactions. Int. J. Epidemiol. 48, 702–712 (2019). source

      Not checked in the free run.

    • cites [77] Tchetgen Tchetgen, E. J., Sun, B. & Walter, S. The GENIUS approach to robust Mendelian randomization inference. Stat. Sci. 36, 443–464 (2019).

      Not checked in the free run.

    • cites [78] Burgess, S., Dudbridge, F. & Thompson, S. G. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Stat. Med. 35, 1880–1906 (2016). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisenone given

    An alternative approach to test interactions is to split allele scores for each exposure into high and low values and compare outcomes across four groups, mimicking a 2x2 factorial randomised trial.

    “An alternative approach is to split the allele scores for each exposure into high and low values and compare outcomes across the resulting four groups by dividing participants up based on their score for each exposure, mimicking a 2x2 factorial randomised trial.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The bias from this situation is equivalent to that caused by pleiotropy, and although unlikely at many independent genetic locations in Mendelian randomization studies with multiple instrumental variables, it should be a consideration in single-instrumental-variable studies.

    “The bias in this situation is equivalent to that caused by pleiotropy (see Fig. 3) and although it is unlikely that this pattern will arise at many independent genetic locations in MR studies with multiple IVs, it should be a consideration in single-IV studies.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationreasoning

    Although colocalization involves an implicit assumption of directionality in its interpretation, the test is not dependent on this assumption and a single pleiotropic variant would satisfy the statistical definition of shared causal variants.

    “Although this involves an implicit assumption of directionality in its interpretation, the test is not dependent on this assumption and indeed a single pleiotropic variant would satisfy the statistical definition of a shared causal variants (Fig 6b).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Instrumental variable conditions need to be assessed for any summary-data Mendelian randomization.

    “As with individual level data analysis, IV conditions need to be assessed for any summary-data MR.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisenone given

    A number of different methods are available to correct for horizontal pleiotropy under different assumptions about the causal structure of that pleiotropy.

    “A number of different methods are available to correct for horizontal pleiotropy — a violation of the third IV condition — under different assumptions about the causal structure of that pleiotropy.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisenone given

    Although the second and third IV conditions cannot be proven to be true, they can sometimes be disproven.

    “Although the second and third IV conditions cannot be proven to be true, they can sometimes be disproven.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Using a few methods selected based on the most plausible assumptions alongside an inverse-variance weighted Mendelian randomization estimation as a sensitivity analysis can determine how robust results are to the assumption that genetic variants have no pleiotropic effects on the outcome under different alternative specifications.

    “Therefore, a few methods should be selected based on the most plausible assumptions for the application in question and used alongside an IVW MR estimation to perform a sensitivity analysis; this can determine how robust MR results are to the assumption that genetic variants have no pleiotropic effects on the outcome under different alternative specifications.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationnone given

    None of the described pleiotropy-robust methods are truly robust to all types of pleiotropy, and each imposes different assumptions on the nature of the pleiotropy and how pleiotropic effects are accounted for.

    “None of the methods described above are truly robust to all types of pleiotropy and each imposes different assumptions on the nature of the pleiotropy and how the pleiotropic effects are accounted for.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationnone given

    Results from bidirectional Mendelian randomization studies should be interpreted with caution; evidence of an effect in both directions could indicate a true bidirectional relationship or be a product of bias from horizontal pleiotropic effects, misspecification of the primary phenotype, or violation of the second instrumental variable condition.

    “Results from bidirectional MR studies should be interpreted with caution. Evidence of an effect in both directions could indicate a true bidirectional relationship between the exposures or be a product of bias from horizontal pleiotropic effects in the variants, misspecification of the primary phenotype, or a violation of the second IV condition owing to confounding of genetic variants and outcome”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    Traditionally, summary-data Mendelian randomization is visualized as a scatter plot plotting associations of the variant and exposure against associations of the variant and outcome; however, this can limit the identification of outliers.

    “Traditionally, summary-data MR is visualized as a scatter plot plotting associations of the variant and exposure against associations of the variant and outcome (Fig 4a, b); however, this can limit the identification of outliers.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationnone given

    Standard Mendelian randomization only provides a single effect estimate, which may not be informative if the effect of the exposure varies in a non-linear way.

    “Standard MR only provides a single effect estimate, which may not be informative if the effect of the exposure varies in a non-linear way — for example, a dose-response curve.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Weighted allele scores provide increased instrument strength and power, although there are cases in which the unweighted approach is preferred.

    “Weighted scores provide increased instrument strength and power, although there are cases in which the unweighted approach is preferred — for example, if the definition of the exposure in the discovery dataset differs from the exposure variable in the estimation data.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [26] Davies, N. M. et al. The many weak instruments problem and Mendelian randomization. Stat. Med. 34, 454–468 (2015). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Many pleiotropy-robust methods have less statistical power than conventional Mendelian randomization, leading to very wide confidence intervals.

    “Furthermore, many methods have less statistical power than conventional MR, leading to very wide confidence intervals.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationnone given

    The inclusion of interaction terms in multivariable Mendelian randomization decreases statistical power for detecting an effect and should be limited to a single interaction.

    “The inclusion of these additional terms decreases the statistical power for detecting an effect and should be limited to a single interaction.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Non-linear Mendelian randomization has supported the J-shaped relationship between BMI and mortality risk, although it has also suggested the J-shape could be caused by the relationship differing for ever-smokers and never-smokers.

    “Non-linear MR has supported this, although it has also suggested that the J-shape could be caused by the relationship between BMI and mortality risk differing for ever-smokers [G] and never-smokers.109”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [109] North, T.-L. et al. Using genetic instruments to estimate interactions in Mendelian randomization studies. Epidemiology 30, e33–e35 (2019). source

      Not checked in the free run.

  • May combine two claims.

    Premisereasoning

    In bidirectional Mendelian randomization, two analyses are conducted on the same pair of phenotypes by reversing the exposure and the outcome, and this method can be used to establish the direction of effect between two variables.

    “In bidirectional MR, two MR analyses are conducted on the same pair of phenotypes by reversing the exposure and the outcome. This method can be used to establish the direction of effect between two variables.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    Including multiple exposures in multivariable Mendelian randomization can reduce power and potentially instrument strength, so the benefit of adding extra exposures must be considered carefully.

    “Multiple, correlated exposures can be included in a multivariable MR; however, including multiple exposures can reduce power and potentially instrument strength and thus the benefit of adding extra exposures must be considered carefully.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Premisenone given

    Ever larger GWAS have provided evidence that hundreds of genetic variants may be associated with many human phenotypes.

    “Ever larger GWAS have now provided evidence that hundreds of genetic variants may be associated with many human phenotypes.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisenone given

    Multivariable Mendelian randomization is an extension of standard Mendelian randomization that includes multiple exposures, predicted by a set of genetic variants used as instruments.

    “Multivariable MR is an extension of standard MR that includes multiple exposures, predicted by a set of genetic variants used as instruments.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    If many SNPs that each only have a small effect on the exposure are being used, combining them into a single score can increase the power of the analysis and reduce the risk of bias from many weak instruments.

    “If many SNPs that each only have a small effect on the exposure are being used, combining them into a single score can increase the power of the analysis and reduce the risk of bias from many weak instruments.27”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [27] Hernán, M. A. & Robins, J. M. Instruments for causal inference: an epidemiologist’s dream? Epidemiology 17, 360–372 (2006). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    Correlation between genetic variants and the outcome can potentially affect most or all genetic variants used as instruments, and it is not easy to correct with current MR methods because most assume that the majority of genetic variants satisfy all of the IV conditions.

    “This correlation between genetic variants and the outcome can potentially affect most (or all) of the genetic variants used as instruments; it is therefore not easy to correct for with current MR methods as most assume that the majority of genetic variants satisfy all of the IV conditions.61”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [61] Davey Smith, G. & Hemani, G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum. Mol. Genet. 23, R89–R98 (2014). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Pre-registration of study protocols has not been widely adopted in observational epidemiology, although it could in principle be applied and help protect against bias such as publication bias against null results or findings that do not fit the anticipated conclusion.

    “Pre-registration of study protocols has not been widely adopted in observational epidemiology, although it could in principle be applied and help protect against bias, such as publication bias against null results or findings that do not fit with the anticipated conclusion.161,162”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [161] Brumpton, B. et al. Avoiding dynastic, assortative mating, and population stratification biases in Mendelian randomization through within-family analyses. Nat. Commun. 11, 3519 (2020). source

      Not checked in the free run.

    • cites [162] Morris, T. T., Davies, N. M., Hemani, G. & Davey Smith, G. Population phenomena inflate genetic associations of complex social traits. Sci. Adv. 6, eaay0328 (2020). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    In addition to physiological consequences, body weight has social consequences such as increasing exposure to stigma and discrimination, and these Mendelian randomization analyses include the consequences of all such mechanisms for hospitalization costs.

    “In addition to physiological consequences, body weight has social consequences such increasing exposure to stigma and discrimination and these MR analyses include the consequences of all such mechanisms for hospitalization costs.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Within-family MR requires data from either pairs of siblings or mother-father-child trios and allows for estimation of causal effects using MR after family-level structure has been taken into account.

    “Within-family MR requires data from either pairs of siblings or mother–father–child trios and allows for estimation of causal effects using MR after family-level structure has been taken into account.164,167”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [164] Davies, N. M. et al. Within family Mendelian randomization studies. Hum. Mol. Genet. 28, R170–R179 (2019). source

      Not checked in the free run.

    • cites [167] Taylor, A. E. et al. Exploring the association of genetic factors with participation in the Avon Longitudinal Study of Parents and Children. Int. J. Epidemiol. 47, 1207–1216 (2018). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationnone given

    Available evidence from Mendelian randomization methods that facilitate estimation in the presence of unobserved confounding, assuming no selection bias, does not support the conclusion that moderate alcohol consumption may lower vascular disease risk and identifies alcohol consumption as a factor linked to increased likelihood of ischemic stroke.

    “Thus, available evidence from MR methods that facilitate estimation in the presence of unobserved confounding — assuming no selection bias — do not support the conclusion that the consumption of a moderate amount of alcohol may lower vascular disease risk and identify alcohol consumption as a factor linked to increased likelihood of ischemic stroke”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: The conclusion about alcohol consumption and vascular disease risk assumes no selection bias.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisenone given

    Open research can increase the robustness of data by allowing greater scrutiny of data and increased error detection by researchers and the wider research community.

    “Open research can increase the robustness of data through allowing greater scrutiny of data and increased error detection by researchers and the wider research community.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Within MR, using methods that make different assumptions (such as regarding pleiotropy) and are therefore subject to different biases can support triangulation, although the least plausible assumption may be shared by many methods, reducing the potential independent insight.

    “Within MR, using methods that make different assumptions (such as those regarding pleiotropy) and are therefore subject to different sources and directions of potential bias can support this approach, although the least plausible assumption may be shared by many methods, reducing the potential independent insight to be gained from comparing studies.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The least plausible assumption may be shared by many methods.

  • May combine two claims.

    Premisecitation

    Bayesian frameworks for colocalization analysis consider GWAS summary statistics for both traits across multiple SNPs in the region around the instrumental variable and assess either the evidence that each variant is jointly causal or consider shared causal variants as one of five competing hypotheses.

    “Alternatively, Bayesian frameworks for colocalization analysis consider GWAS summary statistics for both traits across multiple SNPs in the region around the IV and assess either the evidence that each variant is jointly causal115 or consider shared causal variants as one of five competing hypotheses.116”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [115] Giambartolomei, C. et al. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 10, e1004383 (2014). source

      Not checked in the free run.

    • cites [116] Wallace, C. Eliciting priors and relaxing the single causal variant assumption in colocalisation analyses. PLoS Genet. 16, e1008720 (2020). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Population stratification, dynastic effects and assortative mating all induce bias by creating an artefactual relation at the population level between the genetic variants and the outcome, violating the second IV condition.

    “However, population stratification, dynastic effects and assortative mating all induce bias by creating an artefactual relation at the population level between the genetic variants and the outcome, violating the second IV condition.65,163–166”
    Quote in the paperSubstantial claim0 of 5 sources checked
    • cites [65] Sargan, J. D. The estimation of economic relationships using instrumental variables. Econometrica 26, 393–415 (1958). source

      Not checked in the free run.

    • cites [163] Minică, C. C., Boomsma, D. I., Dolan, C. V., de Geus, E. & Neale, M. C. Empirical comparisons of multiple Mendelian randomization approaches in the presence of assortative mating. Int. J. Epidemiol. 49, 1185–1193 (2020). source

      Not checked in the free run.

    • cites [164] Davies, N. M. et al. Within family Mendelian randomization studies. Hum. Mol. Genet. 28, R170–R179 (2019). source

      Not checked in the free run.

    • cites [165] Minică, C. C., Dolan, C. V., Boomsma, D. I., de Geus, E. & Neale, M. C. Extending causality tests with genetic instruments: an integration of Mendelian randomization with the classical twin design. Behav. Genet. 48, 337–349 (2018). source

      Not checked in the free run.

    • cites [166] Howe, L. J. et al. Within-sibship genome-wide association analyses decrease bias in estimates of direct genetic effects. Nat. Genet. (in the press).

      Not checked in the free run.

    • Rests on an assumption: Genetic variants are generally fixed at conception.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationnone given

    Mendelian randomization studies are themselves not free from issues of bias or lack of power; evidence from Mendelian randomization studies for the presence or absence of an effect should be compared with results from studies with other potential sources of bias.

    “However, it must be noted that MR studies are themselves not free from issues of bias or lack of power; evidence from MR studies for the presence or absence of an effect should be compared with results from studies with other potential sources of bias.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Using genetic variants associated with higher BMI as instruments in an individual-level Mendelian randomization study, higher BMI increased hospital costs with little evidence for non-linearity in this effect.

    “Using genetic variants associated with higher BMI as instruments in an individual-level MR study to estimate the effect of BMI on hospitalization costs,147 the authors found that higher BMI increased hospital costs with little evidence for non-linearity in this effect.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [147] Holmes, M. V. Human genetics and drug development. N. Engl. J. Med. 380, 1076–1079 (2019). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    This suggests that for children who are overweight, losing weight in adulthood can still effectively lower risk of type 2 diabetes and coronary artery disease, and a metabolically unhealthy childhood can potentially be offset by healthy lifestyle approaches adopted in adulthood.

    “This suggests that for children that are overweight, losing weight in their adulthood can still effectively lower risk of T2D and CAD and in this case a metabolically unhealthy childhood can potentially be offset by healthy lifestyle approaches adopted in adulthood.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Study participation is heritable and influenced by a number of different traits, and large studies such as the UK Biobank have high levels of selection in those who participate.

    “Study participation has shown to be heritable and is influenced by a number of different traits, and large studies such as the UK Biobank have been shown to have high levels of selection in those who participate.31,170–172”
    Quote in the paperSubstantial claim0 of 4 sources checked
    • cites [31] Davey Smith, G. Epigenesis for epidemiologists: does evo-devo have implications for population health research and practice? Int. J. Epidemiol. 41, 236–247 (2012). source

      Not checked in the free run.

    • cites [170] Smit, R. A., Trompet, S., Dekkers, O. M., Jukema, J. W. & le Cessie, S. Survival bias in Mendelian randomization studies: a threat to causal inference. Epidemiology 30, 813 (2019). source

      Not checked in the free run.

    • cites [171] Schooling, C. M. et al. Use of multivariable Mendelian randomization to address biases due to competing risk before recruitment. Front. Genet. 11, 610852 (2020). source

      Not checked in the free run.

    • cites [172] Vansteelandt, S., Dukes, O. & Martinussen, T. Survivor bias in Mendelian randomization analysis. Biostatistics 19, 426–443 (2017). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    Gene expression and protein are often instrumented with a single genetic variant, so colocalization can be particularly useful in Mendelian randomization studies of these exposures and can be used to attempt to falsify the second instrumental variable condition.

    “Gene expression and protein are often instrumented with a single genetic variant and so colocalization can be particularly useful in MR studies of these exposures;42 in these settings colocalization can be used to attempt to falsify the second IV condition.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [42] Bound, J., Jaeger, D. A. & Baker, R. M. Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak. J. Am. Stat. Assoc. 90, 443–450 (1995).

      Not checked in the free run.

  • May combine two claims.

    Interpretationnone given

    Where data already exists, Mendelian randomization studies are relatively cheap to conduct compared to a large randomized controlled trial and can provide additional evidence to direct which studies are worth follow up with randomized controlled trials.

    “Indeed, where data already exists, MR studies are relatively cheap to conduct — particularly compared to a large RCT — and can provide additional evidence that can be used to direct which studies are worth follow up with RCTs.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Triangulation of evidence from multiple methodologies subject to different sources and directions of bias can help identify bias in MR studies, and alignment of results across methodologies can improve confidence in an initial causal interpretation.

    “Triangulation of evidence from multiple methodologies — using different methodologies that are subject to different sources and directions of potential bias — can help to identify bias in MR studies.6,7,155 Alignment of results across these different methodologies can improve confidence in an initial causal interpretation.”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [6] Munafò, M. R., Higgins, J. P. & Davey Smith, G. Triangulating evidence through the inclusion of genetically informed designs. Cold Spring Harb. Perspect. Med. 11, a040659 (2021). source

      Not checked in the free run.

    • cites [7] Lawlor, D. A., Tilling, K. & Davey Smith, G. Triangulation in aetiological epidemiology. Int. J. Epidemiol. 45, 1866–1886 (2017).

      Not checked in the free run.

    • cites [155] Davies, N. M. et al. Multivariable two-sample Mendelian randomization estimates of the effects of intelligence and education on health. eLife 8, e43990 (2019). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    The J-shaped association between alcohol consumption and cardiovascular disease is artefactual; the apparent protective effect of moderate alcohol consumption is not causal.

    “An example is the J-shaped association between alcohol consumption and cardiovascular disease; there is now consensus that this apparent protective effect of moderate levels of consumption is artefactual, as discussed above.130”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [130] Goldstein, J. L. & Brown, M. S. A century of cholesterol and coronaries: from plaques to genes to statins. Cell 161, 161–172 (2015). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    In epidemiological research, findings may be replicated in settings with nearly identical sources of bias, and such replication provides little independent confirmation, so even highly consistent replicated findings may not reflect true causal effects.

    “In the context of epidemiological research, one area of concern is that findings may be replicated in settings with nearly identical sources of bias. Data with such replication provide little independent confirmation of the initial result and thus even highly consistent replicated findings may not reflect true causal effects.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    MR studies have shown that smoking causes lung cancer, but this is not useful therapeutically following the onset of the disease as smoking cessation is not a useful treatment once lung cancer has developed.

    “For example, although MR studies have shown that smoking causes lung cancer186, this is not useful therapeutically following the onset of the disease as smoking cessation is not a useful treatment once lung cancer has developed.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [186] Davey Smith, G., Paternoster, L. & Relton, C. When will Mendelian randomization become relevant for clinical practice and public health? JAMA 317, 589–591 (2017). source

      Not checked in the free run.

    • Rests on an assumption: Smoking cessation is not a useful treatment once lung cancer has developed.

  • May combine two claims.

    Interpretationreasoning

    Increasing size of GWAS datasets can introduce problems: smaller and smaller effect sizes are being identified as significant, and such variants may affect the trait of interest through an upstream phenotype that might in turn influence the outcomes under investigation.

    “Although the increasing size of GWAS datasets appears to be positive for MR studies, it can also introduce problems; smaller and smaller effect sizes are being identified as significant in GWAS and it is increasingly likely that such variants affect the trait of interest through an upstream phenotype that might in turn influence the outcomes under investigation.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    The original conclusion that MR offers a more robust approach to understanding the effect of some modifiable exposures on health outcomes than much conventional observational epidemiology remains unchanged.

    “When initially presented, it was concluded that “[MR] offers a more robust approach to understanding the effect of some modifiable exposures on health outcomes than does much conventional observational epidemiology”1 and that where possible, RCTs should follow to establish the effects of interventions. This conclusion remains unchanged, although moving towards formal triangulation of all pertinent”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [1] Davey Smith, G. & Ebrahim, S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int. J. Epidemiol. 32, 1–22 (2003). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Interpretationcitation

    The situation with MR is moving towards the one seen in the meta-analysis literature, with the mass production of redundant, misleading, and conflicted publications.

    “The situation with MR is now moving towards the one seen in the meta-analysis literature, with the mass production of redundant, misleading, and conflicted publications.183”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [183] Zhou, W. et al. Causal relationships between body mass index, smoking and lung cancer: univariable and multivariable Mendelian randomization. Int. J. Cancer 148, 1077–1086 (2021). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Premisecitation

    Two-sample MR studies are relatively easy to conduct and can contain clear errors.

    “These are relatively easy to conduct — perhaps too easy — and they can contain clear errors as discussed and demonstrated elsewhere.81”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [81] Bowden, J. et al. Improving the visualization, interpretation and analysis of two-sample summary data Mendelian randomization via the radial plot and radial regression. Int. J. Epidemiol. 47, 1264–1278 (2018). source

      Not checked in the free run.

  • May combine two claims.

    Premisereasoning

    As GWAS of CRP and vitamin D increased in size, multiple variants that primarily influence adiposity were identified, and adiposity is a confounder of the observational association of these exposures with health outcomes.

    “For example, as the GWAS of CRP and vitamin D increased in size, multiple variants that primarily influence adiposity were identified — with adiposity being a confounder of the observational association of these exposures with health outcomes.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Adiposity is a confounder of the observational association of CRP and vitamin D with health outcomes.

  • May combine two claims.

    Premisereasoning

    MR uses genetic variants that are fixed across the life course to estimate the lifetime effects of the exposure of interest, which introduces a potential limitation in the form of canalization.

    “Finally, MR uses genetic variants that are fixed across the life course to estimate the lifetime effects of the exposure of interest. This introduces a potential limitation in the form of canalization, which refers to a natural tendency for the suppression of phenotypic variation among individuals despite contrasting genotypes.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: MR uses genetic variants that are fixed across the life course to estimate the lifetime effects of the exposure of interest.

  • May combine two claims.

    Premisecitation

    Papers reporting two-sample MR have grown rapidly over recent years and now constitute a large majority of published studies.

    “Papers reporting two-sample MR have grown rapidly over recent years and now constitute a large majority of published studies.8,81”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [8] Richmond, R. C. & Davey Smith, G. Mendelian randomization: concepts and scope. Cold Spring Harb. Perspect. Med. https://doi.org/10.1101/cshperspect.a040501 (2022). source

      Not checked in the free run.

    • cites [81] Bowden, J. et al. Improving the visualization, interpretation and analysis of two-sample summary data Mendelian randomization via the radial plot and radial regression. Int. J. Epidemiol. 47, 1264–1278 (2018). source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Although MR has been a powerful tool for confirming or discovering factors that cause disease, it does not often identify therapeutic targets.

    “This means that although MR has been a powerful tool for confirming or discovering factors that cause disease, it does not often identify therapeutic targets.185”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [185] Kim, Y.-I. Role of folate in colon cancer development and progression. J. Nutr. 133, 3731S–3739S (2003). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    The onset and progression of Crohn's disease are associated with different genetic variants, indicating that different risk factors play a role in onset and development.

    “For example, the onset and progression of Crohn’s disease are associated with different genetic variants, indicating that different risk factors play a role in onset and development.187”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [187] Ye, T., Shao, J. & Kang, H. Debiased inverse-variance weighted estimator in two-sample summary-data Mendelian randomization. Ann. Stat. 49, 2079–2100 (2021).

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    This bias can influence the generalizability of MR findings and exacerbate existing inequity in medical research.

    “This can influence the generalizability of MR findings and exacerbate existing inequity in medical research.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Most contemporary MR studies rely on available GWAS data and suffer from considerable bias with respect to representativeness of populations according to geography and ancestry.

    “As most contemporary MR studies rely on available GWAS data, they unfortunately suffer from considerable bias with respect to representativeness of populations according to geography and ancestry.184”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [184] Lee, J. C. et al. Genome-wide association study identifies distinct genetic contributions to prognosis and susceptibility in Crohn’s disease. Nat. Genet. 49, 262–268 (2017). source

      Not checked in the free run.

    • Rests on an assumption: Most contemporary MR studies rely on available GWAS data.

  • May combine two claims.

    Interpretationcitation

    All forms of selection biases could bias MR estimates, so careful assessment of the potential for selection into the sample or samples used in an MR study is important.

    “All forms of selection biases could bias MR estimates and so careful assessment of the potential for selection into the sample or samples used in an MR study is important.65”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [65] Sargan, J. D. The estimation of economic relationships using instrumental variables. Econometrica 26, 393–415 (1958). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Most studies recruit survivors of the original birth cohorts, so all participants must have survived to observe the outcome; selection on surviving genetic make-up and the outcome or a competing risk effectively applies covariable adjustment on survival into the estimates.

    “In addition, most studies recruit survivors of the original birth cohorts. This means all participants must have survived in order to observe whether they get the outcome of interest. Selection of participants on surviving their genetic make-up and the outcome of interest or a competing risk of the outcome effectively applies covariable adjustment on survival into the estimates.173–175”
    Quote in the paperSubstantial claim0 of 3 sources checked
    • cites [173] Hernán, M. A. Invited commentary: selection bias without colliders. Am. J. Epidemiol. 185, 1048–1050 (2017). source

      Not checked in the free run.

    • cites [174] Mahmoud, O., Dudbridge, F., Davey Smith, G., Munafo, M. & Tilling, K. Slope-Hunter: a robust method for index-event bias correction in genome-wide association studies of subsequent traits. Nat. Commun. (in the press).

      Not checked in the free run.

    • cites [175] Waddington, C. H. Canalization of development and the inheritance of acquired characters. Nature 150, 563–565 (1942). source

      Not checked in the free run.

    • Rests on an assumption: Most studies recruit survivors of the original birth cohorts.

  • May combine two claims.

    Interpretationreasoning

    The resulting bias can be accounted for through multivariable MR if the upstream factor is known; however, in many cases it is not, so bias will remain undetected.

    “The resulting bias can be accounted for through multivariable MR if the upstream factor is known; however, in many cases it is not and thus bias will remain undetected.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Adiposity is a confounder of the observational association of CRP and vitamin D with health outcomes.

  • May combine two claims.

    Interpretationcitation

    This form of selection bias is likely particularly problematic for studies of harmful exposures on disease outcomes that occur in later life and will be least evident in studies where the exposure does not affect survival to recruitment.

    “This form of selection bias is likely to be particularly problematic for studies of harmful exposures on disease outcomes that occur in later life and will be least evident in studies where the exposure does not affect survival to recruitment.176”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [176] Debat, V. & David, P. Mapping phenotypes: canalization, plasticity and developmental stability. Trends Ecol. Evol. 16, 555–561 (2001). source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Case-only study designs may be severely compromised by collider bias, which must be taken into account in data analysis.

    “Importantly, case-only study designs may be severely compromised by collider bias[[G] 56,64, which must be taken into account in data analysis185.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [56] Hernán, M. A., Hernández-Díaz, S. & Robins, J. M. A structural approach to selection bias. Epidemiology 15, 615–625 (2004). source

      Not checked in the free run.

    • cites [64] Hughes, R. A., Davies, N. M., Davey Smith, G. & Tilling, K. Selection bias when estimating average treatment effects using one-sample instrumental variable analysis. Epidemiology 30, 350–357 (2019). source

      Not checked in the free run.

    • cites [185] Kim, Y.-I. Role of folate in colon cancer development and progression. J. Nutr. 133, 3731S–3739S (2003). source

      Not checked in the free run.

68 of 68

  • The paper relies on this without support.

    There is an urgent need for MR tools because many causal questions in health research cannot be adequately answered with conventional observational study designs or are not amenable to evaluation with RCTs for logistical or ethical reasons.

    “There is an urgent need for these tools as many causal questions in health research cannot be adequately answered with conventional observational study designs or are not amenable to evaluation with RCT’s for logistical or ethical reasons.”

  • The authors state this as an assumption.

    Genetic variants provide a source of exogenous variation in the exposure and can act as an instrumental variable.

    “MR is based on the assumption that genetic variants provide a source of such exogenous variation in the exposure and can therefore act as an instrumental variable.1”

  • The authors state this as an assumption.

    The relationship between the exposure and the outcome is linear across different values of the exposure for the vast majority of MR estimation methods.

    “Additionally, the vast majority of MR estimation methods (with non-linear MR30 being the notable exception) impose the assumption that the relationship between the exposure and the outcome is linear across different values of the exposures.”

  • The authors state this as an assumption.

    Changes in genetic variation are equivalent in their effects to changes in the exposure through environmental or pharmaceutical manipulation (gene-environment equivalence).

    “Finally, data used in MR additionally require the assumption that changes in genetic variation are equivalent in their effects to changes in the exposure through environmental or pharmaceutical manipulation — a concept known as gene-environment equivalence.32”

  • The authors state this as an assumption.

    For two-sample MR estimation, the two samples are from the same underlying population, or more narrowly that the association between the genetic variants and exposure is the same in both samples.

    “An important assumption for two-sample MR estimation is that the two samples are from the same underlying population, or more narrowly that the association between the genetic variants and exposure is the same in both samples, although that exposure may not be measured or reported in the sample included in the outcome dataset.39”

  • The authors state this as an assumption.

    Overidentification tests rely on the assumption that each IV estimates the same causal effect.

    “They also rely on the assumption that each IV estimates the same causal effect, which may not be true for complex traits where different genetic variants potentially act as genetic instruments for different aspects of the trait.”

  • The authors state this as an assumption.

    sisVIVE and adaptive LASSO assume that multiple IVs are available and that a majority or plurality of the proposed IVs are valid.

    “These methods assume that multiple IVs are available and that a majority or plurality of the proposed IVs are valid.”

  • The authors state this as an assumption.

    For inverse-variance weighted estimation, the genetic variants are independent of each other.

    “One important assumption for IVW estimation is that the genetic variants are independent of each other.41”

  • The authors state this as an assumption.

    MR Egger analysis assumes that across all instruments, the magnitude of the pleiotropic effect is unrelated to the strength of the association between the genetic variant and the phenotype of interest (the InSIDE assumption).

    “MR Egger analysis assumes that across all instruments, the magnitude of the pleiotropic effect is unrelated to the strength of the association between the genetic variant and the phenotype of interest (known as the InSIDE assumption)”

  • The authors state this as an assumption.

    Multivariable Mendelian randomization assumes that pleiotropic pathways operate through known and well-measured phenotypes that are also included in the estimation.

    “multivariable MR assumes pleiotropic pathways operate through known and well-measured phenotypes that are also included in the estimation.”

  • The authors state this as an assumption.

    For multivariable Mendelian randomization, each exposure must be robustly predicted by the instruments, conditional on the other exposures included in the estimation (MVIV1).

    “each exposure must be robustly predicted by the instruments, conditional on the other exposures included in the estimation (Multivariable instrumental variable condition 1, or MVIV1)”

  • The authors state this as an assumption.

    For multivariable Mendelian randomization, there must be no confounders of the outcome and any of the instruments (MVIV2).

    “there must be no confounders of the outcome and any of instruments (MVIV2)”

  • The authors state this as an assumption.

    For multivariable Mendelian randomization, none of the instruments can have an effect on the outcome that does not act through at least one of the exposures (MVIV3).

    “none of the instruments can have an effect on the outcome that doesn’t act through at least one of the exposures (MVIV3)”

  • The authors state this as an assumption.

    Bayesian colocalization strategies assume that summary data exist for multiple variants in a region, with sufficient density such that any causal variant or variants for an outcome and exposure are likely to be included in the set of variants studied.

    “the latter assume summary data exist for multiple variants in a region, with sufficient density such that any causal variant or variants for an outcome and exposure are likely to be included in the set of variants studied.”

  • The authors state this as an assumption.

    The conclusion about alcohol consumption and vascular disease risk assumes no selection bias.

    “assuming no selection bias”

  • The paper relies on this without support.

    Genetic variants are generally fixed at conception.

    “As genetic variants are generally fixed at conception, it is not intuitively clear how confounding of the instrument and the outcome can occur in MR studies.”

  • The paper relies on this without support.

    Most studies recruit survivors of the original birth cohorts.

    “In addition, most studies recruit survivors of the original birth cohorts.”

  • The paper relies on this without support.

    MR uses genetic variants that are fixed across the life course to estimate the lifetime effects of the exposure of interest.

    “Finally, MR uses genetic variants that are fixed across the life course to estimate the lifetime effects of the exposure of interest.”

  • The paper relies on this without support.

    Most contemporary MR studies rely on available GWAS data.

    “As most contemporary MR studies rely on available GWAS data, they unfortunately suffer from considerable bias with respect to representativeness of populations according to geography and ancestry.184”

  • The paper relies on this without support.

    Adiposity is a confounder of the observational association of CRP and vitamin D with health outcomes.

    “with adiposity being a confounder of the observational association of these exposures with health outcomes.”

  • The paper relies on this without support.

    Smoking cessation is not a useful treatment once lung cancer has developed.

    “as smoking cessation is not a useful treatment once lung cancer has developed.”

  • The paper relies on this without support.

    The underlying magnitude of the effect in drug-target MR is very small.

    “using a genetic instrument for treatment as part of a drug-target MR means that the underlying magnitude of the effect on which potential genetic effect modifiers are investigated is very small”

  • The paper relies on this without support.

    The least plausible assumption may be shared by many methods.

    “although the least plausible assumption may be shared by many methods”
23 of 23

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Bidirectional MR inference reverses the reported direction of effect

    Major

    The stated pattern-dementia-associated variants influencing hearing loss, but hearing-loss-associated variants not influencing dementia-supports dementia affecting hearing loss, not the stated conclusion that hearing loss causes dementia.

  • IVW regression description mislabels the outcome association

    Minor

    The section defines π as the variant-exposure association and Γ as the variant-outcome association, but then calls π the variant-outcome association in its description of the regression.

2 of 2

Cites

1 retracted

Flagged first, then by citations

Research works printed but not linked: 10

  1. Glymour, M. M. Natural experiments and instrumental variable analyses in social epidemiology. Methods Soc. Epidemiol. 1, 429 (2006).
  2. Zhao, Q., Wang, J., Spiller, W., Bowden, J. & Small, D. S. Two-sample instrumental variable analyses using heterogeneous samples. Stat. Sci. 34, 317–333 (2019).
  3. Bound, J., Jaeger, D. A. & Baker, R. M. Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak. J. Am. Stat. Assoc. 90, 443–450 (1995).
  4. Mounier, N. & Kutalik, Z. Correction for sample overlap, winner’s curse and weak instrument bias in two-sample Mendelian Randomization. Preprint at bioRxiv https://doi.org/10.1101/2021.03.26.437168 (2021). · DOI
  5. Wooldridge, J. M. Econometric Analysis of Cross Section and Panel Data (MIT Press, 2010).
  6. Tchetgen Tchetgen, E. J., Sun, B. & Walter, S. The GENIUS approach to robust Mendelian randomization inference. Stat. Sci. 36, 443–464 (2019).
  7. Sallis, H. Triangulation protocol; intergenerational effects of parental substance use on child substance use and mental health outcomes. Preprint at https://osf.io/s6jv4/ (2021).
  8. Brumpton, B. et al. Avoiding dynastic, assortative mating, and population stratification biases in Mendelian randomization through within-family analyses. Nat. Commun. 11, 3519 (2020). · DOI · Paper page
  9. Mahmoud, O., Dudbridge, F., Davey Smith, G., Munafo, M. & Tilling, K. Slope-Hunter: a robust method for index-event bias correction in genome-wide association studies of subsequent traits. Nat. Commun. (in the press).
  10. Ye, T., Shao, J. & Kang, H. Debiased inverse-variance weighted estimator in two-sample summary-data Mendelian randomization. Ann. Stat. 49, 2079–2100 (2021).

Cited by

Sorted by citations

Related

Authors
Sanderson, E., Glymour, M. M., Holmes, M. V., Kang, H., Morrison, J., Munafò, M. R., Palmer, T., Schooling, C. M., Wallace, C., Zhao, Q., & Davey Smith, G.
Affiliations
  • University of Bristol (GB)
  • University of California, San Francisco (US)
  • University of Wisconsin–Madison (US)
  • University of Michigan (US)
  • City University of New York (US)
  • University of Cambridge (GB)
Corresponding author
Sanderson, Eleanor
ORCID
Venue
Nature Reviews Methods Primers
Publisher
Springer Science and Business Media LLC
Published
2022-02-10
Volume
2
Issue
1
ISSN
2662-8449 (ISSN-L)
DOI
10.1038/s43586-021-00092-5
PMC
PMC7614635
OpenAlex
W4211103501
Semantic Scholar
a18b5dabdb938b4c3b9f909a34a109d7744cc5a8
Cited by
2,229
References
222
Citations by year
2026: 450 · 2025: 782 · 2024: 688 · 2023: 243 · 2022: 66
FWCI
69.62
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Reviews Methods Primers (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
  • Explore Bristol Research (acceptedVersion, other-oa)
  • Bristol Research (University of Bristol) (submittedVersion, other-oa)
Funders
  • Wellcome Trust (107881/Z/15/Z, WT107881)
  • British Heart Foundation (FS/18/23/33512)
  • Medical Research Council (MC_UU_00011/7, MC_UU_00002/4, MC_UU_00011/3, MC_UU_00011/1)
  • National Institute on Aging (R01AG057869)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
  • Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
Keywords
Mendelian randomization, causal inference, genetic instruments, unmeasured confounding, Mendelian inheritance, triangulation, genomic data

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-10.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-10.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review