A multiple mediator analysis approach to quantify the effects of the ADH1B and ALDH2 genes on hepatocellular carcinoma risk

Shih, Stephannie; Huang, Yue; Yang, Hwai I.

doi:10.1002/gepi.22120

Cited by 19 publications

(32 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also developed another method in this form for survival outcomes; however, the previously derived method cannot be directly applied to dichotomous outcomes with continuous mediators . Although the proposed method overcomes the above limitations, the method may not achieve satisfied results with smaller potential outcome (ie, the sample size of Y =1 is much smaller than the sample size of Y =0) due to the problem of the asymptotic attainability .…”

Section: Discussionmentioning

confidence: 99%

“…This highlights the importance of developing multiple mediator models for sequentially ordered mediators to account for confounders between both mediators and addressing the inability to separate specific paths from the mediator to the outcome . Regardless of exposure‐induced confounding, path‐specific effects (PSEs) can still be identified to understand the mediation effects and have been applied in survival analyses, continuous outcomes, and dichotomous outcomes with dichotomous mediators . Nonetheless, closed‐form results are only available for dichotomous mediators.…”

Section: Introductionmentioning

confidence: 99%

“…1,5,6,19 Regardless of exposure-induced confounding, path-specific effects (PSEs) can still be identified to understand the mediation effects and have been applied in survival analyses, continuous outcomes, and dichotomous outcomes with dichotomous mediators. 16,[20][21][22][23] Nonetheless, closed-form results are only available for dichotomous mediators. Weighting approaches have been proposed in this context, but may require bootstrapping to estimate valid standard errors.…”

mentioning

confidence: 99%

See 2 more Smart Citations

A mediation analysis for a nonrare dichotomous outcome with sequentially ordered multiple mediators

Lai

Shih

Huang

et al. 2020

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

Mediation analyses can help us to understand the biological mechanism in which an exposure or treatment affects an outcome. Single mediator analyses have been used in various applications, but may not be appropriate for analyzing intricate mechanisms involving multiple mediators that affect each other. Thus, in this article, we studied multiple sequentially ordered mediators for a dichotomous outcome and presented the identifiability assumptions for the path‐specific effects on the outcome, that is, the effect of an exposure on the outcome mediated by a specific set of mediators. We proposed a closed‐form estimator for the path‐specific effects by modeling the dichotomous outcome using a probit model. Asymptotic variance of the proposed estimator is derived and can be approximated via delta method or bootstrapping. Simulations under a finite sample showed the validity of our method in capturing the path‐specific effects when the probability of each potential counterfactual outcome is not small and demonstrated the utility of a computationally efficient alternative to bootstrapping for calculating variance. The method is applied to investigate the effects of polycystic ovarian syndrome on live birth rates mediated by estradiol levels and the number of oocytes retrieved in a large electronic in vitro fertilization database. We implemented the method into an R package SOMM, which is available at https://github.com/roqe/SOMM.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

A mediation analysis for a nonrare dichotomous outcome with sequentially ordered multiple mediators

Lai

Shih

Huang

et al. 2020

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

show abstract

“…al. [49]. Fig 8C shows an edge from Y to W with direction probability 0.72, which is in the 402 opposite direction from how it was simulated.…”

mentioning

confidence: 88%

“…We also investigated the utility of BNs for performing causal inference in a binary trait 783 setting. Discrete binary data was simulated for 5000 individuals using four models 784 considered in recent work by Shih et al [49], in the context of quantifying the effects 785 of alcohol consumption and high alanine transaminase levels on hepatocellular 786 carcinoma. We used the same graph (Fig 7) and parameter settings ( (binomial), reflecting the fact that they followed a discrete distribution.…”

mentioning

confidence: 99%

Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data

Howey

Shin

Relton

et al. 2019

Preprint

View full text Add to dashboard Cite

Mendelian randomization (MR) is an increasingly popular causal inference tool used in genetic epidemiology. But it can have limitations for evaluating simultaneous causal relationships in complex data sets that include, for example, multiple genetic predictors and multiple potential risk factors associated with the same genetic variant. Here we use real and simulated data to investigate Bayesian network analysis (BN) as an alternative approach. A Bayesian network describes the conditional dependencies/independencies of variables using a graphical model (a directed acyclic graph) and its accompanying joint probability. In real data, we found BN inferred simultaneous causal relationships that confirmed the individual causal relationships suggested by bi-directional MR, while allowing for the existence of potential horizontal pleiotropy (that would violate MR assumptions). In simulated data, BN with two directional anchors (mimicking genetic instruments) had greater power for a fixed type 1 error than bi-directional MR, while BN with a single directional anchor performed better than or as well as bi-directional MR. Both BN and MR could be adversely affected by violations of their underlying assumptions (such as genetic confounding due to unmeasured horizontal pleiotropy). BN with no directional anchor generated inference that was no better than by chance, emphasizing the importance of directional anchors in BN (as in MR). Under highly pleiotropic simulated scenarios, BN outperformed both MR (and its recent extensions) and two recently-proposed alternative approaches: a multi-SNP mediation intersection-union test (SMUT) and a latent causal variable (LCV) test. We conclude that BN is a useful complementary method to MR for performing causal inference in complex data sets such as those generated from modern "omics" technologies Author summaryMendelian randomization (MR) is a popular method for inferring causal relationships between variables (such as between an intermediate biological factor and a disease outcome). However, MR relies on a number of assumptions that may be hard to verify, and it is not ideally suited to comparing different underlying causal scenarios. Here we September 19, 2019 1/52 propose the use of an alternative method, Bayesian network analysis (BN), as a complementary tool to MR. We use real and simulated data to investigate the performance of MR, BN and several other recently-proposed methods, and find that BN performs as well as, or better than, the other methods, particularly under complex scenarios. We conclude that BN is a useful complementary method to MR for performing causal inference in complex data sets. 35 been applied to identify innovative drug targets in early stage drug development or to 36 discover novel risk factors at the molecular level by scanning "omics" data 37 systemically [8-10]. An advantage of MR is that individual-level data are not 38 necessarily required; inference can be performed on the basis of summary statistics 39 measuring the relationship between the genetic ins...

show abstract

Cigarette smoking increases the risk of nasopharyngeal carcinoma through the elevated level of IgA antibody against Epstein‐Barr virus capsid antigen: A mediation analysis

et al. 2020

Self Cite

View full text Add to dashboard Cite

Background The study aims are to evaluate the associations between nasopharyngeal carcinoma (NPC) risk and cigarette smoking and to explore the effects of cigarette smoking on Epstein‐Barr virus (EBV) infection for NPC risk. Methods 1235 male NPC cases and 1262 hospital‐based male controls matched to cases were recruited across six collaborative hospitals between 2010 and 2014. Using a standardized questionnaire, information on cigarette smoking and other potential risk factors for NPC was obtained. Blood was collected and used for anti‐EBV VCA IgA and anti‐EBV EA‐EBNA1 IgA testing using standard methods. Unconditional logistic regression analysis was used to estimate odds ratio (OR) with 95% confidence interval (CI) for each risk factor after adjusting for confounders. Results 63.6% of cases and 44.0% of controls reported ever smoking cigarettes. After full adjustment, current smokers had a significant 1.60‐fold (95% CI = 1.30‐1.97) and former smokers a borderline significant 1.27‐fold (95% CI = 1.00‐1.60) increased NPC risk compared to never smokers. NPC risk increased with increasing duration, intensity, and pack‐years of cigarette smoking but not with age at smoking initiation. Among controls, anti‐EBV VCA IgA seropositivity rate was higher in current smokers than never smokers (14.0% vs 8.4%; OR = 1.82; 95% CI = 1.19‐2.79). Mediation analyses showed that more than 90% of the cigarette smoking effect on NPC risk is mediated through anti‐EBV VCA IgA. Conclusion This study confirms the association between long‐term cigarette smoking and NPC and demonstrates that current smoking is associated with seropositivity of anti‐EBV VCA IgA antibodies.

show abstract

A multiple mediator analysis approach to quantify the effects of the ADH1B and ALDH2 genes on hepatocellular carcinoma risk

Cited by 19 publications

References 35 publications

A mediation analysis for a nonrare dichotomous outcome with sequentially ordered multiple mediators

A mediation analysis for a nonrare dichotomous outcome with sequentially ordered multiple mediators

Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data

Cigarette smoking increases the risk of nasopharyngeal carcinoma through the elevated level of IgA antibody against Epstein‐Barr virus capsid antigen: A mediation analysis

Contact Info

Product

Resources

About