Causal mediation analysis investigates the mechanism linking exposure and outcome. Dealing with the impact of unobserved confounders among exposure, mediator and outcome is an issue of great concern. Moreover, when multiple mediators exist, this causal pathway intertwines with other causal pathways, rendering it difficult to estimate the path-specific effects. In this study, we propose a method (PSE-MR) to identify and estimate path-specific effects of an exposure (e.g. education) on an outcome (e.g. osteoarthritis risk) through multiple causally ordered and non-ordered mediators (e.g. body mass index and pack-years of smoking) using summarized genetic data, when the sequential ignorability assumption is violated. Specifically, PSE-MR requires a specific rank condition in which the number of instrumental variables is larger than the number of mediators. Furthermore, we illustrate the utility of PSE-MR by providing guidance for practitioners and exploring the mediation effects of body mass index and pack-years of smoking in the causal pathways from education to osteoarthritis risk. Additionally, the results of simulation reveal that the causal estimates of path-specific effects are almost unbiased with good coverage and Type I error properties. Also, we summarize the least number of instrumental variables for the specific number of mediators to achieve 80% power.
MilesBCHShpitserIKankiP,et al.On semiparametric estimation of a path-specific effect in the presence of mediator-outcome confounding. Biometrika2020; 107: 159–172.
3.
ImaiKKeeleLYamamotoT. Identification, inference and sensitivity analysis for causal mediation effects. Stat Sci2010; 25: 51–71.
4.
FulcherIRShiXTchetgen TchetgenEJ. Estimation of natural indirect effects robust to unmeasured confounding and mediator measurement error. Epidemiology2019; 30: 825–834.
5.
MccandlessLCSomersJM. Bayesian sensitivity analysis for unmeasured confounding in causal mediation analysis. Stat Methods Med Res 2019; 28: 515–531.
6.
MacKinnonDP. Contrasts in multiple mediator models. Multivariate applications in substance use research: New methods for new questions, 141–160. 2000.
7.
AvinCShpitserIPearlJ. Identifiability of path-specific effects. 2005.
8.
ImaiKYamamotoT. Identification and sensitivity analysis for multiple causal mechanisms: Revisiting evidence from framing experiments. Polit Anal2013; 21: 141–171.
VanderWeeleTVansteelandtS. Mediation analysis with multiple mediators. Epidemiol Method2014; 2: 95–115.
11.
TchetgenEJTShpitserI. Semiparametric theory for causal mediation analysis: Efficiency bounds, multiple robustness, and sensitivity analysis. Ann Stat2012; 40: 1816–1845.
12.
LuoPGengZ. Causal mediation analysis for survival outcome with unobserved mediator–outcome confounders. Comput Stat Data Anal2016; 93: 336–347.
13.
VanderWeeleTJChibaY. Sensitivity analysis for direct and indirect effects in the presence of exposure-induced mediator-outcome confounders. Epidemiol Biostat Public Health2014; 11: e9027.
DingPVanderweeleTJ. Sharp sensitivity bounds for mediation under unmeasured mediator-outcome confounding. Biometrika. 2016;103:483–490.
16.
BurgessSSmallDSThompsonSG. A review of instrumental variable estimators for Mendelian randomization. Stat Methods Med Res2017; 26: 2333–2355.
17.
BurgessSButterworthAThompsonSG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol2013; 37: 658–665.
18.
TikkanenEGustafssonSKnowlesJW,et al.Body composition and atrial fibrillation: A Mendelian randomization study. Eur Heart J2019; 40: 1277–1282.
19.
BurgessSThompsonSG. Multivariable Mendelian randomization: The use of pleiotropic genetic variants to estimate causal effects. Am J Epidemiol2015; 181: 251–260.
20.
ReesJMWoodAMBurgessS. Extending the MR-Egger method for multivariable Mendelian randomization to correct for both measured and unmeasured pleiotropy. Stat Med2017; 36: 4705–4718.
21.
BurgessSThompsonDJReesJM,et al.Dissecting causal pathways using Mendelian randomization with summarized genetic data: application to age at menarche and risk of breast cancer. Genetics2017; 207: 481–487.
22.
HawkerGA. Osteoarthritis is a serious disease. Clin Exp Rheumatol2019; 37: 3–6.
23.
GillDKarhunenVMalikR,et al.Cardiometabolic traits mediating the effect of education on osteoarthritis risk: A Mendelian randomization study. Osteoarthritis Cartilage2021; 29: 365–371.
24.
WangHBaiJHeB,et al.Osteoarthritis and the risk of cardiovascular disease: A meta-analysis of observational studies. Sci Rep2016; 6: 1–7.
25.
CarterARGillDDaviesNM,et al.Understanding the consequences of education inequality on cardiovascular disease: Mendelian randomisation study. Br Med J2019; 365: l1855.
26.
TillmannTVaucherJOkbayA, et al.Education and coronary heart disease: Mendelian randomisation study. Br Med J2017: 358: j3542.
27.
KimTJRoeslerNMvon dem KnesebeckO. Causation or selection – examining the relation between education and overweight/obesity in prospective observational studies: A meta-analysis. Obesity Rev: Off J Int Assoc Study Obesity2017; 18: 660–672.
28.
de WalqueD. Does education affect smoking behaviors? Evidence using the Vietnam draft as an instrument for college education. J Health Econ2007; 26: 877–895.
29.
TaylorAERichmondRCPalviainenT, et al.The effect of body mass index on smoking behaviour and nicotine metabolism: A Mendelian randomization study. Hum Mol Genet2019; 28: 1322–1330.
30.
ShpitserI. Counterfactual graphical models for longitudinal mediation analysis with unobserved confounding. Cogn Sci2013; 37: 1011–1035.
31.
TaguriMFeatherstoneJChengJ. Causal mediation analysis with multiple causally non-ordered mediators. Stat Methods Med Res2018; 27: 3–19.
32.
BurgessSDudbridgeFThompsonSG. Combining information on multiple instrumental variables in Mendelian randomization: Comparison of allele score and summarized data methods. Stat Med2016; 35: 1880–1906.
33.
BowdenJDavey SmithGBurgessS. Mendelian randomization with invalid instruments: Effect estimation and bias detection through Egger regression. Int J Epidemiol2015; 44: 512–525.
34.
AngristJDKruegerAB. Split-sample instrumental variables estimates of the return to schooling. J Bus Econ Stat1995; 13: 225–235.
35.
MartensEPPestmanWRde BoerA,et al.Instrumental variables: Application and limitations. Epidemiology2006; 17: 260–267.
36.
LiuQYinRZieglerK,et al.Quantum walks: The mean first detected transition time. Phys Rev Res2020; 2: 033113.
37.
StaigerDStockJH. Instrumental variables regression with weak instruments (No. t0151). National Bureau of Economic Research. 1994.
38.
SwansonSAVanderWeeleTJ. E-values for Mendelian randomization. Epidemiology2020; 31: e23–e24.
39.
LeeJJWedowROkbayA, et al.Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals. Nat Genet2018; 50: 1112–1121.
40.
HoffmannTJChoquetHYinJ, et al.A large multiethnic genome-wide association study of adult body mass index identifies novel loci. Genetics2018; 210: 499–515.
41.
TachmazidouIHatzikotoulasKSouthamL, et al.Identification of new therapeutic targets for osteoarthritis through genome-wide analyses of UK Biobank data. Nat Genet2019; 51: 230–236.
42.
ReesJFoleyCNBurgessS. Factorial Mendelian randomization: Using genetic variants to assess interactions. Int J Epidemiol2020; 49: 1147–1158.
43.
VerbanckMChenCYNealeB,et al.Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet2018; 50: 693–698.
44.
DubéCELiuSHDribanJB,et al.The relationship between smoking and knee osteoarthritis in the osteoarthritis initiative. Osteoarthritis Cartilage2016; 24: 465–472.
45.
FelsonDTZhangY. Smoking and osteoarthritis: A review of the evidence and its implications. Osteoarthritis Cartilage2015; 23: 331–333.
46.
VanderWeeleTJVansteelandtSRobinsJM. Effect decomposition in the presence of an exposure-induced mediator-outcome confounder. Epidemiology2014; 25: 300–306.
47.
MittintyMNLynchJWForbesAB,et al.Effect decomposition through multiple causally nonordered mediators in the presence of exposure–induced mediator–outcome confounding. Stat Med2019; 38: 5085–5102.
48.
VanderWeeleTJVansteelandtS. Conceptual issues concerning mediation, interventions and composition. Stat Interface2009; 2: 457–468.
49.
ValeriLVanderWeeleTJ. Mediation analysis allowing for exposure–mediator interactions and causal interpretation: Theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods2013; 18: 137–150.
50.
VanderWeeleTJ. A unification of mediation and interaction: A four-way decomposition. Epidemiology2014; 25: 749–761.
51.
LohWWMoerkerkeBLoeysT,et al.Non-linear mediation analysis with high-dimensional mediators whose causal structure is unknown. arXiv preprint arXiv:2001.07147. 2020.
52.
TchetgenEJTVanderWeele,TJ. On identification of natural direct effects when a confounder of the mediator is directly affected by exposure. Epidemiology2014; 25: 282–291.
53.
VansteelandtSLinderMVandenbergheS,et al.Mediation analysis of time-to-event endpoints accounting for repeatedly measured mediators subject to time-varying confounding. Stat Med2019; 38: 4828–4840.
54.
ZhengWvan der LaanM. Longitudinal mediation analysis with time-varying mediators and exposures, with application to survival outcomes. J Causal Inference2017; 5: 20160006.
55.
HouLLiHYuY, et al.Exploring the causal pathway from bilirubin to cvd and diabetes in the uk biobank cohort study: Observational findings and Mendelian randomization studies. Atherosclerosis2020; 320: 112-121.
56.
HouLXuMYuY,et al.Exploring the causal pathway from ischemic stroke to atrial fibrillation: A network Mendelian randomization study. Mol Med2020; 26: 7.
57.
LabrecqueJASwansonSA. Commentary: Mendelian randomization with multiple exposures: The importance of thinking about time. Int J Epidemiol2020; 49: 1158–1162.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
