Comparison of variance estimators for meta-analysis of instrumental variable estimates

Abstract Background: Mendelian randomization studies perform instrumental variable (IV) analysis using genetic IVs. Results of individual Mendelian randomization studies can be pooled through meta-analysis. We explored how different variance estimators influence the meta-analysed IV estimate. Methods: Two versions of the delta method (IV before or after pooling), four bootstrap estimators, a jack-knife estimator and a heteroscedasticity-consistent (HC) variance estimator were compared using simulation. Two types of meta-analyses were compared, a two-stage meta-analysis pooling results, and a one-stage meta-analysis pooling datasets. Results: Using a two-stage meta-analysis, coverage of the point estimate using bootstrapped estimators deviated from nominal levels at weak instrument settings and/or outcome probabilities ≤ 0.10. The jack-knife estimator was the least biased resampling method, the HC estimator often failed at outcome probabilities ≤ 0.50 and overall the delta method estimators were the least biased. In the presence of between-study heterogeneity, the delta method before meta-analysis performed best. Using a one-stage meta-analysis all methods performed equally well and better than two-stage meta-analysis of greater or equal size. Conclusions: In the presence of between-study heterogeneity, two-stage meta-analyses should preferentially use the delta method before meta-analysis. Weak instrument bias can be reduced by performing a one-stage meta-analysis.

[1]  O. Klungel,et al.  Adjusting for Confounding in Early Postlaunch Settings: Going Beyond Logistic Regression Models , 2016, Epidemiology.

[2]  Stephen Burgess,et al.  Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods , 2015, Statistics in medicine.

[3]  Saskia le Cessie,et al.  Mendelian randomization studies: a review of the approaches used and the quality of reporting. , 2015, International journal of epidemiology.

[4]  A. Hoes,et al.  Exploring interaction effects in small samples increases rates of false-positive and false-negative findings: results from a systematic review and simulation study. , 2014, Journal of clinical epidemiology.

[5]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[6]  S. Burgess.  Identifying the odds ratio estimated by a two‐stage instrumental variable analysis with a logistic regression model , 2013, Statistics in medicine.

[7]  Tom R. Gaunt,et al.  Population Genomics of Cardiometabolic Traits: Design of the University College London-London School of Hygiene and Tropical Medicine-Edinburgh-Bristol (UCLEB) Consortium , 2013, PLoS ONE.

[8]  B. Pierce,et al.  Efficient Design for Mendelian Randomization Studies: Subsample and 2-Sample Instrumental Variable Estimators , 2013, American journal of epidemiology.

[9]  Daniela M. Witten,et al.  An Introduction to Statistical Learning: with Applications in R , 2013 .

[10]  A. Hoes,et al.  Differences in interaction and subgroup-specific effects were observed between randomized and nonrandomized studies in three empirical examples. , 2013, Journal of clinical epidemiology.

[11]  J. Vandenbroucke.  Why do the results of randomised and observational studies differ? , 2011, BMJ : British Medical Journal.

[12]  D. Lawlor,et al.  American Journal of Epidemiology Practice of Epidemiology Instrumental Variable Estimation of Causal Risk Ratios and Causal Odds Ratios in Mendelian Randomization Analyses , 2022 .

[13]  S. Thompson,et al.  Avoiding bias from weak instruments in Mendelian randomization studies. , 2011, International journal of epidemiology.

[14]  Marcus Richards,et al.  Cohort Profile: Updating the cohort profile for the MRC National Survey of Health and Development: a new clinic-based data collection for ageing research , 2011, International journal of epidemiology.

[15]  Vanessa Didelez,et al.  Assumptions of IV methods for observational epidemiology , 2010, 1011.0595.

[16]  M Alan Brookhart,et al.  American Journal of Epidemiology Practice of Epidemiology Instrumental Variable Analysis for Estimation of Treatment Effects with Dichotomous Outcomes , 2022 .

[17]  Paul J Rathouz,et al.  Two-stage residual inclusion estimation: addressing endogeneity in health econometric modeling. , 2008, Journal of health economics.

[18]  George Davey Smith,et al.  Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology , 2008, Statistics in medicine.

[19]  Maria L. Rizzo,et al.  Statistical Computing with R , 2007 .

[20]  D. Lawlor,et al.  Re: Estimation of bias in nongenetic observational studies using "Mendelian triangulation" by Bautista et al. , 2007, Annals of epidemiology.

[21]  Patrick Royston,et al.  The design of simulation studies in medical statistics , 2006, Statistics in medicine.

[22]  Georg Heinze,et al.  A comparative investigation of methods for logistic regression with separated or nearly separated data , 2006, Statistics in medicine.

[23]  A. Zeileis.  Object-oriented Computation of Sandwich Estimators , 2006 .

[24]  J. Robins,et al.  Instruments for Causal Inference: An Epidemiologist's Dream? , 2006, Epidemiology.

[25]  Wiebe R. Pestman,et al.  Instrumental Variables: Application and Limitations , 2006, Epidemiology.

[26]  Jan P Vandenbroucke,et al.  When are observational studies as credible as randomised trials? , 2004, The Lancet.

[27]  S. Thompson,et al.  Quantifying heterogeneity in a meta‐analysis , 2002, Statistics in medicine.

[28]  J. S. Long,et al.  Using Heteroscedasticity Consistent Standard Errors in the Linear Regression Model , 2000 .

[29]  J. Concato,et al.  Randomized, controlled trials, observational studies, and the hierarchy of research designs. , 2000, The New England journal of medicine.

[30]  J Carpenter,et al.  Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. , 2000, Statistics in medicine.

[31]  Anthony C. Davison,et al.  Bootstrap Methods and Their Application , 1998 .

[32]  S Greenland,et al.  Invited commentary: a critical look at some popular meta-analytic methods. , 1994, American journal of epidemiology.

[33]  M. Parmar,et al.  Meta-analysis of the literature or of individual patient data: is there a difference? , 1993, The Lancet.

[34]  P. Elwood,et al.  Plasma triglyceride and high density lipoprotein cholesterol as predictors of ischaemic heart disease in British men , 1992, British heart journal.

[35]  Gower Street,et al.  Health inequalities among British civil servants: the Whitehall II study , 1991, The Lancet.

[36]  R. Prescott,et al.  Edinburgh Artery Study: prevalence of asymptomatic and symptomatic peripheral arterial disease in the general population. , 1991, International journal of epidemiology.

[37]  S J Pocock,et al.  British Regional Heart Study: cardiovascular risk factors in middle-aged men in 24 towns. , 1981, British medical journal.