TY - JOUR
T1 - Comparison of variance estimators for metaanalysis of instrumental variable estimates
AU - Schmidt, A. F.
AU - Hingorani, A. D.
AU - Jefferis, B. J.
AU - White, J.
AU - Dudbridge, F.
AU - Groenwold, R. H. H.
AU - Ben-Shlomo, Y.
AU - Chaturvedi, N.
AU - Engmann, J.
AU - Hughes, A.
AU - Humphries, S.
AU - Hypponen, E.
AU - Kivimaki, M.
AU - Kuh, D.
AU - Kumari, M.
AU - Menon, U.
AU - Morris, R.
AU - Power, C.
AU - Price, J.
AU - Wannamethee, G.
AU - Whincup, P.
PY - 2016/12/1
Y1 - 2016/12/1
N2 - 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 twostagemeta- 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 metaanalysis 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.
AB - 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 twostagemeta- 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 metaanalysis 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.
UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85017177402&origin=inward
UR - https://www.ncbi.nlm.nih.gov/pubmed/27591262
U2 - https://doi.org/10.1093/ije/dyw123
DO - https://doi.org/10.1093/ije/dyw123
M3 - Article
C2 - 27591262
SN - 0300-5771
VL - 45
SP - 1975
EP - 1986
JO - International journal of epidemiology
JF - International journal of epidemiology
IS - 6
ER -