Premium
Systematically missing confounders in individual participant data meta‐analysis of observational cohort studies
Author(s) -
Daniel Jakson,
Ian R. White,
John B. Kostis,
Andrew Wilson,
Aaron R. Folsom,
Kunpeng Wu,
Lloyd E. Chambless,
U. Benderley,
Uri Goldbourt,
Johann Willeit,
Sophia J. Kiechl,
John Yarnell,
P. M. Sweetnam,
P. C. Elmwood,
Mary Cushman,
Bruce M. Psaty,
Russell P. Tracy,
Anne TybjærgHansen,
F. Haverkate,
Moniek P.M. de Maat,
Simon G. Thompson,
F. G. R. Fowkes,
A. J. Lee,
Fraser Smith,
Veikko Salomaa,
Kennet Harald,
Vesa Rasi,
Elina Vahtera,
Pekka Jousilahti,
RB D’Agostino,
Kannel Wb,
Paul W. Wilson,
Geoffrey H. Tofler,
Daniel Levy,
Roberto Marchioli,
Franco Valagussa,
Annika Rosengren,
L. Wilhemsen,
Georgios Lappas,
Henry Eriksson,
P. Cremer,
Dorothea Nagel,
J. David Curb,
Buenos Rodríguez,
Katsuhiko Yano,
J. T. Salonen,
Kristiiyyssönen,
Tomi Pekka Tuomainen,
Bo Hedblad,
Gunnar Engström,
Göran Berglund,
Hannelore Loewel,
Wolfgang Köenig,
Hans Werner Hense,
T W Meade,
J. A. Copper,
Bianca De Stavola,
C. Knottenbelt,
Grant Miller,
Jayne Cooper,
Kenneth A. Bauer,
R D Rosenberg,
Shinichi Sato,
Akihiko Kitamura,
Yusuke Naito,
Hiroyasu Iso,
T. Palosou,
Pierre Ducimetière,
Philippe Amouyel,
Dominique Arveiler,
Alun Evans,
Jean Ferrières,
I. Juhan-Vague,
Annie Bingham,
Helmut Schulte,
Gerd Assmann,
Bernard Cantin,
Benoı̂t Lamarche,
Jean Després,
G. R. Dagenais,
Hugh TunstallPedoe,
G.D.O. Lowe,
Mark Woodward,
Yatir BenShlomo,
George Davey Smith,
Vittorio Palmieri,
Jeun Liang Yeh,
Alicja R. Rudnicka,
Paul Brennan,
Paul M. Ridker,
Francesco Rodeghiero,
Alberto Tosetto,
John Shepherd,
Donna Lowe,
Ian Ford,
Michele Robertson,
Eric J. Brunner,
Martin J. Shipley,
E. J.M. Fesken,
Emanuele Di Angelantonio
Publication year - 2009
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.3540
Subject(s) - confounding , observational study , bivariate analysis , medicine , cohort study , cohort , meta analysis , demography , incidence (geometry) , propensity score matching , statistics , mathematics , geometry , sociology
One difficulty in performing meta‐analyses of observational cohort studies is that the availability of confounders may vary between cohorts, so that some cohorts provide fully adjusted analyses while others only provide partially adjusted analyses. Commonly, analyses of the association between an exposure and disease either are restricted to cohorts with full confounder information, or use all cohorts but do not fully adjust for confounding. We propose using a bivariate random‐effects meta‐analysis model to use information from all available cohorts while still adjusting for all the potential confounders. Our method uses both the fully adjusted and the partially adjusted estimated effects in the cohorts with full confounder information, together with an estimate of their within‐cohort correlation. The method is applied to estimate the association between fibrinogen level and coronary heart disease incidence using data from 154 012 participants in 31 cohorts. † One hundred and ninety‐nine participants from the original 154 211 withdrew their consent and have been removed from this analysis. Copyright © 2009 John Wiley & Sons, Ltd.