Systematically Missing Confounders in Individual Participant Data Meta-analysis of Observational Cohort Studies

Overview

Journal Stat Med

Publisher Wiley

Specialty Public Health

Date 2009 Feb 18

PMID 19222087

Citations 24

Authors

Dan Jackson

Ian White

J B Kostis

A C Wilson

A R Folsom

K Wu

L Chambless

M Benderly

U Goldbourt

J Willeit

S Kiechl

J W G Yarnell

P M Sweetnam

P C Elwood

M Cushman

B M Psaty

R P Tracy

A Tybjaerg-Hansen

F Haverkate

M P M De Maat

F G R Fowkes

A J Lee

F B Smith

R DAgostino

W B Kannel

P W F Wilson

G Tofler

D Levy

R Marchioli

F Valagussa

A Rosengren

L Wilhelmsen

G Lappas

H Eriksson

P Cremer

D Nagel

J D Curb

B Rodriguez

K Yano

J T Salonen

K Nyyssonen

T-P Tuomainen

B Hedblad

G Engstrom

G Berglund

H Loewel

W Koenig

H W Hense

B De Stavola

G J Miller

K A Bauer

R D Rosenberg

S Sato

A Kitamura

Y Naito

H Iso

V Salomaa

K Harald

V Rasi

E Vahtera

P Jousilahti

T Palosuo

P Ducimetiere

P Amouyel

D Arveiler

A E Evans

J Ferrieres

I Juhan-Vague

A Bingham

H Schulte

G Assmann

B Cantin

B Lamarche

J-P Despres

G R Dagenais

H Tunstall-Pedoe

M Woodward

Y Ben-Shlomo

G Davey Smith

V Palmieri

J L Yeh

T W Meade

A Rudnicka

P Brennan

C Knottenbelt

J A Cooper

P Ridker

F Rodeghiero

A Tosetto

J Shepherd

I Ford

M Robertson

E Brunner

M Shipley

E J M Feskens

E Di Angelantonio

S Kaptoge

S Lewington

G D O Lowe

N Sarwar

S G Thompson

M Walker

S Watson

I R White

A M Wood

J Danesh

Affiliations

Soon will be listed here.

Abstract

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

Citing Articles

Perioperative Factors Associated With Postoperative Delirium in Patients Undergoing Noncardiac Surgery: An Individual Patient Data Meta-Analysis.

Sadeghirad B, Dodsworth B, Gelsomino N, Goettel N, Spence J, Buchan T JAMA Netw Open. 2023; 6(10):e2337239.

PMID: 37819663 PMC: 10568362. DOI: 10.1001/jamanetworkopen.2023.37239.

Combining Multiple Observational Data Sources to Estimate Causal Effects.

Yang S, Ding P J Am Stat Assoc. 2020; 115(531):1540-1554.

PMID: 33088006 PMC: 7571608. DOI: 10.1080/01621459.2019.1609973.

A Bayesian multivariate meta-analysis of prevalence data.

Siegel L, Rudser K, Sutcliffe S, Markland A, Brubaker L, Gahagan S Stat Med. 2020; 39(23):3105-3119.

PMID: 32510638 PMC: 7571488. DOI: 10.1002/sim.8593.

Evidence synthesis in prognosis research.

Debray T, de Jong V, Moons K, Riley R Diagn Progn Res. 2019; 3:13.

PMID: 31338426 PMC: 6621956. DOI: 10.1186/s41512-019-0059-4.

The role of secondary outcomes in multivariate meta-analysis.

Copas J, Jackson D, White I, Riley R J R Stat Soc Ser C Appl Stat. 2018; 67(5):1177-1205.

PMID: 30344346 PMC: 6193545. DOI: 10.1111/rssc.12274.

References

Tudur Smith C, Williamson P, Marson A . Investigating heterogeneity in an individual patient data meta-analysis of time to event outcomes. Stat Med. 2005; 24(9):1307-19. DOI: 10.1002/sim.2050. View

Riley R, Thompson J, Abrams K . An alternative model for bivariate random-effects meta-analysis when the within-study correlations are unknown. Biostatistics. 2007; 9(1):172-86. DOI: 10.1093/biostatistics/kxm023. View

Steyerberg E, Eijkemans M, van Houwelingen J, Lee K, Habbema J . Prognostic models based on literature and individual patient data in logistic regression analysis. Stat Med. 2000; 19(2):141-60. DOI: 10.1002/(sici)1097-0258(20000130)19:2<141::aid-sim334>3.0.co;2-o. View

Higgins J, Whitehead A, Turner R, Omar R, Thompson S . Meta-analysis of continuous outcome data from individual patients. Stat Med. 2001; 20(15):2219-41. DOI: 10.1002/sim.918. View

Hardy R, Thompson S . A likelihood approach to meta-analysis with random effects. Stat Med. 1996; 15(6):619-29. DOI: 10.1002/(SICI)1097-0258(19960330)15:6<619::AID-SIM188>3.0.CO;2-A. View

Williamson P, Tudur Smith C, Hutton J, Marson A . Aggregate data meta-analysis with time-to-event outcomes. Stat Med. 2002; 21(22):3337-51. DOI: 10.1002/sim.1303. View

Whitehead A, Omar R, Higgins J, Savaluny E, Turner R, Thompson S . Meta-analysis of ordinal outcomes using individual patient data. Stat Med. 2001; 20(15):2243-60. DOI: 10.1002/sim.919. View

Danesh J, Lewington S, Thompson S, Lowe G, Collins R, Kostis J . Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: an individual participant meta-analysis. JAMA. 2005; 294(14):1799-809. DOI: 10.1001/jama.294.14.1799. View

Ishak K, Platt R, Joseph L, Hanley J . Impact of approximating or ignoring within-study covariances in multivariate meta-analyses. Stat Med. 2007; 27(5):670-86. DOI: 10.1002/sim.2913. View

10.

Riley R, Simmonds M, Look M . Evidence synthesis combining individual patient data and aggregate data: a systematic review identified current practice and possible methods. J Clin Epidemiol. 2007; 60(5):431-9. DOI: 10.1016/j.jclinepi.2006.09.009. View

11.

Riley R, Abrams K, Lambert P, Sutton A, Thompson J . An evaluation of bivariate random-effects meta-analysis for the joint synthesis of two correlated outcomes. Stat Med. 2006; 26(1):78-97. DOI: 10.1002/sim.2524. View

12.

Turner R, Omar R, Yang M, Goldstein H, Thompson S . A multilevel model framework for meta-analysis of clinical trials with binary outcomes. Stat Med. 2000; 19(24):3417-32. DOI: 10.1002/1097-0258(20001230)19:24<3417::aid-sim614>3.0.co;2-l. View

13.

Blettner M, Sauerbrei W, Schlehofer B, Scheuchenpflug T, Friedenreich C . Traditional reviews, meta-analyses and pooled analyses in epidemiology. Int J Epidemiol. 1999; 28(1):1-9. DOI: 10.1093/ije/28.1.1. View

14.

Ford I, Norrie J, Ahmadi S . Model inconsistency, illustrated by the Cox proportional hazards model. Stat Med. 1995; 14(8):735-46. DOI: 10.1002/sim.4780140804. View

15.

Danesh J, Erqou S, Walker M, Thompson S, Tipping R, Ford C . The Emerging Risk Factors Collaboration: analysis of individual data on lipid, inflammatory and other markers in over 1.1 million participants in 104 prospective studies of cardiovascular diseases. Eur J Epidemiol. 2007; 22(12):839-69. DOI: 10.1007/s10654-007-9165-7. View

16.

Wood A, White I, Thompson S, Lewington S, Danesh J . Regression dilution methods for meta-analysis: assessing long-term variability in plasma fibrinogen among 27,247 adults in 15 prospective studies. Int J Epidemiol. 2006; 35(6):1570-8. DOI: 10.1093/ije/dyl233. View

17.

Steyerberg E, Eijkemans M, Harrell Jr F, Habbema J . Prognostic modelling with logistic regression analysis: a comparison of selection and estimation methods in small data sets. Stat Med. 2000; 19(8):1059-79. DOI: 10.1002/(sici)1097-0258(20000430)19:8<1059::aid-sim412>3.0.co;2-0. View

18.

Riley R, Abrams K, Sutton A, Lambert P, Thompson J . Bivariate random-effects meta-analysis and the estimation of between-study correlation. BMC Med Res Methodol. 2007; 7:3. PMC: 1800862. DOI: 10.1186/1471-2288-7-3. View

19.

Stewart L, Parmar M . Meta-analysis of the literature or of individual patient data: is there a difference?. Lancet. 1993; 341(8842):418-22. DOI: 10.1016/0140-6736(93)93004-k. View

20.

Riley R, Sutton A, Abrams K, Lambert P . Sensitivity analyses allowed more appropriate and reliable meta-analysis conclusions for multiple outcomes when missing data was present. J Clin Epidemiol. 2004; 57(9):911-24. DOI: 10.1016/j.jclinepi.2004.01.018. View