fibrinogen: Fibrinogen Studies Collaboration

Description Usage Format Details Note Source Examples

Description

The Fibrinogen Studies Collaboration is a meta-analysis of individual data on 154,012 adults from 31 prospective cohort studies with information on plasma fibrinogen and major disease outcomes. The dataset reports a subset of the results of a first-stage analysis consisting of the log-hazard ratio of coronary heart disease for categories of levels of fibrinogen versus a baseline category.

Usage

1

Format

A data frame with 31 observations on the following 15 variables:

Details

The published analysis adopted a fixed-effects model on 10 categories of fibrinogen (Fibrinogen Studies Collaboration 2004, 2005). Here a subset of the results of the first-stage analysis is reported, namely the log-hazard ratio for 4 categories and associated (co)variance terms, ordered as the lower triangular elements of the (co)variance matrix taken by column. Details on the first-stage model and the second-stage meta-analysis are provided in White (2009) and Jackson and colleagues (2010).

Note

The data provide an example of application of multivariate meta-analysis for multi-parameter association, where a relationship is defined by functions specified by several coefficients. In this case, the coefficients refer to log-hazard ratio for strata of the original variable versus a baseline category. A general overview of the application of multivariate meta-analysis in this setting is provided by Gasparrini and colleagues (2012).

Source

Fibrinogen Studies Collaboration (2004). Collaborative meta-analysis of prospective studies of plasma fibrinogen and cardiovascular disease. European Journal of Cardiovascular Prevention and Rehabilitation. 11:9–17.

Fibrinogen Studies Collaboration (2005). Plasma fibrinogen level and the risk of major cardiovascular diseases and nonvascular mortality: an individual participant meta-analysis. Journal of the American Medical Association. 294:1799–1809.

White IR (2009). Multivariate random-effects meta-analysis. Stata Journal. 9(1):40–56.

Jackson D, White IR, Thompson SG (2010). Extending DerSimonian and Laird's methodology to perform multivariate random effects meta-analyses. Statistics in Medicine. 29(12):1282–1297.

Sera F, Armstrong B, Blangiardo M, Gasparrini A (2019). An extended mixed-effects framework for meta-analysis.Statistics in Medicine. 2019;38(29):5429-5444. [Freely available here].

Gasparrini A, Armstrong B, Kenward MG (2012). Multivariate meta-analysis for non-linear and other multi-parameter associations. Statistics in Medicine. 31(29):3821–3839. [Freely available here].

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
### REPRODUCE THE RESULTS IN WHITE (2009) AND JACKSON ET AL. (2010)

# INSPECT THE DATA
head(fibrinogen)

# REML MODEL
y <- as.matrix(fibrinogen[2:5])
S <- as.matrix(fibrinogen[6:15])
model <- mvmeta(y,S)

# SUMMARIZE THE RESULTS
print(summary(model),digits=3)
round(model$Psi,3)

Example output

This is mvmeta 0.4.7. For an overview type: help('mvmeta-package').
  cohort           b2          b3         b4        b5      V_2_2      V_2_3
1      1  0.252366632  0.53170693  0.9464425 1.4009348 0.03591865 0.02508728
2      2 -0.184167668 -0.03180776  0.1185954 0.5671058 0.34832495 0.18939747
3      3  0.001030372 -0.52864528 -0.3390995 0.4163348 0.37536496 0.24996307
4      4  0.065784343  0.18355742  0.4065832 0.6450171 0.05759551 0.04175191
5      5  0.077638909  0.40645304  0.5442160 1.0878220 0.10126193 0.05600454
6      6 -0.113321237  0.45640695  0.4555633 0.8752530 0.06492682 0.02341051
       V_2_4      V_2_5      V_3_3      V_3_4      V_3_5      V_4_4      V_4_5
1 0.02512949 0.02513940 0.03256974 0.02523587 0.02525483 0.03103106 0.02537686
2 0.19161053 0.19551039 0.34446433 0.20458534 0.20916429 0.40842286 0.21454701
3 0.24960472 0.24948616 0.32252783 0.25055113 0.25082681 0.28301525 0.25098649
4 0.04176119 0.04178248 0.05280399 0.04183426 0.04188637 0.04838277 0.04196585
5 0.05614832 0.05622534 0.08338879 0.05661783 0.05676502 0.07697710 0.05714733
6 0.02353066 0.02394760 0.05397056 0.02376115 0.02423115 0.06393713 0.02441860
       V_5_5
1 0.03166101
2 0.43007851
3 0.25984231
4 0.05083219
5 0.06780010
6 0.07775327
Call:  mvmeta(formula = y ~ 1, S = S)

Multivariate random-effects meta-analysis
Dimension: 4
Estimation method: REML

Fixed-effects coefficients
    Estimate  Std. Error      z  Pr(>|z|)  95%ci.lb  95%ci.ub     
b2     0.162       0.075  2.143     0.032     0.014     0.309    *
b3     0.393       0.084  4.690     0.000     0.229     0.557  ***
b4     0.562       0.087  6.460     0.000     0.392     0.733  ***
b5     0.897       0.091  9.909     0.000     0.720     1.075  ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 

Between-study random-effects (co)variance components
	Structure: General positive-definite
    Std. Dev   Corr              
b2     0.227     b2     b3     b4
b3     0.286  0.990              
b4     0.308  0.974  0.997       
b5     0.327  0.706  0.801  0.848

Multivariate Cochran Q-test for heterogeneity:
Q = 187.883 (df = 120), p-value = 0.000
I-square statistic = 36.1%

31 studies, 124 observations, 4 fixed and 10 random-effects parameters
 logLik      AIC      BIC  
-79.489  186.978  226.003  

      b2    b3    b4    b5
b2 0.052 0.064 0.068 0.053
b3 0.064 0.082 0.088 0.075
b4 0.068 0.088 0.095 0.086
b5 0.053 0.075 0.086 0.107

mvmeta documentation built on Dec. 10, 2019, 5:07 p.m.