dat.nielweise2008.RdResults from 18 studies comparing the risk of catheter-related bloodstream infection when using anti-infective-treated versus standard catheters for total parenteral nutrition or chemotherapy.
dat.nielweise2008The data frame contains the following columns:
| study | numeric | study number |
| authors | character | study authors |
| year | numeric | publication year |
| x1i | numeric | number of CRBSIs in patients receiving an anti-infective catheter |
| t1i | numeric | total number of catheter days for patients receiving an anti-infective catheter |
| x2i | numeric | number of CRBSIs in patients receiving a standard catheter |
| t2i | numeric | total number of catheter days for patients receiving a standard catheter |
The use of a central venous catheter may lead to a catheter-related bloodstream infection (CRBSI), which in turn increases the risk of morbidity and mortality. Anti-infective-treated catheters have been developed that are meant to reduce the risk of CRBSIs. Niel-Weise et al. (2008) conducted a meta-analysis of studies comparing infection risk when using anti-infective-treated versus standard catheters for total parenteral nutrition or chemotherapy. The results from 9 such studies are included in this dataset.
The dataset was used in the article by Stijnen et al. (2010) to illustrate various generalized linear mixed-effects models for the meta-analysis of incidence rates and incidence rate ratios (see ‘References’).
Niel-Weise, B. S., Stijnen, T., & van den Broek, P. J. (2008). Anti-infective-treated central venous catheters for total parenteral nutrition or chemotherapy: A systematic review. Journal of Hospital Infection, 69(2), 114–123. https://doi.org/10.1016/j.jhin.2008.02.020
Stijnen, T., Hamza, T. H., & Ozdemir, P. (2010). Random effects meta-analysis of event outcome in the framework of the generalized linear mixed model with applications in sparse data. Statistics in Medicine, 29(29), 3046–3067. https://doi.org/10.1002/sim.4040
medicine, incidence rates, generalized linear models
### copy data into 'dat' and examine data
dat <- dat.nielweise2008
dat
#> study authors year x1i t1i x2i t2i
#> 1 1 Bong et al. 2003 7 1344 11 1988
#> 2 2 Ciresi et al. 1996 8 1600 8 1461
#> 3 3 Hanna et al. 2004 3 12012 14 10962
#> 4 4 Harter et al. 2002 6 1536 10 1503
#> 5 5 Jaeger et al. 2001 1 370 1 483
#> 6 6 Jaeger et al. 2005 1 729 8 913
#> 7 7 Logghe et al. 1997 17 6760 15 6840
#> 8 8 Ostendorf et al. 2005 3 1107 7 1015
#> 9 9 Pemberton et al. 1996 2 320 3 440
### load metafor package
library(metafor)
### standard (inverse-variance) random-effects model
res <- rma(measure="IRR", x1i=x1i, t1i=t1i, x2i=x2i, t2i=t2i, data=dat)
print(res, digits=3)
#>
#> Random-Effects Model (k = 9; tau^2 estimator: REML)
#>
#> tau^2 (estimated amount of total heterogeneity): 0.094 (SE = 0.212)
#> tau (square root of estimated tau^2 value): 0.306
#> I^2 (total heterogeneity / total variability): 20.88%
#> H^2 (total variability / sampling variability): 1.26
#>
#> Test for Heterogeneity:
#> Q(df = 8) = 9.698, p-val = 0.287
#>
#> Model Results:
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.396 0.227 -1.747 0.081 -0.841 0.048 .
#>
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
predict(res, transf=exp, digits=2)
#>
#> pred ci.lb ci.ub pi.lb pi.ub
#> 0.67 0.43 1.05 0.32 1.42
#>
### random-effects conditional Poisson model
res <- rma.glmm(measure="IRR", x1i=x1i, t1i=t1i, x2i=x2i, t2i=t2i, data=dat, model="CM.EL")
print(res, digits=3)
#>
#> Random-Effects Model (k = 9; tau^2 estimator: ML)
#> Model Type: Conditional Model with Exact Likelihood
#>
#> tau^2 (estimated amount of total heterogeneity): 0.123
#> tau (square root of estimated tau^2 value): 0.350
#> I^2 (total heterogeneity / total variability): 25.68%
#> H^2 (total variability / sampling variability): 1.35
#>
#> Tests for Heterogeneity:
#> Wld(df = 8) = 9.698, p-val = 0.287
#> LRT(df = 8) = 11.602, p-val = 0.170
#>
#> Model Results:
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.476 0.238 -2.004 0.045 -0.942 -0.010 *
#>
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
predict(res, transf=exp, digits=2)
#>
#> pred ci.lb ci.ub pi.lb pi.ub
#> 0.62 0.39 0.99 0.27 1.42
#>