jarbes (version 2.5.0)

tyge_2011: Mortality data from the tigecycline meta-analysis (Curcio & Verde, 2011)

Description

Mortality data from the tigecycline meta-analysis (Curcio & Verde, 2011)

Format

A data frame with 15 randomized clinical trials and 5 variables:

author_year

Character string identifying the study.

event_e

Number of deaths in the tigecycline treatment group.

n_e

Total number of patients in the tigecycline group.

event_c

Number of deaths in the comparator treatment group.

n_c

Total number of patients in the comparator group.

RR

Relative risk calculated using the 0.5 correction when event is 0.

TE

Treatment effect as log(RR).

seTE

Standard error of the TE.

Details

The dataset is typically analyzed using the **log relative risk**

$$TE_i = \log(RR_i)$$

where

$$RR_i = \frac{event_e / n_e}{event_c / n_c}$$

and the approximate sampling variance

$$seTE_i^2 = \frac{1}{event_e} - \frac{1}{n_e} + \frac{1}{event_c} - \frac{1}{n_c}$$

The data were used to illustrate how **profile likelihood methods can evaluate the support for different values of the between-study heterogeneity parameter** in a random-effects meta-analysis.

References

Curcio D., Verde P.E. (2011). Comment on: Efficacy and safety of tigecycline: a systematic review and meta-analysis. Journal of Antimicrobial Chemotherapy.

Examples

data(tyge_2011)

# compute log relative risks
TE <- log((tyge_2011$event_e / tyge_2011$n_e) /
          (tyge_2011$event_c / tyge_2011$n_c))

seTE <- sqrt(
  1 / tyge_2011$event_e - 1 / tyge_2011$n_e +
  1 / tyge_2011$event_c - 1 / tyge_2011$n_c
)

head(data.frame(TE, seTE))