dosresmeta (version 1.1)

gl: Greenland and Longnecker method

Description

Approximating covariance matrix according to the method described by Greenland and Longnecker (1992).

Usage

gl(cases, n, logrr, v, study, id, data)

Arguments

cases

The variable (or vector) reporting the number of cases, regardless of the type of study.

n

The variable(or vector) reporting the number of the size (for each exposure) of the study for case-control and cumulative incidence data, the Person-Times (for each exposure) for incidence rate data.

logrr

The variable (or vector) reporting the log relative risks for each exposure

v

The variable (or vector) reporting the variance of published log relative risks. If not already in the dataset, it needs to be calculated by the user.

study

the variable (or vector) or value reporting the type of study. The values for the variable must be "cc", "ir", "ci" for case-control, incidence-rate and cumulative-incidence data respectively.

id

an optional variable (or vector) reporting the id of the studies.

data

an optional dataframe containing the variables needed to approximate the covariance matrix.

Value

• A matrix reporting the pseudo-counts originated from the pubblished data.

Details

No further details required.

References

Greenland, S., Longnecker, M. P. (1992). Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. American journal of epidemiology, 135(11), 1301-1309.

Examples

data(cc_ex)
cc_ex$v<-((cc_ex$logub-cc_ex$loglb)/(2*qnorm(.975)))^2
gl(case,n,logrr,v,study="cc",data=cc_ex)