doBucher: Perform all possible indirect comparisons for a given data...

In RichardBirnie/mautils: Utilities to facilitate an automated workflow for meta-analysis

Description

Perform all possible indirect comparisons for a given data set

Usage

doBucher(comparisons, direct, effectType = "all", continuous,
  backtransf = FALSE)

Arguments

comparisons	A data frame with four columns: StudyName, study, comparator, treatment. Describes the treatment comparisons present in the dataset. study, comparator and treatment must be numbers. For example, study = 4, comparator=1, treatment=2 represents the comparison of treatment 2 vs treatment 1 in study 4
direct	A data frame containing the results of direct head-to-head meta-analysis for the treatment comparisons of interest if only one study is available for a given comparison then the result of that study should be used. This data frame can be created by using doDirectMeta and extractDirectRes in that order
effectType	Character string indicating what type of results are required. Default is 'all' which will return both fixed effect and random effect results. Alternatives are 'Fixed' or 'Random' (Case sensitive) if only one set of results is required
continuous	Logical (TRUE/FALSE) indicating whether the effect measure is continuous (mean difference) or a ratio measure (odds ratio, hazard ratio etc)
backtransf	Logical indicating whether the results should be exponentiated or not. If abTE and cbTE are on the log scale (e.g. log hazard ratio) set this to TRUE to return the exponentiated results (e.g. hazard ratio). If TRUE this will return both the log estimates and the exponentiated estimates

Details

This function performs indirect meta-analysis for all possible comparisons in a given data set. This function takes a set of treatment comparisons from one or more studies and identifies all possible indirect comparisons where two treatments can be connected via a common comparator. If there is more than one way to connect two treatments then all possible variations are calculated. This function calls bucher internally to calculate the treatment effects

The inputs for this function will usually be the results from direct meta-analysis for a given set of treatments. The recommended workflow is to use doDirectMeta to perform head to head meta-analysis for a given set of treatments, extract the results as a data frame using extractDirectRes then use that data frame to provide the inputs for this function.

@return A data frame with the following columns:

• Intervention The name of the intervention

• Comparator The name of the comparator

• Common The name of the common treatment linking intervention and comparator

• Effect The type of effect measure. Takes the value of the effect argument

• Model The type of model. Fixed effect or Random Effects.

• log.TE.ind The treatment effect on log scale, e.g. log OR

• log.lower.ind, log.upper.ind The upper and lower 95% confidence intervals for the log treatment effect

• se.log.TE.ind The standard error for the log treatment effect

• TE.ind, lower.ind, upper.ind The treatment effect with lower and upper confidence intervals backtransformed to a linear scale

• n.studies The number of studies included in the analysis

• Studies The names of the studies included in the analysis

See Also

bucher, doDirectMeta, extractDirectRes

RichardBirnie/mautils documentation built on July 12, 2019, 8:56 p.m.