netmeta (version 3.6-0)

netcomparison: Calculate comparison effects of two arbitrary complex interventions in component network meta-analysis

Description

Calculate comparison effects of two arbitrary complex interventions (i.e., combinations of several components) in component network meta-analysis.

Usage

netcomparison(
  x,
  treat1,
  treat2,
  common = x$common,
  random = x$random,
  level = x$level.ma,
  nchar.comps = x$nchar.comps,
  backtransf = x$backtransf,
  warn.deprecated = gs("warn.deprecated"),
  ...
)
# S3 method for netcomparison
print(
  x,
  common = x$common,
  random = x$random,
  backtransf = x$backtransf,
  nchar.comps = x$nchar.comps,
  digits = gs("digits"),
  digits.stat = gs("digits.stat"),
  digits.pval = gs("digits.pval"),
  scientific.pval = gs("scientific.pval"),
  zero.pval = gs("zero.pval"),
  JAMA.pval = gs("JAMA.pval"),
  big.mark = gs("big.mark"),
  legend = gs("legend"),
  warn.deprecated = gs("warn.deprecated"),
  ...
)

Value

A list is returned by the function netcomparison with the following elements:

comparison

Comparison.

TE.common, TE.random

A vector of comparison effects (common and random effects model).

seTE.common, seTE.random

A vector with corresponding standard errors (common and random effects model).

lower.common, lower.random

A vector with lower confidence limits for comparisons (common and random effects model).

upper.common, upper.random

A vector with upper confidence limits for comparisons (common and random effects model).

statistic.common, statistic.random

A vector with z-values for the overall effect of comparisons (common and random effects model).

pval.common, pval.random

A vector with p-values for the overall effect of comparisons (common and random effects model).

trts

Treatments included in comparisons.

comps

Components included in comparisons.

treat1, treat2

A defined above.

common, random

A defined above.

level, nchar.comps, backtransf, x

A defined above.

B.matrix

B matrix.

C.matrix

C matrix.

Arguments

x

An object of class netcomb or netcomparison (print function).

treat1

A character vector defining the first complex intervention(s).

treat2

A character vector defining the second complex intervention(s).

common

A logical indicating whether results for common effects model should be conducted.

random

A logical indicating whether results for random effects model should be conducted.

level

The level used to calculate confidence intervals for combinations of components.

nchar.comps

A numeric defining the minimum number of characters used to create unique names for components (see Details).

backtransf

A logical indicating whether printed results should be back transformed. If backtransf=TRUE, results for sm="OR" are printed as odds ratios rather than log odds ratios.

warn.deprecated

A logical indicating whether warnings should be printed if deprecated arguments are used.

...

Additional arguments (to catch deprecated arguments).

digits

Minimal number of significant digits, see print.default.

digits.stat

Minimal number of significant digits for z-value of test for overall effect, see print.default.

digits.pval

Minimal number of significant digits for p-values, see print.default.

scientific.pval

A logical specifying whether p-values should be printed in scientific notation, e.g., 1.2345e-01 instead of 0.12345.

zero.pval

A logical specifying whether p-values should be printed with a leading zero.

JAMA.pval

A logical specifying whether p-values for test of combination effect should be printed according to JAMA reporting standards.

big.mark

A character used as thousands separator.

legend

A logical indicating whether a legend should be printed.

Author

Guido Schwarzer

Details

R functions netcomb and discomb calculate effects for individual components and complex interventions present in the component network meta-analysis (CNMA). This function can be used to calculate the effect for comparisons of two arbitrary complex interventions defined by arguments treat1 and treat2.

All complex interventions occuring in the network are considered for the first complex intervention if argument treat1 is missing. The reference group defined in the (C)NMA is used as second complex intervention if argument treat2 is missing. The first complex intervention in the (C)NMA is used if the reference group is not defined.

The following matrices are needed to calculate comparison effects of arbitrary complex interventions, (Rücker et al., 2020, Section 3.2):

• B matrix describing how comparisons are composed by complex intervetions,

• C matrix describing how the complex interventions are composed by the components.

Internally, both matrices are constructed based on arguments x, treat1 and treat2.

By default, component names are not abbreviated in printouts. However, in order to get more concise printouts, argument nchar.comps can be used to define the minimum number of characters for abbreviated component names (see abbreviate, argument minlength). R function treats is utilised internally to create abbreviated component names.

References

Rücker G, Petropoulou M, Schwarzer G (2020): Network meta-analysis of multicomponent interventions. Biometrical Journal, 62, 808--21

See Also

netcomb, discomb, netcomplex

Examples

# Only consider studies including Face-to-face PST (to reduce
# runtime of example)
#
face <- subset(Linde2016, id %in% c(16, 24, 49, 118))

# Use pairwise() to transform data to comparison-based format
#
pw <- pairwise(treat = int,
  event = resp, n = n, studlab = paste(author, year),
  data = face, reference = "plac", sm = "OR")

# Conduct random effects network meta-analysis
#
nma1 <- netmeta(pw, common = FALSE)

# Additive model for treatment components (with placebo as inactive
# treatment)
#
cnma1 <- netcomb(nma1, inactive = "placebo")

# Result for comparison Face-to-face PST vs TCA
netcomparison(cnma1, "Face-to-face PST", "TCA", nchar.comps = 4)
netcomparison(cnma1, "F", "T", nchar.comps = 4)

# Result for comparison Face-to-face PST vs TCA + Placebo
netcomparison(cnma1, "Face-to-face PST", "TCA + Plac", nchar.comps = 4)

# \donttest{
# Forest plots
#
t1 <- c("F + TCA", "F + Plac", "SSRI + Plac + TCA")
t2 <- c("UC", "Plac", "UC")
#
netcomparison(cnma1, t1, t2)
#
forest(netcomparison(cnma1, t1, t2))
forest(netcomparison(cnma1, t1, t2), nchar.comps = 4)
forest(netcomparison(cnma1, c("F", "TCA"), "UC"), nchar.comps = 4)

# Artificial example
#
treat1 <- rep("A", 3)
treat2 <- c("B+C", "A+C", "C+D")
TE <- c(0, 1, 0)
seTE <- rep(1, 3)
# Conduct (C)NMA
nma2 <- netmeta(TE, seTE, treat1, treat2, random = FALSE)
cnma2 <- netcomb(nma2)

# Result for comparison A vs B + D
netcomparison(cnma2, "A", "B + D")
# Same results
netcomparison(cnma2, "A", "B+D")
netcomparison(cnma2, "A", "D+B")
netcomparison(cnma2, "a", "d+b")

# Generated B matrix
netcomparison(cnma2, "A", "B + D")$C.matrix
# Generated B matrix
netcomparison(cnma2, "A", "B + D")$B.matrix
# }