heatplot.crossnma function

Heat Plot

Produces a heat plot that contain point estimates of relative effects for all possible pairs of treatments along with credible intervals obtained with the quantile method.

## S3 method for class 'crossnma'
heatplot(
  x,
  median = TRUE,
  backtransf = x$model$backtransf,
  seq = NULL,
  low.colour = "red",
  mid.colour = "white",
  high.colour = "springgreen4",
  cov1.value = NULL,
  cov2.value = NULL,
  cov3.value = NULL,
  size = 6,
  size.trt = 20,
  size.axis = 12,
  digits = gs("digits.forest"),
  exp = backtransf,
  ...
)

Arguments

• x: An object created with crossnma.

• median: A logical indicating whether to use the median (default) or mean to measure relative treatment effects.

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

• seq: A vector of treatment names (character) representing the order in which to display these treatments.

• low.colour: A string indicating the colour of low relative treatment effects for the heat plot (e.g odds ratio of ~0.5)

• mid.colour: A string indicating the colour of null relative treatment effects for the heat plot (e.g odds ratio of ~1.0).

• high.colour: A string indicating the colour of high relative treatment effects for the heat plot (e.g odds ratio of ~2.0).

• cov1.value: The participant covariate value of cov1

  for which to report the results. Must be specified for network meta-regression and when individual participant dataset is used in the analysis. For dichotomous covariates, a character of the level (used in the data) should be indicated.

• cov2.value: The participant covariate value of cov2

  for which to report the results. Must be specified for network meta-regression and when individual participant dataset is used in the analysis. For dichotomous covariates, a character of the level (used in the data) should be indicated.

• cov3.value: The participant covariate value of cov3

  for which to report the results. Must be specified for network meta-regression and when individual participant dataset is used in the analysis. For dichotomous covariates, a character of the level (used in the data) should be indicated.

• size: The size of cell entries with the relative treatment effect and 95% credible intervals.

• size.trt: The size of treatment names placed on the top and left of the plot.

• size.axis: The size of labels on the top and left of the plot

• digits: The number of digits to be used when displaying the results.

• exp: Deprecated argument (replaced by backtransf).

• ...: Additional arguments (ignored at the moment).

Returns

League heat plot, where a color scale is used to represent the values of relative treatment effects.

Examples

## Not run:

# We conduct a network meta-analysis assuming a random-effects
# model. The data comes from randomized-controlled trials and
# non-randomized studies (combined naively)
head(ipddata) # participant-level data
stddata # study-level data

# Create a JAGS model
mod <- crossnma.model(treat, id, relapse, n, design,
  prt.data = ipddata, std.data = stddata,
  reference = "A", trt.effect = "random", method.bias = "naive")

# Fit JAGS model
set.seed(1909)
fit <- crossnma(mod)

# Create a heat plot
heatplot(fit)
## End(Not run)

See Also

crossnma

Author(s)

Tasnim Hamza hamza.a.tasnim@gmail.com