set_agd_arm {multinma}

Set up arm-based aggregate data

Description

Set up a network containing arm-based aggregate data (AgD), such as event counts or mean outcomes on each arm. Multiple data sources may be combined once created using combine_network().

Usage

set_agd_arm(
  data,
  study,
  trt,
  y = NULL,
  se = NULL,
  r = NULL,
  n = NULL,
  E = NULL,
  sample_size = NULL,
  trt_ref = NULL,
  trt_class = NULL
)

Arguments

data

a data frame

study

column of data specifying the studies, coded using integers, strings, or factors

trt

column of data specifying treatments, coded using integers, strings, or factors

y

column of data specifying a continuous outcome

se

column of data specifying the standard error for a continuous outcome

r

column of data specifying a binary or Binomial outcome count

n

column of data specifying Binomial outcome numerator

E

column of data specifying the total time at risk for Poisson outcomes

sample_size

column of data giving the sample size in each arm. Optional, see details.

trt_ref

reference treatment for the network, as a single integer, string, or factor. If not specified, a reasonable well-connected default will be chosen (see details).

trt_class

column of data specifying treatment classes, coded using integers, strings, or factors. By default, no classes are specified.

Details

By default, trt_ref = NULL and a network reference treatment will be chosen that attempts to maximise computational efficiency and stability. If an alternative reference treatment is chosen and the model runs slowly or has low effective sample size (ESS) this may be the cause - try letting the default reference treatment be used instead. Regardless of which treatment is used as the network reference at the model fitting stage, results can be transformed afterwards: see the trt_ref argument of relative_effects() and predict.stan_nma().

The sample_size argument is optional, but when specified:

• Enables automatic centering of predictors (center = TRUE) in nma() when a regression model is given for a network combining IPD and AgD

• Enables production of study-specific relative effects, rank probabilities, etc. for studies in the network when a regression model is given

• Nodes in plot.nma_data() may be weighted by sample size

If a Binomial outcome is specified and sample_size is omitted, n will be used as the sample size by default. If a Multinomial outcome is specified and sample_size is omitted, the sample size will be determined automatically from the supplied counts by default.

All arguments specifying columns of data accept the following:

• A column name as a character string, e.g. study = "studyc"

• A bare column name, e.g. study = studyc

• dplyr::mutate() style semantics for inline variable transformations, e.g. study = paste(author, year)

Value

An object of class nma_data

See Also

set_ipd() for individual patient data, set_agd_contrast() for contrast-based aggregate data, and combine_network() for combining several data sources in one network.

print.nma_data() for the print method displaying details of the network, and plot.nma_data() for network plots.

Examples

# Set up network of smoking cessation data
head(smoking)

smk_net <- set_agd_arm(smoking,
                       study = studyn,
                       trt = trtc,
                       r = r,
                       n = n,
                       trt_ref = "No intervention")

# Print details
smk_net


# Plot network
plot(smk_net)