multinma (version 0.1.3)

nma_data-class: The nma_data class

Description

The nma_data class contains the data for a NMA in a standard format, created using the functions set_ipd(), set_agd_arm(), set_agd_contrast(), or combine_network(). The sub-class mlnmr_data is created by the function add_integration(), and further contains numerical integration points for the aggregate data.

Arguments

Details

Objects of class nma_data have the following components:

agd_arm

data from studies with aggregate data (arm format)

agd_contrast

data from studies with aggregate data (contrast format)

ipd

data from studies with individual patient data

treatments

treatment coding factor for entire network

classes

treatment class coding factor (same length as treatments for entire network)

studies

study coding factor for entire network

outcome

outcome type for each data source, named list

The agd_arm, agd_contrast, and ipd components are tibbles with the following columns:

.study

study (as factor)

.trt

treatment (as factor)

.trtclass

treatment class (as factor), if specified

.y

continuous outcome

.se

standard error (continuous)

.r

event count (discrete)

.n

event count denominator (discrete, agd_arm only)

.E

time at risk (discrete)

.surv

event/censoring time, of type Surv (time-to-event)

.sample_size

sample size (agd_* only)

...

other columns (typically covariates) from the original data frame

Objects of class mlnmr_data additionally have components:

n_int

number of numerical integration points

int_names

names of covariates with numerical integration points

int_cor

correlation matrix for covariates used to generate numerical integration points

The agd_arm and agd_contrast tibbles have additional list columns with prefix .int_, one for each covariate, which contain the numerical integration points nested as length-n_int vectors within each row.

See Also

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