In n8thangreen/CEdecisiontree: Cost-effectiveness decision tree analysis

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

Introduction

Although the CEdecisiontree package allows us to do calculations using different values separately, the main purpose is to carry out cost-effectiveness analyses. This is performed calling the dectree() function for cost and health inputs using the wrapper run_cedectree().

Example

library(CEdecisiontree)
library(purrr)
library(tibble)

For a single outcome type

tree_dat <- 
  tribble(
    ~from, ~to, ~vals, ~prob, 
    1,  2,   10,   0.7, 
    1,  3,   NA,   0.3, 
    2,  4,  100,   0.1, 
    2,  5,   NA,   0.9, 
    3,  6,  100,   0.9, 
    3,  7,   NA,   0.1)

The function dectree() requires the dataframe defining the tree above. Optionally, it can also take PSA distributions on probabilities and values, a PSA sample size n, and a list of groups of nodes state_list. It then returns expected values for point estimates, and for PSA if supplied. Also, pathway joint probabilities are returned if states are provided.

dectree(tree_dat,
        state_list = list(all = c(4,5,6,7)))

For cost-effectiveness analysis including PSA

# run_cedectree(tree_dat)

n8thangreen/CEdecisiontree documentation built on Sept. 13, 2022, 5:25 a.m.