linCC: Computes Lin's concordance correlation between 2 distinct...

In agaye/ceeComp: Assessing agreement between cost-effectiveness estimates from differentes sources

Description

The function takes 2 datasets and a vector of wilingness-to-pay values. The dataset MUST have 4 columns ID, RESPONSE, COST and TREATMENT which hold respectively unique subject identifiers, treatment outcome, treatment cost and 2 treatment arms names. If no values are for wilingness-to-pay, 101 default values ranging from 0 to 500,000 are used to estimate the Lins's concordance correlation between the 2 datasets.

Usage

linCC(data1 = NULL, data2 = NULL, will2pay = NULL, ccThreshold = 0.4,
  CI = 0.95, treatResponse = "beneficial")

Arguments

data1	a dataset with 4 columns holding respectively, ID, treatment outcome, treatment cost and treatment arm.
data2	a dataset with 4 columns holding respectively, ID, treatment outcome, treatment cost and treatment arm.
will2pay	a numeric vector of willingness-to-pay thresholds.
ccThreshold	concordance correlation cut-off, default value is 0.40.
CI	confidence interval.
treatResponse	a character, default is beneficial i.e. the treatment resulted in beneficial response; otherwise harmful, the treatement resulted in harmful outcome.

Details

to be written

Value

a list which holds the below items: cccNB Lin's concordance correlation between net benefits. secccNB standard error of the concordance correlation. lclcccNB lower limit of the concordance correlation. uclcccNB upper limit of the concordance correlation. pvaluecccNB1 pvalue of the concordance correlation for the 1st dataset pvaluecccNB2 pvalue of the concordance correlation for the 2nd dataset

Author(s)

Amadou Gaye & Felix Achana

Examples

{

# load examples datasets
data(dataset1); data(dataset2)

# compute concordance correlation using the default willingness-to-pay thresholds
results <- linCC(data1=dataset1, data2=dataset2, ccThreshold=0.40)

}

agaye/ceeComp documentation built on May 10, 2019, 7:32 a.m.