kpi_fn_: KPI summary functions

In CTU-Bern/kpitools: Tools for creating key performance indicator reports for clinical trial

KPI summary functions

Description

These functions are not intended to be run as they are. They are intended to be passed as arguments to the kpi or kpi_calc functions. They summarize the data in the appropriate manner for the type of KPI. For example, the kpi_fn_prop counts the number of cases and total number of observations then calculates a proportion. kpi_fn_median simply calculates the median of the observations.

Usage

kpi_fn_n(.data)

kpi_fn_prop(.data)

kpi_fn_perc(.data)

kpi_fn_median(.data)

kpi_fn_mean(.data)

kpi_fn_iqr(.data)

kpi_fn_min(.data)

kpi_fn_max(.data)

kpi_fn_missing(.data)

Arguments

.data

data frame

Details

Functions should accept a dataframe with a var variable and return a dataframe with stat (other variables are optional, although an N variable allows for compatibility with downstream functions). All provided functions return stat, n_nonmiss and N, with some also returning n.

See the examples passing custom functions.

Examples

# mean
kpi(mtcars, "mpg", kpi_fn = kpi_fn_mean)
# median
kpi(mtcars, "mpg", kpi_fn = kpi_fn_median)
# interquartile range
kpi(mtcars, "mpg", kpi_fn = kpi_fn_iqr)
# minimum
kpi(mtcars, "mpg", kpi_fn = kpi_fn_min)
# maximum
kpi(mtcars, "mpg", kpi_fn = kpi_fn_max)
# proportion
kpi(mtcars, "am", kpi_fn = kpi_fn_prop)
# percentage
kpi(mtcars, "am", kpi_fn = kpi_fn_perc)
# number/sum
kpi(mtcars, "am", kpi_fn = kpi_fn_n)

CTU-Bern/kpitools documentation built on Aug. 19, 2024, 4:58 a.m.