Treatment Compliance Table

In metalite.sl: Subject-Level Analysis Using 'metalite'

knitr::opts_chunk$set(
  comment = "#>",
  collapse = TRUE,
  out.width = "100%",
  dpi = 150
)

library(metalite)
library(metalite.sl)
library(dplyr)

Create Treatment Compliance Table

The objective of this tutorial is to generate a production-ready Treatment Compliance specification analyses.

This report produces a table that contains a summary of treatment compliance information. The report consists of a treatment compliance category section and a treatment compliance statistics section. To accomplish this using metalite.sl, three essential functions are required:

• prepare_trt_compliance ():this function is a wrapper function of prepare_sl_summary () which prepares data for treatment compliance analysis.

• format_trt_compliance(): prepare analysis outdata with proper format.

• rtf_trt_compliance(): transfer output dataset to RTF table.

An example output:

knitr::include_graphics("pdf/treatment0compliance.pdf")

Example data

Within metalite.sl, we utilized the ADSL datasets from the metalite package to create an illustrative dataset. The metadata structure remains consistent across all analysis examples within metalite.sl. To calculate treatment compliance percent and treatment compliance range, we utilized adex dataset. Additional information can be accessed on the metalite package website.

Build a metadata

adsl <- r2rtf::r2rtf_adsl

adex <- metalite.ae::metalite_ae_adex

adex1 <- adex |>
  filter(EXNUMDOS > 0 & !(is.na(AENDY))) |>
  group_by(USUBJID) |>
  slice(n()) |>
  select(USUBJID, AENDY) |>
  rename(ADURN = AENDY)

adsl <- merge(adsl, adex1, by = "USUBJID")

adsl <- adsl |>
  mutate(
    TRTDUR = as.numeric(TRTDUR),
    ADURN = as.numeric(ADURN),
    CMPLPCT = round((ADURN / TRTDUR) * 100, 2)
  )

adsl <- adsl |>
  mutate(
    CMPLRNG = case_when(
      CMPLPCT >= 0 & CMPLPCT <= 20 ~ "0% to <=20%",
      CMPLPCT > 20 & CMPLPCT <= 40 ~ ">20% to <=40%",
      CMPLPCT > 40 & CMPLPCT <= 60 ~ ">40% to <=60%",
      CMPLPCT > 60 & CMPLPCT <= 80 ~ ">60% to <=80%",
      CMPLPCT > 80 ~ ">80%"
    ),
    CMPLRNGN = case_when(
      CMPLPCT >= 0 & CMPLPCT <= 20 ~ 1,
      CMPLPCT > 20 & CMPLPCT <= 40 ~ 2,
      CMPLPCT > 40 & CMPLPCT <= 60 ~ 3,
      CMPLPCT > 60 & CMPLPCT <= 80 ~ 4,
      CMPLPCT > 80 ~ 5
    )
  )

head(adsl)

adsl$TRTA <- adsl$TRT01A
adsl$TRTA <- factor(adsl$TRTA,
  levels = c("Placebo", "Xanomeline Low Dose", "Xanomeline High Dose"),
  labels = c("Placebo", "Low Dose", "High Dose")
)

plan <- plan(
  analysis = "trt_compliance", population = "apat",
  observation = "apat", parameter = "CMPLRNG;CMPLPCT"
)

meta <- meta_adam(
  population = adsl,
  observation = adsl
) |>
  define_plan(plan = plan) |>
  define_population(
    name = "apat",
    group = "TRTA",
    subset = quote(SAFFL == "Y"),
    var = c("USUBJID", "TRTA", "SAFFL", "CMPLPCT", "CMPLRNG")
  ) |>
  metalite::define_parameter(
    name = "CMPLPCT",
    var = "CMPLPCT",
    label = "Treatment Compliance Percent",
  ) |>
  metalite::define_parameter(
    name = "CMPLRNG",
    var = "CMPLRNG",
    label = "Treatment Compliance Range",
  ) |>
  define_analysis(
    name = "trt_compliance",
    title = "Summary of Treatment Compliance",
    label = "treatment compliance table"
  ) |>
  meta_build()

Click to show the output

meta

Analysis preparation

The function prepare_trt_compliance() is written to prepare data for treatment compliance analysis.The function takes four arguments:

meta is metadata object created by metalite and it contains data from ADSL. Analysis, Population, and Parameter arguments are used to subset and process the meta data. They have default values, which rely on the meta data object.

The function assign default value Analysis to trt_compliance, Population to the population value associated with the trt_compliance analysis in meta plan, and parameter to the parameter(s) associated with the trt_compliance analysis in meta$plan.

However, the user can also manually specify the analysis, population, and parameter values when calling the function, if they want to override the default values.

In the body of the function, it calls another function prepare_sl_summary with the same meta, analysis, population, and parameter arguments. prepare_sl_summary takes the meta data, subsets it based on the analysis, population, and parameter values, and then calculates and returns a summary of the relevant data.

The result of prepare_sl_summary is then returned as the result of prepare_trt_compliance.

The resulting output of the function prepare_trt_compliance() comprises a collection of raw datasets for analysis and reporting.

outdata <- prepare_trt_compliance(meta)

Click to show the output

outdata

• parameter: parameter name

outdata$parameter

• n: number of participants in population

outdata$n

The resulting dataset contains frequently used statistics, with variables indexed according to the order specified in outdata$group.

outdata$group

• char_n: number of participants completed vs not completed in each parameter

outdata$char_n

• char_var : name of parameter

outdata$char_var

• char_prop : proportion of subject with treatment compliance

outdata$char_prop

Format output

Once the raw analysis results are obtained, the format_trt_compliance() function can be employed to prepare the outdata,ensuring its compatibility with production-ready RTF tables.

outdata <- outdata |> format_trt_compliance()

Click to show the output

outdata$tbl

Additional statistics

By using the display argument, we can choose specific statistics to include.

tbl <- outdata |> format_trt_compliance(display_stat = c("mean", "sd", "median", "range"), display_col = c("n", "prop", "total"))

Click to show the output

tbl$tbl

RTF tables

The last step is to prepare the RTF table using rtf_trt_compliance.

outdata |>
  format_trt_compliance() |>
  rtf_trt_compliance(
    "Source: [CDISCpilot: adam-adsl]",
    path_outtable = "outtable/treatment0compliance.rtf"
  )

knitr::include_graphics("pdf/treatment0compliance.pdf")

metalite.sl documentation built on April 3, 2025, 8:52 p.m.