rbmiUtils (version 0.3.0)

summarise_missingness: Summarise Missing Data Patterns

Description

Tabulates missing outcome data by visit and treatment group, and classifies each subject's missing data pattern as complete, monotone, or intermittent.

Usage

summarise_missingness(data, vars)

Value

A list with three components:

by_visit

A tibble with columns: visit, group, n, n_miss, pct_miss

patterns

A tibble with columns: subjid, group, pattern ("complete", "monotone", or "intermittent"), dropout_visit (NA if not monotone)

summary

A tibble with columns: group, n_subjects, n_complete, n_monotone, n_intermittent

Arguments

data

A data.frame containing the analysis dataset with one row per subject-visit combination.

vars

A vars object as created by rbmi::set_vars().

See Also

• rbmi::draws() for imputation after reviewing missingness patterns

• validate_data() to check data before imputation

• prepare_data_ice() to create intercurrent event data from flags

Examples

library(rbmi)

dat <- data.frame(
  USUBJID = factor(rep(c("S1", "S2", "S3", "S4"), each = 3)),
  AVISIT = factor(rep(c("Week 4", "Week 8", "Week 12"), 4),
                  levels = c("Week 4", "Week 8", "Week 12")),
  TRT = factor(rep(c("Placebo", "Drug A"), each = 6)),
  CHG = c(1, 2, 3, 1, NA, NA, 1, 2, NA, 1, NA, 2)
)

vars <- set_vars(
  subjid = "USUBJID",
  visit = "AVISIT",
  group = "TRT",
  outcome = "CHG"
)

result <- summarise_missingness(dat, vars)
print(result$by_visit)
print(result$patterns)
print(result$summary)