Getting Started

kstools provides practical helpers for clinical and data-science workflows: descriptive summaries, pooled estimators from multiple imputations, ROC comparison, SDTM/SUPP operations, and data-loading tools.

Installation

# local source install from package root
install.packages(".", repos = NULL, type = "source")

# from GitHub
remotes::install_github("crow16384/kstools")

Descriptive Summary

library(kstools)

set.seed(42)
adlb <- data.frame(
  trt = rep(c("Drug A", "Placebo"), each = 50),
  visit = rep(c("Week 4", "Week 12"), each = 25, times = 2),
  aval = c(rnorm(50, 28, 8), rnorm(50, 32, 9))
)

ks_summary(adlb, aval, trt, visit)
#> # A tibble: 4 × 12
#>   trt     visit       n  mean    sd median    q1    q3   min   max    cv NUMDEC
#>   <chr>   <chr>   <int> <dbl> <dbl>  <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>  <int>
#> 1 Drug A  Week 12    25  25.9  7.58   25.1  21.7  31.7  8.69  39.6  29.2      4
#> 2 Drug A  Week 4     25  29.5 10.5    27.5  25.7  38.4  6.75  46.3  35.4      4
#> 3 Placebo Week 12    25  32.5  8.05   32.8  24.3  37.9 18.9   45.6  24.7      4
#> 4 Placebo Week 4     25  33.3  8.74   34.6  28.7  38.1  5.06  46.2  26.3      4

Pooled Estimate Across Imputations

est <- c(0.345, 0.303, 0.298, 0.378)
err <- c(0.056, 0.101, 0.099, 0.034)

mianalyze(est, err)

Variance Information (4 Imputations)

Between	Within	Total	DF
0.001426	0.006074	0.007856	58.2727

Missing Data Diagnostics

Relative Increase in Variance	Fraction Missing Information	Relative Efficiency
0.293488	0.252131	0.940705

Parameter Estimates (4 Imputations)

Estimate	Std Error	LCL	UCL	DF
0.331	0.088634	0.153597	0.508403	58.2727

Hypothesis Test

Minimum	Maximum	Theta0	Statistic	PValue
0.298	0.378	0	3.734455	0.00043

Compare ROC Areas

set.seed(123)
n <- 120
y <- rbinom(n, 1, 0.45)
df <- data.frame(
  y = y,
  x1 = y + rnorm(n, sd = 0.8),
  x2 = 0.7 * y + rnorm(n, sd = 0.9),
  x3 = 0.5 * y + rnorm(n, sd = 1.0)
)

roc_out <- ks_roc(df, var = c("x1", "x2", "x3"), response = "y")
roc_out$areas
#>   predictor  roc_area  std_error       lcl       ucl
#> 1        x1 0.9043210 0.02660878 0.8521687 0.9564732
#> 2        x2 0.6787318 0.04894766 0.5827961 0.7746674
#> 3        x3 0.5953984 0.05244706 0.4926041 0.6981928
roc_out$contrasts
#>   contrast   estimate  std_error        lcl       ucl chi_square      p_value
#> 1     Row1 0.22558923 0.05958660  0.1088016 0.3423768   14.33308 1.531501e-04
#> 2     Row2 0.30892256 0.05739691  0.1964267 0.4214184   28.96825 7.357444e-08
#> 3     Row3 0.08333333 0.07003717 -0.0539370 0.2206037    1.41573 2.341076e-01

SDTM Supplemental Merge

dm <- data.frame(
  STUDYID = "S1",
  USUBJID = c("01", "02", "03"),
  AGE = c(34L, 52L, 47L),
  stringsAsFactors = FALSE
)

suppdm <- data.frame(
  STUDYID  = "S1",
  USUBJID  = c("01", "02", "03"),
  IDVAR    = "",
  IDVARVAL = "",
  QNAM     = "RACE2",
  QLABEL   = "Race (detail)",
  QVAL     = c("Asian", "White", "Asian"),
  stringsAsFactors = FALSE
)

ks_supp_merge(dm, suppdm)
#>   STUDYID USUBJID AGE RACE2
#> 1      S1      01  34 Asian
#> 2      S1      02  52 White
#> 3      S1      03  47 Asian

Next Steps

See the mianalyze vignette for Rubin’s rules, Barnard-Rubin edf, and SAS PROC MIANALYZE comparisons.
See the clinical-workflows vignette for end-to-end examples.
Explore inst/examples for runnable scripts by function.