Many clinical tables require statistics presented on a table with
certain decimal precision. In some situations the used decimal precision
is different for different parameters and/or variables included in the
table. In the rtables and nest framework
formats can be specified, typically in the format or
.formats argument of the used afun or
analyze call.
For more information on tabulation, formats and formatting functions, see the vignettes “Tabulation” and “Formatting Functions”.
The purpose in this vignette is to provide some examples on how to
utilize the new junco functions to facilitate adjustment of
decimal precision.
The following table summarizes the 3 J&J conventions for mean, SD/SE, median and range that may be used. In the table below, “X” indicates the number of decimal places to which the raw values are captured or, in the case of derived data, the number of decimal places to which the derived variable is to be expressed.
| Default Conventions | Alternative 1 | Alternative 2 | |
|---|---|---|---|
| Individual (raw/derived) values | X | X | X |
| Mean | X+1 | X+1 | X |
| SD or SE | X+2 | X+1 | X |
| Median | X+1 | X+1 | X |
| Min, max | X, X | X, X | X, X |
These rules have been used to set up a d-style specification as a
string of type “xx.dx”. In the conversion process, the integer value
d will be replaced by d times x,
e.g. with d = 1 xx.dx is converted into the format
specification string xx.xx.
Below is an example on how to set up a d-style based specification manually.
junco_def_d <- c(
"mean" = "xx.dx",
"mean_sd" = "xx.dx (xx.dxx)",
"mean_se" = "xx.dx (xx.dxx)",
"range" = "(xx.d, xx.d)",
"count_fraction" = jjcsformat_count_fraction
)The 3 following d-style based format specifications are available in junco:
junco_def_d_all, used for setup according to default
J&J precision rulesjunco_def_d_alt1, used for setup according to J&J
precision rules alternative 1junco_def_d_alt2, used for setup according to J&J
precision rules alternative 2All 3 contain formatting defaults for all statistics defined in
tern/junco (a total of 140 statistics), also
these that do not follow d-based specification setup (such as
count_fraction, ztest_pval, …). For about a
third of all statistics (53 out of 140) the specifications in
junco_def_d_all are defined with the d-style setup. See
further in this vignette a note on tools get_fmt_details
and compare_fmt_specs that can be used to investigate
format specifications setups.
d_style <- vapply(junco_def_d_all, FUN = function(fmt){
!is.function(fmt) && is.character(fmt) && grepl("xx.d", fmt, fixed = TRUE)},
FUN.VALUE = logical(1))
non_d_style <- junco_def_d_all[!d_style]
d_style <- junco_def_d_all[d_style]
names(d_style)
#> [1] "adj_mean_se" "adj_mean_ci" "adj_mean_est_ci"
#> [4] "cv" "diff" "diff_ci"
#> [7] "diff_est_ci" "diff_mean_se" "diff_mean_ci"
#> [10] "diff_mean_est_ci" "diff_means_est" "diff_means_se"
#> [13] "diff_means_est_se" "diff_means_ci" "diff_means_est_ci"
#> [16] "geom_sd" "geom_se" "geom_mean_sd"
#> [19] "geom_mean_se" "geom_cv" "geom_mean"
#> [22] "geom_mean_ci_3d" "iqr" "quantiles_upper"
#> [25] "quantiles_lower" "lsmean" "lsmean_ci"
#> [28] "lsmean_diff" "lsmean_diff_with_ci" "lsmean_diff_ci"
#> [31] "lsmean_se" "mad" "mean"
#> [34] "mean_sd" "mean_se" "mean_ci_3d"
#> [37] "mean_diff_with_ci" "median" "median_ci"
#> [40] "median_ci_3d" "median_range" "quantiles"
#> [43] "range" "sd" "se"
#> [46] "geom_mean_ci" "max" "mean_ci"
#> [49] "mean_sdi" "mean_sei" "min"
#> [52] "range_censor" "range_event"
d_style[["adj_mean_se"]]
#> [1] "xx.dx (xx.dxx)"The default d-style based input specification is
junco_def_d_all, and when utilizing these specs with
d = 1 it would result in utilizing the junco
defaults defined in junco_default_formats. One exception is
the tern statistic lr_stat_df from
method_group coxph_pairwise, for which the default format
is not a valid formatter format. This is out of scope for
current vignette.
fmt_1 <- fmt_spec_single_d(d = 1, fmt_d_def = junco_def_d_all)
fmt_1[["lr_stat_df"]] <- NULL
# get_fmt_details tool applied to a format specification vector
fmt_details <- get_fmt_details(fmt_1, recursive = FALSE, as_tibble = TRUE)
fmt_details[fmt_details$stat == "adj_mean_se", c("stat", "fun_fact", "str", "str_formatters")]
#> # A tibble: 1 × 4
#> stat fun_fact str str_formatters
#> <chr> <chr> <chr> <lgl>
#> 1 adj_mean_se "" xx.xx (xx.xxx) TRUE
# to compare against junco_default_formats, except lr_stat_df
junco_default_formats_adj <- junco_default_formats
junco_default_formats_adj[["lr_stat_df"]] <- NULL
# compare_fmt_specs tool to compare 2 format specification vectors
comp <- compare_fmt_specs(fmt_1, junco_default_formats)
#> All common stats specs identical
#> More details are available in created object.The tools get_fmt_details and
compare_fmt_specs used in the above code snippet are useful
for understanding how the formatting specifications are constructed.
These tools are out of scope of current vignette, users can refer to the
help pages for basic information on these.
| Function | Core Purpose | Covered in Example |
|---|---|---|
fmt_spec_single_d |
Convert a d-style based format specification into
formatters format specification with 1 specific value of
d |
Example BDS-type 2 |
fmt_spec_df_d |
Convert a tibble with d-style based format specification into
formatters format specification with value of
d coming from column on input tibble |
Example BDS-type 1 |
fmt_spec_var_d |
Convert a variable based vector with d-style based format
specification into formatters format specification |
Example demographic |
This example demonstrates the standard pattern for a table with multiple variables, each with a specific precision.
set.seed(101)
f2n <- function(x) as.numeric(x) - 1
df <- ex_adsl
df <- df |>
mutate(newvar = rnorm(nrow(df), mean = 50 + 3 * f2n(ARMCD) + 4 * AGE, sd = 20))
# a mix of categorical and continuous variables
vars_all <- c("AGE", "SEX", "BMRKR1", "newvar")
# set all variables to d = 0
vars_d <- rep(0, length(vars_all))
names(vars_d) <- vars_all
# set some variables to a different d value
vars_d["BMRKR1"] <- 2
vars_d["newvar"] <- 1
# conversion of d-style format to valid format specification
vars_fmt <- fmt_spec_var_d(vars_d,
stats_in = NULL,
fmt_d_def = junco_def_d,
fmt_d_in = NULL
)
lyt <- basic_table() |>
split_cols_by("ARMCD") |>
analyze(
vars = vars_all,
afun = tern::a_summary,
# note .formats = "default" in extra_args to afun
extra_args = list(
.stats = c("n", "mean_sd", "range", "count_fraction"),
.formats = "default"
),
# note the usage of format argument to analyze call
format = vars_fmt,
section_div = " "
)
rslt <- build_table(lyt, df)
rslt
#> ARM A ARM B ARM C
#> ————————————————————————————————————————————————————————————————————————
#> AGE
#> n 134 134 132
#> Mean (SD) 33.8 (6.55) 35.4 (7.90) 35.4 (7.72)
#> Min - Max (21, 50) (21, 62) (20, 69)
#>
#> SEX
#> n 134 134 132
#> F 79 (59.0%) 77 (57.5%) 66 (50.0%)
#> M 51 (38.1%) 55 (41.0%) 60 (45.5%)
#> U 3 (2.2%) 2 (1.5%) 4 (3.0%)
#> UNDIFFERENTIATED 1 (0.7%) 0 2 (1.5%)
#>
#> BMRKR1
#> n 134 134 132
#> Mean (SD) 5.970 (3.5541) 5.699 (3.3067) 5.618 (3.4940)
#> Min - Max (0.41, 17.67) (0.65, 14.24) (0.17, 21.39)
#>
#> newvar
#> n 134 134 132
#> Mean (SD) 183.69 (33.496) 195.98 (35.029) 196.31 (34.609)
#> Min - Max (100.7, 277.7) (133.3, 297.4) (121.0, 347.5)# set precision d on a parameter by parameter basis
df_d <- tribble(
~PARAMCD, ~d,
"ALT", 2,
"CRP", 3,
"IGA", 1
)
# convert d-based setup into format specification
# a new column fmt_d is created with format specifications
df_d <- fmt_spec_df_d(df_d,
d_column = "d",
fmt_column = "fmt_d",
stats_in = NULL,
fmt_d_def = junco_def_d_all,
fmt_d_in = NULL
)
df <- ex_adlb |>
filter(AVISIT %in% c("BASELINE", "WEEK 1 DAY 8", "WEEK 2 DAY 15"))
# join with dataframe with format specs
df <- df |>
left_join(df_d)
#> Joining with `by = join_by(PARAMCD)`
lyt <- basic_table() |>
split_cols_by("ARMCD") |>
split_rows_by("PARAMCD", split_fun = drop_split_levels) |>
split_rows_by("AVISIT", split_fun = drop_split_levels) |>
analyze(
vars = "AVAL",
afun = tern::a_summary,
# note .formats = "default" in extra_args to afun
extra_args = list(
.stats = c("n", "mean_se", "range"),
.formats = "default"
),
# note the usage of formats_var argument to analyze call
formats_var = "fmt_d"
)
rslt <- build_table(lyt, df, alt_counts_df = ex_adsl)
rslt
#> ARM A ARM B ARM C
#> ———————————————————————————————————————————————————————————————————————————
#> ALT
#> BASELINE
#> n 134 134 132
#> Mean (SE) 49.572 (0.7189) 50.306 (0.7207) 50.902 (0.6808)
#> Min - Max 23.97, 70.85 26.22, 79.12 27.63, 67.45
#> WEEK 1 DAY 8
#> n 134 134 132
#> Mean (SE) 48.602 (0.6878) 50.442 (0.6866) 51.107 (0.6781)
#> Min - Max 27.71, 64.64 21.68, 67.51 29.75, 71.40
#> WEEK 2 DAY 15
#> n 134 134 132
#> Mean (SE) 49.407 (0.7314) 50.245 (0.7369) 48.494 (0.6253)
#> Min - Max 24.34, 71.06 24.38, 71.07 26.22, 63.35
#> CRP
#> BASELINE
#> n 134 134 132
#> Mean (SE) 48.9467 (0.81520) 50.0803 (0.68029) 50.2079 (0.73192)
#> Min - Max 28.959, 75.232 29.912, 73.354 30.455, 72.251
#> WEEK 1 DAY 8
#> n 134 134 132
#> Mean (SE) 51.8606 (0.69774) 50.2698 (0.76090) 50.2549 (0.81518)
#> Min - Max 28.466, 72.897 26.920, 67.335 27.779, 71.177
#> WEEK 2 DAY 15
#> n 134 134 132
#> Mean (SE) 49.7480 (0.71440) 50.8726 (0.64128) 49.1180 (0.71787)
#> Min - Max 29.015, 69.077 30.697, 67.695 20.630, 70.521
#> IGA
#> BASELINE
#> n 134 134 132
#> Mean (SE) 49.93 (0.643) 49.74 (0.716) 50.30 (0.788)
#> Min - Max 33.7, 65.9 25.8, 71.5 26.0, 70.0
#> WEEK 1 DAY 8
#> n 134 134 132
#> Mean (SE) 49.41 (0.699) 51.12 (0.707) 50.75 (0.732)
#> Min - Max 29.3, 69.3 31.2, 74.5 26.2, 71.0
#> WEEK 2 DAY 15
#> n 134 134 132
#> Mean (SE) 50.73 (0.679) 49.88 (0.721) 49.38 (0.652)
#> Min - Max 33.3, 71.9 27.9, 69.8 29.4, 65.1fmt_2 <- fmt_spec_single_d(d = 2, fmt_d_def = junco_def_d_all)
df <- ex_adlb |>
filter(AVISIT %in% c("BASELINE", "WEEK 1 DAY 8", "WEEK 2 DAY 15"))
lyt <- basic_table() |>
split_cols_by("ARMCD") |>
split_rows_by("PARAMCD", split_fun = drop_split_levels) |>
split_rows_by("AVISIT", split_fun = drop_split_levels) |>
analyze(
vars = "AVAL",
afun = tern::a_summary,
extra_args = list(
.stats = c("n", "mean_se", "range"),
.formats = fmt_2
)
)
rslt <- build_table(lyt, df, alt_counts_df = ex_adsl)
rslt
#> ARM A ARM B ARM C
#> —————————————————————————————————————————————————————————————————————
#> ALT
#> BASELINE
#> n 134 134 132
#> Mean (SE) 49.572 (0.7189) 50.306 (0.7207) 50.902 (0.6808)
#> Min - Max 23.97, 70.85 26.22, 79.12 27.63, 67.45
#> WEEK 1 DAY 8
#> n 134 134 132
#> Mean (SE) 48.602 (0.6878) 50.442 (0.6866) 51.107 (0.6781)
#> Min - Max 27.71, 64.64 21.68, 67.51 29.75, 71.40
#> WEEK 2 DAY 15
#> n 134 134 132
#> Mean (SE) 49.407 (0.7314) 50.245 (0.7369) 48.494 (0.6253)
#> Min - Max 24.34, 71.06 24.38, 71.07 26.22, 63.35
#> CRP
#> BASELINE
#> n 134 134 132
#> Mean (SE) 48.947 (0.8152) 50.080 (0.6803) 50.208 (0.7319)
#> Min - Max 28.96, 75.23 29.91, 73.35 30.45, 72.25
#> WEEK 1 DAY 8
#> n 134 134 132
#> Mean (SE) 51.861 (0.6977) 50.270 (0.7609) 50.255 (0.8152)
#> Min - Max 28.47, 72.90 26.92, 67.34 27.78, 71.18
#> WEEK 2 DAY 15
#> n 134 134 132
#> Mean (SE) 49.748 (0.7144) 50.873 (0.6413) 49.118 (0.7179)
#> Min - Max 29.02, 69.08 30.70, 67.70 20.63, 70.52
#> IGA
#> BASELINE
#> n 134 134 132
#> Mean (SE) 49.927 (0.6430) 49.736 (0.7162) 50.305 (0.7883)
#> Min - Max 33.72, 65.91 25.82, 71.47 26.04, 69.99
#> WEEK 1 DAY 8
#> n 134 134 132
#> Mean (SE) 49.406 (0.6986) 51.116 (0.7073) 50.754 (0.7322)
#> Min - Max 29.33, 69.31 31.19, 74.52 26.15, 70.95
#> WEEK 2 DAY 15
#> n 134 134 132
#> Mean (SE) 50.726 (0.6786) 49.880 (0.7207) 49.377 (0.6518)
#> Min - Max 33.33, 71.90 27.90, 69.82 29.39, 65.05Using the same data as in 2 prior examples, now with a different afun, to present statistics in columns rather than in rows.
df <- ex_adlb |>
filter(AVISIT %in% c("BASELINE", "WEEK 1 DAY 8", "WEEK 2 DAY 15")) |>
mutate(AVISIT = droplevels(AVISIT))
# join with dataframe with format specs
df <- df |>
left_join(df_d)
#> Joining with `by = join_by(PARAMCD)`
mysplitfun <- make_split_fun(
post = list(stats_in_cols_setup())
)
lyt <- basic_table() |>
split_cols_by_multivar(
c("AVAL", "BASE", "CHG"),
varlabels = c("Visit Values", " ", "Change from Baseline")
) |>
split_cols_by("STUDYID", split_fun = mysplitfun) |>
split_rows_by("PARAM", split_fun = drop_split_levels) |>
split_rows_by("ARMCD", labels_var = "ARM") |>
analyze("AVISIT", afun = column_stats, extra_args = list(
exclude_visits = toupper(c("Baseline", "Screening")),
.formats = "default",
formats_var = "fmt_d"
))
rslt <- build_table(lyt, df, round_type = "sas")
rslt
#> Visit Values Change from Baseline
#> N Mean SD SE Med Min Max Base Mean (SD) N Mean SD SE Med Min Max
#> ——————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————————
#> Alanine Aminotransferase Measurement
#> A: Drug X
#> BASELINE 134 49.572 8.3223 0.7189 49.571 23.97 70.85
#> WEEK 1 DAY 8 134 48.602 7.9620 0.6878 48.418 27.71 64.64 49.572 (8.3223) 134 -0.970 11.7006 1.0108 -1.022 -25.05 39.39
#> WEEK 2 DAY 15 134 49.407 8.4665 0.7314 48.335 24.34 71.06 49.572 (8.3223) 134 -0.165 12.0056 1.0371 0.075 -26.31 34.35
#> B: Placebo
#> BASELINE 134 50.306 8.3429 0.7207 50.155 26.22 79.12
#> WEEK 1 DAY 8 134 50.442 7.9484 0.6866 50.179 21.68 67.51 50.306 (8.3429) 134 0.135 12.0710 1.0428 -1.449 -24.58 34.81
#> WEEK 2 DAY 15 134 50.245 8.5308 0.7369 49.971 24.38 71.07 50.306 (8.3429) 134 -0.061 12.5263 1.0821 -0.741 -36.75 28.78
#> C: Combination
#> BASELINE 132 50.902 7.8217 0.6808 50.774 27.63 67.45
#> WEEK 1 DAY 8 132 51.107 7.7903 0.6781 50.804 29.75 71.40 50.902 (7.8217) 132 0.205 10.9217 0.9506 0.051 -27.38 31.08
#> WEEK 2 DAY 15 132 48.494 7.1836 0.6253 49.217 26.22 63.35 50.902 (7.8217) 132 -2.407 10.9704 0.9549 -0.964 -28.56 31.29
#> C-Reactive Protein Measurement
#> A: Drug X
#> BASELINE 134 48.9467 9.43665 0.81520 49.5531 28.959 75.232
#> WEEK 1 DAY 8 134 51.8606 8.07696 0.69774 51.2317 28.466 72.897 48.9467 (9.43665) 134 2.9139 12.47670 1.07782 2.9400 -28.226 41.194
#> WEEK 2 DAY 15 134 49.7480 8.26973 0.71440 50.4409 29.015 69.077 48.9467 (9.43665) 134 0.8013 12.55250 1.08437 0.4837 -34.076 30.432
#> B: Placebo
#> BASELINE 134 50.0803 7.87487 0.68029 50.1184 29.912 73.354
#> WEEK 1 DAY 8 134 50.2698 8.80810 0.76090 51.2478 26.920 67.335 50.0803 (7.87487) 134 0.1895 11.57918 1.00029 0.6008 -31.276 37.223
#> WEEK 2 DAY 15 134 50.8726 7.42330 0.64128 51.5064 30.697 67.695 50.0803 (7.87487) 134 0.7923 10.69335 0.92376 0.8225 -22.154 27.164
#> C: Combination
#> BASELINE 132 50.2079 8.40910 0.73192 49.8365 30.455 72.251
#> WEEK 1 DAY 8 132 50.2549 9.36565 0.81518 50.6980 27.779 71.177 50.2079 (8.40910) 132 0.0471 13.07231 1.13780 -0.3029 -32.046 30.125
#> WEEK 2 DAY 15 132 49.1180 8.24771 0.71787 48.6852 20.630 70.521 50.2079 (8.40910) 132 -1.0899 11.12421 0.96824 -1.4121 -30.236 19.810
#> Immunoglobulin A Measurement
#> A: Drug X
#> BASELINE 134 49.93 7.443 0.643 48.69 33.7 65.9
#> WEEK 1 DAY 8 134 49.41 8.087 0.699 49.78 29.3 69.3 49.93 (7.443) 134 -0.52 10.683 0.923 -0.14 -28.5 21.6
#> WEEK 2 DAY 15 134 50.73 7.856 0.679 49.78 33.3 71.9 49.93 (7.443) 134 0.80 11.517 0.995 1.36 -25.8 32.3
#> B: Placebo
#> BASELINE 134 49.74 8.291 0.716 49.28 25.8 71.5
#> WEEK 1 DAY 8 134 51.12 8.187 0.707 50.52 31.2 74.5 49.74 (8.291) 134 1.38 11.241 0.971 1.55 -25.4 35.7
#> WEEK 2 DAY 15 134 49.88 8.342 0.721 50.67 27.9 69.8 49.74 (8.291) 134 0.14 11.656 1.007 0.94 -29.5 27.6
#> C: Combination
#> BASELINE 132 50.30 9.057 0.788 49.74 26.0 70.0
#> WEEK 1 DAY 8 132 50.75 8.413 0.732 51.11 26.2 71.0 50.30 (9.057) 132 0.45 12.899 1.123 0.46 -29.7 31.0
#> WEEK 2 DAY 15 132 49.38 7.489 0.652 49.57 29.4 65.1 50.30 (9.057) 132 -0.93 12.044 1.048 -0.21 -32.8 28.9In order to utilize rtables::analyze() with arguments
format or formats_var, the return of the used
afun (typically either a call from rcell or
in_rows) should have no formatting specification,
i.e. rcell(x, format = NULL, ...) or
in_rows(.list = x, .formats = NULL, ...).
The tern afuns, such as a_summary() have
the argument .formats to control formats, when these are
set to NULL, the default formats will be applied. In order
to have the afun to return an rcell/in_rows
value without formats, the .formats argument should be set to
"default". The function
tern::get_formats_from_stats will then remove the formats
prior to the call to rcell/in_rows. See also
details section of ?tern::get_formats_from_stats.
As long as the (custom) afun adheres to the above principle of
processing a .formats argument, the
rtables::analyze() arguments format and
formats_var can be utilized for this varying decimal setup
approach.