A nice image.

A nice image.


The {gtsummary} package provides an elegant and flexible way to create publication-ready and reproducible analytical tables. see https://cran.r-project.org/web/packages/gtsummary/index.html.


library(knitr)
library(gtsummary)
library(tidyverse)



Step:1

Univariate Analysis



Open data trial 

head(trial)
## # A tibble: 6 x 8
##   trt      age marker stage grade response death ttdeath
##   <chr>  <dbl>  <dbl> <fct> <fct>    <int> <int>   <dbl>
## 1 Drug A    23  0.16  T1    II           0     0    24  
## 2 Drug B     9  1.11  T2    I            1     0    24  
## 3 Drug A    31  0.277 T1    II           0     0    24  
## 4 Drug A    NA  2.07  T3    III          1     1    17.6
## 5 Drug A    51  2.77  T4    III          1     1    16.4
## 6 Drug B    39  0.613 T4    I            0     1    15.6
# make dataset with a few variables to summarize
trial2 <- trial %>% 
  select(trt, age, grade, response)

head(trial2)
## # A tibble: 6 x 4
##   trt      age grade response
##   <chr>  <dbl> <fct>    <int>
## 1 Drug A    23 II           0
## 2 Drug B     9 I            1
## 3 Drug A    31 II           0
## 4 Drug A    NA III          1
## 5 Drug A    51 III          1
## 6 Drug B    39 I            0



Use tbl_summary(data) 

# summarize the data with 'gtsummary' package
table1 <- tbl_summary(trial2,
  missing = "no" # don't list missing data separately
  )
table1
Characteristic N = 2001
Chemotherapy Treatment
Drug A 98 (49%)
Drug B 102 (51%)
Age, yrs 47 (38, 57)
Grade
I 68 (34%)
II 68 (34%)
III 64 (32%)
Tumor Response 61 (32%)

1 Statistics presented: n (%); median (IQR)



Step:2A

bivariate Analysis



Use tbl_summary(data,by=var) 

table2 <- tbl_summary(
  trial2,
  by = trt, # split table by group
  missing = "no" # don't list missing data separately
) %>%
  add_n() %>% # add column with total number of non-missing observations
  add_p() %>% # test if there's difference between groups
  bold_labels()
table2
Characteristic N Drug A, N = 981 Drug B, N = 1021 p-value2
Age, yrs 189 46 (37, 59) 48 (39, 56) 0.7
Grade 200 0.9
I 35 (36%) 33 (32%)
II 32 (33%) 36 (35%)
III 31 (32%) 33 (32%)
Tumor Response 193 28 (29%) 33 (34%) 0.6

1 Statistics presented: median (IQR); n (%)

2 Statistical tests performed: Wilcoxon rank-sum test; chi-square test of independence

Step:2A

bivariate Analysis(Customize Syntax)



Use tbl_summary(data,by=var) 

table3 <- tbl_summary(
  trial2,
  by = trt, # split table by group
  statistic = list(  # Add statistic
      all_continuous() ~ "{mean} ({sd})", 
      all_categorical() ~ "{n} / {N} ({p}%)"),
  missing = "no", # don't list missing data separately
  label = age ~ "Patient Age" # Change level
) %>%
  add_n() %>% # add column with total number of non-missing observations
  add_p(test = list(all_continuous() ~ "t.test",
                    response ~ "fisher.test"))%>% 
  bold_labels()

table3
Characteristic N Drug A, N = 981 Drug B, N = 1021 p-value2
Patient Age 189 47 (15) 47 (14) 0.8
Grade 200 0.9
I 35 / 98 (36%) 33 / 102 (32%)
II 32 / 98 (33%) 36 / 102 (35%)
III 31 / 98 (32%) 33 / 102 (32%)
Tumor Response 193 28 / 95 (29%) 33 / 98 (34%) 0.5

1 Statistics presented: mean (SD); n / N (%)

2 Statistical tests performed: t-test; chi-square test of independence; Fisher's exact test


Step:3

Regression Analysis



Use tbl_regression(model) 

mod1 <- glm(response ~ trt + age + grade, trial, family = binomial)

t1 <- tbl_regression(mod1, exponentiate = TRUE)
t1
Characteristic OR1 95% CI1 p-value
Chemotherapy Treatment
Drug A — —
Drug B 1.13 0.60, 2.13 0.7
Age, yrs 1.02 1.00, 1.04 0.10
Grade
I — —
II 0.85 0.39, 1.85 0.7
III 1.01 0.47, 2.15 >0.9

1 OR = Odds Ratio, CI = Confidence Interval



Step:4

Side-by-side Regression Models



Use tbl_merge() 

library(survival)

t1 <- tbl_regression(mod1, exponentiate = TRUE)
# build survival model table
t2 <-
  coxph(Surv(ttdeath, death) ~ trt + grade + age, trial) %>%
  tbl_regression(exponentiate = TRUE)

# merge tables 
tbl_merge_ex1 <-
  tbl_merge(
    tbls = list(t1, t2),
    tab_spanner = c("**Tumor Response**", "**Time to Death**")
  )
tbl_merge_ex1

Characteristic Tumor Response Time to Death
OR1 95% CI1 p-value HR1 95% CI1 p-value
Chemotherapy Treatment
Drug A — — — —
Drug B 1.13 0.60, 2.13 0.7 1.30 0.88, 1.92 0.2
Age, yrs 1.02 1.00, 1.04 0.10 1.01 0.99, 1.02 0.3
Grade
I — — — —
II 0.85 0.39, 1.85 0.7 1.21 0.73, 1.99 0.5
III 1.01 0.47, 2.15 >0.9 1.79 1.12, 2.86 0.014

1 OR = Odds Ratio, CI = Confidence Interval, HR = Hazard Ratio



Reference

  1. gtsummary http://www.danieldsjoberg.com/gtsummary/?fbclid=IwAR09Yg3e2ZYtYdskNGlME5bHq5Ytv_wTUlYkWTMv8HRRxpyvnupOBmfVTAo
  2. gtsummary-presentation-rladieshttps://margarethannum.github.io/gtsummary-presentation-rladies/#1