Update in R package moonBook

Keon-Woong Moon

2022-01-05

Function “mytable”

Function “mytable”” produce table for descriptive analysis easily. It is most useful to make table to describe baseline characteristics common in medical research papers.

Basic Usage

If you are unfamiliar to package moonBook and mytable function, please see the R package moonBook vignette at: https://CRAN.R-project.org/package=moonBook/vignettes/moonBook.html

Explore a data.frame

You can use mytable function to explore a data.frame.

require(moonBook)
require(ztable)
require(magrittr)
options(ztable.type="html")

mytable(acs)

                 Descriptive Statistics                
———————————————————————————————————————————————————————— 
                    Mean ± SD or %   N   Missing (%)
———————————————————————————————————————————————————————— 
   age                  63.3 ± 11.7  857    0  ( 0.0%)
  sex                                857    0  ( 0.0%)
    - Female           287  (33.5%)                   
    - Male             570  (66.5%)                   
  cardiogenicShock                   857    0  ( 0.0%)
    - No               805  (93.9%)                   
    - Yes                52  (6.1%)                   
  entry                              857    0  ( 0.0%)
    - Femoral          312  (36.4%)                   
    - Radial           545  (63.6%)                   
  Dx                                 857    0  ( 0.0%)
    - NSTEMI           153  (17.9%)                   
    - STEMI            304  (35.5%)                   
    - Unstable Angina  400  (46.7%)                   
   EF                    55.8 ± 9.6  723  134  (15.6%)
   height               163.2 ± 9.1  764   93  (10.9%)
   weight               64.8 ± 11.4  766   91  (10.6%)
   BMI                   24.3 ± 3.3  764   93  (10.9%)
  obesity                            857    0  ( 0.0%)
    - No               567  (66.2%)                   
    - Yes              290  (33.8%)                   
   TC                  185.2 ± 47.8  834   23  ( 2.7%)
   LDLC                116.6 ± 41.1  833   24  ( 2.8%)
   HDLC                 38.2 ± 11.1  834   23  ( 2.7%)
   TG                  125.2 ± 90.9  842   15  ( 1.8%)
  DM                                 857    0  ( 0.0%)
    - No               553  (64.5%)                   
    - Yes              304  (35.5%)                   
  HBP                                857    0  ( 0.0%)
    - No               356  (41.5%)                   
    - Yes              501  (58.5%)                   
  smoking                            857    0  ( 0.0%)
    - Ex-smoker        204  (23.8%)                   
    - Never            332  (38.7%)                   
    - Smoker           321  (37.5%)                   
———————————————————————————————————————————————————————— 

You can use formula without grouping variable(s).

mytable(~age+sex,data=acs)

           Descriptive Statistics           
————————————————————————————————————————————— 
           Mean ± SD or %   N  Missing (%)
————————————————————————————————————————————— 
   age         63.3 ± 11.7  857  0  ( 0.0%)
  sex                       857  0  ( 0.0%)
    - Female  287  (33.5%)                 
    - Male    570  (66.5%)                 
————————————————————————————————————————————— 

Compress an object of class mytable

You can compress mytable. If rows dealing with categorical variables have two unique values, it can be printed in a single row rather than three rows.

mytable(Dx~sex,data=acs)

             Descriptive Statistics by 'Dx'             
————————————————————————————————————————————————————————— 
              NSTEMI       STEMI    Unstable Angina   p  
              (N=153)     (N=304)       (N=400)    
————————————————————————————————————————————————————————— 
 sex                                                0.012
   - Female 50 (32.7%)  84 (27.6%)    153 (38.2%)        
   - Male   103 (67.3%) 220 (72.4%)   247 (61.8%)        
————————————————————————————————————————————————————————— 
mytable(Dx~sex,data=acs) %>% compress

            Descriptive Statistics by 'Dx'            
——————————————————————————————————————————————————————— 
            NSTEMI       STEMI    Unstable Angina   p  
            (N=153)     (N=304)       (N=400)    
——————————————————————————————————————————————————————— 
 sex:Male 103 (67.3%) 220 (72.4%)   247 (61.8%)   0.012
——————————————————————————————————————————————————————— 

The default representative group is the second group. If you want the first group to being representative group, please use the no argument.

mytable(Dx~sex,data=acs) %>% compress(no=1)

            Descriptive Statistics by 'Dx'            
——————————————————————————————————————————————————————— 
              NSTEMI     STEMI    Unstable Angina   p  
             (N=153)    (N=304)       (N=400)    
——————————————————————————————————————————————————————— 
 sex:Female 50 (32.7%) 84 (27.6%)   153 (38.2%)   0.012
——————————————————————————————————————————————————————— 

Sometimes it is more simple to omit the representative group name. You can do this by set the add.label argument FALSE.

mytable(Dx~cardiogenicShock+DM+obesity+HBP,data=acs) %>% compress

                  Descriptive Statistics by 'Dx'                 
—————————————————————————————————————————————————————————————————— 
                        NSTEMI      STEMI    Unstable Angina   p  
                       (N=153)     (N=304)       (N=400)    
—————————————————————————————————————————————————————————————————— 
 cardiogenicShock:Yes 4 ( 2.6%)  48 (15.8%)     0 ( 0.0%)    0.000
 DM:Yes               57 (37.3%) 96 (31.6%)    151 (37.8%)   0.209
 obesity:Yes          47 (30.7%) 95 (31.2%)    148 (37.0%)   0.186
 HBP:Yes              91 (59.5%) 154 (50.7%)   256 (64.0%)   0.002
—————————————————————————————————————————————————————————————————— 
mytable(Dx~cardiogenicShock+DM+obesity+HBP,data=acs) %>% compress(add.label=FALSE)

                Descriptive Statistics by 'Dx'               
—————————————————————————————————————————————————————————————— 
                    NSTEMI      STEMI    Unstable Angina   p  
                   (N=153)     (N=304)       (N=400)    
—————————————————————————————————————————————————————————————— 
 cardiogenicShock 4 ( 2.6%)  48 (15.8%)     0 ( 0.0%)    0.000
 DM               57 (37.3%) 96 (31.6%)    151 (37.8%)   0.209
 obesity          47 (30.7%) 95 (31.2%)    148 (37.0%)   0.186
 HBP              91 (59.5%) 154 (50.7%)   256 (64.0%)   0.002
—————————————————————————————————————————————————————————————— 

You can print mytable object in ‘html5’ or ‘LaTex’ format with ztable.

mytable(Dx~cardiogenicShock+DM+obesity+HBP,data=acs) %>% compress(add.label=FALSE) %>% ztable
NSTEMI STEMI Unstable Angina p
(N=153) (N=304) (N=400)
cardiogenicShock 4 ( 2.6%) 48 (15.8%) 0 ( 0.0%) < 0.001
DM 57 (37.3%) 96 (31.6%) 151 (37.8%) 0.209
obesity 47 (30.7%) 95 (31.2%) 148 (37.0%) 0.186
HBP 91 (59.5%) 154 (50.7%) 256 (64.0%) 0.002

Delete Rows of an object of class mytable

You can delete rows of an object of class mytable.

mytable(sex~Dx,data=acs)

         Descriptive Statistics by 'sex'         
—————————————————————————————————————————————————— 
                       Female       Male       p  
                       (N=287)     (N=570)  
—————————————————————————————————————————————————— 
 Dx                                          0.012
   - NSTEMI          50 (17.4%)  103 (18.1%)      
   - STEMI           84 (29.3%)  220 (38.6%)      
   - Unstable Angina 153 (53.3%) 247 (43.3%)      
—————————————————————————————————————————————————— 

If you want to delete the second row, use the deleteRows() function.

mytable(sex~Dx,data=acs) %>% deleteRows(2)

         Descriptive Statistics by 'sex'         
—————————————————————————————————————————————————— 
                       Female       Male       p  
                       (N=287)     (N=570)  
—————————————————————————————————————————————————— 
 Dx                                          0.012
   - STEMI           84 (29.3%)  220 (38.6%)      
   - Unstable Angina 153 (53.3%) 247 (43.3%)      
—————————————————————————————————————————————————— 

You can delete rows of an object of class cbind.mytable.

mytable(sex+HBP~age+Dx,data=acs) %>% deleteRows(3)

              Descriptive Statistics stratified by 'sex' and 'HBP'             
———————————————————————————————————————————————————————————————————————————————— 
                                   Male                          Female            
                     ————————————————————————————— ————————————————————————————— 
                         No          Yes       p       No          Yes       p  
                       (N=273)     (N=297)          (N=83)      (N=204)       
———————————————————————————————————————————————————————————————————————————————— 
 age                 57.5 ± 11.3 63.5 ± 10.4 0.000 68.0 ± 12.3 69.0 ± 10.0 0.526
 Dx                                          0.016                         0.084
   - STEMI           122 (44.7%) 98 (33.0%)        28 (33.7%)  56 (27.5%)       
   - Unstable Angina 108 (39.6%) 139 (46.8%)       36 (43.4%)  117 (57.4%)      
———————————————————————————————————————————————————————————————————————————————— 

Methods for categorical variables

You can select method for categorical variables with catMethod argument. Possible values are :

You can see which tests are used if you set show.all argument of mytable TRUE.

mytable(obesity~HBP,data=acs,catMethod=1,show.all=TRUE)

                                Descriptive Statistics by 'obesity'                                
———————————————————————————————————————————————————————————————————————————————————————————————————— 
             No          Yes       p   sig  p1   p2  p3      class              ptest             N 
           (N=567)     (N=290)  
———————————————————————————————————————————————————————————————————————————————————————————————————— 
 HBP                             0.034 **  0.034    2.000 categorical Pearson's Chi-squared test 857
   - No  250 (44.1%) 106 (36.6%)                                                                    
   - Yes 317 (55.9%) 184 (63.4%)                                                                    
———————————————————————————————————————————————————————————————————————————————————————————————————— 

For formatted numbers: addComma()

Sometimes, you want to display formatted numbers. For example, 1234.5 can be printed as 1,234.5. You can do this using addComma() function

data(diamonds,package="ggplot2")
mytable(diamonds) %>% addComma

                Descriptive Statistics               
—————————————————————————————————————————————————————— 
               Mean ± SD or %      N    Missing (%)
—————————————————————————————————————————————————————— 
   carat               0.8 ± 0.5  53,940  0  ( 0.0%)
  cut                             53,940  0  ( 0.0%)
    - Fair          1610  (3.0%)                    
    - Good          4906  (9.1%)                    
    - Very Good   12082  (22.4%)                    
    - Premium     13791  (25.6%)                    
    - Ideal       21551  (40.0%)                    
  color                           53,940  0  ( 0.0%)
    - D            6775  (12.6%)                    
    - E            9797  (18.2%)                    
    - F            9542  (17.7%)                    
    - G           11292  (20.9%)                    
    - H            8304  (15.4%)                    
    - I            5422  (10.1%)                    
    - J             2808  (5.2%)                    
  clarity                         53,940  0  ( 0.0%)
    - I1             741  (1.4%)                    
    - SI2          9194  (17.0%)                    
    - SI1         13065  (24.2%)                    
    - VS2         12258  (22.7%)                    
    - VS1          8171  (15.1%)                    
    - VVS2          5066  (9.4%)                    
    - VVS1          3655  (6.8%)                    
    - IF            1790  (3.3%)                    
   depth              61.7 ± 1.4  53,940  0  ( 0.0%)
   table              57.5 ± 2.2  53,940  0  ( 0.0%)
   price         3932.8 ± 3989.4  53,940  0  ( 0.0%)
   x                   5.7 ± 1.1  53,940  0  ( 0.0%)
   y                   5.7 ± 1.1  53,940  0  ( 0.0%)
   z                   3.5 ± 0.7  53,940  0  ( 0.0%)
—————————————————————————————————————————————————————— 

Also you can print this in ‘html5’ or ‘LaTex’ format with ztable.

mytable(cut~.,data=diamonds) %>% addComma %>% ztable
Fair Good Very Good Premium Ideal p
(N= 1,610) (N= 4,906) (N=12,082) (N=13,791) (N=21,551)
carat 1.0 ± 0.5 0.8 ± 0.5 0.8 ± 0.5 0.9 ± 0.5 0.7 ± 0.4 < 0.001
color < 0.001
    D 163 (10.1%) 662 (13.5%) 1,513 (12.5%) 1,603 (11.6%) 2,834 (13.2%)
    E 224 (13.9%) 933 (19%) 2,400 (19.9%) 2,337 (16.9%) 3,903 (18.1%)
    F 312 (19.4%) 909 (18.5%) 2,164 (17.9%) 2,331 (16.9%) 3,826 (17.8%)
    G 314 (19.5%) 871 (17.8%) 2,299 (19%) 2,924 (21.2%) 4,884 (22.7%)
    H 303 (18.8%) 702 (14.3%) 1,824 (15.1%) 2,360 (17.1%) 3,115 (14.5%)
    I 175 (10.9%) 522 (10.6%) 1,204 (10%) 1,428 (10.4%) 2,093 (9.7%)
    J 119 (7.4%) 307 (6.3%) 678 (5.6%) 808 (5.9%) 896 (4.2%)
clarity < 0.001
    I1 210 (13%) 96 (2%) 84 (0.7%) 205 (1.5%) 146 (0.7%)
    SI2 466 (28.9%) 1,081 (22%) 2,100 (17.4%) 2,949 (21.4%) 2,598 (12.1%)
    SI1 408 (25.3%) 1,560 (31.8%) 3,240 (26.8%) 3,575 (25.9%) 4,282 (19.9%)
    VS2 261 (16.2%) 978 (19.9%) 2,591 (21.4%) 3,357 (24.3%) 5,071 (23.5%)
    VS1 170 (10.6%) 648 (13.2%) 1,775 (14.7%) 1,989 (14.4%) 3,589 (16.7%)
    VVS2 69 (4.3%) 286 (5.8%) 1,235 (10.2%) 870 (6.3%) 2,606 (12.1%)
    VVS1 17 (1.1%) 186 (3.8%) 789 (6.5%) 616 (4.5%) 2,047 (9.5%)
    IF 9 (0.6%) 71 (1.4%) 268 (2.2%) 230 (1.7%) 1,212 (5.6%)
depth 64.0 ± 3.6 62.4 ± 2.2 61.8 ± 1.4 61.3 ± 1.2 61.7 ± 0.7 < 0.001
table 59.1 ± 3.9 58.7 ± 2.9 58.0 ± 2.1 58.7 ± 1.5 56.0 ± 1.2 < 0.001
price 4,358.8 ± 3,560.4 3,928.9 ± 3,681.6 3,981.8 ± 3,935.9 4,584.3 ± 4,349.2 3,457.5 ± 3,808.4 < 0.001
x 6.2 ± 1.0 5.8 ± 1.1 5.7 ± 1.1 6.0 ± 1.2 5.5 ± 1.1 < 0.001
z 4.0 ± 0.7 3.6 ± 0.7 3.6 ± 0.7 3.6 ± 0.7 3.4 ± 0.7 < 0.001