# ClinReport Vignette 1: Get Started

## Get started

library(ClinReport)
library(officer)
library(flextable)
library(dplyr)
library(reshape2)
library(nlme)
library(emmeans)
library(car)

# We will use fake data
data(data)
#>    y_numeric y_logistic y_poisson   baseline   VAR GROUP TIMEPOINT SUBJID
#> 1 -0.4203490          1         5 -0.4203490 Cat 1     A        D0 Subj 1
#> 2 -0.1570941          1         5 -0.1570941 Cat 2     A        D0 Subj 1
#> 3         NA          0         3 -2.0853720 Cat 2     A        D0 Subj 1
#> 4 -0.4728527          0         5 -0.4728527 Cat 1     A        D0 Subj 1
#> 5 -0.8651713          1         4 -0.8651713 Cat 1     A        D0 Subj 1
#> 6 -1.5476907          1         3 -1.5476907 Cat 1     A        D0 Subj 1

Create a statistical output for a quantitative response and two explicative variables. For example a treatment group and a time variable corresponding to the visits of a clinical trial.

For that we use the report.quanti() function:

tab1=report.quanti(data=data,y="y_numeric",
x1="GROUP",x2="TIMEPOINT",at.row="TIMEPOINT",
subjid="SUBJID")

tab1
#>    TIMEPOINT Statistics      A (N=30)      B (N=21)      C (N=17)
#> 1         D0          N            30            20            16
#> 2         D0  Mean (SD)   -0.93(0.86)   -0.67(1.09)   -1.19(0.92)
#> 3         D0     Median         -0.82         -0.69         -1.26
#> 4         D0    [Q1;Q3] [-1.59;-0.16] [-1.39;-0.06] [-1.62;-0.83]
#> 5         D0  [Min;Max]  [-2.34;0.36]  [-2.44;2.10]  [-2.99;0.66]
#> 6         D0    Missing             1             1             0
#> 7
#> 8         D1          N            30            20            16
#> 9         D1  Mean (SD)    1.83(1.04)    4.17(1.28)    4.98(0.69)
#> 10        D1     Median          1.78          4.19          5.08
#> 11        D1    [Q1;Q3] [ 0.94; 2.54] [ 3.23; 4.92] [ 4.58; 5.46]
#> 12        D1  [Min;Max]  [ 0.11;3.88]  [ 1.48;6.19]  [ 3.80;6.23]
#> 13        D1    Missing             1             0             0
#> 14
#> 15        D2          N            30            20            16
#> 16        D2  Mean (SD)    1.97(1.17)    4.04(0.89)    4.90(1.36)
#> 17        D2     Median          1.66          4.19          5.06
#> 18        D2    [Q1;Q3] [ 1.23; 2.86] [ 3.62; 4.36] [ 4.34; 5.20]
#> 19        D2  [Min;Max]  [-0.18;4.36]  [ 2.03;5.63]  [ 2.39;7.96]
#> 20        D2    Missing             1             1             0
#> 21
#> 22        D3          N            30            20            16
#> 23        D3  Mean (SD)    1.78(1.17)    3.81(0.94)    5.07(1.12)
#> 24        D3     Median          1.78          3.63          5.22
#> 25        D3    [Q1;Q3] [ 0.93; 2.42] [ 3.13; 4.44] [ 4.11; 5.66]
#> 26        D3  [Min;Max]  [-0.16;3.90]  [ 2.46;6.01]  [ 3.16;7.37]
#> 27        D3    Missing             0             1             1
#> 28
#> 29        D4          N            30            20            16
#> 30        D4  Mean (SD)    1.83(0.85)    3.80(0.95)    5.17(1.03)
#> 31        D4     Median          1.67          3.83          4.88
#> 32        D4    [Q1;Q3] [ 1.26; 2.32] [ 3.12; 4.42] [ 4.69; 5.50]
#> 33        D4  [Min;Max]  [ 0.38;3.97]  [ 2.31;5.41]  [ 3.24;6.96]
#> 34        D4    Missing             1             1             1
#> 35
#> 36        D5          N            30            20            16
#> 37        D5  Mean (SD)    2.27(1.20)    3.64(1.19)    4.43(0.98)
#> 38        D5     Median          2.50          3.86          4.57
#> 39        D5    [Q1;Q3] [ 1.77; 3.21] [ 2.59; 4.60] [ 3.44; 4.97]
#> 40        D5  [Min;Max]  [-1.19;4.31]  [ 0.91;5.12]  [ 2.95;6.54]
#> 41        D5    Missing             0             0             0

The at.row argument is used to space the results between each visit and the subjid argument is used to add in the columns header the total number of subjects randomized by treatment group.

Generally we want also the corresponding graphics. So you can use the specific plot function to print the corresponding graphic of your table:

g1=plot(tab1,title="The title that you want to display")
print(g1)

You can modify the plot by using the following arguments of the plot.desc() function:

args(ClinReport:::plot.desc)
#> function (x, ..., title = "", ylim = NULL, xlim = NULL, xlab = "",
#>     ylab = "", legend.label = "Group", add.sd = F, add.ci = F,
#>     size.title = 10, add.line = T)
#> NULL

Then we can use the report.doc() function which use the flextable package to format the output into a flextable object, ready to export to Microsoft Word with the officer package.

The table will look like this (we can have a preview in HTML, just to check):

report.doc(tab1,title="Quantitative statistics (2 explicative variables)",
colspan.value="Treatment group", init.numbering =T )            
 Output 1: Quantitative statistics (2 explicative variables) Treatment group TIMEPOINT Statistics A (N=30) B (N=21) C (N=17) D0 N 30 20 16 Mean (SD) -0.93(0.86) -0.67(1.09) -1.19(0.92) Median -0.82 -0.69 -1.26 [Q1;Q3] [-1.59;-0.16] [-1.39;-0.06] [-1.62;-0.83] [Min;Max] [-2.34;0.36] [-2.44;2.10] [-2.99;0.66] Missing 1 1 0 D1 N 30 20 16 Mean (SD) 1.83(1.04) 4.17(1.28) 4.98(0.69) Median 1.78 4.19 5.08 [Q1;Q3] [ 0.94; 2.54] [ 3.23; 4.92] [ 4.58; 5.46] [Min;Max] [ 0.11;3.88] [ 1.48;6.19] [ 3.80;6.23] Missing 1 0 0 D2 N 30 20 16 Mean (SD) 1.97(1.17) 4.04(0.89) 4.90(1.36) Median 1.66 4.19 5.06 [Q1;Q3] [ 1.23; 2.86] [ 3.62; 4.36] [ 4.34; 5.20] [Min;Max] [-0.18;4.36] [ 2.03;5.63] [ 2.39;7.96] Missing 1 1 0 D3 N 30 20 16 Mean (SD) 1.78(1.17) 3.81(0.94) 5.07(1.12) Median 1.78 3.63 5.22 [Q1;Q3] [ 0.93; 2.42] [ 3.13; 4.44] [ 4.11; 5.66] [Min;Max] [-0.16;3.90] [ 2.46;6.01] [ 3.16;7.37] Missing 0 1 1 D4 N 30 20 16 Mean (SD) 1.83(0.85) 3.80(0.95) 5.17(1.03) Median 1.67 3.83 4.88 [Q1;Q3] [ 1.26; 2.32] [ 3.12; 4.42] [ 4.69; 5.50] [Min;Max] [ 0.38;3.97] [ 2.31;5.41] [ 3.24;6.96] Missing 1 1 1 D5 N 30 20 16 Mean (SD) 2.27(1.20) 3.64(1.19) 4.43(0.98) Median 2.50 3.86 4.57 [Q1;Q3] [ 1.77; 3.21] [ 2.59; 4.60] [ 3.44; 4.97] [Min;Max] [-1.19;4.31] [ 0.91;5.12] [ 2.95;6.54] Missing 0 0 0

All output numbers will be increased automatically after each call of the function report.doc().

You can restart the numbering of the outputs by using init.numbering=T argument in report.doc() function.

Finally, we add those results to a rdocx object:

doc=read_docx()
doc=report.doc(tab1,title="Quantitative statistics (2 explicative variables)",
colspan.value="Treatment group",doc=doc,init.numbering=T)
doc=body_add_gg(doc, value = g1, style = "centered" )

Write the doc to a docx file:

file=paste(tempfile(),".docx",sep="")
print(doc, target =file)

#Open it
#shell.exec(file)

## The different outputs

### Qualitative tables

An example of qualitative statistics with one explicative variable

tab=report.quali(data=data,y="y_logistic",
x1="VAR",total=T,subjid="SUBJID")

report.doc(tab,title="Qualitative table with two variables",
colspan.value="A variable") 
 Output 2: Qualitative table with two variables A variable Levels Statistics Cat 1 (N=65) Cat 2 (N=63) Total (N=128) 0 n (column %) 100(48.08%) 86(45.74%) 186(46.97%) 1 n (column %) 103(49.52%) 97(51.60%) 200(50.51%) Missing n(%) 5(2.40%) 5(2.66%) 10(2.53%)

An example of qualitative statistics with two explicative variables

tab=report.quali(data=data,y="y_logistic",
x1="GROUP",x2="TIMEPOINT",at.row="TIMEPOINT",
total=T,subjid="SUBJID")

report.doc(tab,title="Qualitative table with two variables",
colspan.value="Treatment group")    
 Output 3: Qualitative table with two variables Treatment group TIMEPOINT Levels Statistics A (N=30) B (N=21) C (N=17) Total (N=68) D0 0 n (column %) 11(36.67%) 11(55.00%) 7(43.75%) 29(43.94%) 1 n (column %) 18(60.00%) 8(40.00%) 7(43.75%) 33(50.00%) Missing n(%) 1(3.33%) 1(5.00%) 2(12.50%) 4(6.06%) D1 0 n (column %) 7(23.33%) 13(65.00%) 8(50.00%) 28(42.42%) 1 n (column %) 21(70.00%) 7(35.00%) 7(43.75%) 35(53.03%) Missing n(%) 2(6.67%) 0(0%) 1(6.25%) 3(4.55%) D2 0 n (column %) 18(60.00%) 7(35.00%) 11(68.75%) 36(54.55%) 1 n (column %) 12(40.00%) 13(65.00%) 5(31.25%) 30(45.45%) Missing n(%) 0(0%) 0(0%) 0(0%) 0(0%) D3 0 n (column %) 11(36.67%) 10(50.00%) 7(43.75%) 28(42.42%) 1 n (column %) 19(63.33%) 10(50.00%) 9(56.25%) 38(57.58%) Missing n(%) 0(0%) 0(0%) 0(0%) 0(0%) D4 0 n (column %) 18(60.00%) 12(60.00%) 6(37.50%) 36(54.55%) 1 n (column %) 12(40.00%) 8(40.00%) 8(50.00%) 28(42.42%) Missing n(%) 0(0%) 0(0%) 2(12.50%) 2(3.03%) D5 0 n (column %) 14(46.67%) 7(35.00%) 8(50.00%) 29(43.94%) 1 n (column %) 15(50.00%) 13(65.00%) 8(50.00%) 36(54.55%) Missing n(%) 1(3.33%) 0(0%) 0(0%) 1(1.52%)

### Quantitative tables

An example of quantitative statistics with one explicative variable

tab=report.quanti(data=data,y="y_numeric",
x1="VAR",total=T,subjid="SUBJID")

report.doc(tab,title="Quantitative table with one explicative variable",
colspan.value="A variable") 
 Output 4: Quantitative table with one explicative variable A variable Statistics Cat 1 (N=65) Cat 2 (N=63) Total (N=128) N 208 188 396 Mean (SD) 2.55(2.18) 2.56(2.23) 2.56(2.20) Median 2.64 2.79 2.71 [Q1;Q3] [0.94;4.36] [1.07;4.19] [1.04;4.33] [Min;Max] [-2.39;6.43] [-2.99;7.96] [-2.99;7.96] Missing 4 6 10

An example of quantitative statistics with two explicative variables

tab=report.quanti(data=data,y="y_numeric",
x1="GROUP",x2="TIMEPOINT",at.row="TIMEPOINT",
total=T,subjid="SUBJID")

report.doc(tab,title="Quantitative table with two explicative variables",
colspan.value="Treatment group")    
 Output 5: Quantitative table with two explicative variables Treatment group TIMEPOINT Statistics A (N=30) B (N=21) C (N=17) Total (N=68) D0 N 30 20 16 66 Mean (SD) -0.93(0.86) -0.67(1.09) -1.19(0.92) -0.92(0.95) Median -0.82 -0.69 -1.26 -0.86 [Q1;Q3] [-1.59;-0.16] [-1.39;-0.06] [-1.62;-0.83] [-1.55;-0.16] [Min;Max] [-2.34;0.36] [-2.44;2.10] [-2.99;0.66] [-2.99;2.10] Missing 1 1 0 2 D1 N 30 20 16 66 Mean (SD) 1.83(1.04) 4.17(1.28) 4.98(0.69) 3.33(1.73) Median 1.78 4.19 5.08 3.57 [Q1;Q3] [ 0.94; 2.54] [ 3.23; 4.92] [ 4.58; 5.46] [ 1.78; 4.91] [Min;Max] [ 0.11;3.88] [ 1.48;6.19] [ 3.80;6.23] [ 0.11;6.23] Missing 1 0 0 1 D2 N 30 20 16 66 Mean (SD) 1.97(1.17) 4.04(0.89) 4.90(1.36) 3.32(1.70) Median 1.66 4.19 5.06 3.57 [Q1;Q3] [ 1.23; 2.86] [ 3.62; 4.36] [ 4.34; 5.20] [ 1.89; 4.44] [Min;Max] [-0.18;4.36] [ 2.03;5.63] [ 2.39;7.96] [-0.18;7.96] Missing 1 1 0 2 D3 N 30 20 16 66 Mean (SD) 1.78(1.17) 3.81(0.94) 5.07(1.12) 3.15(1.75) Median 1.78 3.63 5.22 3.15 [Q1;Q3] [ 0.93; 2.42] [ 3.13; 4.44] [ 4.11; 5.66] [ 1.80; 4.39] [Min;Max] [-0.16;3.90] [ 2.46;6.01] [ 3.16;7.37] [-0.16;7.37] Missing 0 1 1 2 D4 N 30 20 16 66 Mean (SD) 1.83(0.85) 3.80(0.95) 5.17(1.03) 3.22(1.66) Median 1.67 3.83 4.88 3.16 [Q1;Q3] [ 1.26; 2.32] [ 3.12; 4.42] [ 4.69; 5.50] [ 1.69; 4.48] [Min;Max] [ 0.38;3.97] [ 2.31;5.41] [ 3.24;6.96] [ 0.38;6.96] Missing 1 1 1 3 D5 N 30 20 16 66 Mean (SD) 2.27(1.20) 3.64(1.19) 4.43(0.98) 3.21(1.45) Median 2.50 3.86 4.57 3.28 [Q1;Q3] [ 1.77; 3.21] [ 2.59; 4.60] [ 3.44; 4.97] [ 2.42; 4.44] [Min;Max] [-1.19;4.31] [ 0.91;5.12] [ 2.95;6.54] [-1.19;6.54] Missing 0 0 0 0

### Mixed Quantitative and Qualitative tables

You can mix qualitative and quantitative outputs.

But itâ€™s only possible for 1 explicative variable, and it should be the same variable for both response:

tab1=report.quanti(data=data,y="y_numeric",
x1="GROUP",subjid="SUBJID",y.label="Y numeric")

tab2=report.quali(data=data,y="y_logistic",
x1="GROUP",subjid="SUBJID",y.label="Y logistic")

report.doc(tab3,title="Mixed Qualitative and Quantitative outputs",
colspan.value="Treatment group")
 Output 6: Mixed Qualitative and Quantitative outputs Treatment group The label of your choice Levels Statistics A (N=30) B (N=21) C (N=17) Y numeric N 180 120 96 Mean (SD) 1.46(1.50) 3.15(2.00) 3.87(2.52) Median 1.59 3.75 4.73 [Q1;Q3] [0.45;2.50] [2.46;4.44] [3.44;5.30] [Min;Max] [-2.34;4.36] [-2.44;6.19] [-2.99;7.96] Missing 4 4 2 Y logistic 0 n (column %) 79(43.89%) 60(50.00%) 47(48.96%) 1 n (column %) 97(53.89%) 59(49.17%) 44(45.83%) Missing n(%) 4(2.22%) 1(0.83%) 5(5.21%)

### Anova model reporting

For the anova table reporting, itâ€™s basically a call to the function xtable_to_flextable(). The function report.doc() just handle the numbering of the output and the header with the title.

# Removing baseline data for the model
data.mod=droplevels(data[data\$TIMEPOINT!="D0",])

mod=lme(y_numeric~baseline+GROUP+TIMEPOINT+GROUP*TIMEPOINT,
random=~1|SUBJID,data=data.mod,na.action=na.omit)

anov3=Anova(mod,3)

report.doc(anov3,title="Mixed Qualitative and Quantitative output")
 Output 7: Mixed Qualitative and Quantitative output Chisq Df Pr(>Chisq) (Intercept) 84.699 1.000 0.000 baseline 1.695 1.000 0.193 GROUP 107.561 2.000 0.000 TIMEPOINT 4.426 4.000 0.351 GROUP:TIMEPOINT 11.671 8.000 0.166

### LS-Means model reporting

LS-means reporting are based on the package emmeans. The function report.lsmeans() enables to format the output:

lsm=emmeans(mod,~GROUP|TIMEPOINT)

tab=report.lsmeans(lsm,x1="GROUP",x2="TIMEPOINT",data=data.mod,
at.row="TIMEPOINT")

report.doc(tab,title="LS-Means example",
colspan.value="Treatment Group")
 Output 8: LS-Means example Treatment Group TIMEPOINT Statistics A B C D1 Estimate (SE) 1.81(0.20) 4.17(0.24) 5.00(0.27) 95% CI [1.41;2.22] [3.69;4.65] [4.46;5.54] P-value <0.001 <0.001 <0.001 D2 Estimate (SE) 1.96(0.20) 4.05(0.25) 4.90(0.27) 95% CI [1.56;2.36] [3.56;4.55] [4.36;5.44] P-value <0.001 <0.001 <0.001 D3 Estimate (SE) 1.79(0.20) 3.79(0.25) 5.08(0.28) 95% CI [1.39;2.18] [3.29;4.28] [4.52;5.63] P-value <0.001 <0.001 <0.001 D4 Estimate (SE) 1.83(0.20) 3.80(0.25) 5.17(0.28) 95% CI [1.43;2.23] [3.31;4.30] [4.62;5.73] P-value <0.001 <0.001 <0.001 D5 Estimate (SE) 2.28(0.20) 3.64(0.24) 4.42(0.27) 95% CI [1.89;2.68] [3.15;4.12] [3.88;4.96] P-value <0.001 <0.001 <0.001

### Pairs and Contrasts of LS-MEans

If you want to report contrast, youâ€™ll have to specify contrast=TRUE in the call to report.lsmeans().

contr=contrast(lsm, "trt.vs.ctrl", ref = "A")

# Now there is just only one explicative variable
# since we make comparison between treatment group
# so there is only x1="TIMEPOINT" in the call

tab.contr=report.lsmeans(lsm=contr,x1="TIMEPOINT",
data=data.mod,contrast=TRUE,at.row="contrast")

report.doc(tab.contr,title="LS-Means contrast example",
colspan.value="Time points")        
 Output 9: LS-Means contrast example Time points contrast Statistics D1 D2 D3 D4 D5 B - A Estimate (SE) 2.36(0.31) 2.10(0.32) 2.00(0.32) 1.97(0.32) 1.35(0.31) 95% CI [1.66;3.06] [1.38;2.81] [1.29;2.71] [1.26;2.68] [0.66;2.05] P-value <0.001 <0.001 <0.001 <0.001 <0.001 C - A Estimate (SE) 3.19(0.34) 2.94(0.34) 3.29(0.34) 3.34(0.34) 2.14(0.33) 95% CI [2.44;3.94] [2.19;3.69] [2.53;4.05] [2.58;4.11] [1.39;2.88] P-value <0.001 <0.001 <0.001 <0.001 <0.001