По тем или иным причинам нам часто нужно экспортировать результаты нашего анализа, проведённого в R (таблицы, графики и т.п.) в текстовый редактор MS Word. Есть несколько методов, вот один из них. Рассматривается пример экспорта результатов регрессионного анализа в Ворд.
Get the results from Cox Regression Analysis
As an example to illustrate this post, I will compute a survival analysis. Survival analysis is statistical methods for analyzing data where the outcome variable is the time until the occurrence of an event. The event can be a occurrence of a disease or death, etc. In R we compute the survival analysis with the survival package. The function for Cox regression analysis is coxph(). I will use the veteran data which come with survival package.
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 |
## Load survival package library(survival) # Load veteran data data(veteran) # Show first 6 rows head(veteran) trt celltype time status karno diagtime age prior 1 1 squamous 72 1 60 7 69 0 2 1 squamous 411 1 70 5 64 10 3 1 squamous 228 1 60 3 38 0 4 1 squamous 126 1 60 9 63 10 5 1 squamous 118 1 70 11 65 10 6 1 squamous 10 1 20 5 49 0 # Data description help(veteran, package="survival") trt: 1=standard 2=test celltype: 1=squamous, 2=smallcell, 3=adeno, 4=large time: survival time status: censoring status karno: Karnofsky performance score (100=good) diagtime: months from diagnosis to randomisation age: in years prior: prior therapy 0=no, 1=yes |
Now let say that we are interested to know the risk of dying (status) from different cell type (celltype) and treatment (trt) when we adjust for other variables (karno, age prior, diagtime).
This is the model:
|
1 2 |
# Fit the COX model fit = coxph(Surv(time, status) ~ age + celltype + prior + karno + diagtime + trt, data=veteran) |
And the output:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
summary(fit) Call: coxph(formula = Surv(time, status) ~ age + celltype + prior + karno + diagtime + trt, data = veteran) n= 137, number of events= 128 coef exp(coef) se(coef) z Pr(>|z|) age -8.706e-03 9.913e-01 9.300e-03 -0.936 0.34920 celltypesmallcell 8.616e-01 2.367e+00 2.753e-01 3.130 0.00175 ** celltypeadeno 1.196e+00 3.307e+00 3.009e-01 3.975 7.05e-05 *** celltypelarge 4.013e-01 1.494e+00 2.827e-01 1.420 0.15574 prior 7.159e-03 1.007e+00 2.323e-02 0.308 0.75794 karno -3.282e-02 9.677e-01 5.508e-03 -5.958 2.55e-09 *** diagtime 8.132e-05 1.000e+00 9.136e-03 0.009 0.99290 trt 2.946e-01 1.343e+00 2.075e-01 1.419 0.15577 --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 exp(coef) exp(-coef) lower .95 upper .95 age 0.9913 1.0087 0.9734 1.0096 celltypesmallcell 2.3669 0.4225 1.3799 4.0597 celltypeadeno 3.3071 0.3024 1.8336 5.9647 celltypelarge 1.4938 0.6695 0.8583 2.5996 prior 1.0072 0.9929 0.9624 1.0541 karno 0.9677 1.0334 0.9573 0.9782 diagtime 1.0001 0.9999 0.9823 1.0182 trt 1.3426 0.7448 0.8939 2.0166 Concordance= 0.736 (se = 0.03 ) Rsquare= 0.364 (max possible= 0.999 ) Likelihood ratio test= 62.1 on 8 df, p=1.799e-10 Wald test = 62.37 on 8 df, p=1.596e-10 Score (logrank) test = 66.74 on 8 df, p=2.186e-11 |
As we see there are “a lot” of results. In manuscript we often report only the Hazard ratio and 95% Confidence interval and only for the variables of interest. For example in this case I am interested for the cell types and treatment. Note: I will not comment for the regression coefficients since is not the aim of this post.
Prepare the table by creating the columns
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 |
# Prepare the columns HR <- round(exp(coef(fit)), 2) CI <- round(exp(confint(fit)), 2) P <- round(coef(summary(fit))[,5], 3) # Names the columns of CI colnames(CI) <- c("Lower", "Higher") # Bind columns together as dataset table2 <- as.data.frame(cbind(HR, CI, P)) table2 HR Lower Higher P age 0.99 0.97 1.01 0.349 celltypesmallcell 2.37 1.38 4.06 0.002 celltypeadeno 3.31 1.83 5.96 0.000 celltypelarge 1.49 0.86 2.60 0.156 prior 1.01 0.96 1.05 0.758 karno 0.97 0.96 0.98 0.000 diagtime 1.00 0.98 1.02 0.993 trt 1.34 0.89 2.02 0.156 |
Select variables of interest from the table
As I mentioned earlier, I am interested only for 2 variables (cell type and treatment). With the code below I will select those variables.
|
1 2 3 4 5 6 7 8 |
# select variables you want to present in table table2 <- table2[c("celltypesmallcell","celltypeadeno","celltypelarge","trt"),] table2 HR Lower Higher P celltypesmallcell 2.37 1.38 4.06 0.002 celltypeadeno 3.31 1.83 5.96 0.000 celltypelarge 1.49 0.86 2.60 0.156 trt 1.34 0.89 2.02 0.156 |
Format the table
In the manuscript we present the confidence intervals within brackets. Therefore, with the code below I will add the brackets.
|
1 2 3 4 5 6 7 8 9 10 11 |
# add brackes and line for later use in table table2$a <- "("; table2$b <- "-"; table2$c <- ")" # order the columns table2 <- table2[,c("HR","a","Lower","b","Higher","c", "P")] table2 HR a Lower b Higher c P celltypesmallcell 2.37 ( 1.38 - 4.06 ) 0.002 celltypeadeno 3.31 ( 1.83 - 5.96 ) 0.000 celltypelarge 1.49 ( 0.86 - 2.60 ) 0.156 trt 1.34 ( 0.89 - 2.02 ) 0.156 |
Finalize the table and make it ready for Microsoft Word
The table is almost ready, now I will merge in one column by using package tidyr with function unite().
|
1 2 3 4 5 6 7 8 9 10 11 |
# Merge all columns in one library(tidyr) table2 = unite(table2, "HR (95%CI)", c(HR, a, Lower, b, Higher, c), sep = "", remove=T) # add space between the estimates of HR and CI table2[,1] <- gsub("\\(", " (", table2[,1]) table2 HR (95%CI) P celltypesmallcell 2.37 (1.38-4.06) 0.002 celltypeadeno 3.31 (1.83-5.96) 0.000 celltypelarge 1.49 (0.86-2.6) 0.156 trt 1.34 (0.89-2.02) 0.156 |
Export Table from R to Microsoft Word
To export table from R to Microsoft Word I will use the function FlexTable() from the package ReporteRs. (Credits to the author in StackOverflow).
|
1 2 3 4 5 6 7 8 9 10 11 12 |
# Load the packages library(ReporteRs) library(magrittr) # The script docx( ) %>% addFlexTable(table2 %>% FlexTable(header.cell.props = cellProperties( background.color = "#003366"), header.text.props = textBold(color = "white"), add.rownames = TRUE ) %>% setZebraStyle(odd = "#DDDDDD", even = "#FFFFFF")) %>% writeDoc(file = "table2.docx") |
This is the table in Microsoft Word:
