Create a table one - To be deprecated

Description

This is a function created to provide characteristics of a study group with an option to stratify by some variable (usually an exposure).

Usage

cida_table1(
  data,
  includeVars,
  stratifyBy,
  group_labels,
  group_label_span,
  caption = NULL,
  footnote = NULL,
  include_total = TRUE,
  exclude_mean = FALSE,
  exclude_missing = FALSE,
  compute_pval = FALSE,
  nonParametricVars = NULL,
  exportWord = FALSE,
  useSciNotation = FALSE
)

Arguments

data A data frame with data
includeVars A vector of variable names you wish to include in the table.
stratifyBy The group by which you wish to stratify your table
group_labels Higher level labels for you stratified groups
group_label_span The span of columns for each group in ‘group_labels’
caption Optional; Adds a caption to the table
footnote Optional; Adds a footnote to the table
include_total Bool; Default TRUE. Includes a column of the summation of all the stratified groups.
exclude_mean Bool; Default FALSE. Excludes the mean for numerical variables.
exclude_missing Bool; Default FALSE. Excludes percentages of missing data and only returns counts
compute_pval Bool; Default FALSE. Computes p-values for ‘includeVars’ and add a p-value column in the table.
nonParametricVars Vector; of the variable names you would like non-parametric testing to be conducted on for p-values returned
exportWord Bool; whether to export the table in a nice format into word
useSciNotation Bool; Use scientific notation for large/small continuous variables

Value

an html table with N and percentages for categorical variables, mean , SD, Median, and Range for numeric variables. Returns p-values if specified.

Examples

library("CIDAtools")

# Synthetic data
df = data.frame(
    `Age` = c(10,12,14,18,20,19,28,33, rep(NA, 4)),
    `Sex` = c(rep("Female", 4), rep("Male", 4), rep(NA, 4)),
    `Smoking Status` = c(rep('Former', 2), rep('Current', 2), rep('Never', 4), rep(NA, 4)),
    `IL-8` = rnorm(12, 35, sd = 7),
    `Group` = c(rep('Control', 4), rep('Heart Disease', 4), rep('Lung Disease', 4)),
     check.names = FALSE
    )
# Table 1 with no p-values
cida_table1(data = df,
            includeVars = c("Age", "Sex", "Smoking Status", "IL-8"),
            stratifyBy = "Group",
            group_labels = c("Group"),
            group_label_span = c(3),
            caption = "TABLE 1",
            footnote = "My table 1",
            include_total = FALSE,
            compute_pval = FALSE,
            nonParametricVars = NULL,
            exportWord = FALSE)
TABLE 1
Group
Control
(N=4)		 Heart Disease
(N=4)		 Lung Disease
(N=4)

My table 1
Age
Mean (SD) 13.5 (3.42) 25.0 (6.68) NA (NA)
Median [Min, Max] 13.0 [10.0, 18.0] 24.0 [19.0, 33.0] NA [NA, NA]
Missing 0 (0%) 0 (0%) 4 (100%)
Sex
Female 4 (100%) 0 (0%) 0 (0%)
Male 0 (0%) 4 (100%) 0 (0%)
Missing 0 (0%) 0 (0%) 4 (100%)
Smoking Status
Current 2 (50.0%) 0 (0%) 0 (0%)
Former 2 (50.0%) 0 (0%) 0 (0%)
Never 0 (0%) 4 (100%) 0 (0%)
Missing 0 (0%) 0 (0%) 4 (100%)
IL-8
Mean (SD) 35.1 (10.5) 38.7 (7.63) 31.7 (5.29)
Median [Min, Max] 38.1 [20.6, 43.5] 41.2 [27.6, 44.7] 31.8 [26.2, 36.9] 
# Table 1 with p-values, no mean, no percent missing
cida_table1(data = df,
            includeVars = c("Age", "Sex", "Smoking Status", "IL-8"),
            stratifyBy = "Group",
            group_labels = c("", "Group", ""),
            group_label_span = c(1, 3, 1),
            caption = "TABLE 1",
            footnote = "My table 1",
            exclude_mean = TRUE,
            exclude_missing = TRUE,
            include_total = TRUE,
            compute_pval = TRUE,
            nonParametricVars = NULL,
            exportWord = FALSE)
TABLE 1
Group
Total
(N=12)		 Control
(N=4)		 Heart Disease
(N=4)		 Lung Disease
(N=4)		 p-value

My table 1

p-values computed as follows:

  Parametric

    Numeric data with 2 groups -- t-test

    Numeric data with more than 2 groups -- ANOVA

  Non-parametric

    Numeric data with 2 groups -- Wilcoxon-test

    Numeric data with more than 2 groups -- Kruskal-Wallis test

  Categorical data with any cell value < 5 -- Fishers exact test

  Categerical data with all cell values >= 5 -- Chi-square test of independence
Age
Median [Min, Max] 18.5 [10.0, 33.0] 13.0 [10.0, 18.0] 24.0 [19.0, 33.0] NA [NA, NA] 0.022
Sex
Female 4 (33.3%) 4 (100%) 0 (0%) 0 (0%) 0.029
Male 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
Smoking Status
Current 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%) 0.029
Former 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%)
Never 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
IL-8
Median [Min, Max] 36.2 [20.6, 44.7] 38.1 [20.6, 43.5] 41.2 [27.6, 44.7] 31.8 [26.2, 36.9] 0.501 
# Table 1 styling the output
# You can also rename the variables like this (name in data = new name)
cida_table1(data = df,
            includeVars = c("Age" = "age",
             "Sex" = "sex",
              "Smoking Status" = "smoking",
               "IL-8"= "Interleukin 8"),
            stratifyBy = "Group",
            group_labels = c("", "Group", ""),
            group_label_span = c(1, 3, 1),
            caption = "TABLE 1",
            footnote = "My table 1",
            exclude_mean = TRUE,
            exclude_missing = TRUE,
            include_total = TRUE,
            compute_pval = TRUE,
            nonParametricVars = c("Age", "IL-8"), # Uses original name
            exportWord = FALSE)
TABLE 1
Group
Total
(N=12)		 Control
(N=4)		 Heart Disease
(N=4)		 Lung Disease
(N=4)		 p-value

My table 1

p-values computed as follows:

  Parametric

    Numeric data with 2 groups -- t-test

    Numeric data with more than 2 groups -- ANOVA

  Non-parametric

    Numeric data with 2 groups -- Wilcoxon-test

    Numeric data with more than 2 groups -- Kruskal-Wallis test

  Categorical data with any cell value < 5 -- Fishers exact test

  Categerical data with all cell values >= 5 -- Chi-square test of independence
age
Median [Min, Max] 18.5 [10.0, 33.0] 13.0 [10.0, 18.0] 24.0 [19.0, 33.0] NA [NA, NA] 0.021
sex
Female 4 (33.3%) 4 (100%) 0 (0%) 0 (0%) 0.029
Male 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
smoking
Current 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%) 0.029
Former 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%)
Never 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
Interleukin 8
Median [Min, Max] 36.2 [20.6, 44.7] 38.1 [20.6, 43.5] 41.2 [27.6, 44.7] 31.8 [26.2, 36.9] 0.39 
# Tables are html so you can customize them using html and css
cida_table1(data = df,
            includeVars = c("Age" = "age",
                            "Sex" =
                            "<span style =
                            'background-color:pink;'>sex</span>", # Highlight pink
                            "Smoking Status" =
                            "<span style='color:purple;
                             font-family:Snell Roundhand;
                             font-size:1.5em;
                             font-weight:900;'>smoking status</span>",
                            "IL-8"=
                            "Interleukin 8&beta;"), # &beta; is html for greek lowercase beta
            stratifyBy = "Group",
            group_labels = c("", "Group", ""),
            group_label_span = c(1, 3, 1),
            caption = "TABLE 1",
            footnote = "My table 1",
            exclude_mean = TRUE,
            exclude_missing = TRUE,
            include_total = TRUE,
            compute_pval = TRUE,
            nonParametricVars = NULL,
            exportWord = FALSE)
TABLE 1
Group
Total
(N=12)		 Control
(N=4)		 Heart Disease
(N=4)		 Lung Disease
(N=4)		 p-value

My table 1

p-values computed as follows:

  Parametric

    Numeric data with 2 groups -- t-test

    Numeric data with more than 2 groups -- ANOVA

  Non-parametric

    Numeric data with 2 groups -- Wilcoxon-test

    Numeric data with more than 2 groups -- Kruskal-Wallis test

  Categorical data with any cell value < 5 -- Fishers exact test

  Categerical data with all cell values >= 5 -- Chi-square test of independence
age
Median [Min, Max] 18.5 [10.0, 33.0] 13.0 [10.0, 18.0] 24.0 [19.0, 33.0] NA [NA, NA] 0.022
sex
Female 4 (33.3%) 4 (100%) 0 (0%) 0 (0%) 0.029
Male 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
smoking status
Current 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%) 0.029
Former 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%)
Never 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
Interleukin 8β
Median [Min, Max] 36.2 [20.6, 44.7] 38.1 [20.6, 43.5] 41.2 [27.6, 44.7] 31.8 [26.2, 36.9] 0.501 
# You can add icons if you like

cida_table1(data = df,
            includeVars = c("Age" = "age", "Sex" = "sex",
            "Smoking Status" =
            "<link rel='stylesheet'
href='https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css'>
            <span>Smoking Status </span>
            <i class='fa fa-medkit'></i>", # Medkit icon
            "IL-8"= "Interleukin 8&beta;"),
            stratifyBy = "Group",
            group_labels = c("", "Group", ""),
            group_label_span = c(1, 3, 1),
            caption = "TABLE 1",
            footnote = "My table 1",
            exclude_mean = TRUE,
            exclude_missing = TRUE,
            include_total = TRUE,
            compute_pval = TRUE,
            nonParametricVars = NULL,
            exportWord = FALSE)
TABLE 1
Group
Total
(N=12)		 Control
(N=4)		 Heart Disease
(N=4)		 Lung Disease
(N=4)		 p-value

My table 1

p-values computed as follows:

  Parametric

    Numeric data with 2 groups -- t-test

    Numeric data with more than 2 groups -- ANOVA

  Non-parametric

    Numeric data with 2 groups -- Wilcoxon-test

    Numeric data with more than 2 groups -- Kruskal-Wallis test

  Categorical data with any cell value < 5 -- Fishers exact test

  Categerical data with all cell values >= 5 -- Chi-square test of independence
age
Median [Min, Max] 18.5 [10.0, 33.0] 13.0 [10.0, 18.0] 24.0 [19.0, 33.0] NA [NA, NA] 0.022
sex
Female 4 (33.3%) 4 (100%) 0 (0%) 0 (0%) 0.029
Male 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
Smoking Status
Current 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%) 0.029
Former 2 (16.7%) 2 (50.0%) 0 (0%) 0 (0%)
Never 4 (33.3%) 0 (0%) 4 (100%) 0 (0%)
Interleukin 8β
Median [Min, Max] 36.2 [20.6, 44.7] 38.1 [20.6, 43.5] 41.2 [27.6, 44.7] 31.8 [26.2, 36.9] 0.501