# All tables examples

install.packages("finalfit")

## 1 Cross tables

Two-way tables are used extensively in healthcare research, e.g. a 2x2 table comparing two factors with two levels each, or table 1 from a typical clinical study or trial

The main functions all take a dependent variable - the outcome - and explanatory variables - predictors or exposures (any number categorical or continuous variables).

### 1.01 Default

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory) -> t
label levels No Yes
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3)
Age <40 years 68 (7.5) 2 (7.4)
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1)
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0)
Yes 166 (18.8) 10 (37.0)

Note, chi-squared warnings will be generated when the expected count in any cell is less than 5. Fisher’s exact test can be used as below, or go straight to a univariable logistic regression, e.g. colon_s %>% finalfit(dependent, explanatory)

### 1.02 Add or edit variable labels

library(finalfit)
library(dplyr)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
mutate(
sex.factor = ff_label(sex.factor, "Gender")
) %>%
summary_factorlist(dependent, explanatory) -> t
label levels No Yes
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3)
Age <40 years 68 (7.5) 2 (7.4)
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Gender Female 432 (47.9) 13 (48.1)
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0)
Yes 166 (18.8) 10 (37.0)

### 1.03 P-value for hypothesis test

Defaults are chi-squared for categorical explanatory variables and an F-test for continuous (aov, analysis of variance). Alternatives can be specified as per below.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE) -> t
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.542
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0) 0.035
Yes 166 (18.8) 10 (37.0)

### 1.04 With Fisher’s exact test

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, p_cat = "fisher") -> t
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.542
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0) 0.026
Yes 166 (18.8) 10 (37.0)

### 1.05 Parametric explanatory variables

Summaries for continuous explanatory variables are mean (standard deviation) with aov statistical test by default. The statistical test can be changed to the Welch t-test when there are two dependent variable levels if desired.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, p_cont_para = "t.test") -> t
#> Warning in chisq.test(age.factor, perfor.factor): Chi-squared approximation
#> may be incorrect
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.586
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0) 0.035
Yes 166 (18.8) 10 (37.0)

### 1.06 Non-parametric explanatory variables

If desired, all continuous explanatory variables can be considered non-parametric. Summaries will be median (interquartile range) and the statistical test is Kruskal-Wallis/Mann-Whitney U.

library(finalfit)
explanatory = c("age", "nodes", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont = "median") -> t
label levels No Yes p
Age (years) Median (IQR) 61.0 (16.0) 60.0 (18.0) 0.578
nodes Median (IQR) 2.0 (4.0) 3.0 (2.0) 0.125
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0) 0.035
Yes 166 (18.8) 10 (37.0)

### 1.07 Select specific non-parametric variables

Many have asked in the past if only particular variables can be considered non-parametric. The argument cont_nonpara can take a vector (e.g. c(1, 2, 3, 4)) of values corresponding to the explanatory variable to specify which should be summarised as a median and be passed to a non-parametric test.

library(finalfit)
explanatory = c("age", "nodes", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont_nonpara = c(2)) -> t
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.542
nodes Median (IQR) 2.0 (4.0) 3.0 (2.0) 0.125
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (81.2) 17 (63.0) 0.035
Yes 166 (18.8) 10 (37.0)

### 1.08 Missing values for the explanatory variables

Always consider summarising missing values when describing your data.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE) -> t
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.542
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (79.3) 17 (63.0) 0.035
Yes 166 (18.4) 10 (37.0)
(Missing) 21 (2.3) 0 (0.0)

### 1.09 Pass missing values to statistical tests

This is a change from the default behaviour introduced in Finalfit 1.0.0. Previously, when missing data was presented it was also considered as a level in the statistical test. This may or may not be desired. Control this now using na_to_p = TRUE to include missing data in test. A message is produced reminding you that you are doing that.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
na_to_p = TRUE) -> t
#> Explanatory variable(s) missing data included in hypothesis test (p-value).
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.542
Age <40 years 68 (7.5) 2 (7.4) 1.000
40-59 years 334 (37.0) 10 (37.0)
60+ years 500 (55.4) 15 (55.6)
Sex Female 432 (47.9) 13 (48.1) 1.000
Male 470 (52.1) 14 (51.9)
Obstruction No 715 (79.3) 17 (63.0) 0.042
Yes 166 (18.4) 10 (37.0)
(Missing) 21 (2.3) 0 (0.0)

### 1.10 Row proportions (rather than column)

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
column = FALSE) -> t
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.9) 58.4 (13.3) 0.542
Age <40 years 68 (97.1) 2 (2.9) 1.000
40-59 years 334 (97.1) 10 (2.9)
60+ years 500 (97.1) 15 (2.9)
Sex Female 432 (97.1) 13 (2.9) 1.000
Male 470 (97.1) 14 (2.9)
Obstruction No 715 (97.7) 17 (2.3) 0.035
Yes 166 (94.3) 10 (5.7)
(Missing) 21 (100.0) 0 (0.0)

### 1.11 Total column

The terms total column was introduced before this function summarised continuous variables. It would be better to be “All data” or something similar, as the continous explanatory variables a summary statistic is produced for all data. However, to keep backwards compatibility we have left it unchanged for now. For producing row totals including continous explanatory variables, see add_row_total below.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont = "median", na_include = TRUE,
column = TRUE, total_col = TRUE) -> t
label levels No Yes Total p
Age (years) Median (IQR) 61.0 (16.0) 60.0 (18.0) 61.0 (16.0) 0.578
Age <40 years 68 (7.5) 2 (7.4) 70 (7.5) 1.000
40-59 years 334 (37.0) 10 (37.0) 344 (37.0)
60+ years 500 (55.4) 15 (55.6) 515 (55.4)
Sex Female 432 (47.9) 13 (48.1) 445 (47.9) 1.000
Male 470 (52.1) 14 (51.9) 484 (52.1)
Obstruction No 715 (79.3) 17 (63.0) 732 (78.8) 0.035
Yes 166 (18.4) 10 (37.0) 176 (18.9)
(Missing) 21 (2.3) 0 (0.0) 21 (2.3)

### 1.12 Row totals with missing

This was introduced to deal with the problem of summarising missing data for continuous variables. By default, it provides the total N for the variable and includes a column enumerating missing values.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
total_col = TRUE,
add_row_total = TRUE) -> t
label Total N Missing N levels No Yes Total p
Age (years) 929 0 Mean (SD) 59.8 (11.9) 58.4 (13.3) 59.8 (11.9) 0.542
Age 929 0 <40 years 68 (7.5) 2 (7.4) 70 (7.5) 1.000
40-59 years 334 (37.0) 10 (37.0) 344 (37.0)
60+ years 500 (55.4) 15 (55.6) 515 (55.4)
Sex 929 0 Female 432 (47.9) 13 (48.1) 445 (47.9) 1.000
Male 470 (52.1) 14 (51.9) 484 (52.1)
Obstruction 908 21 No 715 (79.3) 17 (63.0) 732 (78.8) 0.035
Yes 166 (18.4) 10 (37.0) 176 (18.9)
(Missing) 21 (2.3) 0 (0.0) 21 (2.3)

### 1.13 Row totals without missing

Remove missing column.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
total_col = TRUE,
include_row_missing_col = FALSE) -> t
label Total N levels No Yes Total p
Age (years) 929 Mean (SD) 59.8 (11.9) 58.4 (13.3) 59.8 (11.9) 0.542
Age 929 <40 years 68 (7.5) 2 (7.4) 70 (7.5) 1.000
40-59 years 334 (37.0) 10 (37.0) 344 (37.0)
60+ years 500 (55.4) 15 (55.6) 515 (55.4)
Sex 929 Female 432 (47.9) 13 (48.1) 445 (47.9) 1.000
Male 470 (52.1) 14 (51.9) 484 (52.1)
Obstruction 908 No 715 (79.3) 17 (63.0) 732 (78.8) 0.035
Yes 166 (18.4) 10 (37.0) 176 (18.9)
(Missing) 21 (2.3) 0 (0.0) 21 (2.3)

### 1.14 Row totals with user-specified column names

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
total_col = TRUE,
row_totals_colname = "N (total)",
row_missing_colname = "N (missing)") -> t
label N (total) N (missing) levels No Yes Total p
Age (years) 929 0 Mean (SD) 59.8 (11.9) 58.4 (13.3) 59.8 (11.9) 0.542
Age 929 0 <40 years 68 (7.5) 2 (7.4) 70 (7.5) 1.000
40-59 years 334 (37.0) 10 (37.0) 344 (37.0)
60+ years 500 (55.4) 15 (55.6) 515 (55.4)
Sex 929 0 Female 432 (47.9) 13 (48.1) 445 (47.9) 1.000
Male 470 (52.1) 14 (51.9) 484 (52.1)
Obstruction 908 21 No 715 (81.2) 17 (63.0) 732 (80.6) 0.035
Yes 166 (18.8) 10 (37.0) 176 (19.4)

### 1.15 Order a variable by total

This is intended for when there is only one explanatory variable.

library(finalfit)
explanatory = c("extent.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont = "median", na_include = TRUE,
column = TRUE, total_col = TRUE, orderbytotal = TRUE) -> t
label levels No Yes Total p
Extent of spread Serosa 736 (81.6) 23 (85.2) 759 (81.7) 0.200
Muscle 105 (11.6) 1 (3.7) 106 (11.4)
Adjacent structures 40 (4.4) 3 (11.1) 43 (4.6)
Submucosa 21 (2.3) 0 (0.0) 21 (2.3)

Column totals can be added, and by default are presented with a row percentage.

explanatory = c("age.factor", "sex.factor")
dependent = "rx.factor"

colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
add_col_totals = TRUE) -> t
label levels Obs Lev Lev+5FU p
Total N (%) 315 (33.9) 310 (33.4) 304 (32.7)
Age <40 years 25 (7.9) 19 (6.1) 26 (8.6) 0.572
40-59 years 124 (39.4) 115 (37.1) 105 (34.5)
60+ years 166 (52.7) 176 (56.8) 173 (56.9)
Sex Female 149 (47.3) 133 (42.9) 163 (53.6) 0.028
Male 166 (52.7) 177 (57.1) 141 (46.4)

### 1.18 Add column totals without proportion.

explanatory = c("age.factor", "sex.factor")
dependent = "rx.factor"

colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
include_col_totals_percent = FALSE)    -> t
label levels Obs Lev Lev+5FU p
Total N 315 310 304
Age <40 years 25 (7.9) 19 (6.1) 26 (8.6) 0.572
40-59 years 124 (39.4) 115 (37.1) 105 (34.5)
60+ years 166 (52.7) 176 (56.8) 173 (56.9)
Sex Female 149 (47.3) 133 (42.9) 163 (53.6) 0.028
Male 166 (52.7) 177 (57.1) 141 (46.4)

### 1.19 Add column totals with user-specified row name and prefix.

explanatory = c("age.factor", "sex.factor")
dependent = "rx.factor"

colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
include_col_totals_percent = FALSE,
col_totals_rowname = "",
col_totals_prefix = "N=")  -> t
label levels Obs Lev Lev+5FU p
N=315 N=310 N=304
Age <40 years 25 (7.9) 19 (6.1) 26 (8.6) 0.572
40-59 years 124 (39.4) 115 (37.1) 105 (34.5)
60+ years 166 (52.7) 176 (56.8) 173 (56.9)
Sex Female 149 (47.3) 133 (42.9) 163 (53.6) 0.028
Male 166 (52.7) 177 (57.1) 141 (46.4)

### 1.20 Label with dependent name

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont = "median", na_include = TRUE,
column = TRUE, total_col = TRUE, add_dependent_label = TRUE) -> t
Dependent: Perforation No Yes Total p
Age (years) Median (IQR) 61.0 (16.0) 60.0 (18.0) 61.0 (16.0) 0.578
Age <40 years 68 (7.5) 2 (7.4) 70 (7.5) 1.000
40-59 years 334 (37.0) 10 (37.0) 344 (37.0)
60+ years 500 (55.4) 15 (55.6) 515 (55.4)
Sex Female 432 (47.9) 13 (48.1) 445 (47.9) 1.000
Male 470 (52.1) 14 (51.9) 484 (52.1)
Obstruction No 715 (79.3) 17 (63.0) 732 (78.8) 0.035
Yes 166 (18.4) 10 (37.0) 176 (18.9)
(Missing) 21 (2.3) 0 (0.0) 21 (2.3)

The dependent name cannot be passed directly to the table intentionally. This is to avoid errors when code is copied and the name is not updated. Change the dependent label using the following. The prefix (“Dependent:”) and any suffix can be altered.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
dplyr::mutate(
perfor.factor = ff_label(perfor.factor, "Perforated cancer")
) %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont = "median", na_include = TRUE,
column = TRUE, total_col = TRUE, add_dependent_label = TRUE, dependent_label_prefix = "") -> t
Perforated cancer No Yes Total p
Age (years) Median (IQR) 61.0 (16.0) 60.0 (18.0) 61.0 (16.0) 0.578
Age <40 years 68 (7.5) 2 (7.4) 70 (7.5) 1.000
40-59 years 334 (37.0) 10 (37.0) 344 (37.0)
60+ years 500 (55.4) 15 (55.6) 515 (55.4)
Sex Female 432 (47.9) 13 (48.1) 445 (47.9) 1.000
Male 470 (52.1) 14 (51.9) 484 (52.1)
Obstruction No 715 (79.3) 17 (63.0) 732 (78.8) 0.035
Yes 166 (18.4) 10 (37.0) 176 (18.9)
(Missing) 21 (2.3) 0 (0.0) 21 (2.3)

### 1.21 Dependent variable with any number of factor levels supported

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "extent.factor"
colon_s %>%
dplyr::mutate(
perfor.factor = ff_label(perfor.factor, "Perforated cancer")
) %>%
summary_factorlist(dependent, explanatory, p = TRUE, cont = "median", na_include = TRUE,
column = TRUE, total_col = TRUE, add_dependent_label = TRUE, dependent_label_prefix = "") -> t
Age (years) Median (IQR) 56.0 (14.0) 61.5 (14.0) 61.0 (16.0) 61.0 (12.5) 61.0 (16.0) 0.334
Age <40 years 2 (9.5) 8 (7.5) 56 (7.4) 4 (9.3) 70 (7.5) 0.338
40-59 years 12 (57.1) 32 (30.2) 285 (37.5) 15 (34.9) 344 (37.0)
60+ years 7 (33.3) 66 (62.3) 418 (55.1) 24 (55.8) 515 (55.4)
Sex Female 13 (61.9) 47 (44.3) 366 (48.2) 19 (44.2) 445 (47.9) 0.483
Male 8 (38.1) 59 (55.7) 393 (51.8) 24 (55.8) 484 (52.1)
Obstruction No 20 (95.2) 88 (83.0) 588 (77.5) 36 (83.7) 732 (78.8) 0.040
Yes 1 (4.8) 13 (12.3) 157 (20.7) 5 (11.6) 176 (18.9)
(Missing) 0 (0.0) 5 (4.7) 14 (1.8) 2 (4.7) 21 (2.3)

### 1.22 Missing data in the dependent

If you are careful to count totals and you know your data, you should realise when there is data missing from the dependent, e.g.:

explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
ff_glimpse(dependent, explanatory)
#> $Continuous #> label var_type n missing_n missing_percent mean sd min #> age Age (years) <dbl> 929 0 0.0 59.8 11.9 18.0 #> quartile_25 median quartile_75 max #> age 53.0 61.0 69.0 85.0 #> #>$Categorical
#>                            label var_type   n missing_n missing_percent
#> mort_5yr        Mortality 5 year    <fct> 915        14             1.5
#> age.factor                   Age    <fct> 929         0             0.0
#> sex.factor                   Sex    <fct> 929         0             0.0
#> obstruct.factor      Obstruction    <fct> 908        21             2.3
#>                 levels_n                                  levels
#> mort_5yr               2            "Alive", "Died", "(Missing)"
#> age.factor             3 "<40 years", "40-59 years", "60+ years"
#> sex.factor             2                        "Female", "Male"
#> obstruct.factor        2                "No", "Yes", "(Missing)"
#>                 levels_count   levels_percent
#> mort_5yr        511, 404, 14 55.0, 43.5,  1.5
#> age.factor      70, 344, 515  7.5, 37.0, 55.4
#> sex.factor          445, 484           48, 52
#> obstruct.factor 732, 176, 21 78.8, 18.9,  2.3

To make sure, a warning is generated when data are dropped from the dependent:

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
total_col = TRUE,
#> Note: dependent includes missing data. These are dropped.
label Total N Missing N levels Alive Died Total p
Total N (%) 511 (55.8) 404 (44.2) 915
Age (years) 915 0 Mean (SD) 59.8 (11.4) 59.9 (12.5) 59.8 (11.9) 0.986
Age 915 0 <40 years 31 (6.1) 36 (8.9) 67 (7.3) 0.020
40-59 years 208 (40.7) 131 (32.4) 339 (37.0)
60+ years 272 (53.2) 237 (58.7) 509 (55.6)
Sex 915 0 Female 243 (47.6) 194 (48.0) 437 (47.8) 0.941
Male 268 (52.4) 210 (52.0) 478 (52.2)
Obstruction 894 21 No 408 (79.8) 312 (77.2) 720 (78.7) 0.219
Yes 89 (17.4) 85 (21.0) 174 (19.0)
(Missing) 14 (2.7) 7 (1.7) 21 (2.3)

You may consider making the missing data explicit.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
mutate(
mort_5yr = forcats::fct_explicit_na(mort_5yr)
) %>%
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
total_col = TRUE,
add_col_totals = TRUE, add_row_totals = TRUE) -> t
label Total N Missing N levels Alive Died (Missing) Total p
Total N (%) 511 (55.0) 404 (43.5) 14 (1.5) 929
Age (years) 929 0 Mean (SD) 59.8 (11.4) 59.9 (12.5) 53.9 (12.7) 59.8 (11.9) 0.185
Age 929 0 <40 years 31 (6.1) 36 (8.9) 3 (21.4) 70 (7.5) 0.018
40-59 years 208 (40.7) 131 (32.4) 5 (35.7) 344 (37.0)
60+ years 272 (53.2) 237 (58.7) 6 (42.9) 515 (55.4)
Sex 929 0 Female 243 (47.6) 194 (48.0) 8 (57.1) 445 (47.9) 0.776
Male 268 (52.4) 210 (52.0) 6 (42.9) 484 (52.1)
Obstruction 908 21 No 408 (79.8) 312 (77.2) 12 (85.7) 732 (78.8) 0.373
Yes 89 (17.4) 85 (21.0) 2 (14.3) 176 (18.9)
(Missing) 14 (2.7) 7 (1.7) 0 (0.0) 21 (2.3)

### 1.23 Directly include missing data in dependent

Rather than making the data explicit in the dataset, you can use na_include_dependent = TRUE to do the same in summary_factorlist().

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
na_include = TRUE, na_include_dependent = TRUE,
total_col = TRUE,
add_col_totals = TRUE, add_row_totals = TRUE) -> t
label Total N Missing N levels Alive Died (Missing) Total p
Total N (%) 511 (55.0) 404 (43.5) 14 (1.5) 929
Age (years) 915 0 Mean (SD) 59.8 (11.4) 59.9 (12.5) 53.9 (12.7) 59.8 (11.9) 0.986
Age 915 0 <40 years 31 (6.1) 36 (8.9) 3 (21.4) 70 (7.5) 0.020
40-59 years 208 (40.7) 131 (32.4) 5 (35.7) 344 (37.0)
60+ years 272 (53.2) 237 (58.7) 6 (42.9) 515 (55.4)
Sex 915 0 Female 243 (47.6) 194 (48.0) 8 (57.1) 445 (47.9) 0.941
Male 268 (52.4) 210 (52.0) 6 (42.9) 484 (52.1)
Obstruction 894 21 No 408 (79.8) 312 (77.2) 12 (85.7) 732 (78.8) 0.219
Yes 89 (17.4) 85 (21.0) 2 (14.3) 176 (18.9)
(Missing) 14 (2.7) 7 (1.7) 0 (0.0) 21 (2.3)

### 1.24 Summarise complete cases

You may wish to see summaries for complete cases across included variables. Rather than selecting including variables and drop_na(), you can pass na_complete_cases = TRUE to summary_factorlist() to do the same.

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
na_complete_cases = TRUE,
total_col = TRUE,
#> Note: dependent includes missing data. These are dropped.
label Total N Missing N levels Alive Died Total p
Total N (%) 497 (55.6) 397 (44.4) 894
Age (years) 894 21 Mean (SD) 59.7 (11.5) 59.9 (12.5) 59.7 (11.9) 0.986
Age 894 21 <40 years 31 (6.2) 35 (8.8) 66 (7.4) 0.020
40-59 years 203 (40.8) 129 (32.5) 332 (37.1)
60+ years 263 (52.9) 233 (58.7) 496 (55.5)
Sex 894 21 Female 237 (47.7) 192 (48.4) 429 (48.0) 0.941
Male 260 (52.3) 205 (51.6) 465 (52.0)
Obstruction 894 21 No 408 (82.1) 312 (78.6) 720 (80.5) 0.219
Yes 89 (17.9) 85 (21.4) 174 (19.5)

### 1.25 Actively dropping missing data (and tidyverse functions that strip attributes)

You may wish to actively remove any rows with missing data, so you are explicit around which data are being used in models. Unfortunately some tidyverse functions silently remove variable attributes (labels). This is complained about then put right. But here is a workaround if it is happening with a variable you wish to use, such as tidyr::drop_na().

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
vlabels = colon_s %>%
extract_variable_label()

colon_s %>%
select(dependent, explanatory) %>%
tidyr::drop_na() %>% # Silently removes attributes
ff_relabel(vlabels) %>% # Relabel
summary_factorlist(dependent, explanatory, p = TRUE, na_include = TRUE,
total_col = TRUE,
#> Note: Using an external vector in selections is ambiguous.
#> ℹ Use all_of(dependent) instead of dependent to silence this message.
#> ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
#> This message is displayed once per session.
#> Note: Using an external vector in selections is ambiguous.
#> ℹ Use all_of(explanatory) instead of explanatory to silence this message.
#> ℹ See <https://tidyselect.r-lib.org/reference/faq-external-vector.html>.
#> This message is displayed once per session.
label Total N Missing N levels Alive Died Total p
Total N (%) 497 (55.6) 397 (44.4) 894
Age (years) 894 0 Mean (SD) 59.7 (11.5) 59.9 (12.5) 59.7 (11.9) 0.791
Age 894 0 <40 years 31 (6.2) 35 (8.8) 66 (7.4) 0.024
40-59 years 203 (40.8) 129 (32.5) 332 (37.1)
60+ years 263 (52.9) 233 (58.7) 496 (55.5)
Sex 894 0 Female 237 (47.7) 192 (48.4) 429 (48.0) 0.894
Male 260 (52.3) 205 (51.6) 465 (52.0)
Obstruction 894 0 No 408 (82.1) 312 (78.6) 720 (80.5) 0.219
Yes 89 (17.9) 85 (21.4) 174 (19.5)

### 1.26 Explanatory variable defaults to factor when ≤5 distinct values

library(finalfit)

# Here, extent is a continuous variable with 4 distinct values.
# Any continuous variable with 5 or fewer unique values is converted silently to factor
# e.g.
explanatory = c("extent")
dependent = "mort_5yr"
colon_s %>%
summary_factorlist(dependent, explanatory) -> t
label levels Alive Died
extent 1 16 (3.1) 4 (1.0)
2 78 (15.3) 25 (6.2)
3 401 (78.5) 349 (86.4)
4 16 (3.1) 26 (6.4)

### 1.27 Keep as continous variable when ≤5 distinct values

library(finalfit)
explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE,
digits = c(1,2,3,4)) -> t
#> Note: dependent includes missing data. These are dropped.
label levels Alive Died p
Age (years) Mean (SD) 59.8 (11.44) 59.9 (12.51) 0.986
Age <40 years 31 (6.0665) 36 (8.9109) 0.020
40-59 years 208 (40.7045) 131 (32.4257)
60+ years 272 (53.2290) 237 (58.6634)
Sex Female 243 (47.5538) 194 (48.0198) 0.941
Male 268 (52.4462) 210 (51.9802)
Obstruction No 408 (82.0926) 312 (78.5894) 0.219
Yes 89 (17.9074) 85 (21.4106)

### 1.28 Stratified crosstables

I’ve been meaning to include support for table stratification for a while. I have delayed for a good reason. Perhaps the most straightforward way to implement stratificiation is with dplyr::group_by(). However, the non-standard evaluation required for multiple strata may confuse as it is not implemented else where in the package.

This translates to whether variable names are passed in quotes or not.

Here is a solution, which while not that pretty, is effective.

Note that tidyverse functions every so often start stripping labels/attributes. Hence the addition of the help function.

library(dplyr)
explanatory = c("age.factor", "sex.factor")
dependent = "perfor.factor"

# Pick option below
split = "rx.factor"
split = c("rx.factor", "node4.factor")

# Piped function to generate stratified crosstabs table
colon_s %>%
group_by(!!! syms(split)) %>% # Looks awkward, but avoids unquoted var names
group_modify(~ summary_factorlist(.x, dependent, explanatory)) %>%
ff_stratify_helper(colon_s) -> t
Treatment >4 positive nodes label levels No Yes
Obs No <40 years 14 (6.3) 0 (0.0)
Obs No 40-59 years 89 (40.3) 3 (42.9)
Obs No 60+ years 118 (53.4) 4 (57.1)
Obs No Female 101 (45.7) 3 (42.9)
Obs No Male 120 (54.3) 4 (57.1)
Obs Yes <40 years 10 (11.8) 1 (50.0)
Obs Yes 40-59 years 31 (36.5) 1 (50.0)
Obs Yes 60+ years 44 (51.8) 0 (0.0)
Obs Yes Female 44 (51.8) 1 (50.0)
Obs Yes Male 41 (48.2) 1 (50.0)
Lev No <40 years 14 (6.5) 0 (0.0)
Lev No 40-59 years 78 (36.3) 3 (50.0)
Lev No 60+ years 123 (57.2) 3 (50.0)
Lev No Female 89 (41.4) 2 (33.3)
Lev No Male 126 (58.6) 4 (66.7)
Lev Yes <40 years 4 (4.7) 1 (25.0)
Lev Yes 40-59 years 33 (38.8) 1 (25.0)
Lev Yes 60+ years 48 (56.5) 2 (50.0)
Lev Yes Female 39 (45.9) 3 (75.0)
Lev Yes Male 46 (54.1) 1 (25.0)
Lev+5FU No <40 years 15 (6.9) 0 (0.0)
Lev+5FU No 40-59 years 72 (33.2) 2 (25.0)
Lev+5FU No 60+ years 130 (59.9) 6 (75.0)
Lev+5FU No Female 115 (53.0) 4 (50.0)
Lev+5FU No Male 102 (47.0) 4 (50.0)
Lev+5FU Yes <40 years 11 (13.9) 0 (NaN)
Lev+5FU Yes 40-59 years 31 (39.2) 0 (NaN)
Lev+5FU Yes 60+ years 37 (46.8) 0 (NaN)
Lev+5FU Yes Female 44 (55.7) 0 (NaN)
Lev+5FU Yes Male 35 (44.3) 0 (NaN)

### 1.29 Digits / decimal places

explanatory = c("age", "age.factor", "sex.factor", "obstruct.factor")
dependent = "perfor.factor"
colon_s %>%
summary_factorlist(dependent, explanatory, p = TRUE, digits = c(1,2,3,4)) -> t
label levels No Yes p
Age (years) Mean (SD) 59.8 (11.91) 58.4 (13.30) 0.542
Age <40 years 68 (7.5388) 2 (7.4074) 1.000
40-59 years 334 (37.0288) 10 (37.0370)
60+ years 500 (55.4324) 15 (55.5556)
Sex Female 432 (47.8936) 13 (48.1481) 1.000
Male 470 (52.1064) 14 (51.8519)
Obstruction No 715 (81.1578) 17 (62.9630) 0.035
Yes 166 (18.8422) 10 (37.0370)

## 2 Model tables with finalfit()

### 2.01 Default

Logistic regression first.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770)

### 2.02 Hide reference levels

Most appropriate when all explanatory variables are continuous or well-known binary variables, such as sex.

library(finalfit)
explanatory = c("age", "sex.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, add_dependent_label = FALSE) %>%
ff_remove_ref() %>%
dependent_label(colon_s, dependent)-> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age (years) Mean (SD) 59.8 (11.4) 59.9 (12.5) 1.00 (0.99-1.01, p=0.986) 1.00 (0.99-1.01, p=0.983)
Sex Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.76-1.27, p=0.888)

### 2.03 Model metrics

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, metrics = TRUE) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770)
 Number in dataframe = 929, Number in model = 894, Missing = 35, AIC = 1230.7, C-statistic = 0.56, H&L = Chi-sq(8) 5.69 (p=0.682)

### 2.04 Model metrics can be applied to all supported base models

library(finalfit)
glm(mort_5yr ~ age.factor + sex.factor + obstruct.factor + perfor.factor, data = colon_s, family = "binomial") %>%
ff_metrics() -> t
 Number in dataframe = 929, Number in model = 894, Missing = 35, AIC = 1230.7, C-statistic = 0.56, H&L = Chi-sq(8) 5.69 (p=0.682)

### 2.05 Reduced model

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
explanatory_multi = c("age.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, explanatory_multi) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.424)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) -
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.26 (0.90-1.76, p=0.176)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) -

### 2.06 Include all models

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
explanatory_multi = c("age.factor", "obstruct.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, explanatory_multi, metrics = TRUE, keep_models = TRUE) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable) OR (multivariable reduced)
Age <40 years 31 (46.3) 36 (53.7) - - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426) 0.81 (0.48-1.36, p=0.424)
Sex Female 243 (55.6) 194 (44.4) - - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902) -
Obstruction No 408 (56.7) 312 (43.3) - - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186) 1.26 (0.90-1.76, p=0.176)
Perforation No 497 (56.0) 391 (44.0) - - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770) -
 Number in dataframe = 929, Number in model = 894, Missing = 35, AIC = 1230.7, C-statistic = 0.56, H&L = Chi-sq(8) 5.69 (p=0.682)
 Number in dataframe = 929, Number in model = 894, Missing = 35, AIC = 1226.8, C-statistic = 0.555, H&L = Chi-sq(8) 0.06 (p=1.000)

### 2.06 Interactions

Interactions can be specified in the normal way. Formatting the output is trickier. At the moment, we have left the default model output. This can be adjusted as necessary.

library(finalfit)
explanatory = c("age.factor*sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.65 (0.32-1.34, p=0.241) 0.66 (0.32-1.36, p=0.258)
60+ years 272 (53.4) 237 (46.6) 0.80 (0.40-1.61, p=0.529) 0.85 (0.42-1.71, p=0.647)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 1.24 (0.47-3.30, p=0.665) 1.17 (0.44-3.15, p=0.752)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.26 (0.90-1.76, p=0.182)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.11 (0.50-2.41, p=0.795)
age.factor40-59 years:sex.factorMale Interaction - - 0.68 (0.23-1.97, p=0.479) 0.74 (0.25-2.18, p=0.588)
age.factor60+ years:sex.factorMale Interaction - - 0.86 (0.30-2.39, p=0.766) 0.89 (0.31-2.51, p=0.822)

### 2.07 Interactions: create interaction variable with two factors

library(finalfit)
#explanatory = c("age.factor*sex.factor", "obstruct.factor", "perfor.factor")
explanatory = c("obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
ff_interaction(age.factor, sex.factor) %>%
finalfit(dependent, c(explanatory, "age.factor__sex.factor")) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.26 (0.90-1.76, p=0.182)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.11 (0.50-2.41, p=0.795)
Age:Sex <40 years|Female 18 (48.6) 19 (51.4) - -
<40 years|Male 13 (43.3) 17 (56.7) 1.24 (0.47-3.30, p=0.665) 1.17 (0.44-3.15, p=0.752)
40-59 years|Female 96 (59.3) 66 (40.7) 0.65 (0.32-1.34, p=0.241) 0.66 (0.32-1.36, p=0.258)
40-59 years|Male 112 (63.3) 65 (36.7) 0.55 (0.27-1.12, p=0.100) 0.57 (0.28-1.18, p=0.129)
60+ years|Female 129 (54.2) 109 (45.8) 0.80 (0.40-1.61, p=0.529) 0.85 (0.42-1.71, p=0.647)
60+ years|Male 143 (52.8) 128 (47.2) 0.85 (0.42-1.69, p=0.638) 0.88 (0.44-1.77, p=0.725)

### 2.08 Dependent name

The dependent name cannot be specified directly intentionally. This is to prevent errors when copying code. Re-label using ff_label(). The dependent prefix and suffix can also be altered.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
dplyr::mutate(
mort_5yr = ff_label(mort_5yr, "5-year mortality")
) %>%
finalfit(dependent, explanatory, dependent_label_prefix = "",
dependent_label_suffix = " (full model)") -> t
5-year mortality (full model) Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770)

### 2.09 Estimate name

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, estimate_name = "Odds ratio") -> t
Dependent: Mortality 5 year Alive Died Odds ratio (univariable) Odds ratio (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770)

### 2.10 Digits / decimal places

Number of digits to round to regression results. (1) estimate, (2) confidence interval limits, (3) p-value. Default is c(2,2,3). Trailing zeros are preserved. Number of decimal places for counts and mean (sd) / median (IQR) not currently supported. Defaults are senisble :)

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, digits = c(3,3,4)) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.542 (0.319-0.918, p=0.0230) 0.574 (0.335-0.978, p=0.0412)
60+ years 272 (53.4) 237 (46.6) 0.750 (0.448-1.250, p=0.2704) 0.810 (0.481-1.360, p=0.4261)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.981 (0.756-1.275, p=0.8886) 0.983 (0.754-1.283, p=0.9023)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.249 (0.896-1.741, p=0.1892) 1.255 (0.896-1.757, p=0.1859)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.180 (0.542-2.553, p=0.6716) 1.122 (0.512-2.442, p=0.7699)

### 2.11 Confidence interval type

One of c("profile", "default") for GLM models (confint.glm()). Note, a little awkwardly, the ‘default’ setting is profile, rather than default. Profile levels are probably a little more accurate. Only go to default if taking a significant length of time for profile, i.e. data is greater than hundreds of thousands of lines.

For glmer/lmer models (confint.merMod()), c("profile", "Wald", "boot"). Not implemented for lm(), coxph() or coxphlist, which use default.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, confint_type = "default") -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.55-2.54, p=0.672) 1.12 (0.52-2.43, p=0.770)

### 2.12 Confidence interval level

Probably never change this :) Note, the p-value is intentionally not included for confidence levels other than 95% to avoid confusion.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, confint_level = 0.90) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.35-0.84) 0.57 (0.37-0.90)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.49-1.15) 0.81 (0.52-1.25)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.79-1.22) 0.98 (0.79-1.23)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.95-1.65) 1.25 (0.95-1.66)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.62-2.25) 1.12 (0.58-2.15)

### 2.13 Confidence interval separation

Some like to avoid the hyphen so as not to confuse with minus sign. Obviously not an issue in logistic regression.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory, confint_sep = " to ") -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32 to 0.92, p=0.023) 0.57 (0.34 to 0.98, p=0.041)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45 to 1.25, p=0.270) 0.81 (0.48 to 1.36, p=0.426)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76 to 1.27, p=0.889) 0.98 (0.75 to 1.28, p=0.902)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90 to 1.74, p=0.189) 1.25 (0.90 to 1.76, p=0.186)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54 to 2.55, p=0.672) 1.12 (0.51 to 2.44, p=0.770)

### 2.14 Remove p-value

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
colon_s %>%
finalfit(dependent, explanatory) %>%
ff_remove_p() -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92) 0.57 (0.34-0.98)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25) 0.81 (0.48-1.36)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27) 0.98 (0.75-1.28)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74) 1.25 (0.90-1.76)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55) 1.12 (0.51-2.44)

### 2.15 Mixed effects random-intercept model

At its simplest, a random-intercept model can be specified using a single quoted variable. In this example, it is the equivalent of quoting random_effect = "(1 | hospital)".

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
random_effect = "hospital"
colon_s %>%
finalfit(dependent, explanatory, random_effect = random_effect,
dependent_label_suffix = " (random intercept)") -> t
Dependent: Mortality 5 year (random intercept) Alive Died OR (univariable) OR (multilevel)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.75 (0.39-1.44, p=0.382)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 1.03 (0.55-1.96, p=0.916)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.80 (0.58-1.11, p=0.180)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.23 (0.82-1.83, p=0.320)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.03 (0.43-2.51, p=0.940)

### 2.16 Mixed effects random-slope model

In the example below, allow the effect of age on outcome to vary by hospital. Note, this specification must have parentheses included.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
random_effect = "(age.factor | hospital)"
colon_s %>%
finalfit(dependent, explanatory, random_effect = random_effect,
dependent_label_suffix = " (random slope: age)") -> t
Dependent: Mortality 5 year (random slope: age) Alive Died OR (univariable) OR (multilevel)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.54 (0.32-0.92, p=0.023) 0.81 (0.37-1.81, p=0.611)
60+ years 272 (53.4) 237 (46.6) 0.75 (0.45-1.25, p=0.270) 1.08 (0.54-2.20, p=0.822)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 0.98 (0.76-1.27, p=0.889) 0.80 (0.58-1.11, p=0.179)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.24 (0.83-1.85, p=0.298)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.02 (0.42-2.48, p=0.967)

### 2.17 Mixed effects random-slope model directly from lme4

Clearly, as models get more complex, parameters such as random effect group variances may require to be extracted directly from model outputs.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
random_effect = "(age.factor | hospital)"
colon_s %>%
lme4::glmer(mort_5yr ~ age.factor + (age.factor | hospital), family = "binomial", data = .) %>%
broom::tidy() -> t

### 2.18 Exclude all missing data in final model from univariable analyses

This can be useful if you want the numbers in the final table to match the final multivariable model. However, be careful to include a full explanation of this in the methods and the reason for exluding the missing data.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = 'mort_5yr'
colon_s %>%
dplyr::select(explanatory, dependent) %>%
na.omit() %>%
finalfit(dependent, explanatory) -> t
Dependent: mort_5yr Alive Died OR (univariable) OR (multivariable)
age.factor <40 years 31 (47.0) 35 (53.0) - -
40-59 years 203 (61.1) 129 (38.9) 0.56 (0.33-0.96, p=0.034) 0.57 (0.34-0.98, p=0.041)
60+ years 263 (53.0) 233 (47.0) 0.78 (0.47-1.31, p=0.356) 0.81 (0.48-1.36, p=0.426)
sex.factor Female 237 (55.2) 192 (44.8) - -
Male 260 (55.9) 205 (44.1) 0.97 (0.75-1.27, p=0.841) 0.98 (0.75-1.28, p=0.902)
obstruct.factor No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
perfor.factor No 483 (55.7) 384 (44.3) - -
Yes 14 (51.9) 13 (48.1) 1.17 (0.54-2.53, p=0.691) 1.12 (0.51-2.44, p=0.770)

### 2.19 Linear regression

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = 'nodes'
colon_s %>%
finalfit(dependent, explanatory) -> t
Dependent: nodes unit value Coefficient (univariable) Coefficient (multivariable)
Age <40 years Mean (sd) 4.7 (4.5) - -
40-59 years Mean (sd) 3.6 (3.3) -1.14 (-2.08 to -0.21, p=0.016) -1.21 (-2.16 to -0.26, p=0.012)
60+ years Mean (sd) 3.6 (3.6) -1.19 (-2.10 to -0.28, p=0.010) -1.25 (-2.18 to -0.33, p=0.008)
Sex Female Mean (sd) 3.7 (3.6) - -
Male Mean (sd) 3.6 (3.6) -0.14 (-0.60 to 0.33, p=0.565) -0.07 (-0.54 to 0.40, p=0.779)
Obstruction No Mean (sd) 3.7 (3.7) - -
Yes Mean (sd) 3.5 (3.2) -0.24 (-0.83 to 0.36, p=0.435) -0.31 (-0.91 to 0.29, p=0.313)
Perforation No Mean (sd) 3.7 (3.6) - -
Yes Mean (sd) 3.9 (2.8) 0.24 (-1.13 to 1.61, p=0.735) 0.28 (-1.09 to 1.66, p=0.686)

### 2.20 Mixed effects random-intercept linear regression

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "nodes"
random_effect = "hospital"
colon_s %>%
finalfit(dependent, explanatory, random_effect = random_effect,
dependent_label_suffix = " (random intercept)") -> t
Dependent: nodes (random intercept) unit value Coefficient (univariable) Coefficient (multilevel)
Age <40 years Mean (sd) 4.7 (4.5) - -
40-59 years Mean (sd) 3.6 (3.3) -1.14 (-2.08 to -0.21, p=0.016) -0.79 (-1.65 to 0.07, p=0.035)
60+ years Mean (sd) 3.6 (3.6) -1.19 (-2.10 to -0.28, p=0.010) -0.98 (-1.81 to -0.14, p=0.011)
Sex Female Mean (sd) 3.7 (3.6) - -
Male Mean (sd) 3.6 (3.6) -0.14 (-0.60 to 0.33, p=0.565) -0.19 (-0.62 to 0.24, p=0.195)
Obstruction No Mean (sd) 3.7 (3.7) - -
Yes Mean (sd) 3.5 (3.2) -0.24 (-0.83 to 0.36, p=0.435) -0.37 (-0.92 to 0.17, p=0.091)
Perforation No Mean (sd) 3.7 (3.6) - -
Yes Mean (sd) 3.9 (2.8) 0.24 (-1.13 to 1.61, p=0.735) 0.23 (-1.01 to 1.48, p=0.357)

### 2.21 Mixed effects random-slope linear regression

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "nodes"
random_effect = "(age.factor | hospital)"
colon_s %>%
finalfit(dependent, explanatory, random_effect = random_effect,
dependent_label_suffix = " (random slope: age)") -> t
Dependent: nodes (random slope: age) unit value Coefficient (univariable) Coefficient (multilevel)
Age <40 years Mean (sd) 4.7 (4.5) - -
40-59 years Mean (sd) 3.6 (3.3) -1.14 (-2.08 to -0.21, p=0.016) -0.76 (-1.73 to 0.22, p=0.065)
60+ years Mean (sd) 3.6 (3.6) -1.19 (-2.10 to -0.28, p=0.010) -0.93 (-1.77 to -0.08, p=0.016)
Sex Female Mean (sd) 3.7 (3.6) - -
Male Mean (sd) 3.6 (3.6) -0.14 (-0.60 to 0.33, p=0.565) -0.19 (-0.62 to 0.24, p=0.196)
Obstruction No Mean (sd) 3.7 (3.7) - -
Yes Mean (sd) 3.5 (3.2) -0.24 (-0.83 to 0.36, p=0.435) -0.34 (-0.88 to 0.21, p=0.112)
Perforation No Mean (sd) 3.7 (3.6) - -
Yes Mean (sd) 3.9 (2.8) 0.24 (-1.13 to 1.61, p=0.735) 0.20 (-1.05 to 1.45, p=0.377)

### 2.22 Cox proportional hazards model (survival / time to event)

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "Surv(time, status)"
colon_s %>%
finalfit(dependent, explanatory) -> t
Dependent: Surv(time, status) all HR (univariable) HR (multivariable)
Age <40 years 70 (100.0) - -
40-59 years 344 (100.0) 0.76 (0.53-1.09, p=0.132) 0.79 (0.55-1.13, p=0.196)
60+ years 515 (100.0) 0.93 (0.66-1.31, p=0.668) 0.98 (0.69-1.40, p=0.926)
Sex Female 445 (100.0) - -
Male 484 (100.0) 1.01 (0.84-1.22, p=0.888) 1.02 (0.85-1.23, p=0.812)
Obstruction No 732 (100.0) - -
Yes 176 (100.0) 1.29 (1.03-1.62, p=0.028) 1.30 (1.03-1.64, p=0.026)
Perforation No 902 (100.0) - -
Yes 27 (100.0) 1.17 (0.70-1.95, p=0.556) 1.08 (0.64-1.81, p=0.785)

### 2.23 Cox proportional hazards model: change dependent label

As above, the dependent label cannot be specfied directly in the model to avoid errors. However, in survival modelling the surivial object specification can be long or awkward. Therefore, here is the work around.

library(finalfit)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "Surv(time, status)"
colon_s %>%
finalfit(dependent, explanatory, add_dependent_label = FALSE) %>%
dplyr::rename("Overall survival" = label) %>%
dplyr::rename(" " = levels) -> t
Overall survival all HR (univariable) HR (multivariable)
Age <40 years 70 (100.0) - -
40-59 years 344 (100.0) 0.76 (0.53-1.09, p=0.132) 0.79 (0.55-1.13, p=0.196)
60+ years 515 (100.0) 0.93 (0.66-1.31, p=0.668) 0.98 (0.69-1.40, p=0.926)
Sex Female 445 (100.0) - -
Male 484 (100.0) 1.01 (0.84-1.22, p=0.888) 1.02 (0.85-1.23, p=0.812)
Obstruction No 732 (100.0) - -
Yes 176 (100.0) 1.29 (1.03-1.62, p=0.028) 1.30 (1.03-1.64, p=0.026)
Perforation No 902 (100.0) - -
Yes 27 (100.0) 1.17 (0.70-1.95, p=0.556) 1.08 (0.64-1.81, p=0.785)

## 3 Model tables manually using ff_merge()

### 3.1 Basic table

Note summary_factorlist() needs argument, fit_id = TRUE.

library(finalfit)
library(dplyr)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"

## Crosstable
colon_s %>%
summary_factorlist(dependent, explanatory, fit_id=TRUE) -> table_1

## Univariable
colon_s %>%
glmuni(dependent, explanatory) %>%
fit2df(estimate_suffix=" (univariable)") -> table_2

## Merge

table_1 %>%
ff_merge(table_2) %>%
select(-c(fit_id, index)) %>%
dependent_label(colon_s, dependent)-> t
Dependent: Mortality 5 year Alive Died OR (univariable)
Age <40 years 31 (6.1) 36 (8.9) -
40-59 years 208 (40.7) 131 (32.4) 0.54 (0.32-0.92, p=0.023)
60+ years 272 (53.2) 237 (58.7) 0.75 (0.45-1.25, p=0.270)
Sex Female 243 (47.6) 194 (48.0) -
Male 268 (52.4) 210 (52.0) 0.98 (0.76-1.27, p=0.889)
Obstruction No 408 (82.1) 312 (78.6) -
Yes 89 (17.9) 85 (21.4) 1.25 (0.90-1.74, p=0.189)
Perforation No 497 (97.3) 391 (96.8) -
Yes 14 (2.7) 13 (3.2) 1.18 (0.54-2.55, p=0.672)

### 3.2 Complex table (all in single pipe)

library(finalfit)
library(dplyr)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
random_effect = "hospital"
dependent = "mort_5yr"

# All in one pipe

colon_s %>%
## Crosstable
summary_factorlist(dependent, explanatory, fit_id=TRUE)  %>%

ff_merge(
glmuni(colon_s, dependent, explanatory) %>%
fit2df(estimate_suffix=" (univariable)")
) %>%

ff_merge(
glmmulti(colon_s, dependent, explanatory) %>%
fit2df(estimate_suffix=" (multivariable)")
) %>%

ff_merge(
glmmixed(colon_s, dependent, explanatory, random_effect) %>%
fit2df(estimate_suffix=" (multilevel)"),
last_merge = TRUE
) %>%
dependent_label(colon_s, dependent) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable) OR (multilevel)
Age <40 years 31 (6.1) 36 (8.9) - - -
40-59 years 208 (40.7) 131 (32.4) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041) 0.75 (0.39-1.44, p=0.382)
60+ years 272 (53.2) 237 (58.7) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426) 1.03 (0.55-1.96, p=0.916)
Sex Female 243 (47.6) 194 (48.0) - - -
Male 268 (52.4) 210 (52.0) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902) 0.80 (0.58-1.11, p=0.180)
Obstruction No 408 (82.1) 312 (78.6) - - -
Yes 89 (17.9) 85 (21.4) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186) 1.23 (0.82-1.83, p=0.320)
Perforation No 497 (97.3) 391 (96.8) - - -
Yes 14 (2.7) 13 (3.2) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770) 1.03 (0.43-2.51, p=0.940)

### 3.3 Other GLM models

#### Poisson

library(finalfit)
library(dplyr)

## Dobson (1990) Page 93: Randomized Controlled Trial :
counts = c(18,17,15,20,10,20,25,13,12)
outcome = gl(3,1,9)
treatment = gl(3,3)

dependent = "counts"
explanatory = c("outcome", "treatment")

glmuni(dependent, explanatory, family = poisson) %>%
fit2df(estimate_name = "Rate ratio (univariable)")

glmmulti(dependent, explanatory, family = poisson) %>%
fit2df(estimate_name = "Rate ratio (multivariable)")

# All in one pipe
## Crosstable
summary_factorlist(dependent, explanatory, cont = "median", fit_id=TRUE)  %>%

ff_merge(fit_uni, estimate_name = "Rate ratio") %>%

ff_merge(fit_multi, estimate_name = "Rate ratio",
last_merge = TRUE) %>%
dependent_label(d.AD, dependent) -> t
Dependent: counts unit value Rate ratio (univariable) Rate ratio (multivariable)
outcome 1 Median (IQR) 20.0 (3.5) - -
2 Median (IQR) 13.0 (3.5) 0.63 (0.42-0.94, p=0.025) 0.63 (0.42-0.94, p=0.025)
3 Median (IQR) 15.0 (4.0) 0.75 (0.51-1.09, p=0.128) 0.75 (0.51-1.09, p=0.128)
treatment 1 Median (IQR) 17.0 (1.5) - -
2 Median (IQR) 20.0 (5.0) 1.00 (0.67-1.48, p=1.000) 1.00 (0.67-1.48, p=1.000)
3 Median (IQR) 13.0 (6.5) 1.00 (0.67-1.48, p=1.000) 1.00 (0.67-1.48, p=1.000)

#### Gamma

library(finalfit)
library(dplyr)

# A Gamma example, from McCullagh & Nelder (1989, pp. 300-2)
clotting <- data.frame(
u = c(5,10,15,20,30,40,60,80,100),
lot1 = c(118,58,42,35,27,25,21,19,18),
lot2 = c(69,35,26,21,18,16,13,12,12))

dependent = "lot1"
explanatory = "log(u)"

fit_uni = clotting %>%
glmuni(dependent, explanatory, family = Gamma) %>%
fit2df(estimate_name = "Coefficient", exp = FALSE, digits = c(3,3,4))

# All in one pipe
clotting %>%
## Crosstable
summary_factorlist(dependent, explanatory, cont = "median", fit_id=TRUE)  %>%

ff_merge(fit_uni, last_merge = TRUE) %>%
dependent_label(colon_s, dependent) -> t
Dependent: lot1 unit value Coefficient
u [5.0,100.0] Median (IQR) 27.0 (21.0) -
NA NA NA NA 0.015 (0.015-0.016, p<0.0001)

### 3.4 Weighted regression

library(finalfit)
library(dplyr)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"
weights = runif(dim(colon_s)[1]) # random just for example

# All in one pipe
colon_s %>%
## Crosstable
summary_factorlist(dependent, explanatory, fit_id=TRUE)  %>%

ff_merge(
glmuni(colon_s, dependent, explanatory, weights = weights, family = quasibinomial) %>%
fit2df(estimate_suffix=" (univariable)")
) %>%

ff_merge(
glmmulti(colon_s, dependent, explanatory, weights = weights, family = quasibinomial) %>%
fit2df(estimate_suffix=" (multivariable)"),
last_merge = TRUE
) %>%
dependent_label(colon_s, dependent) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (6.1) 36 (8.9) - -
40-59 years 208 (40.7) 131 (32.4) 0.60 (0.36-1.00, p=0.050) 0.62 (0.37-1.04, p=0.071)
60+ years 272 (53.2) 237 (58.7) 0.97 (0.59-1.59, p=0.894) 1.02 (0.62-1.69, p=0.942)
Sex Female 243 (47.6) 194 (48.0) - -
Male 268 (52.4) 210 (52.0) 1.06 (0.82-1.38, p=0.657) 1.05 (0.80-1.37, p=0.716)
Obstruction No 408 (82.1) 312 (78.6) - -
Yes 89 (17.9) 85 (21.4) 1.09 (0.78-1.52, p=0.606) 1.10 (0.79-1.55, p=0.573)
Perforation No 497 (97.3) 391 (96.8) - -
Yes 14 (2.7) 13 (3.2) 1.41 (0.69-2.94, p=0.345) 1.39 (0.67-2.93, p=0.374)

### 3.5 Using base R functions

Note ff_formula() convenience function to make multivariable formula (y ~ x1 + x2 + x3 etc.) from a dependent and explanatory vector of names.

library(finalfit)
library(dplyr)
explanatory = c("age.factor", "sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"

# All in one pipe

colon_s %>%
## Crosstable
summary_factorlist(dependent, explanatory, fit_id=TRUE)  %>%

ff_merge(
glmuni(colon_s, dependent, explanatory) %>%
fit2df(estimate_suffix=" (univariable)")
) %>%

ff_merge(
glm(
ff_formula(dependent, explanatory), data = colon_s, family = "binomial", weights = NULL
) %>%
fit2df(estimate_suffix=" (multivariable)"),
last_merge = TRUE
) %>%
dependent_label(colon_s, dependent) -> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (6.1) 36 (8.9) - -
40-59 years 208 (40.7) 131 (32.4) 0.54 (0.32-0.92, p=0.023) 0.57 (0.34-0.98, p=0.041)
60+ years 272 (53.2) 237 (58.7) 0.75 (0.45-1.25, p=0.270) 0.81 (0.48-1.36, p=0.426)
Sex Female 243 (47.6) 194 (48.0) - -
Male 268 (52.4) 210 (52.0) 0.98 (0.76-1.27, p=0.889) 0.98 (0.75-1.28, p=0.902)
Obstruction No 408 (82.1) 312 (78.6) - -
Yes 89 (17.9) 85 (21.4) 1.25 (0.90-1.74, p=0.189) 1.25 (0.90-1.76, p=0.186)
Perforation No 497 (97.3) 391 (96.8) - -
Yes 14 (2.7) 13 (3.2) 1.18 (0.54-2.55, p=0.672) 1.12 (0.51-2.44, p=0.770)

### 3.6 Edit table rows

This can be done as any dataframe would be edited.

library(finalfit)
library(dplyr)
explanatory = c("age.factor*sex.factor", "obstruct.factor", "perfor.factor")
dependent = "mort_5yr"

# Run model for term test
fit <- glm(
ff_formula(dependent, explanatory),
data=colon_s, family = binomial
)

# Not run
#term_test <- survey::regTermTest(fit, "age.factor:sex.factor")

# Run final table with results of term test
colon_s %>%
finalfit(dependent, explanatory) %>%
rbind(c(
"age.factor:sex.factor (overall)",
"Interaction",
"-",
"-",
"-",
paste0("p = 0.775")
))-> t
Dependent: Mortality 5 year Alive Died OR (univariable) OR (multivariable)
Age <40 years 31 (46.3) 36 (53.7) - -
40-59 years 208 (61.4) 131 (38.6) 0.65 (0.32-1.34, p=0.241) 0.66 (0.32-1.36, p=0.258)
60+ years 272 (53.4) 237 (46.6) 0.80 (0.40-1.61, p=0.529) 0.85 (0.42-1.71, p=0.647)
Sex Female 243 (55.6) 194 (44.4) - -
Male 268 (56.1) 210 (43.9) 1.24 (0.47-3.30, p=0.665) 1.17 (0.44-3.15, p=0.752)
Obstruction No 408 (56.7) 312 (43.3) - -
Yes 89 (51.1) 85 (48.9) 1.25 (0.90-1.74, p=0.189) 1.26 (0.90-1.76, p=0.182)
Perforation No 497 (56.0) 391 (44.0) - -
Yes 14 (51.9) 13 (48.1) 1.18 (0.54-2.55, p=0.672) 1.11 (0.50-2.41, p=0.795)
age.factor40-59 years:sex.factorMale Interaction - - 0.68 (0.23-1.97, p=0.479) 0.74 (0.25-2.18, p=0.588)
age.factor60+ years:sex.factorMale Interaction - - 0.86 (0.30-2.39, p=0.766) 0.89 (0.31-2.51, p=0.822)
age.factor:sex.factor (overall) Interaction - - - p = 0.775

### 3.7 Base model + individual explanatory variables

This was an email enquiry about how to build on a base model. The example request was in a survival context.

This has been updated August 2019. We have left the original example of building the table from scratch as a comparison.

ff_permute() allows combinations of variables to be built on a base model. See options on help page to,

• include the base model and/or the full model,
• present the permuted variables at the top or bottom of the table,
• produce separate model tables, or the default which is a single table.
library(dplyr)
mydata = colon_s
explanatory_base = c("age.factor", "sex.factor")
explanatory_permute = c("obstruct.factor", "perfor.factor", "node4.factor")
dependent = "Surv(time, status)"

mydata %>%
ff_permute(dependent, explanatory_base, explanatory_permute) %>%
rename("Overall survival" = Dependent: Surv(time, status), # optional tidying
n (%) = "all") -> t
Overall survival n (%) HR (univariable) HR (multivariable) 1 HR (multivariable) 2 HR (multivariable) 3 HR (multivariable) 4 HR (multivariable) 5
Age <40 years 70 (100.0) - - - - - -
40-59 years 344 (100.0) 0.76 (0.53-1.09, p=0.132) 0.76 (0.53-1.08, p=0.129) 0.79 (0.55-1.13, p=0.198) 0.76 (0.53-1.08, p=0.127) 0.85 (0.59-1.22, p=0.379) 0.90 (0.63-1.30, p=0.590)
60+ years 515 (100.0) 0.93 (0.66-1.31, p=0.668) 0.93 (0.66-1.31, p=0.660) 0.98 (0.69-1.40, p=0.931) 0.92 (0.65-1.31, p=0.656) 1.09 (0.77-1.55, p=0.615) 1.19 (0.83-1.69, p=0.346)
Sex Female 445 (100.0) - - - - - -
Male 484 (100.0) 1.01 (0.84-1.22, p=0.888) 1.02 (0.85-1.23, p=0.847) 1.02 (0.85-1.24, p=0.803) 1.02 (0.85-1.22, p=0.854) 1.04 (0.87-1.26, p=0.647) 1.05 (0.87-1.27, p=0.597)
Obstruction No 732 (100.0) - - - - - -
Yes 176 (100.0) 1.29 (1.03-1.62, p=0.028) - 1.31 (1.04-1.64, p=0.022) - - 1.35 (1.07-1.70, p=0.011)
Perforation No 902 (100.0) - - - - - -
Yes 27 (100.0) 1.17 (0.70-1.95, p=0.556) - - 1.18 (0.70-1.97, p=0.535) - 1.16 (0.69-1.94, p=0.581)
>4 positive nodes No 674 (100.0) - - - - - -
Yes 255 (100.0) 2.60 (2.15-3.14, p<0.001) - - - 2.64 (2.18-3.19, p<0.001) 2.68 (2.21-3.26, p<0.001)
library(finalfit)
library(dplyr)

mydata = colon_s
base_explanatory = c("age.factor", "sex.factor")
explanatory = c("obstruct.factor", "perfor.factor", "node4.factor")
dependent = "Surv(time, status)"

mydata %>%
# Counts
summary_factorlist(dependent, c(base_explanatory,
explanatory),
column = TRUE,
fit_id = TRUE) %>%

# Univariable
ff_merge(
coxphuni(mydata, dependent, c(base_explanatory, explanatory)) %>%
fit2df(estimate_suffix = " (Univariable)")
) %>%

# Base
ff_merge(
coxphmulti(mydata, dependent, base_explanatory) %>%
fit2df(estimate_suffix = " (Base model)")
) %>%

# Model 1
ff_merge(
coxphmulti(mydata, dependent, c(base_explanatory, explanatory[1])) %>%
fit2df(estimate_suffix = " (Model 1)")
) %>%

# Model 2
ff_merge(
coxphmulti(mydata, dependent, c(base_explanatory, explanatory[2])) %>%
fit2df(estimate_suffix = " (Model 2)")
) %>%

# Model 3
ff_merge(
coxphmulti(mydata, dependent, c(base_explanatory, explanatory[3])) %>%
fit2df(estimate_suffix = " (Model 3)")
) %>%

# Full
ff_merge(
coxphmulti(mydata, dependent, c(base_explanatory, explanatory)) %>%
fit2df(estimate_suffix = " (Full)"),
last_merge = TRUE
) %>%

# Tidy-up
rename("Overall survival" = label) %>%
rename(" " = levels) %>%
rename(n (%) = all) -> t
Overall survival n (%) HR (Univariable) HR (Base model) HR (Model 1) HR (Model 2) HR (Model 3) HR (Full)
Age <40 years 70 (7.5) - - - - - -
40-59 years 344 (37.0) 0.76 (0.53-1.09, p=0.132) 0.76 (0.53-1.08, p=0.129) 0.79 (0.55-1.13, p=0.198) 0.76 (0.53-1.08, p=0.127) 0.85 (0.59-1.22, p=0.379) 0.90 (0.63-1.30, p=0.590)
60+ years 515 (55.4) 0.93 (0.66-1.31, p=0.668) 0.93 (0.66-1.31, p=0.660) 0.98 (0.69-1.40, p=0.931) 0.92 (0.65-1.31, p=0.656) 1.09 (0.77-1.55, p=0.615) 1.19 (0.83-1.69, p=0.346)
Sex Female 445 (47.9) - - - - - -
Male 484 (52.1) 1.01 (0.84-1.22, p=0.888) 1.02 (0.85-1.23, p=0.847) 1.02 (0.85-1.24, p=0.803) 1.02 (0.85-1.22, p=0.854) 1.04 (0.87-1.26, p=0.647) 1.05 (0.87-1.27, p=0.597)
Obstruction No 732 (80.6) - - - - - -
Yes 176 (19.4) 1.29 (1.03-1.62, p=0.028) - 1.31 (1.04-1.64, p=0.022) - - 1.35 (1.07-1.70, p=0.011)
Perforation No 902 (97.1) - - - - - -
Yes 27 (2.9) 1.17 (0.70-1.95, p=0.556) - - 1.18 (0.70-1.97, p=0.535) - 1.16 (0.69-1.94, p=0.581)
>4 positive nodes No 674 (72.6) - - - - - -
Yes 255 (27.4) 2.60 (2.15-3.14, p<0.001) - - - 2.64 (2.18-3.19, p<0.001) 2.68 (2.21-3.26, p<0.001)

## 4 Support for complex survey structures via library(survey)

### 4.1 Linear regression

Examples taken from survey::svyglm() help page.

library(survey)
library(dplyr)

data(api)
dependent = "api00"
explanatory = c("ell", "meals", "mobility")

# Label data frame
apistrat = apistrat %>%
mutate(
api00 = ff_label(api00, "API in 2000 (api00)"),
ell = ff_label(ell, "English language learners (percent)(ell)"),
meals = ff_label(meals, "Meals eligible (percent)(meals)"),
mobility = ff_label(mobility, "First year at the school (percent)(mobility)"),
sch.wide = ff_label(sch.wide, "School-wide target met (sch.wide)")
)

# Linear example
dependent = "api00"
explanatory = c("ell", "meals", "mobility")

# Stratified design
dstrat = svydesign(id=~1,strata=~stype, weights=~pw, data=apistrat, fpc=~fpc)

# Univariable fit
fit_uni = dstrat %>%
svyglmuni(dependent, explanatory) %>%
fit2df(estimate_suffix = " (univariable)")

# Multivariable fit
fit_multi = dstrat %>%
svyglmmulti(dependent, explanatory) %>%
fit2df(estimate_suffix = " (multivariable)")

# Pipe together
apistrat %>%
summary_factorlist(dependent, explanatory, fit_id = TRUE) %>%
ff_merge(fit_uni) %>%
ff_merge(fit_multi, last_merge = TRUE) %>%
dependent_label(apistrat, dependent) -> t
Dependent: API in 2000 (api00) unit value Coefficient (univariable) Coefficient (multivariable)
English language learners (percent)(ell) [0.0,84.0] Mean (sd) 652.8 (121.0) -3.73 (-4.35–3.11, p<0.001) -0.48 (-1.25-0.29, p=0.222)
Meals eligible (percent)(meals) [0.0,100.0] Mean (sd) 652.8 (121.0) -3.38 (-3.71–3.05, p<0.001) -3.14 (-3.70–2.59, p<0.001)
First year at the school (percent)(mobility) [1.0,99.0] Mean (sd) 652.8 (121.0) -1.43 (-3.30-0.44, p=0.137) 0.23 (-0.54-1.00, p=0.567)

### 4.2 Binomial example

Note model family needs specified and exponentiation set to TRUE if desired.

library(survey)
library(dplyr)

data(api)
dependent = "sch.wide"
explanatory = c("ell", "meals", "mobility")

# Label data frame
apistrat = apistrat %>%
mutate(
api00 = ff_label(api00, "API in 2000 (api00)"),
ell = ff_label(ell, "English language learners (percent)(ell)"),
meals = ff_label(meals, "Meals eligible (percent)(meals)"),
mobility = ff_label(mobility, "First year at the school (percent)(mobility)"),
sch.wide = ff_label(sch.wide, "School-wide target met (sch.wide)")
)

# Univariable fit
fit_uni = dstrat %>%
svyglmuni(dependent, explanatory, family = "quasibinomial") %>%
fit2df(exp = TRUE, estimate_name = "OR", estimate_suffix = " (univariable)")

# Multivariable fit
fit_multi = dstrat %>%
svyglmmulti(dependent, explanatory, family = "quasibinomial") %>%
fit2df(exp = TRUE, estimate_name = "OR", estimate_suffix = " (multivariable)")

# Pipe together
apistrat %>%
summary_factorlist(dependent, explanatory, fit_id = TRUE) %>%
ff_merge(fit_uni) %>%
ff_merge(fit_multi, last_merge = TRUE) %>%
dependent_label(apistrat, dependent) -> t
Dependent: School-wide target met (sch.wide) No Yes OR (univariable) OR (multivariable)
English language learners (percent)(ell) Mean (SD) 22.5 (19.3) 20.5 (20.0) 1.00 (0.98-1.01, p=0.715) 1.00 (0.97-1.02, p=0.851)
Meals eligible (percent)(meals) Mean (SD) 46.0 (29.1) 44.7 (29.0) 1.00 (0.99-1.01, p=0.968) 1.00 (0.98-1.01, p=0.732)
First year at the school (percent)(mobility) Mean (SD) 13.9 (8.6) 17.2 (13.0) 1.06 (1.00-1.12, p=0.049) 1.06 (1.00-1.13, p=0.058)