Standardised mean differences for categorical variables

Question

Hi I would like to ask how CreateTableOne in R calculates standardised mean differences for variables with multiple categories.

I used matchit to conduct propensity score matching, and then used CreateTableOne to create a summary table to check the balances between the treatment and control groups. This is the code that I wrote:

matched_outcome <- matchit(treatment ~ age_group + gender, data = mydata, method = "nearest")
matched_data <- match.data(matched_outcome)

xvars <- c("age_group", "gender")
matched_table1 <- CreateTableOne(vars = xvars, strata = "treatment", data = matched_data, test = F)
print(matched_table1, smd = T)

                   Stratified by treatment    Stratified by treatment   SMD
                              0                          1
n                            656                        656
age_group (%)                                                           0.048
   19-49                     28 (4.3)                   25 (3.8)
   50-64                     63 (9.6)                   69 (10.5)
   65-74                    110 (16.8)                 110 (16.8)
   75-84                    195 (29.7)                 202 (30.8)
   85+                      259 (39.5)                 249 (38.0)
   Under 19                   1 (0.2)                    1 (0.2)
gender = Male (%)           324 (49.4)                 322 (49.1)       0.006

I understand how the SMD (standardised mean difference) for gender was calculated since it's binary. However, I don't know how the SMD was calculated for age_group. I checked the documentation for CreateTableOne. The documentation only says that "......standardized mean differences for all pairwise comparisons are calculated".

I would be really grateful if someone can help me to understand how the this is calculated. Thanks in advance.