Skip to contents

Check proportional hazards assumption for Cox models

Source: R/check_ph.R
check_ph.Rd

Test the proportional hazards assumption for fitted Cox proportional hazards models using Schoenfeld residuals.

Usage

check_ph(
  model,
  transform = c("km", "rank", "identity"),
  alpha = 0.05,
  format = c("flextable", "gt", "tibble")
)

Arguments

model

A fitted coxph model or a cox_reg() object.

transform

Time transformation passed to survival::cox.zph(). One of "km", "rank", or "identity".

alpha

Significance level used for the simple interpretation column.

format

Output format. One of "flextable" (default), "gt", or "tibble".

Value

A tibble, gt_tbl, or flextable with Schoenfeld residual proportional hazards tests. The table contains:

Model

Model name. For a direct coxph object, this is "cox_model".

Term

Model term or "GLOBAL".

Chi.square

Chi-square statistic.

df

Degrees of freedom.

p.value

Test p-value.

Interpretation

Simple screening interpretation using alpha.

Details

check_ph() is a diagnostic aid for Cox models. A small p-value suggests possible evidence against the proportional hazards assumption for a term or for the global model test. This should be interpreted with plots, clinical knowledge, follow-up patterns, and modelling purpose.

Examples

lung_data <- data_lungcancer
lung_data$trt <- factor(lung_data$trt, levels = c(1, 2),
                        labels = c("Standard", "Test"))

cox_fit <- cox_reg(
  data = lung_data,
  time = time,
  event = status,
  exposures = c(trt, celltype, age)
)

check_ph(cox_fit)

Proportional hazards check
Model
Term
Test
Chi-square
df
p-value
Interpretation
trt
trt
Term
3.54
1
0.060
No evidence of PH violation
trt
GLOBAL
Global
3.54
1
0.060
No evidence of PH violation
celltype
celltype
Term
8.87
3
0.031
Possible PH violation
celltype
GLOBAL
Global
8.87
3
0.031
Possible PH violation
age
age
Term
1.67
1
0.196
No evidence of PH violation
age
GLOBAL
Global
1.67
1
0.196
No evidence of PH violation
Screening aid only. Small p-values suggest possible non-proportional hazards; interpret with Schoenfeld residual plots, follow-up pattern, clinical context, and model purpose. alpha = 0.05; transform = km.
check_ph(cox_fit, format = tibble)
#> # A tibble: 6 × 7
#>   Model    Term     Test   Chi.square    df p.value Interpretation             
#>   <chr>    <chr>    <chr>       <dbl> <dbl>   <dbl> <chr>                      
#> 1 trt      trt      Term         3.54     1  0.0600 No evidence of PH violation
#> 2 trt      GLOBAL   Global       3.54     1  0.0600 No evidence of PH violation
#> 3 celltype celltype Term         8.87     3  0.0310 Possible PH violation      
#> 4 celltype GLOBAL   Global       8.87     3  0.0310 Possible PH violation      
#> 5 age      age      Term         1.67     1  0.196  No evidence of PH violation
#> 6 age      GLOBAL   Global       1.67     1  0.196  No evidence of PH violation