sta310r

sta310r collects the data sets that accompany Collett (2023), Modelling Survival Data in Medical Research (4th edition), together with the computer exercises used in the STA310 course at the University of Gothenburg. The package focuses on the data sets, their documentation and the exercise vignettes.

Installation

The recommended way is from R-universe, which provides pre-built binaries for Windows, macOS and Linux, so no compiler toolchain is needed:

install.packages(
  "sta310r",
  repos = c(
    "https://eribul.r-universe.dev",
    "https://cloud.r-project.org"
  ),
  dependencies = TRUE
)

You can also install the development version directly from GitHub with pak (this builds from source, so Windows users need Rtools installed):

pak::pak("eribul/sta310r")

Name conflicts

The package includes six data sets with names also used by the survival package, for data sets that are not the same:

• bladder
• cgd
• kidney
• myeloma
• ovarian
• rats

Attach sta310r last (library(sta310r) AFTER library(survival)), or use sta310r::xxx explicitly if both packages are attached. You may also just refer to the datasets by their aliases as described below.

Data sets by example

Each data set that is used in a worked example of Collett (2023) has an alias dX.Y, where X.Y is the number of the example. For example, d3.7 is the myeloma data used in Example 3.7, and d8.4 is the baseline cirrhosis data. The aliases are ordinary data objects: d3.7 works after library(sta310r), and ?d3.7 opens the help page of the data set. See vignette("datasets-by-exercise").

Note that 2 data sets are used in example 8.4 (hence no aliases used there).

library(sta310r)
dataset_by_exercise |>
  dplyr::filter(!is.na(chapter)) |>
  dplyr::mutate(chapter = paste(chapter, chapter_title, sep = ". ")) |>
  dplyr::select(chapter, example, alias, dataset, n_rows, n_variables) |>
  gt::gt(groupname_col = "chapter", id = "datasets-by-example")

example	alias	dataset	n_rows	n_variables
Survival analysis
1.1	d1.1	iud	18	2
1.2	d1.2	breast	45	3
1.3	d1.3	myeloma	48	10
1.4	d1.4	prostate	38	8
Some non-parametric procedures
2.1	d2.1	pulmonary	11	1
2.2	d2.2	myeloma	48	10
2.3	d2.3	iud	18	2
2.4	d2.4	iud	18	2
2.5	d2.5	iud	18	2
2.6	d2.6	myeloma	48	10
2.7	d2.7	iud	18	2
2.8	d2.8	iud	18	2
2.9	d2.9	iud	18	2
2.10	d2.10	iud	18	2
2.11	d2.11	breast	45	3
2.12	d2.12	breast	45	3
2.13	d2.13	breast	45	3
2.14	d2.14	breast	45	3
2.15	d2.15	melanoma	30	4
2.16	d2.16	melanoma	30	4
The Cox regression model
3.1	d3.1	breast	45	3
3.2	d3.2	myeloma	48	10
3.3	d3.3	breast	45	3
3.4	d3.4	hyper	36	4
3.5	d3.5	myeloma	48	10
3.6	d3.6	prostate	38	8
3.7	d3.7	myeloma	48	10
3.8	d3.8	myeloma	48	10
3.9	d3.9	myeloma	48	10
3.10	d3.10	hyper	36	4
3.11	d3.11	hyper	36	4
3.12	d3.12	prostate	38	8
3.13	d3.13	hyper	36	4
3.14	d3.14	hyper	36	4
3.15	d3.15	myeloma	48	10
3.16	d3.16	prostate	38	8
3.17	d3.17	myeloma	48	10
3.19	d3.19	myeloma	48	10
3.20	d3.20	myeloma	48	10
Model checking in the Cox regression model
4.1	d4.1	dialysis	13	5
4.2	d4.2	dialysis	13	5
4.3	d4.3	dialysis	13	5
4.4	d4.4	dialysis	13	5
4.5	d4.5	myeloma	48	10
4.6	d4.6	dialysis	13	5
4.7	d4.7	dialysis	13	5
4.8	d4.8	myeloma	48	10
4.9	d4.9	myeloma	48	10
4.10	d4.10	dialysis	13	5
4.11	d4.11	dialysis	13	5
4.12	d4.12	dialysis	13	5
Parametric regression models
5.1	d5.1	iud	18	2
5.2	d5.2	iud	18	2
5.3	d5.3	iud	18	2
5.4	d5.4	iud	18	2
5.5	d5.5	breast	45	3
5.6	d5.6	breast	45	3
5.7	d5.7	breast	45	3
5.8	d5.8	breast	45	3
5.9	d5.9	numerical_illustration	37	4
5.10	d5.10	hyper	36	4
5.11	d5.11	ovarian	26	7
5.12	d5.12	ovarian	26	7
5.13	d5.13	breast	45	3
5.14	d5.14	breast	45	3
5.15	d5.15	prostate	38	8
5.16	d5.16	myeloma	48	10
5.17	d5.17	breast	45	3
Flexible parametric models
6.1	d6.1	hyper	36	4
6.4	d6.4	hyper	36	4
6.5	d6.5	breast_rfs	686	11
6.6	d6.6	breast_rfs	686	11
Model checking in parametric models
7.1	d7.1	ovarian	26	7
7.2	d7.2	ovarian	26	7
7.3	d7.3	ovarian	26	7
7.4	d7.4	ovarian	26	7
Time-dependent variables
8.2	d8.2	bmt_leukaemia	23	8
8.3	d8.3	ovarian	26	7
8.4	d8.4	cirrhosis	12	6
8.4	NA	cirrhosis_cp	54	7
8.4	NA	cirrhosis_lbr	42	3
8.5	d8.5	aortic	988	11
8.6	d8.6	aortic	988	11
8.7	d8.7	aortic	988	11
8.8	d8.8	aortic	988	11
8.9	d8.9	aortic	988	11
Interval-censored survival data
9.1	d9.1	ulcer	43	6
9.4	d9.4	ulcer	43	6
9.5	d9.5	ulcer	43	6
9.6	d9.6	ulcer	43	6
9.7	d9.7	ulcer	43	6
Frailty models
10.1	d10.1	breast	45	3
10.2	d10.2	breast	45	3
10.3	d10.3	lung_transplant	196	7
10.4	d10.4	kidney	434	7
Non-proportional hazards and institutional comparisons
11.1	d11.1	ulcer	43	6
11.2	d11.2	gastric	90	4
11.3	d11.3	gastric	90	4
11.4	d11.4	kidney_centres	1439	9
11.5	d11.5	kidney_centres	1439	9
11.6	d11.6	kidney_centres	1439	9
11.7	d11.7	kidney_centres	1439	9
11.8	d11.8	kidney_centres	1439	9
Competing risks
12.1	d12.1	liver	1761	7
12.2	d12.2	liver	1761	7
12.3	d12.3	liver	1761	7
12.4	d12.4	liver	1761	7
12.5	d12.5	liver	1761	7
12.6	d12.6	liver	1761	7
12.7	d12.7	mice	181	3
Multiple events and event history modelling
13.2	d13.2	rats	254	4
13.3	d13.3	tamoxifen	641	18
13.4	d13.4	bmt_outcome	2204	9
Dependent censoring
14.1	d14.1	liver_wait	281	7
14.3	d14.3	liver_wait	281	7
Sample size requirements for a survival study
15.1	d15.1	hepatitis	44	3
15.2	d15.2	hepatitis	44	3
15.3	d15.3	hepatitis	44	3
Bayesian survival analysis
16.1	d16.1	iud	18	2
16.2	d16.2	iud	18	2
16.3	d16.3	iud	18	2
16.4	d16.4	breast	45	3
16.5	d16.5	breast	45	3
16.6	d16.6	breast	45	3
16.7	d16.7	help	447	7

Computer exercises

The course exercises are shipped as vignettes. The first exercise covers censoring, the survivor and hazard functions, Kaplan–Meier, Nelson–Aalen and the log-rank family, and ships with a solutions vignette:

vignette("cs1-survival-data")
vignette("cs1-solutions")