Package 'collett' reference manual

Title:	Datasets from "Modelling Survival Data in Medical Research" by Collett
Description:	Datasets for the book entitled "Modelling Survival Data in Medical Research" by Collett (2023) <doi:10.1201/9781003282525>. The datasets provide extensive examples of time-to-event data.
Authors:	Mark Clements [aut, cre] (ORCID: <https://orcid.org/0000-0003-4518-5670>), Enoch Yi-Tung Chen [ctb] (ORCID: <https://orcid.org/0000-0003-2448-708X>)
Maintainer:	Mark Clements <[email protected]>
License:	MIT + file LICENSE
Version:	0.1.3
Built:	2026-06-02 09:40:54 UTC
Source:	https://github.com/mclements/collett

Chronic active hepatitis

Description

Clinical trial of 44 patients with chronic active hepatitis randomised to either the drug prednisolone or an untreated control group.

Usage

active_hepatitis
active_hepatitis

Format

A data frame with 44 rows and 3 variables:

treatment: integer treatment (1=prednisolone, 2=control)
time: integer survival time from admission to study (months)
status: integer event indicator (1=event, 0=right censored)

Details

See Collett (2023)

Prognosis for women with breast cancer

Description

For female breast cancer patients from Middlesex Hospital. The dataset includes the result of staining using Helix pomatia agglutinin (HPA).

Usage

bcancer
bcancer

Format

A data frame with 45 rows and 3 variables:

stain: integer for negative staining (=1) or positive staining (=2)
time: integer time in months for survival
status: integer for status at end of follow-up (0=censored, 1=death)

Details

For details about the study design, see Leathem and Brooks (1987).

The dataset is described in Example 1.2 and Table 1.2 (Collett, 2023, pages 6-7).

References

Leathem AJ, Brooks S. Predictive value of lectin binding on breast-cancer recurrence and survival. The Lancet. 1987 May 9;329(8541):1054-6. doi:10.1016/S0140-6736(87)90482-X

Examples

library(survival)
plot(survfit(Surv(time,status)~stain, data=bcancer), col=1:2, xlab="Survival time (months)",
ylab="Survival")
legend("topright", legend=c("Negative staining","Positive staining"), col=1:2, lty=1,
bty="n")
library(survival)
plot(survfit(Surv(time,status)~stain, data=bcancer), col=1:2, xlab="Survival time (months)",
ylab="Survival")
legend("topright", legend=c("Negative staining","Positive staining"), col=1:2, lty=1,
bty="n")

Recurrence of bladder cancer

Description

Placebo controlled trial of bladder cancer patients randomised to thiopeta or to placebo

Usage

bladder
bladder

Format

A data frame with 86 rows and 6 variables:

patient: integer patient number (1-86)
time: integer survival time in months
status: integer status of patient (0=censored, 1=recurrence)
treat: integer treatment group (1=placebo, 2=thiotepa)
init: integer initial number of tumours
size: integer diameter of larger initial tumour in cm

Details

See Collett (2023)

Bone marrow transplantation

Description

A study of 37 patients with leukaemia in complete remission who received a non-depleted allogenic bone marrow transplant.

Usage

bone_marrow
bone_marrow

Format

A data frame with 37 rows and 9 variables:

patient: integer patient number (1-37)
time: integer survival time in days
status: integer status of patient (0=alive, 1=dead)
rage: integer age of patient in years
dage: integer age of donor in years
type: integer type of leukaemia (1=AML, 2=ALL, 3=CML)
preg: integer Donor pregnancy (0=no, 1=yes)
index: double index of cell-lymphocyte reactions
gvhd: integer graft-versus-host disease (0=no, 1=yes)

Details

See Collett (2023)

Patient outcome following bone marrow transplantation

Description

Patient outcome following bone marrow transplantation

Usage

bone_marrow_tx
bone_marrow_tx

Format

A data frame with 2204 rows and 9 variables:

id: integer patient id
leukaemia: character type of leukaemia (CML,ALL,AML)
age: character age group of patient in years (<=20, 21-40, >40))
match: integer indicator for whether there was a donor gender match (0=no, 1=yes)
tcell: integer indicator for whether there was T-cell depletion (1=yes, n=no)
ptime: integer time to platelet recovery (days)
pcens: integer event indicator for platelet recovery (1=event, 0=censored)
rdtime: integer time to relapse of death (days)
rdcens: integer event indicator for relapse or death (1=event, 0=censored)

Details

See Collett (2023)

Recurrence free survival in breast cancer patients

Description

Recurrence free survival in breast cancer patients

Usage

breast_rfs
breast_rfs

Format

A data frame with 686 rows and 11 variables:

id: integer patient id
treat: integer hormonal treatment (0=no tamoxifen, 1=tamoxifen)
age: integer patient age (years)
men: integer menopausal status (1=premenopausal, 2=postmenopausal)
size: integer tumour size (mm)
grade: integer tumour grade (1,2,3)
nodes: integer number of positive pymph nodes
prog: integer progesterone receptor status (femtomoles)
oest: integer oestrogen receptor status (femtomoles)
time: integer recurrence-free survival time (days)
status: integer event indicator (0=censored, 1=relapse or death)

Details

See Collett (2023)

Datasets

Description

The datasets are based on the official .zip file. A table for the dataset names and file names sorted by file name is here:

Dataset name	File name
--------------------	-----------------
illustration	"A numerical illustration.dat"
leukaemia	"Bone marrow transplantation in the treatment of leukaemia.dat"
bone_marrow	"Bone marrow transplantation.dat"
ovarian	"Chemotherapy in ovarian cancer patients.dat"
active_hepatitis	"Chronic active hepatitis.dat"
granulomatous	"Chronic granulomatous disease.dat"
tamoxifen	"Clinical trial of tamoxifen in breast cancer patients.dat"
prostatic	"Comparison of two treatments for prostatic cancer.dat"
kidneytx	"Comparisons between kidney transplant centres.dat"
liverbase	"Data from a cirrhosis study (baseline).dat"
liver_counting	"Data from a cirrhosis study (in counting process format).dat"
lbrdata0	"Data from a cirrhosis study (lbr data).dat"
HELP	"Health evaluation and linkage to primary care.dat"
dialysis	"Infection in patients on dialysis.dat"
bone_marrow_tx	"Patient outcome following bone marrow transplantation.dat"
bcancer	"Prognosis for women with breast cancer.dat"
pulmonary	"Pulmonary metastasis.dat"
breast_rfs	"Recurrence free survival in breast cancer patients.dat"
ulcer	"Recurrence of an ulcer.dat"
bladder	"Recurrence of bladder cancer.dat"
mammary	"Recurrence of mammary tumours in female rats.dat"
valve	"Survival following aortic valve replacement.dat"
tplant	"Survival following kidney transplantation.dat"
ducks	"Survival of black ducks.dat"
mice	"Survival of laboratory mice.dat"
liver	"Survival of liver transplant recipients.dat"
myeloma	"Survival of multiple myeloma patients.dat"
lung	"Survival of patients registered for a lung transplant.dat"
gcancer	"Survival of patients with gastric cancer.dat"
melanoma	"Survival times of patients with melanoma .dat"
livertx	"Time to death while waiting for a liver transplant.dat"
IUD	"Time to discontinuation of the use of an IUD.dat"
kidney	"Treatment of hypernephroma.dat"

And now sorted by the dataset names:

Dataset name	File name
--------------------	-----------------
active_hepatitis	"Chronic active hepatitis.dat"
bcancer	"Prognosis for women with breast cancer.dat"
bladder	"Recurrence of bladder cancer.dat"
bone_marrow	"Bone marrow transplantation.dat"
bone_marrow_tx	"Patient outcome following bone marrow transplantation.dat"
breast_rfs	"Recurrence free survival in breast cancer patients.dat"
dialysis	"Infection in patients on dialysis.dat"
ducks	"Survival of black ducks.dat"
gcancer	"Survival of patients with gastric cancer.dat"
granulomatous	"Chronic granulomatous disease.dat"
HELP	"Health evaluation and linkage to primary care.dat"
illustration	"A numerical illustration.dat"
IUD	"Time to discontinuation of the use of an IUD.dat"
kidney	"Treatment of hypernephroma.dat"
kidneytx	"Comparisons between kidney transplant centres.dat"
lbrdata0	"Data from a cirrhosis study (lbr data).dat"
leukaemia	"Bone marrow transplantation in the treatment of leukaemia.dat"
liver	"Survival of liver transplant recipients.dat"
liver_counting	"Data from a cirrhosis study (in counting process format).dat"
liverbase	"Data from a cirrhosis study (baseline).dat"
livertx	"Time to death while waiting for a liver transplant.dat"
lung	"Survival of patients registered for a lung transplant.dat"
mammary	"Recurrence of mammary tumours in female rats.dat"
melanoma	"Survival times of patients with melanoma .dat"
mice	"Survival of laboratory mice.dat"
myeloma	"Survival of multiple myeloma patients.dat"
ovarian	"Chemotherapy in ovarian cancer patients.dat"
prostatic	"Comparison of two treatments for prostatic cancer.dat"
pulmonary	"Pulmonary metastasis.dat"
tamoxifen	"Clinical trial of tamoxifen in breast cancer patients.dat"
tplant	"Survival following kidney transplantation.dat"
ulcer	"Recurrence of an ulcer.dat"
valve	"Survival following aortic valve replacement.dat"

As an alternative to using the R datasets, the collett_data function allows for reading from the original .dat files that are stored in the package.

Usage

collett_data(name)
collett_data(name)

Arguments

name

Character string with the original filename

Value

A data-frame

Author(s)

Maintainer: Mark Clements [email protected] (ORCID)

Other contributors:

Enoch Yi-Tung Chen [email protected] (ORCID) [contributor]

Source

https://s3-eu-west-1.amazonaws.com/s3-euw1-ap-pe-ws4-cws-documents.ri-prod/9781032252858/Data%20sets%20from%20Modelling%20Survival%20Data%20in%20Medical%20Research%2C%204th%20edition.zip

Examples

head(collett_data("A numerical illustration.dat"))
## which is equivalent to: head(illustration)
head(collett_data("A numerical illustration.dat"))
## which is equivalent to: head(illustration)

Infection in patients on dialysis

Description

Time from dialysis to infection for patients with diseases of the kidney.

Usage

dialysis
dialysis

Format

A data frame with 13 rows and 5 variables:

patient: integer patient id
time: integer time to infection (days)
status: integer event indicator (0=censored, 1=infection)
age: integer age in years
sex: integer sex of the patient (1=male, 2=female)

Details

See Collett (2023)

Survival of black ducks

Description

Black ducks, Anas rubripes, were followed the US Fish and Wildlife Service.

Usage

ducks
ducks

Format

A data frame with 50 rows and 6 variables:

duck: integer duck indicator
time: integer survival time in days
status: integer status of bird (0=alive or missing, 1=dead)
age: integer age group (0=hatch-year bird, 1=bird aged >= 1 year)
weight: integer weight of bird in g
length: integer length of wing in mm

Details

See Collett (2023)

Survival of patients with gastric cancer

Description

Survival of patients with gastric cancer

Usage

gcancer
gcancer

Format

A data frame with 90 rows and 4 variables:

patient: integer patient id
time: integer survival time in days
status: integer event indicator (0=censored, 1=dead)
treat: integer treatment arm (0=chemotherapy alone, 1=chemotherapy and radiotherapy)

Details

See Collett (2023)

Chronic granulomatous disease

Description

Trial comparing interferon with a placebo.

Usage

granulomatous
granulomatous

Format

A data frame with 128 rows and 12 variables:

patient: integer patient number (1-128)
time: integer time to first infection (days)
status: integer status of patient (0=censored, 1=infection)
centre: integer treatment centre; see Collett (2023, page 504)
treat: integer treatment group (0=placebo, 1=interferon)
age: integer age in years
sex: integer sex (1=male, 2=female)
height: double height in cm
weight: double weight in kg
pattern: integer pattern of inheritance (1=X-linked, 2=autosomal recessive)
cort: integer use of corticosteroids at trial entry (1=used, 2=not used)
anti: integer Use of antibiotics at trial entry (1=used, 2=not used)

Details

See Collett (2023)

Health evaluation and linkage to primary care

Description

A clinical trial for patients in a residential detoxification programme. Patients were randomised to either get a referral to a HELP clinic or not.

Usage

HELP
HELP

Format

A data frame with 447 rows and 7 variables:

subject: integer subject id
days: integer time to linkage to primary care in days
status: integer event indicator (0=no linkage, 1=linkage)
age: integer age of patient in years
gender: integer gender of the patient (0=female, 1=male)
housing: integer Homelessness status (0=homeless, 1=housed)
linkage: integer assistance to linking to healthcare (0=no, 1=yes)

Details

Collett (2023) defines this dataset as "help", however that leads to issues with using R's help system. We have changed the dataset name to "HELP". Moreover, the book uses the variables "Time" an d"Help", whereas the dataset includes variables "days" and "linkage", respectively.

A numerical illustration

Description

Artificial data on patient survival classified according to factors a and b

Usage

illustration
illustration

Format

A data frame with 37 rows and 4 variables:

a: integer factor a
b: integer factor b
time: integer event time
status: integer event status (1=event, 0=right censored)

Details

See Collett (2023).

Time to discontinuation of the use of an IUD

Description

A very simple dataset showing potential right censoring for time to discontinuation of the use of an IUD.

Usage

IUD
IUD

Format

A data frame with 18 rows and 2 variables:

time: integer Time in weeks to discontinuation of the use of an IUD
status: integer Indicator for whether the IUD was discontinued: 0=No, 1=Yes

Details

These data are reported in Table 1.1 (Collett, 2023, page 6).

Treatment of hypernephroma

Description

This study was undertaken at the University of Oklahoma Health Sciences Center to investigate survival among 36 patients with a kidney tumour (hypernephroma). Standard tangent included chemotherapy and immunotherapy, with some patients also having a nephrectomy, or surgical removal of the kidney. For further details, see Lee and Wang (2013).

Usage

kidney
kidney

Format

A data frame with 36 rows and 4 variables:

nephrectomy: integer indicator for nephrectomy (0=No; 1=Yes)
age: integer age group (1=<60; 2=60-70; 3=>70
time: integer for the follow-up time in months
status: integer for status at the end of follow-up (1=died; 0=censored)

References

Lee ET, Wang J. Statistical Methods for Survival Data Analysis. New York, NY: John Wiley & Sons; 2013, fourth edition. https://www.wiley.com/en-sg/Statistical+Methods+for+Survival+Data+Analysis%252C+4th+Edition-p-9781118095027

Comparisons between kidney transplant centres

Description

Transplant survival rates by recipients of organs from deceased donors. No event was defined as being alive with a functioning graft at the last known follow-up.

Usage

kidneytx
kidneytx

Format

A data frame with 1439 rows and 9 variables:

patient: integer patient id
centre: integer transplant centre (1-8)
tsurv: integer transplant survival time (days)
tcens: integer event indicator (0=censored, 1=transplant failure)
dage: integer donor age (years)
dtype: integer donor type (0=deceased following brain death, 1=circulatory death)
rage: integer recipient age (years)
diab: integer diabetic status (0=absent, 1=present)
cit: double cold ischaemic time (hours)

Details

See Collett (2023). Thirty-five patients had tsurv==0 (that is, the transplanted kidney did not function).

Data from a cirrhosis study (lbr data)

Description

DATASET_DESCRIPTION

Usage

lbrdata0
lbrdata0

Format

A data frame with 42 rows and 3 variables:

patient: integer patient id
time: integer date of measurement (days)
lbr: double log bilirubin level

Details

See Collett (2023)

Bone marrow transplantation in the treatment of leukaemia

Description

Bone marrow transplantation in the treatment of leukaemia

Usage

leukaemia
leukaemia

Format

A data frame with 23 rows and 8 variables:

patient: integer patient id
time: integer survival time in days
status: integer event indicator (0=alive, 1=dead)
group: integer disease group (1=ALL, 2=low-risk AML, 3=high-risk AML)
page: integer age of patient in years
dage: integer age of donor in years
precovery: integer platelet recovery indicator (0=no, 1=yes)
ptime: character time in days to return of platelets to normal level (if precovery=1)

Details

See Collett (2023). Note that ptime will need conversion:).

Survival of liver transplant recipients

Description

Survival of liver transplant recipients

Usage

liver
liver

Format

A data frame with 1761 rows and 7 variables:

patient: integer patient id
age: integer patient age in years
gender: integer patient gender (1=male, 2=female)
disease: integer primary disease (1=PBC, 2=PSC, 3=ALD)
time: integer time to event (days)
status: integer cof>0
cof: integer cause of graft failure (0=functioning graft, 1=rejection, 2=thrombosis, 3=recurrent disease, 4=other)

Details

See Collett (2023)

Data from a cirrhosis study (in counting process format)

Description

Artificial data

Usage

liver_counting
liver_counting

Format

A data frame with 54 rows and 7 variables:

patient: integer patient id
start: integer start time (days)
stop: integer stop time (days)
status: integer event indicator (0=censored, 1=uncensored)
treat: integer treatment group (0=placebo, 1=Liverol)
age: integer age of the patient at start of study (years)
lbrt: double logarithm of bilirubin level

Details

See Collett (2023). Note that the variable for log of bilirubin differs to that for "liverbase".

Data from a cirrhosis study (baseline)

Description

Articial data

Usage

liverbase
liverbase

Format

A data frame with 12 rows and 6 variables:

patient: integer patient id
time: integer survival time in days
status: integer event indicator (0=censored, 1=uncensored)
age: integer age of the patient (years)
treat: integer treatment group (0=placebo, 1=Liverol)
lbr: double logarithm of bilirubin level

Details

See Collett (2023)

Time to death while waiting for a liver transplant

Description

Investigate the time on the liver transplantation list.

Usage

livertx
livertx

Format

A data frame with 281 rows and 7 variables:

patient: integer patient id
time: integer time on the list
status: integer event indicator (0=censored, including having a transplant, 1=died on the list)
age: integer patient age in years
gender: integer patient gender (1=male, 0=female)
bmi: double body mass index (kg/m^2)
ukeld: integer UK endstage liver disease score

Details

See Collett (2023). A higher UKELD is associated with worse disease severity.

Survival of patients registered for a lung transplant

Description

Survival of patients registered for a lung transplant

Usage

lung
lung

Format

A data frame with 196 rows and 7 variables:

patient: integer patient id
time: integer time from registration to the earlist of removal from list, last known follow-up date, 30 April 2012, or death (days)
status: integer event indicator (0=censored, 1=dead)
age: integer age in years
gender: integer gender (1=male, 2=female)
bmi: double body mass index
disease: integer disease (1=COPD, 2=fibrosis, 3=suppurative, 4=other)

Details

See Collett (2023)

Recurrence of mammary tumours in female rats

Description

This is an animal experiment to compare the use of retinyl acetate (related to vitamin A) across the study (treatment) to treatment with retinyl acetate to 60 days and then no further treatment (control). The female rats all had mammary tumours.

Usage

mammary
mammary

Format

A data frame with 254 rows and 4 variables:

rat: integer id for each rat
treatment: integer treatment arm indicator (1=treatment, 0=control)
time: double follow-up time (days)
status: integer recurrence indicator (0=no, 1=yes)

Details

See Collett (2023)

Survival times of patients with melanoma

Description

Comparing two immunotherapy treatments for patients with melanoma

Usage

melanoma
melanoma

Format

A data frame with 30 rows and 4 variables:

age: integer age group (1=21-44, 2=41-60, 3=61+)
treatment: integer treatment arm (1=BCG, 2=C. parvum)
time: integer survival time (months)
status: integer event indicator (0=censored, 1=dead)

Details

See Collett (2023)

Survival of laboratory mice

Description

Laboratory study of survival for two groups of mice exposed to radiation.

Usage

mice
mice

Format

A data frame with 181 rows and 3 variables:

environment: integer type of environment (1=standard, 2=germ-free)
causeofdeath: integer cause of death (1=thymic lymphoma, 2=reticulum cell sarcoma, 3=other causes)
time: integer survival time (days)

Details

See Collett (2023). Note that are no censored event times.

Survival of multiple myeloma patients

Description

Patients diagnosed with multiple myeloma who were diagnosed and treated with alkylating agents at West Virginia University Medical Center for ages 50-80 years.

Usage

myeloma
myeloma

Format

A data frame with 48 rows and 10 variables:

patient: integer for a patient identifier
time: integer survival time in months
status: integer for status at follow-up (0=Alive, 1=Dead)
age: integer age at diagnosis in years
sex: integer for sex of the patient (1=male, 2=female)
bun: integer level of blood urea nitrogen at diagnosis (unit assumed to be mg/dL based on the normal range for adults reported by https://en.wikipedia.org/wiki/Blood_urea_nitrogen)
ca: integer serum calcium at diagnosis in mg/dL
hb: double for serum hemoglobin level at diagnosis in g/dL (equivalently, grams per 100 mL)
pcells: integer percent of plasma cells in the bone marrow at diagnosis
protein: integer indicator for whether or not the Bence-Jones protein was present in the urine at diagnosis (0=absent, 1=present)

Details

Krall et al (1975) did not provide the units for all of these measurements. In their analyses, they used some data transformations: log(bun). Collett (2023) converted data from Krall et al (1975): BUN is reported by Krall and colleagues as X1=log(BUN), however the log base and unit is unclear; Krall and colleagues reported for 65 individuals, including those younger than 50 and older than 80.

References

Krall JM, Uthoff VA, Harley JB. A step-up procedure for selecting variables associated with survival. Biometrics. 1975 Mar 1:49-57. doi:10.2307/2529709

Examples

## To be completed.

## To be completed.

Chemotherapy in ovarian cancer patients

Description

Trial for treatment of ovarian cancer patients comparing cyclophosphamide alone with cyclophosphamide combined with adriamycin.

Usage

ovarian
ovarian

Format

A data frame with 26 rows and 7 variables:

patient: integer identifer
time: integer survival time from randomisation in days
status: integer event indicator (0=right censored, 1=event)
treat: integer treatment (1=single, 2=combined)
age: integer age of patients in years
rdisease: integer extent of residual disease (1=incomplete, 2=complete)
perf: integer performance status (1=good, 2=poor)

Details

See Collett (2023)

Comparison of two treatments for prostatic cancer

Description

Randomised controlled trial from the Veteran's Administration Cooperative Urological Research Group. Includes patients who had stage III cancers and were randomised to placebo or daily oral treatment with 1.0 mg of diethylstilbesterol (DES).

Usage

prostatic
prostatic

Format

A data frame with 38 rows and 8 variables:

patient: integer patient identifier
treatment: integer treatment indicator (1=placebo; 2=daily treatment with 1.0 mg of diethylstilbesterol (DES))
time: integ er survival time from trial entry to end of follow-up in months
status: integer for follow-up status (0=alive or died from other causes, 1=died from prostate cancer
age: integer age at trial entry in years
shb: double serum hemoglobin at trial entry in g/dL
size: integer size of the primary tumour in cm^3
index: integer Gleason index based on histopathology

Details

TBC.

References

Andrews DF, Herzberg AM. Data: a collection of problems from many fields for the student and research worker. Springer Series in Statistics; Springer New York, NY; 1985. doi:10.1007/978-1-4612-5098-2

Pulmonary metastasis

Description

A very simple dataset with no censoring

Usage

pulmonary
pulmonary

Format

A data frame with 11 rows and 1 variables:

time: integer survival time from pulmonary metastasis to death in months

Details

See Collett (2023)

Simulated data

Description

Simulated data with left truncated follow-up and potentially right censored outcomes.

Usage

simdata
simdata

Format

A data frame with 30 observations and 8 variables:

id: integer index each individual
trt: numeric for whether treated (1=treated; 0=not treated)
age: integer for age in years
entry_time: numeric for year of entry
observed_duration: numeric for years that an individual was observed
status: integer for status at the end of follow-up (1=event, 0=censored)
event_calendar_time: numeric hypothetical event time in calendar time
stop_calendar_time: numeric end of follow-up in calendar time

Examples

## Simulate 30 individuals survival based on Weibull distribution
set.seed(13579)
n <- 30

## Randomly assign treatment groups (15 each)
trt <- sample(c(1, 0), n, replace = TRUE)

## Randomly assign integer age 50-80 to each individual
age <- sample(50:80, n, replace = TRUE)

## Simulate true event times based on Weibull distribution
true_shape <- 3
true_scale <- 8
true_times <- rweibull(n, shape = true_shape, scale = true_scale)

## Simulate right censoring times based on exponential distribution
censoring_rate <- 0.1
censoring_times <- rexp(n, rate = censoring_rate)

## Random entry times
entry_time <- runif(n, min = 2000, max = 2010)

## Convert durations (time-on-study) to calendar times
event_calendar_time <- entry_time + true_times
censor_calendar_time <- entry_time + censoring_times

## Study end time (Administrative Censoring)
study_end_time <- 2012
study_censor_calendar_time <- rep(study_end_time, n)

## Determine the calendar time when observation ends
stop_calendar_time <- pmin(event_calendar_time, censor_calendar_time,
                           study_censor_calendar_time)

## Calculate the observed duration
observed_duration <- stop_calendar_time - entry_time

## Create tied data
observed_duration <- round(observed_duration * 2) / 2

## Determine the final status (1 if event, 0 if censored)
status <- as.integer(event_calendar_time <= pmin(censor_calendar_time, study_censor_calendar_time))

## Create a data frame
simdata <- data.frame(
  id = 1:n,
  trt, ## Treatment group
  age, ## Age at diagnosis
  entry_time, ## When they entered (Calendar time)
  observed_duration, ## Time-on-study
  status, ## Event status (1=event, 0=censored)
  event_calendar_time, ## Hypothetical event time (Calendar time)
  stop_calendar_time ## When observation ended (Calendar time)
)
## Save the data frame
## save(simdata, file = "~/src/R/collett/data/simdata.rda")
## Simulate 30 individuals survival based on Weibull distribution
set.seed(13579)
n <- 30

## Randomly assign treatment groups (15 each)
trt <- sample(c(1, 0), n, replace = TRUE)

## Randomly assign integer age 50-80 to each individual
age <- sample(50:80, n, replace = TRUE)

## Simulate true event times based on Weibull distribution
true_shape <- 3
true_scale <- 8
true_times <- rweibull(n, shape = true_shape, scale = true_scale)

## Simulate right censoring times based on exponential distribution
censoring_rate <- 0.1
censoring_times <- rexp(n, rate = censoring_rate)

## Random entry times
entry_time <- runif(n, min = 2000, max = 2010)

## Convert durations (time-on-study) to calendar times
event_calendar_time <- entry_time + true_times
censor_calendar_time <- entry_time + censoring_times

## Study end time (Administrative Censoring)
study_end_time <- 2012
study_censor_calendar_time <- rep(study_end_time, n)

## Determine the calendar time when observation ends
stop_calendar_time <- pmin(event_calendar_time, censor_calendar_time,
                           study_censor_calendar_time)

## Calculate the observed duration
observed_duration <- stop_calendar_time - entry_time

## Create tied data
observed_duration <- round(observed_duration * 2) / 2

## Determine the final status (1 if event, 0 if censored)
status <- as.integer(event_calendar_time <= pmin(censor_calendar_time, study_censor_calendar_time))

## Create a data frame
simdata <- data.frame(
  id = 1:n,
  trt, ## Treatment group
  age, ## Age at diagnosis
  entry_time, ## When they entered (Calendar time)
  observed_duration, ## Time-on-study
  status, ## Event status (1=event, 0=censored)
  event_calendar_time, ## Hypothetical event time (Calendar time)
  stop_calendar_time ## When observation ended (Calendar time)
)
## Save the data frame
## save(simdata, file = "~/src/R/collett/data/simdata.rda")

Utilities

Description

Given a data-frame with an "s" step function, expand the data-frame to include the steps.

Given a survfit object, return a data-frame

Given a summary.survfit object, return a data-frame

Calculates the linear predictor. Typically, the tt function should include an intercept term (see the examples below). Note that spline terms assume that the x argument is multiplicative; moreover, the additional arguments are not passed. For other types of tt terms, the x is passed directly to the tt function together with other arguments.

Usage

step_s(
  data,
  x,
  y,
  ymin,
  ymax,
  group,
  add_origin = TRUE,
  x_origin = 0,
  y_origin = 1
)

## S3 method for class 'survfit'
as.data.frame(x, row.names, optional, type = c("expanded", "plain"), ...)

## S3 method for class 'summary.survfit'
as.data.frame(x, row.names, optional, type = c("expanded", "plain"), ...)

predict_coxph_tt(object, newdata, type = "lp", se.fit = FALSE, ...)

predict_coxph_tv(object, data, id)

plot_coxph_functional(
  formula,
  data,
  x = NULL,
  pch = 19,
  ylab = "Martingale residual for null model",
  smoother = c("loess", "lm"),
  smoother.formula = resi ~ xi,
  smoother.args = list(),
  points.args = list(),
  ...
)
step_s(
  data,
  x,
  y,
  ymin,
  ymax,
  group,
  add_origin = TRUE,
  x_origin = 0,
  y_origin = 1
)

## S3 method for class 'survfit'
as.data.frame(x, row.names, optional, type = c("expanded", "plain"), ...)

## S3 method for class 'summary.survfit'
as.data.frame(x, row.names, optional, type = c("expanded", "plain"), ...)

predict_coxph_tt(object, newdata, type = "lp", se.fit = FALSE, ...)

predict_coxph_tv(object, data, id)

plot_coxph_functional(
  formula,
  data,
  x = NULL,
  pch = 19,
  ylab = "Martingale residual for null model",
  smoother = c("loess", "lm"),
  smoother.formula = resi ~ xi,
  smoother.args = list(),
  points.args = list(),
  ...
)

Arguments

data

a dataset for evaluation of the coxph model

x

a numeric vector for the smoother (defaults to the 401 values between the range)

y

name of the y-variable (required)

ymin

name of the ymin variable (optional)

ymax

name of the ymax variable (optional)

group

name of a grouping variable (optional)

add_origin

logical for whether to add an origin to the start of each group

x_origin

double for the value of x at the origin

y_origin

double for the value of y, ymin and ymax at the origin

row.names

not used (in generic signature)

optional

not used (in generic signature)

type

a character for the type of prediction (currently only the linear predictor for the tt argument)

...

other arguments to pass to the plot function

object

a coxph object

newdata

a data-frame used for the predictions

se.fit

a logical for whether to return the standard errors

id

a character for the subject id

formula

a formula with a Surv on the lhs and a single variable on the rhs

pch

an integer for the pch argument in the plot for the residuals

ylab

a character for the ylab argument in the plot

smoother

a character for the name of the smoother

smoother.formula

a formula for the smoother in terms of resi and xi

smoother.args

a list of arguments to pass to the smoother function

points.args

a list of arguments to pass to the points function

Value

expanded data-frame with the same names

a vector of fitted values (when se.fit=FALSE) or a data-frame with fitted and se.fit columns (when se.fit=TRUE)

an update of data with survival probabilities

invisible plot return

Examples

  step_s(data.frame(g=c(1,1), a=1:2, b=4:5), a, b)
  step_s(data.frame(g=c(2,2), a=3:4, b=6:7), a, b)
  step_s(data.frame(g=c(1,1,2,2), a=1:4, b=4:7), a, b, group=g)
  library(survival)
  library(tinyplot)
  sfit1 = survfit(Surv(time,status)~rx, data=survival::colon,
                  subset=etype==1)
  with(as.data.frame(sfit1),
       tinyplot::plt(surv~time|strata,ymin=lower,ymax=upper,type="ribbon"))
  library(survival)
  library(tinyplot)
  sfit1 = survfit(Surv(time,status)~rx, data=survival::colon,
                  subset=etype==1)
  with(as.data.frame(sfit1, type="expanded"),
       tinyplot::plt(surv~time|strata,ymin=lower,ymax=upper,type="ribbon"))
 library(splines)
 library(tinyplot)
 fit1 = coxph(Surv(time,status)~tt(treat),data=breast_rfs,
              tt=function(x,t,...) x*cbind(1,t))
 fit2 = coxph(Surv(time,status)~tt(treat),data=breast_rfs,
              tt=function(x,t,...) x*ns(t,df=4,intercept=TRUE))
 times = seq(0,2500,len=301L)
 nd = data.frame(treat=1,time=times)
 df1 = transform(predict_coxph_tt(fit1,nd,se.fit=TRUE),
                 lower=exp(fitted-1.96*se.fit),
                 upper=exp(fitted+1.96*se.fit),
                 fitted=exp(fitted),
                 model="linear",times=times)
 df2 = transform(predict_coxph_tt(fit2,nd,se.fit=TRUE),
                 lower=exp(fitted-1.96*se.fit),
                 upper=exp(fitted+1.96*se.fit),
                 fitted=exp(fitted),
                 model="ns",times=times)
 with(rbind(df1,df2),
      plt(fitted ~ times | model, ymin=lower, ymax=upper, type="ribbon",
          xlab="Time since diagnosis (days)",
          ylab="Hazard ratio comparing treated with untreated"))
 with(subset(breast_rfs,status==1), rug(time))
library(survival)
liver = transform(collett::liverbase, lbr=NULL)
liver = tmerge(liver, liver, id=patient, status=event(time,status))
liver = tmerge(liver,
               rbind(with(collett::liverbase, data.frame(patient,tstart=0,lbr)),
                     with(collett::lbrdata0, data.frame(patient,tstart=time,lbr))),
           id=patient, lbr = tdc(tstart,lbr))
fit3 = coxph(Surv(tstart,tstop,status)~lbr+treat,liver)
predict_coxph_tv(fit3,data=subset(liver,patient %in% c(1,7)),
                 id="patient")
library(survival)
par(mfrow=c(2,2))
plot_coxph_functional(Surv(time,status)~hb, data=collett::myeloma,
                      xlab="Value of Hb")
plot_coxph_functional(Surv(time,status)~bun, data=collett::myeloma,
                      xlab="Value of Bun")
plot_coxph_functional(Surv(time,status)~log(bun), data=collett::myeloma,
                      xlab="Value of log Bun")
step_s(data.frame(g=c(1,1), a=1:2, b=4:5), a, b)
  step_s(data.frame(g=c(2,2), a=3:4, b=6:7), a, b)
  step_s(data.frame(g=c(1,1,2,2), a=1:4, b=4:7), a, b, group=g)
  library(survival)
  library(tinyplot)
  sfit1 = survfit(Surv(time,status)~rx, data=survival::colon,
                  subset=etype==1)
  with(as.data.frame(sfit1),
       tinyplot::plt(surv~time|strata,ymin=lower,ymax=upper,type="ribbon"))
  library(survival)
  library(tinyplot)
  sfit1 = survfit(Surv(time,status)~rx, data=survival::colon,
                  subset=etype==1)
  with(as.data.frame(sfit1, type="expanded"),
       tinyplot::plt(surv~time|strata,ymin=lower,ymax=upper,type="ribbon"))
 library(splines)
 library(tinyplot)
 fit1 = coxph(Surv(time,status)~tt(treat),data=breast_rfs,
              tt=function(x,t,...) x*cbind(1,t))
 fit2 = coxph(Surv(time,status)~tt(treat),data=breast_rfs,
              tt=function(x,t,...) x*ns(t,df=4,intercept=TRUE))
 times = seq(0,2500,len=301L)
 nd = data.frame(treat=1,time=times)
 df1 = transform(predict_coxph_tt(fit1,nd,se.fit=TRUE),
                 lower=exp(fitted-1.96*se.fit),
                 upper=exp(fitted+1.96*se.fit),
                 fitted=exp(fitted),
                 model="linear",times=times)
 df2 = transform(predict_coxph_tt(fit2,nd,se.fit=TRUE),
                 lower=exp(fitted-1.96*se.fit),
                 upper=exp(fitted+1.96*se.fit),
                 fitted=exp(fitted),
                 model="ns",times=times)
 with(rbind(df1,df2),
      plt(fitted ~ times | model, ymin=lower, ymax=upper, type="ribbon",
          xlab="Time since diagnosis (days)",
          ylab="Hazard ratio comparing treated with untreated"))
 with(subset(breast_rfs,status==1), rug(time))
library(survival)
liver = transform(collett::liverbase, lbr=NULL)
liver = tmerge(liver, liver, id=patient, status=event(time,status))
liver = tmerge(liver,
               rbind(with(collett::liverbase, data.frame(patient,tstart=0,lbr)),
                     with(collett::lbrdata0, data.frame(patient,tstart=time,lbr))),
           id=patient, lbr = tdc(tstart,lbr))
fit3 = coxph(Surv(tstart,tstop,status)~lbr+treat,liver)
predict_coxph_tv(fit3,data=subset(liver,patient %in% c(1,7)),
                 id="patient")
library(survival)
par(mfrow=c(2,2))
plot_coxph_functional(Surv(time,status)~hb, data=collett::myeloma,
                      xlab="Value of Hb")
plot_coxph_functional(Surv(time,status)~bun, data=collett::myeloma,
                      xlab="Value of Bun")
plot_coxph_functional(Surv(time,status)~log(bun), data=collett::myeloma,
                      xlab="Value of log Bun")

Clinical trial of tamoxifen in breast cancer patients

Description

Clinical trial for breast cancer patients comparing combined tamoxifen and radiotherapy with tamoxifen alone.

Usage

tamoxifen
tamoxifen

Format

A data frame with 641 rows and 18 variables:

id: integer patient identifier
treat: integer treatment group (0=tamoxifen+radiotherapy, 1=tamoxifen)
age: integer patient age at study entry (years)
size: double tumour size (cm)
hist: integer tumour histology (1=ductal, 2=lobular, 3=medullary, 4=mixed, 5=other)
hr: integer hormone receptor level (0=negative, 1=positive)
hb: integer Haemoglobin level (g/l)
andis: integer axillary relapse (0=no, 1=yes)
lsurv: integer time to local relapse or last follow-up (days)
ls: integer local relapse (0=no, 1=yes))
asurv: integer time to axillary relapse or last follow-up (days)
as: integer axillary relapse (0=no, 1=yes)
dsurv: integer Time to distant relapse or last follow-up (days)
ds: integer distant relapse (0=no, 1=yes)
msurv: integer time to second malignancy or last follow-up (days)
ms: integer second malignancy (0=no, 1=yes)
tsurv: integer time from randomisation to death or last follow-up (days)
ts: integer status at last follw-up (0=alive, 1=dead)

Details

See Collett (2023)

Survival following kidney transplantation

Description

Survival following kidney transplantation

Usage

tplant
tplant

Format

A data frame with 434 rows and 7 variables:

patient: integer patient id
donor: integer donoe id
time: integer survival time in days
status: integer event indicator (0=censored, 1=graft failure or death with a functioning graft)
age: integer patient age (years)
diabetes: integer diabetes status (0=absent, 1=present)
cit: double cold ischaemic time, the time in hours between retrieval of the kidney from the donor and the transplantation

Details

See Collett (2023)

Recurrence of an ulcer

Description

A double-blind trial comparing two treatments for ulcers. Data from Belgium.

Usage

ulcer
ulcer

Format

A data frame with 43 rows and 6 variables:

patient: integer patient id
age: integer age at the end of the trial in years
duration: integer duration of verified disease (1: <5 years, 2: >=5 years
treatment: integer treatment arm (1=A,2=B)
time: integer time since last visit (months)
result: integer result of the last visit (1=no ulcer detected, 2=ulcer detected)

Details

See Collett (2023)

Survival following aortic valve replacement

Description

Patients following an aortic valve replacement are measured for left ventricular mass index (LVMI).

Usage

valve
valve

Format

A data frame with 988 rows and 11 variables:

id: integer patient id
futime: double total follow-up time from date of surgery (years)
status: integer event indicator (0=censored, 1=death)
time: double time of LVMI measurement after surgery (years)
lvmi: double standardised LVMI
age: integer age of patient in years
sex: integer sex of patient (0=male, 1=female)
redo: integer previous cardiac surgery (0=no, 1=yes)
emerg: integer operative urgency (0=elective, 1=urgent or emergency)
dm: integer preoperative diabetes mellitus (0=no, 1=yes)
type: integer type of valve (1=human tissue, 2=porcine tissue)

Details

See Collett (2023)

Package 'collett'

Help Index

Chronic active hepatitis

Description

Usage

Format

Details

Prognosis for women with breast cancer

Description

Usage

Format

Details

References

Examples

Recurrence of bladder cancer

Description

Usage

Format

Details

Bone marrow transplantation

Description

Usage

Format

Details

Patient outcome following bone marrow transplantation

Description

Usage

Format

Details

Recurrence free survival in breast cancer patients

Description

Usage

Format

Details

Datasets

Description

Usage

Arguments

Value

Author(s)

Source

See Also

Examples

Infection in patients on dialysis

Description

Usage

Format

Details

Survival of black ducks

Description

Usage

Format

Details

Survival of patients with gastric cancer

Description

Usage

Format

Details

Chronic granulomatous disease

Description

Usage

Format

Details

Health evaluation and linkage to primary care

Description

Usage

Format

Details

A numerical illustration

Description

Usage

Format

Details

Time to discontinuation of the use of an IUD

Description

Usage

Format

Details

Treatment of hypernephroma

Description