Data from 2204 leukaemia patients who received a bone marrow transplant, with times to platelet recovery and to relapse or death (Collett, 2023, Example 13.4).
Format
A data frame with 2204 rows and 9 variables:
- id
Patient identifier.
- leukaemia
Disease type:
"AML","ALL"or"CML".- age
Age group:
"<=20","21-40"or">40".- match
Donor-recipient gender match: 0 = no, 1 = yes.
- tcell
T-cell depletion: 0 = no, 1 = yes.
- ptime
Time from transplantation to platelet recovery, in days.
- pcens
Platelet recovery indicator: 0 = censored, 1 = recovery.
- rdtime
Time from transplantation to relapse or death, in days.
- rdcens
Relapse-or-death indicator: 0 = censored, 1 = event.
Source
Collett, D. (2023) Modelling Survival Data in Medical Research, 4th ed. Chapman & Hall/CRC, Example 13.4 (Table 13.9). Data from the EBMT Registry, see Putter, Fiocco and Geskus (2007).
Details
The data come from the European Group for Blood and Marrow Transplantation (EBMT) Registry, and are the version used by Putter, Fiocco and Geskus (2007). Following the transplant a patient may either recover platelet function and later relapse or die, or relapse or die before platelet recovery. The data therefore describe a three-state model with transition rates for transplantation to platelet recovery, transplantation to relapse or death, and platelet recovery to relapse or death.
Of the 2204 patients, 1169 experienced platelet recovery and 458 relapsed
or died before their platelet count had returned to normal; a further 383
patients relapsed or died after platelet recovery. Note that the ptime
of a patient who relapsed before recovering platelets is censored at the
time of relapse when the first transition is modelled. To fit the model the
data are expanded into one row per possible transition per patient
(Example 13.4).
Aliases
The data set is available under the alias d13.4; dX.Y is the alias for
the data set used in Example X.Y of Collett (2023). See
dataset_by_exercise for the complete index.
Examples
# Collett (2023), Chapter 17 style -----------------------------------
# Example 13.4: the two endpoints of the three-state model
# 1 -> 2: platelet recovery
survival::coxph(
survival::Surv(ptime, pcens) ~ leukaemia + age + match + tcell,
data = bmt_outcome
)
#> Call:
#> survival::coxph(formula = survival::Surv(ptime, pcens) ~ leukaemia +
#> age + match + tcell, data = bmt_outcome)
#>
#> coef exp(coef) se(coef) z p
#> leukaemiaAML 0.04399 1.04497 0.07789 0.565 0.5722
#> leukaemiaCML -0.25403 0.77567 0.08297 -3.062 0.0022
#> age<=20 0.16313 1.17719 0.07905 2.064 0.0390
#> age>40 0.07137 1.07398 0.06776 1.053 0.2922
#> match 0.04745 1.04859 0.06661 0.712 0.4762
#> tcell 0.43133 1.53931 0.08040 5.365 8.11e-08
#>
#> Likelihood ratio test=56.99 on 6 df, p=1.835e-10
#> n= 2204, number of events= 1169
# 1 -> 3: relapse or death before recovery
survival::coxph(
survival::Surv(rdtime, rdcens) ~ leukaemia + age + match + tcell,
data = bmt_outcome
)
#> Call:
#> survival::coxph(formula = survival::Surv(rdtime, rdcens) ~ leukaemia +
#> age + match + tcell, data = bmt_outcome)
#>
#> coef exp(coef) se(coef) z p
#> leukaemiaAML -0.21088 0.80988 0.09972 -2.115 0.0345
#> leukaemiaCML -0.06129 0.94055 0.09850 -0.622 0.5338
#> age<=20 -0.17669 0.83804 0.10710 -1.650 0.0990
#> age>40 0.37699 1.45789 0.07609 4.955 7.25e-07
#> match 0.03522 1.03585 0.07922 0.445 0.6566
#> tcell 0.18945 1.20858 0.09523 1.989 0.0467
#>
#> Likelihood ratio test=49.74 on 6 df, p=5.294e-09
#> n= 2204, number of events= 841
# Modern tools --------------------------------------------------------
# Transition-specific hazards are fitted by stratifying on the transition
if (requireNamespace("mstate", quietly = TRUE)) {
tmat <- mstate::transMat(
list(c(2, 3), 3, c()),
names = c("transplant", "recovery", "relapse")
)
tmat
}
#> to
#> from transplant recovery relapse
#> transplant NA 1 2
#> recovery NA NA 3
#> relapse NA NA NA