Three data frames, ndmm_ipd, ndmm_agd, and ndmm_agd_covs containing
(simulated) individual patient data (IPD) from three studies and aggregate
data (AgD) from two studies on newly diagnosed multiple myeloma. The outcome
of interest is progression-free survival after autologous stem cell
transplant. The IPD studies in ndmm_ipd provide event/censoring times and
covariate values for each individual. The AgD studies provide reconstructed
event/censoring times from digitized Kaplan-Meier curves in ndmm_agd and
covariate summaries in ndmm_agd_covs, obtained from published trial
reports. The data are constructed to resemble those used by
Leahy2019;textualmultinma.
ndmm_ipdndmm_agd
ndmm_agd_covs
The individual patient data are contained in a data frame ndmm_ipd
with 1325 rows, one per individual, and 10 variables:
study name
treatment name
event/censoring time
censoring indicator (0 = censored, 1 = event)
age (years)
ISS stage 3 (0 = no, 1 = yes)
complete or very good partial response (0 = no, 1 = yes)
male sex (0 = no, 1 = yes)
The reconstructed Kaplan-Meier data for the aggregate studies are
contained in a data frame ndmm_agd with 2819 rows and 6 variables:
study name
treatment name
event/censoring time
censoring indicator (0 = censored, 1 = event)
The covariate summaries extracted from published reportes for the
aggregate studies are contained in a data frame ndmm_agd_covs with 4
rows, one per study arm, and 15 columns:
study name
treatment name
sample size in each arm
summary statistics for age (years)
proportion of participants with ISS stage 3
proportion of participants with complete or very good partial response
proportion of male participants