Landmark analysis of survival

Subjects failing or censored before the landmark are dropped and the clock is reset there, which removes the immortal-time bias that arises from conditioning on a post-baseline event.

Usage

Landmark(time, event, landmark_time, X = NULL, group = NULL, alpha = 0.05)

morie_survival_landmark(
  time,
  event,
  landmark_time,
  X = NULL,
  group = NULL,
  alpha = 0.05
)

Arguments

time

observed follow-up times.

event

event indicator, 1 = event, 0 = censored.

landmark_time

the landmark at which the clock is reset.

X

optional covariate matrix.

group

optional grouping vector.

alpha

significance level for the interval.

Value

list with the landmark survival estimate, the number retained and dropped, and the group comparison when supplied.

References

Anderson, J. R., Cain, K. C. and Gelber, R. D. (1983). Analysis of survival by tumor response. Journal of Clinical Oncology 1(11), 710-719.

Examples

Landmark(time = c(1, 2, 3, 4, 5, 6, 7, 8), event = c(0, 1, 0, 1, 1, 0, 1, 0), landmark_time = 5L)
#> $landmark
#> [1] 5
#> 
#> $n_original
#> [1] 8
#> 
#> $alpha
#> [1] 0.05
#> 
#> $km_lower
#> [1] 0
#> 
#> $km_upper
#> [1] 1
#> 
#> $n_retained
#> [1] 3
#> 
#> $n_dropped
#> [1] 5
#> 
#> $kept_index
#> [1] 6 7 8
#> 
#> $time
#> [1] 1 2 3
#> 
#> $event
#> [1] 0 1 0
#> 
#> $km_time
#> [1] 2
#> 
#> $km_surv
#> [1] 0.5
#> 
#> $conditional_on_surviving_to_landmark
#> [1] TRUE
#> 
#> $method
#> [1] "landmark analysis; the clock is reset at the landmark and earlier subjects are dropped"
#>