I am attempting to replicate a previously published method using an example dataset. From the bolded line (///KM on original cohort (naive analysis)) the code works, and generates a point estimate of survival and restricted mean survival time. However, I wish to also obtain the 95% confidence intervals via bootstrapping (100 replications) for these endpoints. How can I modify this code to accomplish this aim? Thank you for your time and effort.
************************************************** *********************************************
* VARIABLES
*id: patient identifier
*fup_obs: observed follow-up time (time to death or 1 year if censored alive)
*death: observed event of interest (all-cause death) 1: dead, 0:alive
*timetosurgery: time to surgery (NA if no surgery)
*surgery: observed treatment 1 if the patient received surgery within 6 month, 0 otherwise
*age: age at diagnosis
*sex: patient's sex
*perf: performance status at diagnosis
*stage: stage at diagnosis
*deprivation: deprivation score
*charlson: Charlson's comorbidity index
*emergency: route to diagnosis
*Dataset: these variables are in a dataset called "SelectedCohort.dta"
************************************************** *********************************************
forval i = 0(1)100 {
/// selected patients
use "SelectedCohort.dta", replace
if `i' > 0 {
bsample
}
/// KM on original cohort (naive analysis)
stset fup_obs, failure(death)
sts generate KM_o_s_`i' = s if surgery == 1
sts generate KM_o_ns_`i' = s if surgery == 0
integ KM_o_s_`i' _t, gen(RMST_o_s_`i')
integ KM_o_ns_`i' _t, gen(RMST_o_ns_`i')
preserve
rename fup_obs fup
keep fup KM_o_s_ KM_o_ns_ RMST_o_s_ RMST_o_ns_
bysort fup (KM_o_s_`i'): replace KM_o_s_`i' = KM_o_s_`i'[1]
bysort fup (KM_o_ns_`i'): replace KM_o_ns_`i' = KM_o_ns_`i'[1]
bysort fup (RMST_o_s_`i'): replace RMST_o_s_`i' = RMST_o_s_`i'[1]
bysort fup (RMST_o_ns_`i'): replace RMST_o_ns_`i' = RMST_o_ns_`i'[1]
duplicates drop
keep if fup >365
sort fup
cap merge fup using "BootstrapResults.dta"
cap drop _m
sort fup
save "BootstrapResults.dta", replace
restore
************************************************** *********************************************
* VARIABLES
*id: patient identifier
*fup_obs: observed follow-up time (time to death or 1 year if censored alive)
*death: observed event of interest (all-cause death) 1: dead, 0:alive
*timetosurgery: time to surgery (NA if no surgery)
*surgery: observed treatment 1 if the patient received surgery within 6 month, 0 otherwise
*age: age at diagnosis
*sex: patient's sex
*perf: performance status at diagnosis
*stage: stage at diagnosis
*deprivation: deprivation score
*charlson: Charlson's comorbidity index
*emergency: route to diagnosis
*Dataset: these variables are in a dataset called "SelectedCohort.dta"
************************************************** *********************************************
forval i = 0(1)100 {
/// selected patients
use "SelectedCohort.dta", replace
if `i' > 0 {
bsample
}
/// KM on original cohort (naive analysis)
stset fup_obs, failure(death)
sts generate KM_o_s_`i' = s if surgery == 1
sts generate KM_o_ns_`i' = s if surgery == 0
integ KM_o_s_`i' _t, gen(RMST_o_s_`i')
integ KM_o_ns_`i' _t, gen(RMST_o_ns_`i')
preserve
rename fup_obs fup
keep fup KM_o_s_ KM_o_ns_ RMST_o_s_ RMST_o_ns_
bysort fup (KM_o_s_`i'): replace KM_o_s_`i' = KM_o_s_`i'[1]
bysort fup (KM_o_ns_`i'): replace KM_o_ns_`i' = KM_o_ns_`i'[1]
bysort fup (RMST_o_s_`i'): replace RMST_o_s_`i' = RMST_o_s_`i'[1]
bysort fup (RMST_o_ns_`i'): replace RMST_o_ns_`i' = RMST_o_ns_`i'[1]
duplicates drop
keep if fup >365
sort fup
cap merge fup using "BootstrapResults.dta"
cap drop _m
sort fup
save "BootstrapResults.dta", replace
restore
