Title: Stata 4, Survival
1Stata 4, Survival
- Hein Stigum
- Presentation, data and programs at
- http//folk.uio.no/heins/
2Agenda
- Kaplan-Meier plots
- Cox regression
- Example
- Age at first intercourse
3Survival data
Outcome
- Unajusted analysis
- Kaplan-Meier
- Regression method
- Cox-regression
4Survival data setup
- Status and time
- generate status!missing(DebutAge)
- generate timeDebutAge
- replace timeAge if status0
- generate time2timeuniform() avoid ties
- Set and describe
- stset time, failure(status1) Set data
- stdes Describe
5Setting the timescale
Time time since diagnosis in years stset
dateexit, failure(dead1) origin(datediag)
scale(365.25)
Time age in years stset dateexit,
failure(dead1) origin(datebth) enter(datediag)
scale(365.25)
6Mathematical functions
- Standard distribution functions
- Time to event T
- Density f(t)
- Cumulative density F(t)
- Survival functions
7Some relationships
8Kaplan-Meier
- Survival function
- Syntax
- sts graph, by(sex) KM survival plot
- sts test sex log-rank test
- stci, p(50) by(sex) time to 50 failure
- sts list, at(5 10 30) survival at time 5,
9Kaplan-Meier, all
sts graph, fail gwood tmin(8) tmax(30) noorigin
Age at 50 failure stci, p(50) 18.4 (18.1,18.8)
10Kaplan-Meier, by sex
sts graph, fail by(gender) tmin(8) tmax(30)
noorigin
Age at 50 failure stci, p(50) by
(gender) Males 18.6 (18.3,19.0) Females
18.1(17.8,18.9)
Log-rank test sts test gender p-value0.3
11Hazards
sts graph, hazard by(gender) width(2)
12Cox regression
- Model
- Syntax
- stcox x1 x2
- Proportional hazard test
- stcox x1 x2, schoenfeld(sc) scaledsch(ssc)
- estat phtest, detail
- estat phtest, plot(x1)
13Full model
stcox gender cohab partfrq
14Proportion hazard test
Save residuals stcox gender cohab partfrq,
schoenfeld(sc) scaledsch(ssc) Test estat
phtest, detail
15Smoothed Schoenfeld residuals
estat phtest, plot(cohab)
16Baseline hazard
stcox gender cohab partfrq, basesurv(bsurv)
basehc(bhaz) stcurve, hazard at(gender1 cohab1
partfrq0) range(8 30) width(1)
17Predicted survival
stcurve, survival at1(gender1 cohab1
partfrq0) at2(gender2 cohab1 partfrq0)
18If proptional hazard fails
- Stratified Cox regression
- Separate analysis on time intervals
- Time dependent covariats
- Additive model
19Some Cox options
- stcox drug age, strata(sex) Stratified
- stcox drug age, shared(family) Frailty
- stcox drug age, tvc(varlist) Timevar cov