topic How do I get the 95% Confidence Limits for multiple data points in proc lifetest? in New SAS User

How do I get the 95% Confidence Limits for multiple data points in proc lifetest?

hpatel3 — Tue, 09 Oct 2018 14:16:22 GMT

Hello,

I'm trying to get 95% confidence limits for 5, 10, and 15 year individually on my on my KM survival curve. I know how to get the bands on my plot using plots=(survival(cb=hw)), but how do I get the specific confidence limit for each of my 3 data points? All I get with my current code are the quartile estimates.

I'm using Proc LifeTest and here is my code by the way. What I don't know is if I'm supposed use an ODS statement, or if there is a step within proc-lifetest to calculate the 95% CIs. and WHERE in my code is this step supposed to go? (fairly new to SAS)

My code is below:

libname Hetal "\\tuftsmc\home\hpatel3\SAS Datasets";
run; 

proc import out=hetal.ES_database datafile="\\tuftsmc\home\hpatel3\SAS Datasets\ES Database Edit.csv" dbms=csv replace;  getnames=yes; datarow=2; 
run;
ods graphics on / attrpriority=color; 
Ods output survivalplot=survplot;
proc lifetest data=hetal.es_database conftype=loglog plot=(s) plots=(survival(cb=hw))
timelist=(5 10) outs=survival_rates reduceout; 
time FU_Total*Status(0);
Title"5 & 10 Year Survival Estimate for ES";
run;

proc sgplot data=survplot noborder nowall;
step x=Time y=Survival / lineattrs=(color=darkblue thickness=2);
xaxis display=(noticks) label="Years" values=(0 to 10 by 2) min=0 max=10 labelattrs=(size=10 weight=bold );
yaxis label='Survival Probability' values=(0 to 1.0 by 0.2) min=0 max=1.0 labelattrs=(size=10 weight=bold);
inset / title="95% CI:10.4476 (5.770, 16.2245) " position=bottomright;
run;

Re: How do I get the 95% Confidence Limits for multiple data points in proc lifetest?

Reeza — Tue, 09 Oct 2018 15:32:21 GMT

Did you check all your output data sets?

Particularly, survival_rates?

@hpatel3 wrote:

Hello,

I'm trying to get 95% confidence limits for 5, 10, and 15 year individually on my on my KM survival curve. I know how to get the bands on my plot using plots=(survival(cb=hw)), but how do I get the specific confidence limit for each of my 3 data points? All I get with my current code are the quartile estimates.

I'm using Proc LifeTest and here is my code by the way. What I don't know is if I'm supposed use an ODS statement, or if there is a step within proc-lifetest to calculate the 95% CIs. and WHERE in my code is this step supposed to go? (fairly new to SAS)

My code is below:
libname Hetal "\\tuftsmc\home\hpatel3\SAS Datasets";
run; 

proc import out=hetal.ES_database datafile="\\tuftsmc\home\hpatel3\SAS Datasets\ES Database Edit.csv" dbms=csv replace;  getnames=yes; datarow=2; 
run;
ods graphics on / attrpriority=color; 
Ods output survivalplot=survplot;
proc lifetest data=hetal.es_database conftype=loglog plot=(s) plots=(survival(cb=hw))
timelist=(5 10) outs=survival_rates reduceout; 
time FU_Total*Status(0);
Title"5 & 10 Year Survival Estimate for ES";
run;

proc sgplot data=survplot noborder nowall;
step x=Time y=Survival / lineattrs=(color=darkblue thickness=2);
xaxis display=(noticks) label="Years" values=(0 to 10 by 2) min=0 max=10 labelattrs=(size=10 weight=bold );
yaxis label='Survival Probability' values=(0 to 1.0 by 0.2) min=0 max=1.0 labelattrs=(size=10 weight=bold);
inset / title="95% CI:10.4476 (5.770, 16.2245) " position=bottomright;
run;

Re: How do I get the 95% Confidence Limits for multiple data points in proc lifetest?

hpatel3 — Wed, 10 Oct 2018 12:32:39 GMT

yes! I have the survival rates and quartile estimates, but not the confidence limits.

Re: How do I get the 95% Confidence Limits for multiple data points in proc lifetest?

Reeza — Wed, 10 Oct 2018 14:58:16 GMT

Hmm...please post a proc contents on the survival_rates data set.

Note that your next step uses the survplot data set not the survival_rates data.

proc contents data=survival_rates;run;

From the documentation the OUTSURV/OUTS data sets include the following, which is what I get in my test. What happens if you run the code below? Does that generate the confidence intervals?

title 'Survival of Males with Angina Pectoris';
data Males;
   keep Freq Years Censored;
   retain Years -.5;
   input fail withdraw @@;
   Years + 1;
   Censored=0;
   Freq=fail;
   output;
   Censored=1;
   Freq=withdraw;
   output;
   datalines;
456   0 226  39 152  22 171  23 135 24 125 107
 83 133  74 102  51  68  42  64  43 45  34  53
 18  33   9  27   6  23   0  30
;

ods graphics on;
proc lifetest data=Males 
              plots=(s,ls,lls,h,p) outs=survival_Rates timelist = (0 to 15 by 3) reduceout;
   time Years*Censored(1) ;
   freq Freq;
   
run;
ods graphics off;

OUTSURV= Data Set

You can specify the OUTSURV= option in the PROC LIFETEST statement to create an output data set that contains the survival estimates. The data set contains the following columns:

any specified BY variables
a numeric variable STRATUM that numbers the strata, if you specify the STRATA statement
any specified STRATA variables, their values coming from either their original values or the midpoints of the stratum intervals if endpoints are used to define strata (semi-infinite intervals are labeled by their finite endpoint)
the GROUP= variables, if you specify the GROUP= option in the STRATA statement
the time variable as specified in the TIME statement. For METHOD=KM, METHOD=BRESLOW, or METHOD=FH, it contains the observed failure or censored times. For the life-table estimates, it contains the lower endpoints of the time intervals.
SURVIVAL, a variable that contains the survivor function estimates
CONFTYPE, a variable that contains the name of the transformation applied to the survival time in the computation of confidence intervals
SDF_LCL, a variable that contains the lower limits of the pointwise confidence intervals for the survivor function
SDF_UCL, a variable that contains the upper limits of the pointwise confidence intervals for the survivor function

@hpatel3 wrote:

yes! I have the survival rates and quartile estimates, but not the confidence limits.