BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
hpatel3
Obsidian | Level 7

I am doing a project where I need to create a KM curve for "freedom from heart arrhythmias at 1, 3, and 5 years post heart surgery" (where everyone in the dataset HAS had the surgery)

 

I've set up and my graph and felt like it was correct (Code and graph below)

However, when I sent it over to my boss, he said the numbers from last time this was done do not match up.

I've racked my brain and played around a lot with proc lifetest, but my numbers don't change too much no matter how I look at it.

 

Does anyone else know what variables COULD change survival rates in this scenario?  Brainstorming ideas welcome.  Because I feel like I've frankly tried everything short of just having my code be disastrously wrong.

 

My numbers, survival at:

1 year: 94%
3 years: 60%

5 years: 25%

 

My bosses numbers:
1 year:  84%
3 years: None
5 years: 64%

 

My variables:

abnormal heart beat after surgery Y (1) / N (0) =  Rec_AF

Living Status Alive (1) / Dead (0)  ** with those who had the event coded as 0) =  Status

Follow up time = tot.fu

 

(variables in data set now used: Gender, age at last FU, time interval till patient had event, type of event, )

 

ods graphics on / attrpriority=color; 
ods output survivalplot=survplot; 
run; 

proc lifetest data=hp.maze_database conftype=loglog plots=survival 
timelist=(1 3 5) outs=surv_rate reduceout alpha=.05; 
time tot_fu*Status(0);
title "AF recurrance after MAZE" ;
run; 

proc sgplot data=survplot noborder nowall; 
step x=time y=survival / lineattrs=(color=darkblue thickness=2);
xaxis label='Years post Maze' values=(0 to 5 by 1) labelattrs=(size=13 weight=bold);
yaxis label='Proportion Who have AF recurrence' values=(0 to 1.0 by 0.2) ;
run;

 

 

 

 

4-18 end curve.PNG

 

1 ACCEPTED SOLUTION

Accepted Solutions
Rick_SAS
SAS Super FREQ

You say, "My bosses numbers:"
1 year:  84%
3 years: None
5 years: 64%

 

Please ask your boss how 64% survived after 5 years when there were zero survivors after 3 years. Survival curves are nonincreasing functions, so either your boss is confused, or you misheard your boss, or you are using different definitions. 

View solution in original post

2 REPLIES 2
Reeza
Super User
Any chance they used Cox regression instead of just KM curves?
Rick_SAS
SAS Super FREQ

You say, "My bosses numbers:"
1 year:  84%
3 years: None
5 years: 64%

 

Please ask your boss how 64% survived after 5 years when there were zero survivors after 3 years. Survival curves are nonincreasing functions, so either your boss is confused, or you misheard your boss, or you are using different definitions. 

sas-innovate-2024.png

 

Time is running out to save with the early bird rate. Register by Friday, March 1 for just $695 - $100 off the standard rate.

 

Check out the agenda and get ready for a jam-packed event featuring workshops, super demos, breakout sessions, roundtables, inspiring keynotes and incredible networking events. 

 

Register now!

What is Bayesian Analysis?

Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 2 replies
  • 626 views
  • 3 likes
  • 3 in conversation