I'd be grateful for your comments:
I've run what should be a simple Kaplan-Meier on both SPSS and SAS (Proc Lifetest). While the life table printouts are identical, SPSS and SAS differ in terms of the median survival time (0.75 and 0.79 yrs respectively) and in its 95% confidence interval {(0.45, 1.05) or (0.33, 2.00)}, the latter being quite a large difference. In each case there were 34 cases, with 3 censored.
Could you advise me as to why there should be these differences ?
Best wishes,
Martin Holt
Medical Statistician
Southern Derbyshire Acute Hospitals NHS Trust
[log in to unmask]
|