Median and CI results are incorrect between lifelines and Proc lifetest procedure from SAS

[indra3007]

Hi, I am comparing the results between the sas procedure PROC LIFETEST and LIFELINES for KM estimates. The results are incorrect. How can I get the correct results?

[CamDavidsonPilon]

You'll have to provide more information: dataset, results, code, etc.

[indra3007]

Here are the results.

Code and Results from python.
df_subset = df[df["TRT01PN"] == 81]
df.to_excel("/Users/invarma/Desktop/streamlit/test.xlsx", index=False)
T = df_subset["AVAL"]
E = df_subset["CNSR"]

# Fit the Kaplan-Meier survival model

kmf = KaplanMeierFitter()
kmf.fit(durations=T, event_observed=E)

# Calculate the median survival time

median = kmf.median_survival_time_
print(median)
Result = 167.0
# Calculate the median survival time confidence interval
median_confidence_interval = median_survival_times(kmf.confidence_interval_)
print(median_confidence_interval)
Result =
KM_estimate_lower_0.95 KM_estimate_upper_0.95
0.5 63.0 188.0

Sas code and results.

data all;
set adtte;
where trt01pn eq 81;
keep subjid trt01pn aval cnsr;
run;
proc lifetest data=all reduceout OUTSURV=ple conftype=loglog stderr;
time aval*cnsr(1);
strata trt01pn;
ods output Quartiles=surve;
run;
Result :
I have attached the screenshot.

Attached the data as well.
test.xlsx