Example 2 PROC ICLIFETEST - Cosmetic Data

This script analyzes the cosmetic effects of two breast cancer treatments (Radiotherapy 'RT' vs Radio-chemotherapy 'RCT'). It creates the data via DATA steps, then uses the ICLIFETEST procedure to estimate survival functions, perform comparisons between groups (strata), and generate survival plots with confidence intervals. The 'impute' option is used to handle interval censoring via multiple imputation.

Data Analysis

Type : INTERNAL_CREATION

Data is created directly in the script via the DATALINES statement. Two tables (RT and RCT) are created then concatenated into the final BCS table.

1 Code Block

DATA STEP Data

Explanation :
Creation of raw datasets for each treatment group. The 'input lTime rTime @code_sas_json/8_SAS_Intro_ReadFile_MultiCol_@@.json;' statement (corrected from the prompt's artifact) allows reading multiple pairs of observations on the same line. The two tables are then stacked into the 'BCS' table.

Copied!

1	DATA RT;
2	INPUT lTime rTime @code_sas_json/8_SAS_Intro_ReadFile_MultiCol_@@.json;
3	trt = ' RT';
4	DATALINES;
5	45 . 25 37 37 .
6	6 10 46 . 0 5
7	0 7 26 40 18 .
8	46 . 46 . 24 .
9	46 . 27 34 36 .
10	7 16 36 44 5 11
11	17 . 46 . 19 35
12	7 14 36 48 17 25
13	37 44 37 . 24 .
14	0 8 40 . 32 .
15	4 11 17 25 33 .
16	15 . 46 . 19 26
17	11 15 11 18 37 .
18	22 . 38 . 34 .
19	46 . 5 12 36 .
20	46 .
21	;
22
23	DATA RCT;
24	INPUT lTime rTime @code_sas_json/8_SAS_Intro_ReadFile_MultiCol_@@.json;
25	trt = 'RCT';
26	DATALINES;
27	8 12 0 5 30 34
28	0 22 5 8 13 .
29	24 31 12 20 10 17
30	17 27 11 . 8 21
31	17 23 33 40 4 9
32	24 30 31 . 11 .
33	16 24 13 39 14 19
34	13 . 19 32 4 8
35	11 13 34 . 34 .
36	16 20 13 . 30 36
37	18 25 16 24 18 24
38	17 26 35 . 16 60
39	32 . 15 22 35 39
40	23 . 11 17 21 .
41	44 48 22 32 11 20
42	14 17 10 35 48 .
43	;
44
45	DATA BCS;
46	SET RT RCT;
47	RUN;

2 Code Block

PROC ICLIFETEST

Explanation :
First survival analysis. The 'plots=survival(cl)' option requests plotting the survival curve with confidence limits. 'impute(seed=1234)' activates imputation for interval-censored data with a fixed random seed for reproducibility. 'strata trt' defines the group variable.

Copied!

1	PROC ICLIFETEST DATA=BCS plots=survival(cl) impute(seed=1234);
2	strata trt;
3	time (lTime, rTime);
4	RUN;

3 Code Block

PROC ICLIFETEST

Explanation :
Second analysis similar to the first, but the graphic option 'nodash' is added so that confidence curves are plotted with solid lines rather than dashed lines.

Copied!

1	PROC ICLIFETEST DATA=BCS plots=survival(cl nodash) impute(seed=1234);
2	strata trt;
3	time (lTime, rTime);
4	RUN;

4 Code Block

PROC ICLIFETEST

Explanation :
Third analysis. The 'strata=panel' option requests separating the plots of each stratum (treatment group) into distinct panels.

Copied!

1	PROC ICLIFETEST DATA=BCS plots=survival(cl strata=panel) impute(seed=1234);
2	strata trt;
3	time (lTime, rTime);
4	RUN;

This material is provided "as is" by We Are Cas. There are no warranties, expressed or implied, as to merchantability or fitness for a particular purpose regarding the materials or code contained herein. We Are Cas is not responsible for errors in this material as it now exists or will exist, nor does We Are Cas provide technical support for it.

Retour à la liste