Hello,
I have run some cumulative risk estimates and produced some life tables,
using -stset- and -sts list-. I wanted to look at the period between
entry at risk and exit from risk for a cancer cohort where failure
(flagged by a variable: censvar) was developing a cancer.
I -stset- the data and calculated the interval between entry and exit in
years (generated variable: intvlexit). I then tabulated this variable
and found the maximum interval (entry to exit to be 48 years). When I
ran -sts list- on this data I set the time range from 0 - 65 as I don't
know in advance what the maximum interval is). I expected that, after
time value = 49, there would be 0 persons entering risk, but in the life
table all Time values from 49 to 65 have 1 record apparently "still at
risk".
The log file is as follows:
------------------------------------------------------------------------
----
log: M:\DaveW\wilms\LifeTable.log
log type: text
opened on: 24 Apr 2006, 16:26:33
. do "C:\DOCUME~1\WINTER~1.INT\LOCALS~1\Temp\STD04000000.tmp"
. /* STATA .do file to produce Life tables & Cum Risk */
. /* version 1 (APR 2006) */
. /* Dave Winter */
.
.
. display "RUNNING CUM-RISK USING TAB-Separated DATA..."
RUNNING CUM-RISK USING TAB-Separated DATA...
.
. quietly {
.
. /* Transform dates to years since B.C. Algorithm allows for same day
and month in year and leap years *
> /
. /* N.B. 28/2/anyyear = 59th day */
.
. gen byte extraDOB = 0
. gen byte extraENT = 0
. gen byte extraEX = 0
. replace extraDOB = 1 if doy(mdy(12,31,birthyr)) == 366 &
doy(mdy(birthmth,birthday,birthyr)) > 59
(302 real changes made)
. replace extraENT = 1 if doy(mdy(12,31,entryyr)) == 366 &
doy(mdy(entrymth,entryday,entryyr)) > 59
(262 real changes made)
. replace extraEX = 1 if doy(mdy(12,31,exityr)) == 366 &
doy(mdy(exitmth,exitday,exityr)) > 59
(39 real changes made)
. gen begyear = ((endate - mdy(1,1,entryyr) - extraENT)/365.25)+entryyr
. gen endyear = ((exdate - mdy(1,1,exityr) - extraEX)/365.25)+exityr
. gen byear = ((bdate - mdy(1,1,birthyr) - extraDOB)/365.25)+birthyr
.
. drop extraDOB extraENT extraEX
.
. /* 'stset' data to enable time in study ('intvlexit') calculated */
. stset endyear, fail(censvar) enter(begyear) origin(byear) id(indexno)
id: indexno
failure event: censvar != 0 & censvar < .
obs. time interval: (endyear[_n-1], endyear]
enter on or after: time begyear
exit on or before: failure
t for analysis: (time-origin)
origin: time byear
------------------------------------------------------------------------
------
1341 total obs.
0 exclusions
------------------------------------------------------------------------
------
1341 obs. remaining, representing
1341 subjects
41 failures in single failure-per-subject data
18467.25 total analysis time at risk, at risk from t = 0
earliest observed entry t = 15
last observed exit t = 63.19714
. gen intvlexit = int(_t - _t0)
. /* stset when time in study has been derived */
. stset intvlexit, fail(censvar)
failure event: censvar != 0 & censvar < .
obs. time interval: (0, intvlexit]
exit on or before: failure
------------------------------------------------------------------------
------
1341 total obs.
21 obs. end on or before enter()
------------------------------------------------------------------------
------
1320 obs. remaining, representing
41 failures in single record/single failure data
17798 total analysis time at risk, at risk from t = 0
earliest observed entry t = 0
last observed exit t = 48
.
. /* tabulation of interval entry to exit */
. tab intvlexit
intvlexit | Freq. Percent Cum.
------------+-----------------------------------
0 | 21 1.57 1.57
1 | 98 7.31 8.87
2 | 44 3.28 12.16
3 | 60 4.47 16.63
4 | 43 3.21 19.84
5 | 46 3.43 23.27
6 | 54 4.03 27.29
7 | 58 4.33 31.62
8 | 70 5.22 36.84
9 | 46 3.43 40.27
10 | 42 3.13 43.40
11 | 50 3.73 47.13
12 | 53 3.95 51.08
13 | 53 3.95 55.03
14 | 49 3.65 58.69
15 | 58 4.33 63.01
16 | 45 3.36 66.37
17 | 52 3.88 70.25
18 | 47 3.50 73.75
19 | 41 3.06 76.81
20 | 32 2.39 79.19
21 | 34 2.54 81.73
22 | 24 1.79 83.52
23 | 35 2.61 86.13
24 | 17 1.27 87.40
25 | 28 2.09 89.49
26 | 14 1.04 90.53
27 | 15 1.12 91.65
28 | 17 1.27 92.92
29 | 17 1.27 94.18
30 | 11 0.82 95.00
31 | 9 0.67 95.67
32 | 6 0.45 96.12
33 | 9 0.67 96.79
34 | 11 0.82 97.61
35 | 4 0.30 97.91
36 | 4 0.30 98.21
37 | 8 0.60 98.81
38 | 4 0.30 99.11
39 | 3 0.22 99.33
41 | 2 0.15 99.48
42 | 3 0.22 99.70
44 | 1 0.07 99.78
45 | 1 0.07 99.85
47 | 1 0.07 99.93
48 | 1 0.07 100.00
------------+-----------------------------------
Total | 1,341 100.00
.
. /* produce life table */
. sts list, failure at (0(1)65)
failure _d: censvar
analysis time _t: intvlexit
Beg. Failure Std.
Time Total Fail Function Error [95% Conf.
Int.]
------------------------------------------------------------------------
----
0 0 0 0.0000 . .
.
1 1320 0 0.0000 . .
.
2 1222 1 0.0008 0.0008 0.0001
0.0058
3 1178 0 0.0008 0.0008 0.0001
0.0058
4 1118 2 0.0026 0.0015 0.0008
0.0081
5 1075 4 0.0063 0.0024 0.0030
0.0132
6 1029 0 0.0063 0.0024 0.0030
0.0132
7 975 3 0.0094 0.0030 0.0050
0.0174
8 917 2 0.0115 0.0033 0.0066
0.0203
9 847 0 0.0115 0.0033 0.0066
0.0203
10 801 3 0.0152 0.0039 0.0092
0.0253
11 759 2 0.0178 0.0043 0.0111
0.0287
12 709 1 0.0192 0.0045 0.0121
0.0305
13 656 1 0.0207 0.0048 0.0132
0.0325
14 603 1 0.0223 0.0050 0.0143
0.0347
15 554 1 0.0241 0.0053 0.0156
0.0371
16 496 2 0.0280 0.0060 0.0184
0.0426
17 451 1 0.0302 0.0064 0.0200
0.0455
18 399 3 0.0375 0.0076 0.0252
0.0556
19 352 0 0.0375 0.0076 0.0252
0.0556
20 311 1 0.0406 0.0082 0.0273
0.0601
21 279 3 0.0509 0.0100 0.0346
0.0747
22 245 1 0.0548 0.0107 0.0373
0.0801
23 221 1 0.0590 0.0115 0.0403
0.0862
24 186 0 0.0590 0.0115 0.0403
0.0862
25 169 1 0.0646 0.0127 0.0439
0.0946
26 141 0 0.0646 0.0127 0.0439
0.0946
27 127 1 0.0720 0.0146 0.0483
0.1066
28 112 0 0.0720 0.0146 0.0483
0.1066
29 95 1 0.0817 0.0174 0.0537
0.1234
30 78 0 0.0817 0.0174 0.0537
0.1234
31 67 0 0.0817 0.0174 0.0537
0.1234
32 58 1 0.0976 0.0232 0.0609
0.1544
33 52 1 0.1149 0.0285 0.0702
0.1852
34 43 0 0.1149 0.0285 0.0702
0.1852
35 32 0 0.1149 0.0285 0.0702
0.1852
36 28 0 0.1149 0.0285 0.0702
0.1852
37 24 1 0.1518 0.0453 0.0835
0.2672
38 16 2 0.2578 0.0805 0.1359
0.4559
39 12 0 0.2578 0.0805 0.1359
0.4559
40 12 0 0.2578 0.0805 0.1359
0.4559
41 9 0 0.2578 0.0805 0.1359
0.4559
42 7 0 0.2578 0.0805 0.1359
0.4559
43 7 0 0.2578 0.0805 0.1359
0.4559
44 4 0 0.2578 0.0805 0.1359
0.4559
45 3 0 0.2578 0.0805 0.1359
0.4559
46 3 0 0.2578 0.0805 0.1359
0.4559
47 2 0 0.2578 0.0805 0.1359
0.4559
48 1 0 0.2578 0.0805 0.1359
0.4559
49 1 0 . . .
.
50 1 0 . . .
.
51 1 0 . . .
.
52 1 0 . . .
.
53 1 0 . . .
.
54 1 0 . . .
.
55 1 0 . . .
.
56 1 0 . . .
.
57 1 0 . . .
.
58 1 0 . . .
.
59 1 0 . . .
.
60 1 0 . . .
.
61 1 0 . . .
.
62 1 0 . . .
.
63 1 0 . . .
.
64 1 0 . . .
.
65 1 0 . . .
.
------------------------------------------------------------------------
----
Note: Failure function is calculated over full data and evaluated at
indicated times; it is not calculated from aggregates shown at
left.
.
. /* sts graph */
.
.
. display "ANALYSIS COMPLETED SUCCESSFULLY!"
ANALYSIS COMPLETED SUCCESSFULLY!
.
.
.
end of do-file
. log close
log: M:\DaveW\wilms\LifeTable.log
log type: text
closed on: 24 Apr 2006, 16:26:48
------------------------------------------------------------------------
----
Have I missed something very straight-forward (and potentially
embarrassing) here?
Best wishes,
Dave
****************************************************
Mr David L. Winter, HNC
I.T. Manager (BCCSS Study)
Department of Public Health & Epidemiology
Room 105, Public Health Building
University of Birmingham
Edgbaston
Birmingham
B15 2TT
UK
tel.: +44 (0)121 414 6766
fax.: +44 (0)121 414 7923
****************************************************
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
