BMT data

The FREQ Procedure

g Frequency Percent Cumulative
Frequency		 Cumulative
Percent
1 38 27.74 38 27.74
2 54 39.42 92 67.15
3 45 32.85 137 100.00

d1 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 56 40.88 56 40.88
1 81 59.12 137 100.00

d2 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 95 69.34 95 69.34
1 42 30.66 137 100.00

d3 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 54 39.42 54 39.42
1 83 60.58 137 100.00

A Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 111 81.02 111 81.02
1 26 18.98 137 100.00

C Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 76 55.47 76 55.47
1 61 44.53 137 100.00

P Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 17 12.41 17 12.41
1 120 87.59 137 100.00

Z3 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 57 41.61 57 41.61
1 80 58.39 137 100.00

Z4 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 49 35.77 49 35.77
1 88 64.23 137 100.00

Z5 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 69 50.36 69 50.36
1 68 49.64 137 100.00

Z6 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 79 57.66 79 57.66
1 58 42.34 137 100.00

Z8 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 92 67.15 92 67.15
1 45 32.85 137 100.00

Z9 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
1 76 55.47 76 55.47
2 17 12.41 93 67.88
3 23 16.79 116 84.67
4 21 15.33 137 100.00

Z10 Frequency Percent Cumulative
Frequency		 Cumulative
Percent
0 97 70.80 97 70.80
1 40 29.20 137 100.00


BMT data

The MEANS Procedure

Variable N Mean Std Dev Minimum Maximum
T1
T2
TA
TC
TP
Z1
Z2
Z7
137
137
137
137
137
137
137
137
839.1605839
782.0291971
718.3065693
466.7372263
49.0656934
28.3649635
28.3284672
275.0875912
727.3710845
741.9281908
742.0735313
636.7459355
153.5894645
9.5577778
10.1779097
364.6558304
1.0000000
1.0000000
1.0000000
1.0000000
0
7.0000000
2.0000000
24.0000000
2640.00
2640.00
2640.00
2640.00
1298.00
52.0000000
56.0000000
2616.00


Model 1

The PHREG Procedure

Model Information
Data Set T.BMT
Dependent Variable T2
Censoring Variable d2
Censoring Value(s) 0
Ties Handling BRESLOW

Number of Observations Read
Number of Observations Used
137
137

Summary of the Number of Event and Censored
Values
Total Event Censored Percent
Censored
137 42 95 69.34

Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Model Fit Statistics
Criterion Without
Covariates		 With
Covariates
-2 LOG L 380.487 379.012
AIC 380.487 385.012
SBC 380.487 390.225

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 1.4753 3 0.6880
Score 1.2862 3 0.7324
Wald 1.2378 3 0.7440

Analysis of Maximum Likelihood Estimates
Variable DF Parameter
Estimate		 Standard
Error		 Chi-Square Pr > ChiSq Hazard
Ratio		 95% Hazard Ratio Confidence
Limits
acute 1 -0.76843 0.73417 1.0955 0.2953 0.464 0.110 1.955
chronic 1 -0.19008 0.41780 0.2070 0.6491 0.827 0.365 1.875
both 1 0.77638 0.98648 0.6194 0.4313 2.174 0.314 15.027


Model 2

The PHREG Procedure

Model Information
Data Set T.BMT
Dependent Variable T2
Censoring Variable d2
Censoring Value(s) 0
Ties Handling BRESLOW

Number of Observations Read
Number of Observations Used
137
137

Summary of the Number of Event and Censored
Values
Total Event Censored Percent
Censored
137 42 95 69.34

Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Model Fit Statistics
Criterion Without
Covariates		 With
Covariates
-2 LOG L 380.487 379.018
AIC 380.487 385.018
SBC 380.487 390.231

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 1.4691 3 0.6894
Score 1.2843 3 0.7329
Wald 1.2371 3 0.7441

Analysis of Maximum Likelihood Estimates
Variable DF Parameter
Estimate		 Standard
Error		 Chi-Square Pr > ChiSq Hazard
Ratio		 95% Hazard Ratio Confidence
Limits
acute 1 -0.76446 0.73415 1.0843 0.2977 0.466 0.110 1.963
chronic 1 -0.19237 0.41769 0.2121 0.6451 0.825 0.364 1.871
both 1 0.76075 0.98649 0.5947 0.4406 2.140 0.310 14.794


Obs with error

Obs g T1 T2 d1 d2 d3 TA A TC C TP P Z1 Z2 Z3 Z4 Z5 Z6 Z7 Z8 Z9 Z10
1 1 269 110 1 1 1 269 0 120 1 27 1 29 20 0 1 1 1 361 0 2 1
2 3 168 168 1 0 1 168 0 200 1 48 1 32 43 0 1 1 1 150 1 2 1


Model 3

The PHREG Procedure

Model Information
Data Set T.BMT
Dependent Variable T2
Censoring Variable d2
Censoring Value(s) 0
Ties Handling BRESLOW

Number of Observations Read
Number of Observations Used
137
137

Summary of the Number of Event and Censored
Values
Total Event Censored Percent
Censored
137 42 95 69.34

Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Model Fit Statistics
Criterion Without
Covariates		 With
Covariates
-2 LOG L 380.487 343.325
AIC 380.487 375.325
SBC 380.487 403.128

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 37.1618 16 0.0020
Score 35.8112 16 0.0031
Wald 30.6130 16 0.0151

Analysis of Maximum Likelihood Estimates
Variable DF Parameter
Estimate		 Standard
Error		 Chi-Square Pr > ChiSq Hazard
Ratio		 95% Hazard Ratio Confidence
Limits
g2 1 -2.04750 0.60335 11.5163 0.0007 0.129 0.040 0.421
g3 1 -0.55461 0.53398 1.0788 0.2990 0.574 0.202 1.636
Z1 1 0.04173 0.03251 1.6481 0.1992 1.043 0.978 1.111
Z2 1 -0.03151 0.02850 1.2223 0.2689 0.969 0.916 1.025
Z3 1 -0.30865 0.35746 0.7455 0.3879 0.734 0.364 1.480
Z4 1 0.37208 0.37589 0.9798 0.3222 1.451 0.694 3.031
Z5 1 0.17324 0.38386 0.2037 0.6518 1.189 0.560 2.523
Z6 1 -0.08540 0.37217 0.0526 0.8185 0.918 0.443 1.904
Z7 1 -0.00101 0.0007852 1.6484 0.1992 0.999 0.997 1.001
Z8 1 1.39900 0.45325 9.5272 0.0020 4.051 1.666 9.849
h2 1 -0.07719 0.59672 0.0167 0.8971 0.926 0.287 2.981
h3 1 0.15453 0.44415 0.1211 0.7279 1.167 0.489 2.787
h4 1 -0.58339 0.58016 1.0112 0.3146 0.558 0.179 1.740
Z10 0 0 . . . . . .
acute 1 -0.61748 0.76421 0.6529 0.4191 0.539 0.121 2.412
chronic 1 -0.29656 0.45051 0.4333 0.5104 0.743 0.307 1.798
both 1 0.58211 1.01518 0.3288 0.5664 1.790 0.245 13.090


Model 4

The PHREG Procedure

Model Information
Data Set T.BMT
Dependent Variable T2
Censoring Variable d2
Censoring Value(s) 0
Ties Handling BRESLOW

Number of Observations Read
Number of Observations Used
137
137

Summary of the Number of Event and Censored
Values
Total Event Censored Percent
Censored
137 42 95 69.34

Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Model Fit Statistics
Criterion Without
Covariates		 With
Covariates
-2 LOG L 380.487 344.608
AIC 380.487 372.608
SBC 380.487 396.935

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 35.8790 14 0.0011
Score 35.5413 14 0.0012
Wald 30.2949 14 0.0069

Analysis of Maximum Likelihood Estimates
Variable DF Parameter
Estimate		 Standard
Error		 Chi-Square Pr > ChiSq Hazard
Ratio		 95% Hazard Ratio Confidence
Limits
g2 1 -2.07054 0.59540 12.0934 0.0005 0.126 0.039 0.405
g3 1 -0.55932 0.54059 1.0705 0.3008 0.572 0.198 1.649
Z1 1 0.03577 0.03192 1.2558 0.2624 1.036 0.974 1.103
Z2 1 -0.03223 0.02761 1.3628 0.2431 0.968 0.917 1.022
Z3 1 -0.22812 0.35220 0.4195 0.5172 0.796 0.399 1.588
Z4 1 0.28374 0.36508 0.6041 0.4370 1.328 0.649 2.716
Z5 1 0.22960 0.37479 0.3753 0.5401 1.258 0.604 2.623
Z6 1 -0.12162 0.37302 0.1063 0.7444 0.885 0.426 1.839
Z7 1 -0.00107 0.0007969 1.7936 0.1805 0.999 0.997 1.000
Z8 1 1.31343 0.45058 8.4971 0.0036 3.719 1.538 8.994
Z10 1 0.22301 0.37310 0.3573 0.5500 1.250 0.602 2.597
acute 1 -0.75555 0.75252 1.0081 0.3154 0.470 0.107 2.053
chronic 1 -0.32199 0.44790 0.5168 0.4722 0.725 0.301 1.743
both 1 0.70084 0.99923 0.4919 0.4831 2.015 0.284 14.286


Model 5 by stepwise selection

The PHREG Procedure

Model Information
Data Set T.BMT
Dependent Variable T2
Censoring Variable d2
Censoring Value(s) 0
Ties Handling BRESLOW

Number of Observations Read
Number of Observations Used
137
137

Summary of the Number of Event and Censored
Values
Total Event Censored Percent
Censored
137 42 95 69.34


The following variable(s) will be included in each model:


acute chronic both g2 g3

Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Model Fit Statistics
Criterion Without
Covariates		 With
Covariates
-2 LOG L 380.487 362.624
AIC 380.487 372.624
SBC 380.487 381.313

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 17.8628 5 0.0031
Score 18.2402 5 0.0027
Wald 16.0647 5 0.0067


Step 1. Variable Z8 is entered. The model contains the following explanatory variables:


acute chronic both g2 g3 Z8

Convergence Status
Convergence criterion (GCONV=1E-8) satisfied.

Model Fit Statistics
Criterion Without
Covariates		 With
Covariates
-2 LOG L 380.487 351.385
AIC 380.487 363.385
SBC 380.487 373.811

Testing Global Null Hypothesis: BETA=0
Test Chi-Square DF Pr > ChiSq
Likelihood Ratio 29.1017 6 <.0001
Score 29.5686 6 <.0001
Wald 26.1462 6 0.0002


Note: No (additional) variables met the 0.05 level for entry into the model.

Analysis of Maximum Likelihood Estimates
Variable DF Parameter
Estimate		 Standard
Error		 Chi-Square Pr > ChiSq Hazard
Ratio		 95% Hazard Ratio Confidence
Limits
acute 1 -0.60540 0.73953 0.6702 0.4130 0.546 0.128 2.326
chronic 1 -0.46433 0.42977 1.1673 0.2800 0.629 0.271 1.459
both 1 0.72959 0.98867 0.5446 0.4605 2.074 0.299 14.402
g2 1 -1.62886 0.52815 9.5116 0.0020 0.196 0.070 0.552
g3 1 -0.27381 0.49201 0.3097 0.5779 0.760 0.290 1.995
Z8 1 1.31004 0.41889 9.7809 0.0018 3.706 1.631 8.424

Summary of Stepwise Selection
Step Variable Number
In 		Score
Chi-Square		 Wald
Chi-Square		 Pr > ChiSq
Entered Removed
1 Z8   6 11.1173 . 0.0009