0% found this document useful (0 votes)
22 views

Probability

1) The document discusses using probability and logistic regression to predict mortality within 14 days. Sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio are calculated for various probability cut-off points. 2) Logistic regression analysis was performed using CRP, ferritin, and IL-6 lab values as predictor variables. The logistic regression model was statistically significant according to the Omnibus tests. However, the Hosmer-Lemeshow test showed poor calibration of predictions across deciles. 3) The logistic regression model correctly classified 90.4% of non-deaths and 27.4% of deaths, for an overall correct classification of 66.2%. All three lab

Uploaded by

firda
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
22 views

Probability

1) The document discusses using probability and logistic regression to predict mortality within 14 days. Sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio are calculated for various probability cut-off points. 2) Logistic regression analysis was performed using CRP, ferritin, and IL-6 lab values as predictor variables. The logistic regression model was statistically significant according to the Omnibus tests. However, the Hosmer-Lemeshow test showed poor calibration of predictions across deciles. 3) The logistic regression model correctly classified 90.4% of non-deaths and 27.4% of deaths, for an overall correct classification of 66.2%. All three lab

Uploaded by

firda
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 11

MENGGUNAKAN PROBABILITY

Nilai Akuras
Nilai Likelihoo Likelihoo
ramal i
Sensitivita Spesifisita ramal d ratio d ratio
negativ
s (Sn) s (Sp) positif positif negatif
Mati Hidup e
(NRP) (LR+) (LR-)
(NRN)
Pro ≥0,3 72 50 59.02 68.85
61.54% 73.40% 75.41% 2.31 0.52
b 6 % %

<0,3 45 138

LANGKAH LANGKAH :

Logistic Regression

Case Processing Summary


Unweighted Casesa N Percent
Selected Cases Included in Analysis 305 100.0
Missing Cases 0 .0
Total 305 100.0
Unselected Cases 0 .0
Total 305 100.0
a. If weight is in effect, see classification table for the total number of cases.

Dependent Variable
Encoding
Original Value Internal Value
Tidak 0
Ya 1
Block 0: Beginning Block

Classification Tablea,b
Predicted
Meninggal 14 hari Percentage
Observed Tidak Ya Correct
Step 0 Meninggal 14 hari Tidak 188 0 100.0
Ya 117 0 .0
Overall Percentage 61.6
a. Constant is included in the model.
b. The cut value is .500

Variables in the Equation


B S.E. Wald df Sig. Exp(B)
Step 0 Constant -.474 .118 16.222 1 .000 .622

Variables not in the Equation


Score df Sig.
Step 0 Variables Lab_CRP 4.348 1 .037
Lab_Feritin 8.676 1 .003
Lab_IL-6 15.481 1 .000
Overall Statistics 23.706 3 .000

Block 1: Method = Enter

Omnibus Tests of Model Coefficients


Chi-square df Sig.
Step 1 Step 32.351 3 .000
Block 32.351 3 .000
Model 32.351 3 .000

Model Summary
Cox & Snell R Nagelkerke R
Step -2 Log likelihood Square Square
1 373.789a .101 .137
a. Estimation terminated at iteration number 5 because parameter
estimates changed by less than .001.
Hosmer and Lemeshow Test
Step Chi-square df Sig.
1 18.341 8 .019

Contingency Table for Hosmer and Lemeshow Test


Meninggal 14 hari = Tidak Meninggal 14 hari = Ya
Observed Expected Observed Expected Total
Step 1 1 22 23.718 9 7.282 31
2 21 22.667 10 8.333 31
3 26 22.053 5 8.947 31
4 26 21.553 5 9.447 31
5 19 20.965 12 10.035 31
6 25 20.270 6 10.730 31
7 16 19.277 15 11.723 31
8 11 17.843 20 13.157 31
9 16 14.237 15 16.763 31
10 6 5.418 20 20.582 26

Classification Tablea
Predicted
Meninggal 14 hari Percentage
Observed Tidak Ya Correct
Step 1 Meninggal 14 hari Tidak 170 18 90.4
Ya 85 32 27.4
Overall Percentage 66.2
a. The cut value is .500

Variables in the Equation


95% C.I.for EXP(B)
B S.E. Wald df Sig. Exp(B) Lower Upper
Step 1a Lab_CRP .002 .003 .392 1 .531 1.002 .997 1.007
Lab_Feritin .000 .000 6.421 1 .011 1.000 1.000 1.000
Lab_IL-6 .002 .001 13.202 1 .000 1.002 1.001 1.003
Constant -1.321 .311 18.038 1 .000 .267
a. Variable(s) entered on step 1: Lab_CRP, Lab_Feritin, Lab_IL-6.
ROC Curve

Case Processing Summary


Meninggal 14 hari Valid N (listwise)
Positivea 117
Negative 188
Larger values of the test result variable(s)
indicate stronger evidence for a positive
actual state.
a. The positive actual state is Ya.

Area Under the Curve


Test Result Variable(s): Predicted probability
Asymptotic 95% Confidence Interval
Area Std. Errora Asymptotic Sig.b Lower Bound Upper Bound
.682 .033 .000 .618 .746
The test result variable(s): Predicted probability has at least one tie between the positive
actual state group and the negative actual state group. Statistics may be biased.
a. Under the nonparametric assumption
b. Null hypothesis: true area = 0.5
Coordinates of the Curve
Test Result Variable(s): Predicted probability
Positive if Greater
Than or Equal Toa Sensitivity 1 - Specificity
.0000000 1.000 1.000
.2206896 1.000 .995
.2251770 1.000 .989
.2263601 .991 .989
.2272666 .991 .984
.2278773 .991 .979
.2285281 .991 .973
.2297211 .991 .968
.2306544 .974 .968
.2312714 .974 .963
.2317878 .974 .957
.2319131 .974 .952
.2322827 .974 .941
.2331676 .974 .936
.2337809 .974 .931
.2341806 .974 .926
.2356509 .974 .920
.2370496 .966 .915
.2379489 .966 .910
.2385842 .957 .910
.2393059 .949 .910
.2408201 .949 .904
.2417761 .949 .899
.2422495 .949 .894
.2440914 .940 .894
.2456738 .940 .888
.2480074 .940 .883
.2519012 .932 .883
.2547489 .923 .883
.2566431 .923 .878
.2575813 .915 .878
.2580019 .915 .872
.2585668 .906 .867
.2592347 .906 .856
.2596400 .906 .851
.2605281 .889 .851
.2612983 .889 .846
.2615016 .889 .840
.2629694 .889 .835
.2650373 .889 .830
.2684028 .880 .830
.2722785 .880 .824
.2744140 .880 .814
.2752957 .880 .809
.2758446 .880 .803
.2768655 .872 .803
.2773843 .863 .803
.2774066 .863 .798
.2781922 .863 .793
.2789760 .855 .793
.2790122 .855 .787
.2794120 .846 .787
.2798065 .846 .782
.2798417 .846 .777
.2801746 .838 .777
.2805229 .838 .771
.2808703 .838 .761
.2813589 .838 .755
.2824897 .838 .750
.2835189 .838 .745
.2839041 .829 .745
.2842229 .821 .745
.2843188 .821 .739
.2848777 .821 .734
.2856961 .821 .729
.2860761 .821 .723
.2861453 .821 .713
.2865731 .812 .713
.2874709 .812 .707
.2884632 .803 .702
.2890060 .803 .697
.2899794 .803 .691
.2916256 .803 .686
.2926559 .803 .681
.2929520 .803 .676
.2930599 .803 .670
.2939344 .803 .665
.2947806 .803 .660
.2948977 .803 .654
.2952632 .803 .649
.2959312 .803 .644
.2966574 .795 .644
.2971094 .795 .638
.2972530 .795 .633
.2973729 .795 .628
.2979290 .795 .622
.2986674 .795 .617
.2989919 .795 .612
.2991842 .795 .606
.3001458 .786 .606
.3013265 .786 .601
.3018131 .786 .590
.3022574 .786 .585
.3029610 .786 .574
.3034143 .786 .569
.3034609 .786 .564
.3038793 .786 .559
.3044562 .786 .553
.3046614 .786 .548
.3048867 .786 .543
.3054435 .786 .537
.3060248 .786 .532
.3068394 .786 .527
.3085802 .786 .521
.3098195 .778 .521
.3099951 .778 .511
.3104930 .769 .511
.3111403 .761 .511
.3114011 .761 .505
.3118565 .761 .495
.3124209 .752 .495
.3126297 .752 .489
.3128381 .752 .484
.3135316 .752 .479
.3148509 .744 .479
.3157210 .744 .473
.3159259 .744 .463
.3171289 .744 .457
.3191647 .744 .452
.3206977 .744 .447
.3214865 .735 .441
.3231049 .735 .436
.3245608 .726 .436
.3248511 .726 .431
.3254047 .718 .431
.3258370 .709 .431
.3266473 .709 .420
.3282038 .701 .420
.3291163 .692 .420
.3294957 .684 .420
.3299259 .675 .420
.3301506 .667 .415
.3304424 .667 .410
.3308155 .667 .404
.3315433 .667 .399
.3322595 .658 .399
.3327398 .650 .399
.3336174 .650 .394
.3344780 .641 .388
.3350947 .641 .383
.3358995 .632 .383
.3363408 .632 .378
.3363994 .632 .372
.3364541 .632 .367
.3364935 .632 .362
.3375735 .632 .356
.3396301 .632 .351
.3409831 .632 .346
.3413750 .624 .340
.3420890 .624 .335
.3430062 .624 .330
.3436581 .624 .324
.3469677 .624 .319
.3498828 .624 .309
.3500696 .624 .303
.3514619 .615 .303
.3528447 .615 .298
.3541938 .615 .293
.3555744 .615 .287
.3573311 .615 .282
.3589887 .615 .277
.3595031 .615 .271
.3600700 .615 .266
.3602833 .607 .266
.3605169 .598 .266
.3610186 .598 .261
.3615578 .590 .261
.3620399 .590 .255
.3625072 .581 .255
.3635858 .573 .255
.3644768 .573 .250
.3646597 .564 .250
.3651796 .564 .245
.3670774 .556 .245
.3693912 .556 .239
.3707323 .547 .234
.3713872 .547 .229
.3736306 .547 .223
.3774295 .547 .218
.3796810 .547 .213
.3807309 .538 .213
.3814361 .530 .213
.3831229 .530 .207
.3845327 .530 .202
.3850282 .530 .197
.3864068 .530 .191
.3876421 .521 .191
.3881344 .513 .191
.3884317 .513 .186
.3892287 .504 .186
.3900694 .496 .186
.3903906 .487 .186
.3929715 .479 .186
.3960061 .479 .181
.3967045 .470 .181
.3972332 .470 .176
.3981658 .462 .176
.3988588 .462 .165
.3990324 .453 .160
.3992090 .444 .160
.4003996 .436 .160
.4023669 .427 .160
.4038520 .427 .154
.4045738 .419 .154
.4087758 .410 .154
.4150214 .402 .154
.4178676 .393 .154
.4199766 .393 .149
.4228590 .376 .149
.4259467 .376 .144
.4284252 .368 .144
.4302917 .359 .144
.4317365 .350 .144
.4324679 .350 .138
.4343111 .342 .138
.4406873 .342 .128
.4472598 .333 .128
.4506028 .333 .122
.4526878 .325 .122
.4547911 .316 .122
.4624892 .308 .122
.4776447 .308 .117
.4882312 .299 .117
.4911487 .291 .117
.4940474 .291 .106
.4965582 .282 .106
.4978584 .274 .106
.4983644 .274 .101
.5008693 .274 .096
.5032785 .265 .096
.5090858 .256 .096
.5167090 .256 .090
.5187448 .256 .085
.5194665 .256 .080
.5205760 .256 .074
.5252620 .256 .069
.5319673 .248 .069
.5372929 .248 .064
.5427622 .239 .064
.5473543 .239 .059
.5498792 .231 .059
.5526721 .222 .059
.5571300 .222 .053
.5622711 .214 .053
.5654506 .214 .048
.5712951 .214 .043
.5810666 .205 .043
.5880458 .197 .043
.5909943 .188 .043
.6012047 .188 .037
.6143755 .179 .037
.6240437 .179 .032
.6309787 .171 .032
.6314334 .162 .032
.6318778 .154 .032
.6375117 .154 .027
.6430191 .154 .021
.6583920 .145 .021
.6736049 .137 .021
.6827480 .128 .021
.6957106 .120 .021
.7046913 .111 .021
.7265437 .103 .021
.7468962 .094 .016
.7680099 .085 .016
.7937953 .077 .016
.8358460 .077 .005
.8730382 .068 .005
.8849341 .060 .005
.8959363 .060 .000
.9079596 .051 .000
.9281086 .043 .000
.9527281 .034 .000
.9667957 .026 .000
.9822829 .017 .000
.9988174 .009 .000
1.0000000 .000 .000
The test result variable(s): Predicted probability has at
least one tie between the positive actual state group
and the negative actual state group.
a. The smallest cutoff value is the minimum observed
test value minus 1, and the largest cutoff value is the
maximum observed test value plus 1. All the other
cutoff values are the averages of two consecutive
ordered observed test values.

Crosstabs

Case Processing Summary


Cases
Valid Missing Total
N Percent N Percent N Percent
Kat_Prob * Luaran 305 100.0% 0 0.0% 305 100.0%

Kat_Prob * Luaran Crosstabulation


Luaran
Mati Hidup Total
Kat_Prob >0,36 Count 72 50 122
% within Kat_Prob 59.0% 41.0% 100.0%
<0,36 Count 45 138 183
% within Kat_Prob 24.6% 75.4% 100.0%
Total Count 117 188 305
% within Kat_Prob 38.4% 61.6% 100.0%

Chi-Square Tests
Asymptotic
Significance (2- Exact Sig. (2- Exact Sig. (1-
Value df sided) sided) sided)
Pearson Chi-Square 36.690a 1 .000
b
Continuity Correction 35.248 1 .000
Likelihood Ratio 36.850 1 .000
Fisher's Exact Test .000 .000
Linear-by-Linear Association 36.570 1 .000
N of Valid Cases 305
a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 46.80.
b. Computed only for a 2x2 table

Risk Estimate
95% Confidence Interval
Value Lower Upper
Odds Ratio for Kat_Prob 4.416 2.696 7.232
(>0,36 / <0,36)
For cohort Luaran = 1.00 2.400 1.789 3.219
For cohort Luaran = 2.00 .543 .432 .683
N of Valid Cases 305

You might also like