Model1 : age sex_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.279
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.592
## Area under the curve: 0.6686
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 41673 5846
## died 18146 6652
##
## Accuracy : 0.6682
## 95% CI : (0.6648, 0.6717)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1648
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.53225
## Specificity : 0.69665
## Pos Pred Value : 0.26825
## Neg Pred Value : 0.87698
## Prevalence : 0.17282
## Detection Rate : 0.09198
## Detection Prevalence : 0.34291
## Balanced Accuracy : 0.61445
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.721799 0.008075 -213.216 < 2e-16 ***
## age 0.657208 0.005948 110.491 < 2e-16 ***
## sex_catMale 0.044462 0.010191 4.363 1.28e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 252169 on 289265 degrees of freedom
## AIC: 252175
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.93 |
1.91, 1.95 |
<0.001 |
sex_catMale |
1.05 |
1.02, 1.07 |
<0.001 |
Model2 : age sex_cat ALBUMIN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.460
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.645
## Area under the curve: 0.7781
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46400 4558
## died 13419 7940
##
## Accuracy : 0.7514
## 95% CI : (0.7482, 0.7546)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.321
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6353
## Specificity : 0.7757
## Pos Pred Value : 0.3717
## Neg Pred Value : 0.9106
## Prevalence : 0.1728
## Detection Rate : 0.1098
## Detection Prevalence : 0.2954
## Balanced Accuracy : 0.7055
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.021156 0.009084 -222.50 <2e-16 ***
## age 0.646368 0.006270 103.09 <2e-16 ***
## sex_catMale 0.165173 0.010947 15.09 <2e-16 ***
## ALBUMIN -0.885629 0.005455 -162.34 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 221723 on 289264 degrees of freedom
## AIC: 221731
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.91 |
1.89, 1.93 |
<0.001 |
sex_catMale |
1.18 |
1.15, 1.21 |
<0.001 |
ALBUMIN |
0.41 |
0.41, 0.42 |
<0.001 |
Model_3 : age sex_cat ALBUMIN COMO_CHF
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.466
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.648
## Area under the curve: 0.783
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46349 4487
## died 13470 8011
##
## Accuracy : 0.7517
## 95% CI : (0.7485, 0.7548)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3238
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6410
## Specificity : 0.7748
## Pos Pred Value : 0.3729
## Neg Pred Value : 0.9117
## Prevalence : 0.1728
## Detection Rate : 0.1108
## Detection Prevalence : 0.2970
## Balanced Accuracy : 0.7079
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.190874 0.010184 -215.13 <2e-16 ***
## age 0.624261 0.006318 98.80 <2e-16 ***
## sex_catMale 0.168705 0.010987 15.36 <2e-16 ***
## ALBUMIN -0.882142 0.005479 -161.00 <2e-16 ***
## COMO_CHFY 0.454742 0.011042 41.19 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 220046 on 289263 degrees of freedom
## AIC: 220056
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.87 |
1.84, 1.89 |
<0.001 |
sex_catMale |
1.18 |
1.16, 1.21 |
<0.001 |
ALBUMIN |
0.41 |
0.41, 0.42 |
<0.001 |
COMO_CHFY |
1.58 |
1.54, 1.61 |
<0.001 |
model_4 : age sex_cat ALBUMIN COMO_CHF COMO_DM_INS
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.466
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.648
## Area under the curve: 0.7832
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46404 4468
## died 13415 8030
##
## Accuracy : 0.7527
## 95% CI : (0.7496, 0.7559)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3259
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6425
## Specificity : 0.7757
## Pos Pred Value : 0.3744
## Neg Pred Value : 0.9122
## Prevalence : 0.1728
## Detection Rate : 0.1110
## Detection Prevalence : 0.2965
## Balanced Accuracy : 0.7091
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.161257 0.011356 -190.32 < 2e-16 ***
## age 0.617511 0.006414 96.28 < 2e-16 ***
## sex_catMale 0.166359 0.010996 15.13 < 2e-16 ***
## ALBUMIN -0.883798 0.005487 -161.07 < 2e-16 ***
## COMO_CHFY 0.461541 0.011107 41.55 < 2e-16 ***
## COMO_DM_INSY -0.064885 0.011169 -5.81 6.26e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 220012 on 289262 degrees of freedom
## AIC: 220024
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.85 |
1.83, 1.88 |
<0.001 |
sex_catMale |
1.18 |
1.16, 1.21 |
<0.001 |
ALBUMIN |
0.41 |
0.41, 0.42 |
<0.001 |
COMO_CHFY |
1.59 |
1.55, 1.62 |
<0.001 |
COMO_DM_INSY |
0.94 |
0.92, 0.96 |
<0.001 |
model_5 : age sex_cat ALBUMIN COMO_CHF COMO_DM_INS COMO_HTN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.467
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.648
## Area under the curve: 0.7839
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46440 4480
## died 13379 8018
##
## Accuracy : 0.753
## 95% CI : (0.7499, 0.7562)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3261
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6415
## Specificity : 0.7763
## Pos Pred Value : 0.3747
## Neg Pred Value : 0.9120
## Prevalence : 0.1728
## Detection Rate : 0.1109
## Detection Prevalence : 0.2959
## Balanced Accuracy : 0.7089
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.907899 0.017931 -106.404 < 2e-16 ***
## age 0.619979 0.006422 96.546 < 2e-16 ***
## sex_catMale 0.161899 0.011005 14.711 < 2e-16 ***
## ALBUMIN -0.879503 0.005492 -160.156 < 2e-16 ***
## COMO_CHFY 0.463878 0.011117 41.727 < 2e-16 ***
## COMO_DM_INSY -0.059215 0.011179 -5.297 1.18e-07 ***
## COMO_HTNY -0.289558 0.016070 -18.018 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 219697 on 289261 degrees of freedom
## AIC: 219711
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.86 |
1.84, 1.88 |
<0.001 |
sex_catMale |
1.18 |
1.15, 1.20 |
<0.001 |
ALBUMIN |
0.41 |
0.41, 0.42 |
<0.001 |
COMO_CHFY |
1.59 |
1.56, 1.63 |
<0.001 |
COMO_DM_INSY |
0.94 |
0.92, 0.96 |
<0.001 |
COMO_HTNY |
0.75 |
0.73, 0.77 |
<0.001 |
model_6 : age sex_cat ALBUMIN COMO_CHF COMO_DM_INS COMO_HTN
CLM_FROM_1year_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.484
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.658
## Area under the curve: 0.8056
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46263 4004
## died 13556 8494
##
## Accuracy : 0.7572
## 95% CI : (0.754, 0.7603)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3479
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6796
## Specificity : 0.7734
## Pos Pred Value : 0.3852
## Neg Pred Value : 0.9203
## Prevalence : 0.1728
## Detection Rate : 0.1175
## Detection Prevalence : 0.3049
## Balanced Accuracy : 0.7265
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.994173 0.018289 -109.04 < 2e-16 ***
## age 0.595808 0.006495 91.73 < 2e-16 ***
## sex_catMale 0.188199 0.011163 16.86 < 2e-16 ***
## ALBUMIN -0.814854 0.005574 -146.19 < 2e-16 ***
## COMO_CHFY 0.417570 0.011282 37.01 < 2e-16 ***
## COMO_DM_INSY -0.071434 0.011339 -6.30 2.98e-10 ***
## COMO_HTNY -0.274014 0.016297 -16.81 < 2e-16 ***
## CLM_FROM_1year_cat 0.462243 0.005877 78.66 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 213184 on 289260 degrees of freedom
## AIC: 213200
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.81 |
1.79, 1.84 |
<0.001 |
sex_catMale |
1.21 |
1.18, 1.23 |
<0.001 |
ALBUMIN |
0.44 |
0.44, 0.45 |
<0.001 |
COMO_CHFY |
1.52 |
1.49, 1.55 |
<0.001 |
COMO_DM_INSY |
0.93 |
0.91, 0.95 |
<0.001 |
COMO_HTNY |
0.76 |
0.74, 0.79 |
<0.001 |
CLM_FROM_1year_cat |
1.59 |
1.57, 1.61 |
<0.001 |
model_7 : age ALBUMIN COMO_CHF
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.464
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.647
## Area under the curve: 0.7819
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46322 4511
## died 13497 7987
##
## Accuracy : 0.751
## 95% CI : (0.7478, 0.7541)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3219
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6391
## Specificity : 0.7744
## Pos Pred Value : 0.3718
## Neg Pred Value : 0.9113
## Prevalence : 0.1728
## Detection Rate : 0.1104
## Detection Prevalence : 0.2971
## Balanced Accuracy : 0.7067
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.093260 0.007819 -267.71 <2e-16 ***
## age 0.622924 0.006315 98.64 <2e-16 ***
## ALBUMIN -0.877384 0.005467 -160.50 <2e-16 ***
## COMO_CHFY 0.453278 0.011033 41.08 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 220283 on 289264 degrees of freedom
## AIC: 220291
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.86 |
1.84, 1.89 |
<0.001 |
ALBUMIN |
0.42 |
0.41, 0.42 |
<0.001 |
COMO_CHFY |
1.57 |
1.54, 1.61 |
<0.001 |
model_8 : age ALBUMIN COMO_CHF IRON_SAT_PERCENT
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.486
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.654
## Area under the curve: 0.7944
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46489 4206
## died 13330 8292
##
## Accuracy : 0.7575
## 95% CI : (0.7544, 0.7606)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3419
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6635
## Specificity : 0.7772
## Pos Pred Value : 0.3835
## Neg Pred Value : 0.9170
## Prevalence : 0.1728
## Detection Rate : 0.1147
## Detection Prevalence : 0.2990
## Balanced Accuracy : 0.7203
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.121043 0.007962 -266.38 <2e-16 ***
## age 0.644748 0.006398 100.77 <2e-16 ***
## ALBUMIN -0.859458 0.005519 -155.74 <2e-16 ***
## COMO_CHFY 0.404469 0.011155 36.26 <2e-16 ***
## IRON_SAT_PERCENT -0.345400 0.005589 -61.80 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 216292 on 289263 degrees of freedom
## AIC: 216302
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.91 |
1.88, 1.93 |
<0.001 |
ALBUMIN |
0.42 |
0.42, 0.43 |
<0.001 |
COMO_CHFY |
1.50 |
1.47, 1.53 |
<0.001 |
IRON_SAT_PERCENT |
0.71 |
0.70, 0.72 |
<0.001 |
model_9 : age ALBUMIN COMO_CHF IRON_SAT_PERCENT HGB
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.506
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.660
## Area under the curve: 0.8008
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46939 4166
## died 12880 8332
##
## Accuracy : 0.7643
## 95% CI : (0.7612, 0.7674)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3538
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6667
## Specificity : 0.7847
## Pos Pred Value : 0.3928
## Neg Pred Value : 0.9185
## Prevalence : 0.1728
## Detection Rate : 0.1152
## Detection Prevalence : 0.2933
## Balanced Accuracy : 0.7257
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.146852 0.008062 -266.30 <2e-16 ***
## age 0.660195 0.006459 102.21 <2e-16 ***
## ALBUMIN -0.759916 0.005785 -131.37 <2e-16 ***
## COMO_CHFY 0.423171 0.011254 37.60 <2e-16 ***
## IRON_SAT_PERCENT -0.292260 0.005645 -51.77 <2e-16 ***
## HGB -0.311487 0.005761 -54.06 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 213317 on 289262 degrees of freedom
## AIC: 213329
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.94 |
1.91, 1.96 |
<0.001 |
ALBUMIN |
0.47 |
0.46, 0.47 |
<0.001 |
COMO_CHFY |
1.53 |
1.49, 1.56 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.74, 0.75 |
<0.001 |
HGB |
0.73 |
0.72, 0.74 |
<0.001 |
model_10 : age ALBUMIN COMO_CHF IRON_SAT_PERCENT
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.499
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.662
## Area under the curve: 0.8132
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46448 3838
## died 13371 8660
##
## Accuracy : 0.762
## 95% CI : (0.7589, 0.7651)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3606
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6929
## Specificity : 0.7765
## Pos Pred Value : 0.3931
## Neg Pred Value : 0.9237
## Prevalence : 0.1728
## Detection Rate : 0.1198
## Detection Prevalence : 0.3046
## Balanced Accuracy : 0.7347
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.180200 0.008258 -264.02 <2e-16 ***
## age 0.620832 0.006460 96.10 <2e-16 ***
## ALBUMIN -0.799336 0.005594 -142.89 <2e-16 ***
## COMO_CHFY 0.362263 0.011306 32.04 <2e-16 ***
## IRON_SAT_PERCENT -0.320349 0.005643 -56.77 <2e-16 ***
## CLM_FROM_1year_cat 0.440702 0.005911 74.56 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 210466 on 289262 degrees of freedom
## AIC: 210478
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.86 |
1.84, 1.88 |
<0.001 |
ALBUMIN |
0.45 |
0.44, 0.45 |
<0.001 |
COMO_CHFY |
1.44 |
1.41, 1.47 |
<0.001 |
IRON_SAT_PERCENT |
0.73 |
0.72, 0.73 |
<0.001 |
CLM_FROM_1year_cat |
1.55 |
1.54, 1.57 |
<0.001 |
model_11 : age,ALBUMIN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.458
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.645
## Area under the curve: 0.777
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46351 4590
## died 13468 7908
##
## Accuracy : 0.7503
## 95% CI : (0.7471, 0.7534)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3182
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6327
## Specificity : 0.7749
## Pos Pred Value : 0.3699
## Neg Pred Value : 0.9099
## Prevalence : 0.1728
## Detection Rate : 0.1094
## Detection Prevalence : 0.2956
## Balanced Accuracy : 0.7038
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.926110 0.006396 -301.1 <2e-16 ***
## age 0.644985 0.006267 102.9 <2e-16 ***
## ALBUMIN -0.880964 0.005443 -161.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 221952 on 289265 degrees of freedom
## AIC: 221958
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.91 |
1.88, 1.93 |
<0.001 |
ALBUMIN |
0.41 |
0.41, 0.42 |
<0.001 |
model_12 : age ALBUMIN IRON_SAT_PERCENT
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.482
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.652
## Area under the curve: 0.791
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46544 4307
## died 13275 8191
##
## Accuracy : 0.7569
## 95% CI : (0.7537, 0.76)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3376
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6554
## Specificity : 0.7781
## Pos Pred Value : 0.3816
## Neg Pred Value : 0.9153
## Prevalence : 0.1728
## Detection Rate : 0.1133
## Detection Prevalence : 0.2968
## Balanced Accuracy : 0.7167
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.973510 0.006598 -299.12 <2e-16 ***
## age 0.665406 0.006354 104.73 <2e-16 ***
## ALBUMIN -0.861807 0.005499 -156.72 <2e-16 ***
## IRON_SAT_PERCENT -0.359041 0.005568 -64.48 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 217592 on 289264 degrees of freedom
## AIC: 217600
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.95 |
1.92, 1.97 |
<0.001 |
ALBUMIN |
0.42 |
0.42, 0.43 |
<0.001 |
IRON_SAT_PERCENT |
0.70 |
0.69, 0.71 |
<0.001 |
model_13 : age ALBUMIN IRON_SAT_PERCENT HGB
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.502
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.658
## Area under the curve: 0.7972
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46922 4218
## died 12897 8280
##
## Accuracy : 0.7633
## 95% CI : (0.7602, 0.7664)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3506
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6625
## Specificity : 0.7844
## Pos Pred Value : 0.3910
## Neg Pred Value : 0.9175
## Prevalence : 0.1728
## Detection Rate : 0.1145
## Detection Prevalence : 0.2928
## Balanced Accuracy : 0.7235
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.991916 0.006667 -298.76 <2e-16 ***
## age 0.681436 0.006414 106.25 <2e-16 ***
## ALBUMIN -0.764150 0.005763 -132.59 <2e-16 ***
## IRON_SAT_PERCENT -0.307330 0.005622 -54.66 <2e-16 ***
## HGB -0.305663 0.005747 -53.19 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 214716 on 289263 degrees of freedom
## AIC: 214726
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.98 |
1.95, 2.00 |
<0.001 |
ALBUMIN |
0.47 |
0.46, 0.47 |
<0.001 |
IRON_SAT_PERCENT |
0.74 |
0.73, 0.74 |
<0.001 |
HGB |
0.74 |
0.73, 0.74 |
<0.001 |
model_14 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.513
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.665
## Area under the curve: 0.8156
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46711 3892
## died 13108 8606
##
## Accuracy : 0.7649
## 95% CI : (0.7618, 0.768)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3635
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6886
## Specificity : 0.7809
## Pos Pred Value : 0.3963
## Neg Pred Value : 0.9231
## Prevalence : 0.1728
## Detection Rate : 0.1190
## Detection Prevalence : 0.3003
## Balanced Accuracy : 0.7347
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.064292 0.007017 -294.18 <2e-16 ***
## age 0.655180 0.006473 101.21 <2e-16 ***
## ALBUMIN -0.714410 0.005832 -122.51 <2e-16 ***
## IRON_SAT_PERCENT -0.287782 0.005672 -50.73 <2e-16 ***
## HGB -0.274668 0.005782 -47.50 <2e-16 ***
## CLM_FROM_1year_cat 0.431525 0.005938 72.67 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 209195 on 289262 degrees of freedom
## AIC: 209207
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.93 |
1.90, 1.95 |
<0.001 |
ALBUMIN |
0.49 |
0.48, 0.50 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.74, 0.76 |
<0.001 |
HGB |
0.76 |
0.75, 0.77 |
<0.001 |
CLM_FROM_1year_cat |
1.54 |
1.52, 1.56 |
<0.001 |
model_15 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
race_cat_4
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.517
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.667
## Area under the curve: 0.818
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46757 3843
## died 13062 8655
##
## Accuracy : 0.7662
## 95% CI : (0.7631, 0.7693)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3671
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6925
## Specificity : 0.7816
## Pos Pred Value : 0.3985
## Neg Pred Value : 0.9241
## Prevalence : 0.1728
## Detection Rate : 0.1197
## Detection Prevalence : 0.3003
## Balanced Accuracy : 0.7371
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.508537 0.035075 -71.518 < 2e-16 ***
## age 0.634762 0.006549 96.931 < 2e-16 ***
## ALBUMIN -0.711977 0.005846 -121.792 < 2e-16 ***
## IRON_SAT_PERCENT -0.282247 0.005681 -49.686 < 2e-16 ***
## HGB -0.286745 0.005817 -49.292 < 2e-16 ***
## CLM_FROM_1year_cat 0.424845 0.005953 71.368 < 2e-16 ***
## race_cat_4Black 0.206247 0.036868 5.594 2.22e-08 ***
## race_cat_4Other -0.037393 0.057833 -0.647 0.518
## race_cat_4White 0.558290 0.035312 15.810 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208205 on 289259 degrees of freedom
## AIC: 208223
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.89 |
1.86, 1.91 |
<0.001 |
ALBUMIN |
0.49 |
0.49, 0.50 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.75, 0.76 |
<0.001 |
HGB |
0.75 |
0.74, 0.76 |
<0.001 |
CLM_FROM_1year_cat |
1.53 |
1.51, 1.55 |
<0.001 |
race_cat_4Black |
1.23 |
1.14, 1.32 |
<0.001 |
race_cat_4Other |
0.96 |
0.86, 1.08 |
0.5 |
race_cat_4White |
1.75 |
1.63, 1.87 |
<0.001 |
model_16 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
race_cat_4 state_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.517
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.667
## Area under the curve: 0.818
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46754 3835
## died 13065 8663
##
## Accuracy : 0.7663
## 95% CI : (0.7632, 0.7694)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3674
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6932
## Specificity : 0.7816
## Pos Pred Value : 0.3987
## Neg Pred Value : 0.9242
## Prevalence : 0.1728
## Detection Rate : 0.1198
## Detection Prevalence : 0.3005
## Balanced Accuracy : 0.7374
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.420995 0.037655 -64.295 < 2e-16 ***
## age 0.637712 0.006582 96.888 < 2e-16 ***
## ALBUMIN -0.712204 0.005848 -121.791 < 2e-16 ***
## IRON_SAT_PERCENT -0.282566 0.005683 -49.725 < 2e-16 ***
## HGB -0.286541 0.005829 -49.158 < 2e-16 ***
## CLM_FROM_1year_cat 0.423348 0.005962 71.005 < 2e-16 ***
## race_cat_4Black 0.162747 0.037702 4.317 1.58e-05 ***
## race_cat_4Other -0.050267 0.057882 -0.868 0.3852
## race_cat_4White 0.525975 0.035798 14.693 < 2e-16 ***
## state_catFIPSNE -0.131690 0.017992 -7.320 2.49e-13 ***
## state_catFIPSS -0.029338 0.014646 -2.003 0.0452 *
## state_catFIPSW -0.108808 0.018206 -5.977 2.28e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208128 on 289256 degrees of freedom
## AIC: 208152
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.89 |
1.87, 1.92 |
<0.001 |
ALBUMIN |
0.49 |
0.48, 0.50 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.75, 0.76 |
<0.001 |
HGB |
0.75 |
0.74, 0.76 |
<0.001 |
CLM_FROM_1year_cat |
1.53 |
1.51, 1.55 |
<0.001 |
race_cat_4Black |
1.18 |
1.09, 1.27 |
<0.001 |
race_cat_4Other |
0.95 |
0.85, 1.06 |
0.4 |
race_cat_4White |
1.69 |
1.58, 1.82 |
<0.001 |
state_catFIPSNE |
0.88 |
0.85, 0.91 |
<0.001 |
state_catFIPSS |
0.97 |
0.94, 1.00 |
0.045 |
state_catFIPSW |
0.90 |
0.87, 0.93 |
<0.001 |
model_17 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
CALCIUM_UNCORRECTED
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.514
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.666
## Area under the curve: 0.8166
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46789 3865
## died 13030 8633
##
## Accuracy : 0.7664
## 95% CI : (0.7633, 0.7695)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3666
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6908
## Specificity : 0.7822
## Pos Pred Value : 0.3985
## Neg Pred Value : 0.9237
## Prevalence : 0.1728
## Detection Rate : 0.1194
## Detection Prevalence : 0.2996
## Balanced Accuracy : 0.7365
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.068786 0.007037 -293.98 <2e-16 ***
## age 0.641270 0.006510 98.50 <2e-16 ***
## ALBUMIN -0.777355 0.006559 -118.51 <2e-16 ***
## IRON_SAT_PERCENT -0.286026 0.005680 -50.35 <2e-16 ***
## HGB -0.278934 0.005788 -48.19 <2e-16 ***
## CLM_FROM_1year_cat 0.429547 0.005944 72.27 <2e-16 ***
## CALCIUM_UNCORRECTED 0.135079 0.006227 21.69 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208722 on 289261 degrees of freedom
## AIC: 208736
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.90 |
1.87, 1.92 |
<0.001 |
ALBUMIN |
0.46 |
0.45, 0.47 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.74, 0.76 |
<0.001 |
HGB |
0.76 |
0.75, 0.77 |
<0.001 |
CLM_FROM_1year_cat |
1.54 |
1.52, 1.55 |
<0.001 |
CALCIUM_UNCORRECTED |
1.14 |
1.13, 1.16 |
<0.001 |
model_18 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
CALCIUM_UNCORRECTED FERRITIN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.514
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.666
## Area under the curve: 0.8167
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46815 3861
## died 13004 8637
##
## Accuracy : 0.7668
## 95% CI : (0.7637, 0.7699)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3674
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6911
## Specificity : 0.7826
## Pos Pred Value : 0.3991
## Neg Pred Value : 0.9238
## Prevalence : 0.1728
## Detection Rate : 0.1194
## Detection Prevalence : 0.2993
## Balanced Accuracy : 0.7368
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.069259 0.007039 -293.96 <2e-16 ***
## age 0.635825 0.006525 97.45 <2e-16 ***
## ALBUMIN -0.768787 0.006587 -116.72 <2e-16 ***
## IRON_SAT_PERCENT -0.306653 0.005885 -52.10 <2e-16 ***
## HGB -0.264919 0.005873 -45.11 <2e-16 ***
## CLM_FROM_1year_cat 0.427847 0.005948 71.93 <2e-16 ***
## CALCIUM_UNCORRECTED 0.126133 0.006264 20.14 <2e-16 ***
## FERRITIN 0.073668 0.005282 13.95 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208529 on 289260 degrees of freedom
## AIC: 208545
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.89 |
1.86, 1.91 |
<0.001 |
ALBUMIN |
0.46 |
0.46, 0.47 |
<0.001 |
IRON_SAT_PERCENT |
0.74 |
0.73, 0.74 |
<0.001 |
HGB |
0.77 |
0.76, 0.78 |
<0.001 |
CLM_FROM_1year_cat |
1.53 |
1.52, 1.55 |
<0.001 |
CALCIUM_UNCORRECTED |
1.13 |
1.12, 1.15 |
<0.001 |
FERRITIN |
1.08 |
1.07, 1.09 |
<0.001 |
model_19 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
CALCIUM_UNCORRECTED FERRITIN PHOSPHORUS
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.514
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.666
## Area under the curve: 0.8168
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46829 3863
## died 12990 8635
##
## Accuracy : 0.767
## 95% CI : (0.7639, 0.77)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3676
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6909
## Specificity : 0.7828
## Pos Pred Value : 0.3993
## Neg Pred Value : 0.9238
## Prevalence : 0.1728
## Detection Rate : 0.1194
## Detection Prevalence : 0.2990
## Balanced Accuracy : 0.7369
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.068959 0.007039 -293.943 < 2e-16 ***
## age 0.629128 0.006731 93.469 < 2e-16 ***
## ALBUMIN -0.763495 0.006715 -113.706 < 2e-16 ***
## IRON_SAT_PERCENT -0.306566 0.005886 -52.088 < 2e-16 ***
## HGB -0.264875 0.005872 -45.111 < 2e-16 ***
## CLM_FROM_1year_cat 0.427795 0.005949 71.916 < 2e-16 ***
## CALCIUM_UNCORRECTED 0.123506 0.006303 19.596 < 2e-16 ***
## FERRITIN 0.072153 0.005295 13.627 < 2e-16 ***
## PHOSPHORUS -0.024473 0.006101 -4.012 6.03e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208513 on 289259 degrees of freedom
## AIC: 208531
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.88 |
1.85, 1.90 |
<0.001 |
ALBUMIN |
0.47 |
0.46, 0.47 |
<0.001 |
IRON_SAT_PERCENT |
0.74 |
0.73, 0.74 |
<0.001 |
HGB |
0.77 |
0.76, 0.78 |
<0.001 |
CLM_FROM_1year_cat |
1.53 |
1.52, 1.55 |
<0.001 |
CALCIUM_UNCORRECTED |
1.13 |
1.12, 1.15 |
<0.001 |
FERRITIN |
1.07 |
1.06, 1.09 |
<0.001 |
PHOSPHORUS |
0.98 |
0.96, 0.99 |
<0.001 |
model_20 : age ALBUMIN IRON_SAT_PERCENT HGB sex_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.504
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.658
## Area under the curve: 0.798
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 47012 4232
## died 12807 8266
##
## Accuracy : 0.7644
## 95% CI : (0.7613, 0.7675)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3518
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6614
## Specificity : 0.7859
## Pos Pred Value : 0.3923
## Neg Pred Value : 0.9174
## Prevalence : 0.1728
## Detection Rate : 0.1143
## Detection Prevalence : 0.2914
## Balanced Accuracy : 0.7236
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.091990 0.009321 -224.44 <2e-16 ***
## age 0.682846 0.006418 106.39 <2e-16 ***
## ALBUMIN -0.768015 0.005773 -133.05 <2e-16 ***
## IRON_SAT_PERCENT -0.306214 0.005624 -54.45 <2e-16 ***
## HGB -0.307903 0.005749 -53.55 <2e-16 ***
## sex_catMale 0.175436 0.011131 15.76 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 214466 on 289262 degrees of freedom
## AIC: 214478
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.98 |
1.95, 2.00 |
<0.001 |
ALBUMIN |
0.46 |
0.46, 0.47 |
<0.001 |
IRON_SAT_PERCENT |
0.74 |
0.73, 0.74 |
<0.001 |
HGB |
0.73 |
0.73, 0.74 |
<0.001 |
sex_catMale |
1.19 |
1.17, 1.22 |
<0.001 |
model_21 : age ALBUMIN IRON_SAT_PERCENT HGB sex_cat
CLM_FROM_1year_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.515
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.666
## Area under the curve: 0.8165
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46834 3914
## died 12985 8584
##
## Accuracy : 0.7663
## 95% CI : (0.7632, 0.7694)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.365
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6868
## Specificity : 0.7829
## Pos Pred Value : 0.3980
## Neg Pred Value : 0.9229
## Prevalence : 0.1728
## Detection Rate : 0.1187
## Detection Prevalence : 0.2983
## Balanced Accuracy : 0.7349
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.178120 0.009667 -225.31 <2e-16 ***
## age 0.656705 0.006479 101.36 <2e-16 ***
## ALBUMIN -0.718428 0.005841 -123.00 <2e-16 ***
## IRON_SAT_PERCENT -0.286247 0.005675 -50.44 <2e-16 ***
## HGB -0.276991 0.005784 -47.89 <2e-16 ***
## sex_catMale 0.198635 0.011268 17.63 <2e-16 ***
## CLM_FROM_1year_cat 0.434514 0.005947 73.07 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208882 on 289261 degrees of freedom
## AIC: 208896
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.93 |
1.90, 1.95 |
<0.001 |
ALBUMIN |
0.49 |
0.48, 0.49 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.74, 0.76 |
<0.001 |
HGB |
0.76 |
0.75, 0.77 |
<0.001 |
sex_catMale |
1.22 |
1.19, 1.25 |
<0.001 |
CLM_FROM_1year_cat |
1.54 |
1.53, 1.56 |
<0.001 |
model_22 : age ALBUMIN IRON_SAT_PERCENT HGB sex_cat
CLM_FROM_1year_cat race_cat_4
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.519
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.668
## Area under the curve: 0.8187
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46818 3826
## died 13001 8672
##
## Accuracy : 0.7673
## 95% CI : (0.7642, 0.7704)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3693
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6939
## Specificity : 0.7827
## Pos Pred Value : 0.4001
## Neg Pred Value : 0.9245
## Prevalence : 0.1728
## Detection Rate : 0.1199
## Detection Prevalence : 0.2997
## Balanced Accuracy : 0.7383
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.612162 0.035756 -73.055 < 2e-16 ***
## age 0.636895 0.006555 97.163 < 2e-16 ***
## ALBUMIN -0.715801 0.005855 -122.255 < 2e-16 ***
## IRON_SAT_PERCENT -0.280975 0.005683 -49.445 < 2e-16 ***
## HGB -0.288197 0.005818 -49.536 < 2e-16 ***
## CLM_FROM_1year_cat 0.427749 0.005961 71.758 < 2e-16 ***
## sex_catMale 0.177662 0.011319 15.695 < 2e-16 ***
## race_cat_4Black 0.219596 0.036910 5.950 2.69e-09 ***
## race_cat_4Other -0.030253 0.057882 -0.523 0.601
## race_cat_4White 0.556048 0.035342 15.733 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 207958 on 289258 degrees of freedom
## AIC: 207978
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.89 |
1.87, 1.92 |
<0.001 |
ALBUMIN |
0.49 |
0.48, 0.49 |
<0.001 |
IRON_SAT_PERCENT |
0.76 |
0.75, 0.76 |
<0.001 |
HGB |
0.75 |
0.74, 0.76 |
<0.001 |
CLM_FROM_1year_cat |
1.53 |
1.52, 1.55 |
<0.001 |
sex_catMale |
1.19 |
1.17, 1.22 |
<0.001 |
race_cat_4Black |
1.25 |
1.16, 1.34 |
<0.001 |
race_cat_4Other |
0.97 |
0.87, 1.09 |
0.6 |
race_cat_4White |
1.74 |
1.63, 1.87 |
<0.001 |
model_23 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
rstate_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.515
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.666
## Area under the curve: 0.8166
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46779 3872
## died 13040 8626
##
## Accuracy : 0.7661
## 95% CI : (0.763, 0.7692)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.366
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6902
## Specificity : 0.7820
## Pos Pred Value : 0.3981
## Neg Pred Value : 0.9236
## Prevalence : 0.1728
## Detection Rate : 0.1193
## Detection Prevalence : 0.2996
## Balanced Accuracy : 0.7361
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.103094 0.014250 -147.588 < 2e-16 ***
## age 0.659309 0.006520 101.118 < 2e-16 ***
## ALBUMIN -0.718992 0.005843 -123.042 < 2e-16 ***
## IRON_SAT_PERCENT -0.286190 0.005677 -50.411 < 2e-16 ***
## HGB -0.276383 0.005799 -47.659 < 2e-16 ***
## CLM_FROM_1year_cat 0.432770 0.005955 72.670 < 2e-16 ***
## sex_catMale 0.201113 0.011276 17.835 < 2e-16 ***
## state_catFIPSNE -0.144626 0.017972 -8.047 8.46e-16 ***
## state_catFIPSS -0.064029 0.014560 -4.398 1.09e-05 ***
## state_catFIPSW -0.136781 0.017890 -7.646 2.08e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208793 on 289258 degrees of freedom
## AIC: 208813
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.93 |
1.91, 1.96 |
<0.001 |
ALBUMIN |
0.49 |
0.48, 0.49 |
<0.001 |
IRON_SAT_PERCENT |
0.75 |
0.74, 0.76 |
<0.001 |
HGB |
0.76 |
0.75, 0.77 |
<0.001 |
CLM_FROM_1year_cat |
1.54 |
1.52, 1.56 |
<0.001 |
sex_catMale |
1.22 |
1.20, 1.25 |
<0.001 |
state_catFIPSNE |
0.87 |
0.84, 0.90 |
<0.001 |
state_catFIPSS |
0.94 |
0.91, 0.97 |
<0.001 |
state_catFIPSW |
0.87 |
0.84, 0.90 |
<0.001 |
model_24 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
sex_cat CALCIUM_UNCORRECTED FERRITIN PHOSPHORUS
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.518
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.667
## Area under the curve: 0.818
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 46929 3881
## died 12890 8617
##
## Accuracy : 0.7681
## 95% CI : (0.765, 0.7712)
## No Information Rate : 0.8272
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.3688
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.6895
## Specificity : 0.7845
## Pos Pred Value : 0.4007
## Neg Pred Value : 0.9236
## Prevalence : 0.1728
## Detection Rate : 0.1192
## Detection Prevalence : 0.2974
## Balanced Accuracy : 0.7370
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.206609 0.009770 -225.864 < 2e-16 ***
## age 0.629356 0.006739 93.387 < 2e-16 ***
## ALBUMIN -0.776165 0.006754 -114.921 < 2e-16 ***
## IRON_SAT_PERCENT -0.305415 0.005889 -51.862 < 2e-16 ***
## HGB -0.267464 0.005874 -45.533 < 2e-16 ***
## CLM_FROM_1year_cat 0.430983 0.005959 72.326 < 2e-16 ***
## sex_catMale 0.239708 0.011400 21.026 < 2e-16 ***
## CALCIUM_UNCORRECTED 0.140538 0.006367 22.074 < 2e-16 ***
## FERRITIN 0.076290 0.005312 14.361 < 2e-16 ***
## PHOSPHORUS -0.022966 0.006098 -3.766 0.000166 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 266319 on 289267 degrees of freedom
## Residual deviance: 208067 on 289258 degrees of freedom
## AIC: 208087
##
## Number of Fisher Scoring iterations: 5
Characteristic |
OR |
95% CI |
p-value |
age |
1.88 |
1.85, 1.90 |
<0.001 |
ALBUMIN |
0.46 |
0.45, 0.47 |
<0.001 |
IRON_SAT_PERCENT |
0.74 |
0.73, 0.75 |
<0.001 |
HGB |
0.77 |
0.76, 0.77 |
<0.001 |
CLM_FROM_1year_cat |
1.54 |
1.52, 1.56 |
<0.001 |
sex_catMale |
1.27 |
1.24, 1.30 |
<0.001 |
CALCIUM_UNCORRECTED |
1.15 |
1.14, 1.17 |
<0.001 |
FERRITIN |
1.08 |
1.07, 1.09 |
<0.001 |
PHOSPHORUS |
0.98 |
0.97, 0.99 |
<0.001 |
model_12 vs model_13 Population (all patients) vs Cases Only (people
who actually died) vs Controls Only (people who survived)
## New
## Standard < 0.2 >= 0.2
## < 0.2 48929 1922
## >= 0.2 2211 19255
## New
## Standard < 0.2 >= 0.2
## < 0.2 3763 544
## >= 0.2 455 7736
## New
## Standard < 0.2 >= 0.2
## < 0.2 45166 1378
## >= 0.2 1756 11519

## Estimate
## NRI 0.013440202
## NRI+ 0.007121139
## NRI- 0.006319063
## Pr(Up|Case) 0.043526964
## Pr(Down|Case) 0.036405825
## Pr(Down|Ctrl) 0.029355222
## Pr(Up|Ctrl) 0.023036159
## Estimate Std.Error Lower Upper
## NRI 0.013440202 0.0026413164 0.008151912 0.018762788
## NRI+ 0.007121139 0.0024823553 0.002031467 0.011883842
## NRI- 0.006319063 0.0009779882 0.004400827 0.008278948
## Pr(Up|Case) 0.043526964 0.0017728315 0.039985528 0.047064479
## Pr(Down|Case) 0.036405825 0.0016822331 0.033153876 0.039770916
## Pr(Down|Ctrl) 0.029355222 0.0007124233 0.027974651 0.030780662
## Pr(Up|Ctrl) 0.023036159 0.0006300763 0.021796019 0.024245110
model_12 vs model_14 Population (all patients) vs Cases Only (people
who actually died) vs Controls Only (people who survived)
## New
## Standard < 0.2 >= 0.2
## < 0.2 46498 4353
## >= 0.2 4105 17361
## New
## Standard < 0.2 >= 0.2
## < 0.2 3241 1066
## >= 0.2 651 7540
## New
## Standard < 0.2 >= 0.2
## < 0.2 43257 3287
## >= 0.2 3454 9821

## Estimate
## NRI 0.035997068
## NRI+ 0.033205313
## NRI- 0.002791755
## Pr(Up|Case) 0.085293647
## Pr(Down|Case) 0.052088334
## Pr(Down|Ctrl) 0.057740852
## Pr(Up|Ctrl) 0.054949096
## Estimate Std.Error Lower Upper
## NRI 0.035997068 0.0036645312 0.0286724661 0.043207831
## NRI+ 0.033205313 0.0033580053 0.0262936635 0.039865206
## NRI- 0.002791755 0.0013640239 0.0001674173 0.005579906
## Pr(Up|Case) 0.085293647 0.0025339040 0.0800431058 0.090046055
## Pr(Down|Case) 0.052088334 0.0019851529 0.0481467043 0.056156489
## Pr(Down|Ctrl) 0.057740852 0.0009729783 0.0558129407 0.059736288
## Pr(Up|Ctrl) 0.054949096 0.0009149237 0.0532179430 0.056725985
model_13 vs model_14 Full Population (all patients) vs Cases Only
(people who actually died) vs Controls Only (people who survived)
## New
## Standard < 0.2 >= 0.2
## < 0.2 47017 4123
## >= 0.2 3586 17591
## New
## Standard < 0.2 >= 0.2
## < 0.2 3286 932
## >= 0.2 606 7674
## New
## Standard < 0.2 >= 0.2
## < 0.2 43731 3191
## >= 0.2 2980 9917

## Estimate
## NRI 0.022556866
## NRI+ 0.026084173
## NRI- -0.003527307
## Pr(Up|Case) 0.074571932
## Pr(Down|Case) 0.048487758
## Pr(Down|Ctrl) 0.049816948
## Pr(Up|Ctrl) 0.053344255
## Estimate Std.Error Lower Upper
## NRI 0.022556866 0.0033188915 0.016371290 0.0293655558
## NRI+ 0.026084173 0.0030529201 0.020342784 0.0322581540
## NRI- -0.003527307 0.0013526816 -0.006118782 -0.0008189014
## Pr(Up|Case) 0.074571932 0.0022470749 0.070259374 0.0789979129
## Pr(Down|Case) 0.048487758 0.0019049600 0.044722943 0.0523012055
## Pr(Down|Ctrl) 0.049816948 0.0009136710 0.048143042 0.0516659237
## Pr(Up|Ctrl) 0.053344255 0.0009326839 0.051392539 0.0551035896
hoslem.test model_12
## cutyhat Var2 Freq_observed
## 1 [0.00108,0.0492] y0 14015
## 2 (0.0492,0.0906] y0 13501
## 3 (0.0906,0.152] y0 12812
## 4 (0.152,0.27] y0 11582
## 5 (0.27,0.982] y0 7909
## 6 [0.00108,0.0492] y1 449
## 7 (0.0492,0.0906] y1 962
## 8 (0.0906,0.152] y1 1651
## 9 (0.152,0.27] y1 2881
## 10 (0.27,0.982] y1 6555
## cutyhat Var2 Freq_expected
## 1 [0.00108,0.0492] yhat0 14041.1738
## 2 (0.0492,0.0906] yhat0 13466.1678
## 3 (0.0906,0.152] yhat0 12749.5108
## 4 (0.152,0.27] yhat0 11536.3725
## 5 (0.27,0.982] yhat0 8030.2462
## 6 [0.00108,0.0492] yhat1 422.8262
## 7 (0.0492,0.0906] yhat1 996.8322
## 8 (0.0906,0.152] yhat1 1713.4892
## 9 (0.152,0.27] yhat1 2926.6275
## 10 (0.27,0.982] yhat1 6433.7538
hoslem.test model_13
## cutyhat Var2 Freq_observed
## 1 [0.00106,0.0471] y0 14020
## 2 (0.0471,0.0874] y0 13537
## 3 (0.0874,0.148] y0 12887
## 4 (0.148,0.269] y0 11601
## 5 (0.269,0.989] y0 7774
## 6 [0.00106,0.0471] y1 444
## 7 (0.0471,0.0874] y1 926
## 8 (0.0874,0.148] y1 1576
## 9 (0.148,0.269] y1 2862
## 10 (0.269,0.989] y1 6690
## cutyhat Var2 Freq_expected
## 1 [0.00106,0.0471] yhat0 14059.6387
## 2 (0.0471,0.0874] yhat0 13505.2362
## 3 (0.0874,0.148] yhat0 12796.7825
## 4 (0.148,0.269] yhat0 11573.4702
## 5 (0.269,0.989] yhat0 7892.0772
## 6 [0.00106,0.0471] yhat1 404.3613
## 7 (0.0471,0.0874] yhat1 957.7638
## 8 (0.0874,0.148] yhat1 1666.2175
## 9 (0.148,0.269] yhat1 2889.5298
## 10 (0.269,0.989] yhat1 6571.9228
hoslem.test model_14
## cutyhat Var2 Freq_observed
## 1 [0.00101,0.0386] y0 14177
## 2 (0.0386,0.0782] y0 13694
## 3 (0.0782,0.146] y0 12893
## 4 (0.146,0.283] y0 11428
## 5 (0.283,0.988] y0 7627
## 6 [0.00101,0.0386] y1 287
## 7 (0.0386,0.0782] y1 769
## 8 (0.0782,0.146] y1 1570
## 9 (0.146,0.283] y1 3035
## 10 (0.283,0.988] y1 6837
## cutyhat Var2 Freq_expected
## 1 [0.00101,0.0386] yhat0 14138.0886
## 2 (0.0386,0.0782] yhat0 13637.7051
## 3 (0.0782,0.146] yhat0 12890.0060
## 4 (0.146,0.283] yhat0 11503.6725
## 5 (0.283,0.988] yhat0 7659.1454
## 6 [0.00101,0.0386] yhat1 325.9114
## 7 (0.0386,0.0782] yhat1 825.2949
## 8 (0.0782,0.146] yhat1 1572.9940
## 9 (0.146,0.283] yhat1 2959.3275
## 10 (0.283,0.988] yhat1 6804.8546