Table 1 without transplant no missing, used for analysis
Data preprocessing
1 year selected variables
Model1 : age sex_cat
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.301
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.447
## Area under the curve: 0.5732
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 4586 1058
## died 25456 8843
##
## Accuracy : 0.3362
## 95% CI : (0.3316, 0.3409)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0251
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8931
## Specificity : 0.1527
## Pos Pred Value : 0.2578
## Neg Pred Value : 0.8125
## Prevalence : 0.2479
## Detection Rate : 0.2214
## Detection Prevalence : 0.8587
## Balanced Accuracy : 0.5229
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.140137 0.008965 -127.178 <2e-16 ***
## age 0.244435 0.006099 40.079 <2e-16 ***
## sex_catMale 0.026604 0.011780 2.258 0.0239 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 177283 on 159776 degrees of freedom
## AIC: 177289
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.28 |
1.26, 1.29 |
<0.001 |
| sex_catMale |
1.03 |
1.00, 1.05 |
0.024 |
Model2 : age sex_cat ALBUMIN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.351
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.459
## Area under the curve: 0.6231
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9273 1818
## died 20769 8083
##
## Accuracy : 0.4345
## 95% CI : (0.4297, 0.4394)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0762
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8164
## Specificity : 0.3087
## Pos Pred Value : 0.2802
## Neg Pred Value : 0.8361
## Prevalence : 0.2479
## Detection Rate : 0.2024
## Detection Prevalence : 0.7223
## Balanced Accuracy : 0.5625
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.200762 0.009174 -130.892 < 2e-16 ***
## age 0.259498 0.006149 42.204 < 2e-16 ***
## sex_catMale 0.086011 0.011949 7.198 6.1e-13 ***
## ALBUMIN -0.332049 0.005811 -57.144 < 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: 178955 on 159778 degrees of freedom
## Residual deviance: 173999 on 159775 degrees of freedom
## AIC: 174007
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.30 |
1.28, 1.31 |
<0.001 |
| sex_catMale |
1.09 |
1.06, 1.12 |
<0.001 |
| ALBUMIN |
0.72 |
0.71, 0.73 |
<0.001 |
Model_3 : age sex_cat ALBUMIN COMO_CHF
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.356
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.461
## Area under the curve: 0.6286
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9859 1835
## died 20183 8066
##
## Accuracy : 0.4488
## 95% CI : (0.4439, 0.4537)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0881
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8147
## Specificity : 0.3282
## Pos Pred Value : 0.2855
## Neg Pred Value : 0.8431
## Prevalence : 0.2479
## Detection Rate : 0.2019
## Detection Prevalence : 0.7072
## Balanced Accuracy : 0.5714
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.291494 0.010279 -125.640 < 2e-16 ***
## age 0.250938 0.006178 40.621 < 2e-16 ***
## sex_catMale 0.089138 0.011966 7.449 9.37e-14 ***
## ALBUMIN -0.328876 0.005825 -56.455 < 2e-16 ***
## COMO_CHFY 0.249572 0.012202 20.454 < 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: 178955 on 159778 degrees of freedom
## Residual deviance: 173584 on 159774 degrees of freedom
## AIC: 173594
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.29 |
1.27, 1.30 |
<0.001 |
| sex_catMale |
1.09 |
1.07, 1.12 |
<0.001 |
| ALBUMIN |
0.72 |
0.71, 0.73 |
<0.001 |
| COMO_CHFY |
1.28 |
1.25, 1.31 |
<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 0.356
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.461
## Area under the curve: 0.6287
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9858 1841
## died 20184 8060
##
## Accuracy : 0.4486
## 95% CI : (0.4437, 0.4535)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0877
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8141
## Specificity : 0.3281
## Pos Pred Value : 0.2854
## Neg Pred Value : 0.8426
## Prevalence : 0.2479
## Detection Rate : 0.2018
## Detection Prevalence : 0.7071
## Balanced Accuracy : 0.5711
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.286066 0.011677 -110.138 < 2e-16 ***
## age 0.249773 0.006290 39.712 < 2e-16 ***
## sex_catMale 0.088793 0.011971 7.417 1.2e-13 ***
## ALBUMIN -0.329298 0.005841 -56.376 < 2e-16 ***
## COMO_CHFY 0.250812 0.012268 20.445 < 2e-16 ***
## COMO_DM_INSY -0.011860 0.012126 -0.978 0.328
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 173583 on 159773 degrees of freedom
## AIC: 173595
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.28 |
1.27, 1.30 |
<0.001 |
| sex_catMale |
1.09 |
1.07, 1.12 |
<0.001 |
| ALBUMIN |
0.72 |
0.71, 0.73 |
<0.001 |
| COMO_CHFY |
1.29 |
1.25, 1.32 |
<0.001 |
| COMO_DM_INSY |
0.99 |
0.96, 1.01 |
0.3 |
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 0.357
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.461
## Area under the curve: 0.6296
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9928 1837
## died 20114 8064
##
## Accuracy : 0.4504
## 95% CI : (0.4456, 0.4553)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0895
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8145
## Specificity : 0.3305
## Pos Pred Value : 0.2862
## Neg Pred Value : 0.8439
## Prevalence : 0.2479
## Detection Rate : 0.2019
## Detection Prevalence : 0.7055
## Balanced Accuracy : 0.5725
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.157069 0.019966 -57.951 < 2e-16 ***
## age 0.250487 0.006291 39.816 < 2e-16 ***
## sex_catMale 0.087540 0.011974 7.311 2.66e-13 ***
## ALBUMIN -0.327848 0.005844 -56.099 < 2e-16 ***
## COMO_CHFY 0.251752 0.012271 20.516 < 2e-16 ***
## COMO_DM_INSY -0.009464 0.012131 -0.780 0.435
## COMO_HTNY -0.146122 0.018429 -7.929 2.21e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 173522 on 159772 degrees of freedom
## AIC: 173536
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.28 |
1.27, 1.30 |
<0.001 |
| sex_catMale |
1.09 |
1.07, 1.12 |
<0.001 |
| ALBUMIN |
0.72 |
0.71, 0.73 |
<0.001 |
| COMO_CHFY |
1.29 |
1.26, 1.32 |
<0.001 |
| COMO_DM_INSY |
0.99 |
0.97, 1.01 |
0.4 |
| COMO_HTNY |
0.86 |
0.83, 0.90 |
<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 0.367
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.6456
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12162 2157
## died 17880 7744
##
## Accuracy : 0.4984
## 95% CI : (0.4934, 0.5033)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.122
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7821
## Specificity : 0.4048
## Pos Pred Value : 0.3022
## Neg Pred Value : 0.8494
## Prevalence : 0.2479
## Detection Rate : 0.1939
## Detection Prevalence : 0.6415
## Balanced Accuracy : 0.5935
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.462660 0.021322 -68.599 < 2e-16 ***
## age 0.246071 0.006320 38.938 < 2e-16 ***
## sex_catMale 0.110311 0.012067 9.142 < 2e-16 ***
## ALBUMIN -0.288106 0.005934 -48.553 < 2e-16 ***
## COMO_CHFY 0.225730 0.012368 18.252 < 2e-16 ***
## COMO_DM_INSY -0.014822 0.012215 -1.213 0.225
## COMO_HTNY -0.137038 0.018559 -7.384 1.54e-13 ***
## CLM_FROM_1year_cat1 0.555915 0.012159 45.721 < 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: 178955 on 159778 degrees of freedom
## Residual deviance: 171400 on 159771 degrees of freedom
## AIC: 171416
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.28 |
1.26, 1.29 |
<0.001 |
| sex_catMale |
1.12 |
1.09, 1.14 |
<0.001 |
| ALBUMIN |
0.75 |
0.74, 0.76 |
<0.001 |
| COMO_CHFY |
1.25 |
1.22, 1.28 |
<0.001 |
| COMO_DM_INSY |
0.99 |
0.96, 1.01 |
0.2 |
| COMO_HTNY |
0.87 |
0.84, 0.90 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.74 |
1.70, 1.79 |
<0.001 |
model_7 : age ALBUMIN COMO_CHF
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.355
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.461
## Area under the curve: 0.6284
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9772 1821
## died 20270 8080
##
## Accuracy : 0.4469
## 95% CI : (0.4421, 0.4518)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.087
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8161
## Specificity : 0.3253
## Pos Pred Value : 0.2850
## Neg Pred Value : 0.8429
## Prevalence : 0.2479
## Detection Rate : 0.2023
## Detection Prevalence : 0.7098
## Balanced Accuracy : 0.5707
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.239777 0.007516 -164.95 <2e-16 ***
## age 0.249718 0.006174 40.45 <2e-16 ***
## ALBUMIN -0.325229 0.005805 -56.03 <2e-16 ***
## COMO_CHFY 0.248379 0.012198 20.36 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 173640 on 159775 degrees of freedom
## AIC: 173648
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.28 |
1.27, 1.30 |
<0.001 |
| ALBUMIN |
0.72 |
0.71, 0.73 |
<0.001 |
| COMO_CHFY |
1.28 |
1.25, 1.31 |
<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 0.357
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.462
## Area under the curve: 0.6323
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 10150 1843
## died 19892 8058
##
## Accuracy : 0.4558
## 95% CI : (0.451, 0.4607)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0942
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8139
## Specificity : 0.3379
## Pos Pred Value : 0.2883
## Neg Pred Value : 0.8463
## Prevalence : 0.2479
## Detection Rate : 0.2017
## Detection Prevalence : 0.6997
## Balanced Accuracy : 0.5759
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.237792 0.007526 -164.46 <2e-16 ***
## age 0.260366 0.006213 41.91 <2e-16 ***
## ALBUMIN -0.317463 0.005831 -54.44 <2e-16 ***
## COMO_CHFY 0.231845 0.012239 18.94 <2e-16 ***
## IRON_SAT_PERCENT -0.116009 0.005997 -19.34 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 173262 on 159774 degrees of freedom
## AIC: 173272
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.30 |
1.28, 1.31 |
<0.001 |
| ALBUMIN |
0.73 |
0.72, 0.74 |
<0.001 |
| COMO_CHFY |
1.26 |
1.23, 1.29 |
<0.001 |
| IRON_SAT_PERCENT |
0.89 |
0.88, 0.90 |
<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 0.360
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.463
## Area under the curve: 0.6349
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 10449 1905
## died 19593 7996
##
## Accuracy : 0.4618
## 95% CI : (0.4569, 0.4667)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0972
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8076
## Specificity : 0.3478
## Pos Pred Value : 0.2898
## Neg Pred Value : 0.8458
## Prevalence : 0.2479
## Detection Rate : 0.2002
## Detection Prevalence : 0.6907
## Balanced Accuracy : 0.5777
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.242642 0.007547 -164.66 <2e-16 ***
## age 0.266402 0.006226 42.79 <2e-16 ***
## ALBUMIN -0.286903 0.006025 -47.62 <2e-16 ***
## COMO_CHFY 0.236998 0.012260 19.33 <2e-16 ***
## IRON_SAT_PERCENT -0.097690 0.006051 -16.14 <2e-16 ***
## HGB -0.122962 0.006199 -19.84 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 172867 on 159773 degrees of freedom
## AIC: 172879
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.31 |
1.29, 1.32 |
<0.001 |
| ALBUMIN |
0.75 |
0.74, 0.76 |
<0.001 |
| COMO_CHFY |
1.27 |
1.24, 1.30 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.92 |
<0.001 |
| HGB |
0.88 |
0.87, 0.90 |
<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 0.366
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.6469
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12167 2163
## died 17875 7738
##
## Accuracy : 0.4983
## 95% CI : (0.4934, 0.5033)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1218
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7815
## Specificity : 0.4050
## Pos Pred Value : 0.3021
## Neg Pred Value : 0.8491
## Prevalence : 0.2479
## Detection Rate : 0.1937
## Detection Prevalence : 0.6412
## Balanced Accuracy : 0.5933
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.515253 0.010123 -149.68 <2e-16 ***
## age 0.254379 0.006238 40.78 <2e-16 ***
## ALBUMIN -0.279342 0.005918 -47.20 <2e-16 ***
## COMO_CHFY 0.208690 0.012328 16.93 <2e-16 ***
## IRON_SAT_PERCENT -0.095877 0.006039 -15.88 <2e-16 ***
## CLM_FROM_1year_cat1 0.537424 0.012179 44.13 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 171288 on 159773 degrees of freedom
## AIC: 171300
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.29 |
1.27, 1.31 |
<0.001 |
| ALBUMIN |
0.76 |
0.75, 0.77 |
<0.001 |
| COMO_CHFY |
1.23 |
1.20, 1.26 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.92 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.71 |
1.67, 1.75 |
<0.001 |
model_11 : age,ALBUMIN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.350
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.459
## Area under the curve: 0.623
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9269 1785
## died 20773 8116
##
## Accuracy : 0.4352
## 95% CI : (0.4304, 0.4401)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0781
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8197
## Specificity : 0.3085
## Pos Pred Value : 0.2809
## Neg Pred Value : 0.8385
## Prevalence : 0.2479
## Detection Rate : 0.2032
## Detection Prevalence : 0.7233
## Balanced Accuracy : 0.5641
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.151270 0.006001 -191.84 <2e-16 ***
## age 0.258276 0.006145 42.03 <2e-16 ***
## ALBUMIN -0.328514 0.005790 -56.74 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 174051 on 159776 degrees of freedom
## AIC: 174057
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.29 |
1.28, 1.31 |
<0.001 |
| ALBUMIN |
0.72 |
0.71, 0.73 |
<0.001 |
model_12 : age ALBUMIN IRON_SAT_PERCENT
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.352
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.460
## Area under the curve: 0.6276
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 9773 1833
## died 20269 8068
##
## Accuracy : 0.4467
## 95% CI : (0.4418, 0.4516)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0863
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8149
## Specificity : 0.3253
## Pos Pred Value : 0.2847
## Neg Pred Value : 0.8421
## Prevalence : 0.2479
## Detection Rate : 0.2020
## Detection Prevalence : 0.7094
## Balanced Accuracy : 0.5701
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.155417 0.006022 -191.86 <2e-16 ***
## age 0.269005 0.006184 43.50 <2e-16 ***
## ALBUMIN -0.319974 0.005819 -54.99 <2e-16 ***
## IRON_SAT_PERCENT -0.123703 0.005977 -20.70 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 173618 on 159775 degrees of freedom
## AIC: 173626
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.31 |
1.29, 1.32 |
<0.001 |
| ALBUMIN |
0.73 |
0.72, 0.73 |
<0.001 |
| IRON_SAT_PERCENT |
0.88 |
0.87, 0.89 |
<0.001 |
model_13 : age ALBUMIN IRON_SAT_PERCENT HGB
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.355
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.462
## Area under the curve: 0.6301
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 10106 1869
## died 19936 8032
##
## Accuracy : 0.4541
## 95% CI : (0.4492, 0.459)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0916
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8112
## Specificity : 0.3364
## Pos Pred Value : 0.2872
## Neg Pred Value : 0.8439
## Prevalence : 0.2479
## Detection Rate : 0.2011
## Detection Prevalence : 0.7002
## Balanced Accuracy : 0.5738
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.158382 0.006037 -191.89 <2e-16 ***
## age 0.275139 0.006197 44.40 <2e-16 ***
## ALBUMIN -0.290083 0.006011 -48.26 <2e-16 ***
## IRON_SAT_PERCENT -0.105884 0.006030 -17.56 <2e-16 ***
## HGB -0.120535 0.006192 -19.46 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 173237 on 159774 degrees of freedom
## AIC: 173247
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.32 |
1.30, 1.33 |
<0.001 |
| ALBUMIN |
0.75 |
0.74, 0.76 |
<0.001 |
| IRON_SAT_PERCENT |
0.90 |
0.89, 0.91 |
<0.001 |
| HGB |
0.89 |
0.88, 0.90 |
<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 0.364
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.467
## Area under the curve: 0.6451
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12136 2152
## died 17906 7749
##
## Accuracy : 0.4978
## 95% CI : (0.4929, 0.5028)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1217
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7826
## Specificity : 0.4040
## Pos Pred Value : 0.3020
## Neg Pred Value : 0.8494
## Prevalence : 0.2479
## Detection Rate : 0.1940
## Detection Prevalence : 0.6423
## Balanced Accuracy : 0.5933
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.438664 0.009179 -156.73 <2e-16 ***
## age 0.267173 0.006223 42.93 <2e-16 ***
## ALBUMIN -0.258759 0.006082 -42.55 <2e-16 ***
## IRON_SAT_PERCENT -0.089474 0.006068 -14.74 <2e-16 ***
## HGB -0.094202 0.006237 -15.10 <2e-16 ***
## CLM_FROM_1year_cat1 0.528415 0.012222 43.24 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 171344 on 159773 degrees of freedom
## AIC: 171356
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.31 |
1.29, 1.32 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.93 |
<0.001 |
| HGB |
0.91 |
0.90, 0.92 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.70 |
1.66, 1.74 |
<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 0.368
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.469
## Area under the curve: 0.6483
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12310 2146
## died 17732 7755
##
## Accuracy : 0.5023
## 95% CI : (0.4974, 0.5073)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1263
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7833
## Specificity : 0.4098
## Pos Pred Value : 0.3043
## Neg Pred Value : 0.8515
## Prevalence : 0.2479
## Detection Rate : 0.1942
## Detection Prevalence : 0.6381
## Balanced Accuracy : 0.5965
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.369192 0.009947 -137.644 < 2e-16 ***
## age 0.256758 0.006285 40.850 < 2e-16 ***
## ALBUMIN -0.258220 0.006092 -42.390 < 2e-16 ***
## IRON_SAT_PERCENT -0.086471 0.006075 -14.235 < 2e-16 ***
## HGB -0.103115 0.006273 -16.438 < 2e-16 ***
## CLM_FROM_1year_cat1 0.522844 0.012237 42.726 < 2e-16 ***
## race_cat_4Asian -0.289848 0.036018 -8.047 8.47e-16 ***
## race_cat_4Black -0.222031 0.014556 -15.254 < 2e-16 ***
## race_cat_4Other -0.288272 0.044669 -6.454 1.09e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 171034 on 159770 degrees of freedom
## AIC: 171052
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.29 |
1.28, 1.31 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.92 |
0.91, 0.93 |
<0.001 |
| HGB |
0.90 |
0.89, 0.91 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.69 |
1.65, 1.73 |
<0.001 |
| race_cat_4Asian |
0.75 |
0.70, 0.80 |
<0.001 |
| race_cat_4Black |
0.80 |
0.78, 0.82 |
<0.001 |
| race_cat_4Other |
0.75 |
0.69, 0.82 |
<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 0.367
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.469
## Area under the curve: 0.6484
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12324 2144
## died 17718 7757
##
## Accuracy : 0.5027
## 95% CI : (0.4978, 0.5077)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1268
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7835
## Specificity : 0.4102
## Pos Pred Value : 0.3045
## Neg Pred Value : 0.8518
## Prevalence : 0.2479
## Detection Rate : 0.1942
## Detection Prevalence : 0.6378
## Balanced Accuracy : 0.5968
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.317146 0.014947 -88.123 < 2e-16 ***
## age 0.258580 0.006315 40.949 < 2e-16 ***
## ALBUMIN -0.258245 0.006095 -42.372 < 2e-16 ***
## IRON_SAT_PERCENT -0.086549 0.006077 -14.241 < 2e-16 ***
## HGB -0.102565 0.006287 -16.314 < 2e-16 ***
## CLM_FROM_1year_cat1 0.519550 0.012255 42.394 < 2e-16 ***
## race_cat_4Asian -0.252697 0.036554 -6.913 4.74e-12 ***
## race_cat_4Black -0.234576 0.014802 -15.848 < 2e-16 ***
## race_cat_4Other -0.262204 0.045008 -5.826 5.69e-09 ***
## state_catFIPSNE -0.099387 0.019454 -5.109 3.24e-07 ***
## state_catFIPSS -0.026432 0.015639 -1.690 0.091 .
## state_catFIPSW -0.116344 0.019314 -6.024 1.70e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 170981 on 159767 degrees of freedom
## AIC: 171005
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.30 |
1.28, 1.31 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.92 |
0.91, 0.93 |
<0.001 |
| HGB |
0.90 |
0.89, 0.91 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.68 |
1.64, 1.72 |
<0.001 |
| race_cat_4Asian |
0.78 |
0.72, 0.83 |
<0.001 |
| race_cat_4Black |
0.79 |
0.77, 0.81 |
<0.001 |
| race_cat_4Other |
0.77 |
0.70, 0.84 |
<0.001 |
| state_catFIPSNE |
0.91 |
0.87, 0.94 |
<0.001 |
| state_catFIPSS |
0.97 |
0.94, 1.00 |
0.091 |
| state_catFIPSW |
0.89 |
0.86, 0.92 |
<0.001 |
model_17 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
CALCIUM_CORRECTED
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.364
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.467
## Area under the curve: 0.6452
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12152 2153
## died 17890 7748
##
## Accuracy : 0.4982
## 95% CI : (0.4933, 0.5031)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.122
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7825
## Specificity : 0.4045
## Pos Pred Value : 0.3022
## Neg Pred Value : 0.8495
## Prevalence : 0.2479
## Detection Rate : 0.1940
## Detection Prevalence : 0.6419
## Balanced Accuracy : 0.5935
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.438543 0.009180 -156.712 <2e-16 ***
## age 0.266126 0.006249 42.589 <2e-16 ***
## ALBUMIN -0.257584 0.006115 -42.121 <2e-16 ***
## IRON_SAT_PERCENT -0.089274 0.006070 -14.708 <2e-16 ***
## HGB -0.094654 0.006242 -15.165 <2e-16 ***
## CLM_FROM_1year_cat1 0.528156 0.012223 43.211 <2e-16 ***
## CALCIUM_CORRECTED 0.010582 0.005804 1.823 0.0683 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 171341 on 159772 degrees of freedom
## AIC: 171355
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.30 |
1.29, 1.32 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.93 |
<0.001 |
| HGB |
0.91 |
0.90, 0.92 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.70 |
1.66, 1.74 |
<0.001 |
| CALCIUM_CORRECTED |
1.01 |
1.00, 1.02 |
0.068 |
model_18 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
CALCIUM_CORRECTED FERRITIN
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.364
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.6454
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12182 2157
## died 17860 7744
##
## Accuracy : 0.4989
## 95% CI : (0.4939, 0.5038)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1225
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7821
## Specificity : 0.4055
## Pos Pred Value : 0.3025
## Neg Pred Value : 0.8496
## Prevalence : 0.2479
## Detection Rate : 0.1939
## Detection Prevalence : 0.6410
## Balanced Accuracy : 0.5938
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.437897 0.009180 -156.629 < 2e-16 ***
## age 0.263589 0.006271 42.030 < 2e-16 ***
## ALBUMIN -0.256843 0.006118 -41.979 < 2e-16 ***
## IRON_SAT_PERCENT -0.097359 0.006302 -15.449 < 2e-16 ***
## HGB -0.089855 0.006320 -14.218 < 2e-16 ***
## CLM_FROM_1year_cat1 0.526598 0.012228 43.066 < 2e-16 ***
## CALCIUM_CORRECTED 0.008373 0.005880 1.424 0.154
## FERRITIN 0.029325 0.006065 4.835 1.33e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 171317 on 159771 degrees of freedom
## AIC: 171333
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.30 |
1.29, 1.32 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.92 |
<0.001 |
| HGB |
0.91 |
0.90, 0.93 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.69 |
1.65, 1.73 |
<0.001 |
| CALCIUM_CORRECTED |
1.01 |
1.00, 1.02 |
0.2 |
| FERRITIN |
1.03 |
1.02, 1.04 |
<0.001 |
model_19 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
CALCIUM_CORRECTED FERRITIN PHOSPHORUS
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.365
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.6466
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12198 2139
## died 17844 7762
##
## Accuracy : 0.4997
## 95% CI : (0.4948, 0.5046)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.124
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7840
## Specificity : 0.4060
## Pos Pred Value : 0.3031
## Neg Pred Value : 0.8508
## Prevalence : 0.2479
## Detection Rate : 0.1943
## Detection Prevalence : 0.6411
## Balanced Accuracy : 0.5950
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.440489 0.009190 -156.742 < 2e-16 ***
## age 0.284681 0.006547 43.481 < 2e-16 ***
## ALBUMIN -0.266163 0.006177 -43.090 < 2e-16 ***
## IRON_SAT_PERCENT -0.098719 0.006307 -15.652 < 2e-16 ***
## HGB -0.086889 0.006334 -13.717 < 2e-16 ***
## CLM_FROM_1year_cat1 0.528995 0.012235 43.237 < 2e-16 ***
## CALCIUM_CORRECTED 0.014869 0.005805 2.561 0.0104 *
## FERRITIN 0.034020 0.006078 5.597 2.18e-08 ***
## PHOSPHORUS 0.072975 0.006324 11.539 < 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: 178955 on 159778 degrees of freedom
## Residual deviance: 171185 on 159770 degrees of freedom
## AIC: 171203
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.33 |
1.31, 1.35 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.89, 0.92 |
<0.001 |
| HGB |
0.92 |
0.91, 0.93 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.70 |
1.66, 1.74 |
<0.001 |
| CALCIUM_CORRECTED |
1.01 |
1.00, 1.03 |
0.010 |
| FERRITIN |
1.03 |
1.02, 1.05 |
<0.001 |
| PHOSPHORUS |
1.08 |
1.06, 1.09 |
<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 0.356
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.462
## Area under the curve: 0.6306
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 10178 1892
## died 19864 8009
##
## Accuracy : 0.4553
## 95% CI : (0.4504, 0.4602)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.0918
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8089
## Specificity : 0.3388
## Pos Pred Value : 0.2873
## Neg Pred Value : 0.8432
## Prevalence : 0.2479
## Detection Rate : 0.2005
## Detection Prevalence : 0.6978
## Balanced Accuracy : 0.5739
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.216411 0.009225 -131.858 <2e-16 ***
## age 0.276706 0.006202 44.617 <2e-16 ***
## ALBUMIN -0.293435 0.006025 -48.705 <2e-16 ***
## IRON_SAT_PERCENT -0.105944 0.006031 -17.568 <2e-16 ***
## HGB -0.123208 0.006198 -19.877 <2e-16 ***
## sex_catMale 0.100838 0.011994 8.407 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 173166 on 159773 degrees of freedom
## AIC: 173178
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.32 |
1.30, 1.33 |
<0.001 |
| ALBUMIN |
0.75 |
0.74, 0.75 |
<0.001 |
| IRON_SAT_PERCENT |
0.90 |
0.89, 0.91 |
<0.001 |
| HGB |
0.88 |
0.87, 0.89 |
<0.001 |
| sex_catMale |
1.11 |
1.08, 1.13 |
<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 0.366
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.646
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12215 2150
## died 17827 7751
##
## Accuracy : 0.4999
## 95% CI : (0.4949, 0.5048)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1238
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7829
## Specificity : 0.4066
## Pos Pred Value : 0.3030
## Neg Pred Value : 0.8503
## Prevalence : 0.2479
## Detection Rate : 0.1941
## Detection Prevalence : 0.6404
## Balanced Accuracy : 0.5947
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.509941 0.011724 -128.795 <2e-16 ***
## age 0.268992 0.006228 43.190 <2e-16 ***
## ALBUMIN -0.262460 0.006094 -43.070 <2e-16 ***
## IRON_SAT_PERCENT -0.089369 0.006069 -14.725 <2e-16 ***
## HGB -0.097162 0.006242 -15.566 <2e-16 ***
## sex_catMale 0.119790 0.012077 9.919 <2e-16 ***
## CLM_FROM_1year_cat1 0.532825 0.012236 43.547 <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: 178955 on 159778 degrees of freedom
## Residual deviance: 171245 on 159772 degrees of freedom
## AIC: 171259
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.31 |
1.29, 1.32 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.93 |
<0.001 |
| HGB |
0.91 |
0.90, 0.92 |
<0.001 |
| sex_catMale |
1.13 |
1.10, 1.15 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.70 |
1.66, 1.75 |
<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 0.369
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.469
## Area under the curve: 0.6489
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12334 2158
## died 17708 7743
##
## Accuracy : 0.5026
## 95% CI : (0.4977, 0.5076)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1262
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7820
## Specificity : 0.4106
## Pos Pred Value : 0.3042
## Neg Pred Value : 0.8511
## Prevalence : 0.2479
## Detection Rate : 0.1939
## Detection Prevalence : 0.6372
## Balanced Accuracy : 0.5963
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.434744 0.012511 -114.674 < 2e-16 ***
## age 0.258815 0.006292 41.134 < 2e-16 ***
## ALBUMIN -0.261594 0.006104 -42.856 < 2e-16 ***
## IRON_SAT_PERCENT -0.086441 0.006075 -14.229 < 2e-16 ***
## HGB -0.105304 0.006276 -16.780 < 2e-16 ***
## CLM_FROM_1year_cat1 0.526906 0.012251 43.011 < 2e-16 ***
## sex_catMale 0.105968 0.012122 8.742 < 2e-16 ***
## race_cat_4Asian -0.288171 0.036027 -7.999 1.26e-15 ***
## race_cat_4Black -0.212321 0.014602 -14.541 < 2e-16 ***
## race_cat_4Other -0.284054 0.044677 -6.358 2.04e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 170957 on 159769 degrees of freedom
## AIC: 170977
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.30 |
1.28, 1.31 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.92 |
0.91, 0.93 |
<0.001 |
| HGB |
0.90 |
0.89, 0.91 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.69 |
1.65, 1.73 |
<0.001 |
| sex_catMale |
1.11 |
1.09, 1.14 |
<0.001 |
| race_cat_4Asian |
0.75 |
0.70, 0.80 |
<0.001 |
| race_cat_4Black |
0.81 |
0.79, 0.83 |
<0.001 |
| race_cat_4Other |
0.75 |
0.69, 0.82 |
<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 0.366
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.6461
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12220 2158
## died 17822 7743
##
## Accuracy : 0.4998
## 95% CI : (0.4949, 0.5047)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1234
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7820
## Specificity : 0.4068
## Pos Pred Value : 0.3029
## Neg Pred Value : 0.8499
## Prevalence : 0.2479
## Detection Rate : 0.1939
## Detection Prevalence : 0.6400
## Balanced Accuracy : 0.5944
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.447654 0.016274 -88.953 < 2e-16 ***
## age 0.270813 0.006267 43.215 < 2e-16 ***
## ALBUMIN -0.262863 0.006097 -43.111 < 2e-16 ***
## IRON_SAT_PERCENT -0.089133 0.006072 -14.679 < 2e-16 ***
## HGB -0.096428 0.006259 -15.406 < 2e-16 ***
## CLM_FROM_1year_cat1 0.529638 0.012254 43.223 < 2e-16 ***
## sex_catMale 0.122035 0.012087 10.096 < 2e-16 ***
## state_catFIPSNE -0.107268 0.019441 -5.518 3.44e-08 ***
## state_catFIPSS -0.050320 0.015550 -3.236 0.00121 **
## state_catFIPSW -0.122754 0.018940 -6.481 9.11e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 178955 on 159778 degrees of freedom
## Residual deviance: 171191 on 159769 degrees of freedom
## AIC: 171211
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.31 |
1.30, 1.33 |
<0.001 |
| ALBUMIN |
0.77 |
0.76, 0.78 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.90, 0.93 |
<0.001 |
| HGB |
0.91 |
0.90, 0.92 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.70 |
1.66, 1.74 |
<0.001 |
| sex_catMale |
1.13 |
1.10, 1.16 |
<0.001 |
| state_catFIPSNE |
0.90 |
0.86, 0.93 |
<0.001 |
| state_catFIPSS |
0.95 |
0.92, 0.98 |
0.001 |
| state_catFIPSW |
0.88 |
0.85, 0.92 |
<0.001 |
model_24 : age ALBUMIN IRON_SAT_PERCENT HGB CLM_FROM_1year_cat
sex_cat CALCIUM_CORRECTED FERRITIN PHOSPHORUS
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 pr_auc binary 0.367
## # A tibble: 1 × 3
## .metric .estimator .estimate
## <chr> <chr> <dbl>
## 1 brier_class binary 0.468
## Area under the curve: 0.6476
## Confusion Matrix and Statistics
##
## Reference
## Prediction alive died
## alive 12259 2161
## died 17783 7740
##
## Accuracy : 0.5007
## 95% CI : (0.4958, 0.5056)
## No Information Rate : 0.7521
## P-Value [Acc > NIR] : 1
##
## Kappa : 0.1241
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.7817
## Specificity : 0.4081
## Pos Pred Value : 0.3033
## Neg Pred Value : 0.8501
## Prevalence : 0.2479
## Detection Rate : 0.1938
## Detection Prevalence : 0.6390
## Balanced Accuracy : 0.5949
##
## 'Positive' Class : died
##
##
## Call:
## NULL
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.518255 0.011775 -128.943 < 2e-16 ***
## age 0.285968 0.006553 43.641 < 2e-16 ***
## ALBUMIN -0.269372 0.006186 -43.544 < 2e-16 ***
## IRON_SAT_PERCENT -0.099596 0.006309 -15.787 < 2e-16 ***
## HGB -0.089704 0.006338 -14.153 < 2e-16 ***
## CLM_FROM_1year_cat1 0.533412 0.012248 43.550 < 2e-16 ***
## sex_catMale 0.130984 0.012206 10.731 < 2e-16 ***
## CALCIUM_CORRECTED 0.022488 0.005957 3.775 0.00016 ***
## FERRITIN 0.038192 0.006091 6.270 3.6e-10 ***
## PHOSPHORUS 0.074064 0.006323 11.714 < 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: 178955 on 159778 degrees of freedom
## Residual deviance: 171069 on 159769 degrees of freedom
## AIC: 171089
##
## Number of Fisher Scoring iterations: 4
| Characteristic |
OR |
95% CI |
p-value |
| age |
1.33 |
1.31, 1.35 |
<0.001 |
| ALBUMIN |
0.76 |
0.75, 0.77 |
<0.001 |
| IRON_SAT_PERCENT |
0.91 |
0.89, 0.92 |
<0.001 |
| HGB |
0.91 |
0.90, 0.93 |
<0.001 |
| CLM_FROM_1year_cat1 |
1.70 |
1.66, 1.75 |
<0.001 |
| sex_catMale |
1.14 |
1.11, 1.17 |
<0.001 |
| CALCIUM_CORRECTED |
1.02 |
1.01, 1.03 |
<0.001 |
| FERRITIN |
1.04 |
1.03, 1.05 |
<0.001 |
| PHOSPHORUS |
1.08 |
1.06, 1.09 |
<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 10435 1171
## >= 0.2 1540 26797
## New
## Standard < 0.2 >= 0.2
## < 0.2 1580 253
## >= 0.2 289 7779
## New
## Standard < 0.2 >= 0.2
## < 0.2 8855 918
## >= 0.2 1251 19018
## Estimate
## NRI 0.007448485
## NRI+ -0.003635996
## NRI- 0.011084482
## Pr(Up|Case) 0.025552974
## Pr(Down|Case) 0.029188971
## Pr(Down|Ctrl) 0.041641702
## Pr(Up|Ctrl) 0.030557220
## Estimate Std.Error Lower Upper
## NRI 0.007448485 0.002722883 0.002293588 0.012799418
## NRI+ -0.003635996 0.002343448 -0.008084741 0.001155687
## NRI- 0.011084482 0.001550883 0.007993854 0.014175318
## Pr(Up|Case) 0.025552974 0.001606399 0.022441516 0.028773413
## Pr(Down|Case) 0.029188971 0.001667243 0.025871333 0.032225275
## Pr(Down|Ctrl) 0.041641702 0.001135158 0.039381627 0.043802409
## Pr(Up|Ctrl) 0.030557220 0.001021138 0.028527693 0.032621678
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 8639 2967
## >= 0.2 5649 22688
## New
## Standard < 0.2 >= 0.2
## < 0.2 1166 667
## >= 0.2 986 7082
## New
## Standard < 0.2 >= 0.2
## < 0.2 7473 2300
## >= 0.2 4663 15606
## Estimate
## NRI 0.04643758
## NRI+ -0.03221897
## NRI- 0.07865655
## Pr(Up|Case) 0.06736693
## Pr(Down|Case) 0.09958590
## Pr(Down|Ctrl) 0.15521603
## Pr(Up|Ctrl) 0.07655948
## Estimate Std.Error Lower Upper
## NRI 0.04643758 0.005032227 0.03676496 0.05706975
## NRI+ -0.03221897 0.004198386 -0.04015489 -0.02339857
## NRI- 0.07865655 0.002787990 0.07318613 0.08397340
## Pr(Up|Case) 0.06736693 0.002506589 0.06229258 0.07208713
## Pr(Down|Case) 0.09958590 0.003102287 0.09341115 0.10576379
## Pr(Down|Ctrl) 0.15521603 0.002115412 0.15101267 0.15930416
## Pr(Up|Ctrl) 0.07655948 0.001573679 0.07368142 0.07959437
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 8984 2991
## >= 0.2 5304 22664
## New
## Standard < 0.2 >= 0.2
## < 0.2 1212 657
## >= 0.2 940 7092
## New
## Standard < 0.2 >= 0.2
## < 0.2 7772 2334
## >= 0.2 4364 15572
## Estimate
## NRI 0.03898909
## NRI+ -0.02858297
## NRI- 0.06757207
## Pr(Up|Case) 0.06635693
## Pr(Down|Case) 0.09493991
## Pr(Down|Ctrl) 0.14526330
## Pr(Up|Ctrl) 0.07769123
## Estimate Std.Error Lower Upper
## NRI 0.03898909 0.004922244 0.02993463 0.04842877
## NRI+ -0.02858297 0.004051782 -0.03648771 -0.02052189
## NRI- 0.06757207 0.002657664 0.06235732 0.07281740
## Pr(Up|Case) 0.06635693 0.002553261 0.06150261 0.07142492
## Pr(Down|Case) 0.09493991 0.002950776 0.08905765 0.10056321
## Pr(Down|Ctrl) 0.14526330 0.002040179 0.14134676 0.14922027
## Pr(Up|Ctrl) 0.07769123 0.001513570 0.07469694 0.08049996
hoslem.test model_12
## cutyhat Var2 Freq_observed
## 1 [0.046,0.181] y0 6785
## 2 (0.181,0.221] y0 6491
## 3 (0.221,0.259] y0 6123
## 4 (0.259,0.309] y0 5709
## 5 (0.309,0.699] y0 4934
## 6 [0.046,0.181] y1 1204
## 7 (0.181,0.221] y1 1497
## 8 (0.221,0.259] y1 1866
## 9 (0.259,0.309] y1 2279
## 10 (0.309,0.699] y1 3055
## cutyhat Var2 Freq_expected
## 1 [0.046,0.181] yhat0 6796.055
## 2 (0.181,0.221] yhat0 6378.333
## 3 (0.221,0.259] yhat0 6077.979
## 4 (0.259,0.309] yhat0 5738.186
## 5 (0.309,0.699] yhat0 5050.824
## 6 [0.046,0.181] yhat1 1192.945
## 7 (0.181,0.221] yhat1 1609.667
## 8 (0.221,0.259] yhat1 1911.021
## 9 (0.259,0.309] yhat1 2249.814
## 10 (0.309,0.699] yhat1 2938.176
hoslem.test model_13
## cutyhat Var2 Freq_observed
## 1 [0.043,0.179] y0 6819
## 2 (0.179,0.219] y0 6487
## 3 (0.219,0.258] y0 6133
## 4 (0.258,0.311] y0 5667
## 5 (0.311,0.729] y0 4936
## 6 [0.043,0.179] y1 1170
## 7 (0.179,0.219] y1 1501
## 8 (0.219,0.258] y1 1856
## 9 (0.258,0.311] y1 2321
## 10 (0.311,0.729] y1 3053
## cutyhat Var2 Freq_expected
## 1 [0.043,0.179] yhat0 6814.883
## 2 (0.179,0.219] yhat0 6392.156
## 3 (0.219,0.258] yhat0 6088.259
## 4 (0.258,0.311] yhat0 5735.201
## 5 (0.311,0.729] yhat0 5011.084
## 6 [0.043,0.179] yhat1 1174.117
## 7 (0.179,0.219] yhat1 1595.844
## 8 (0.219,0.258] yhat1 1900.741
## 9 (0.258,0.311] yhat1 2252.799
## 10 (0.311,0.729] yhat1 2977.916
hoslem.test model_14
## cutyhat Var2 Freq_observed
## 1 [0.0385,0.163] y0 6983
## 2 (0.163,0.21] y0 6491
## 3 (0.21,0.263] y0 6086
## 4 (0.263,0.328] y0 5619
## 5 (0.328,0.727] y0 4863
## 6 [0.0385,0.163] y1 1006
## 7 (0.163,0.21] y1 1497
## 8 (0.21,0.263] y1 1903
## 9 (0.263,0.328] y1 2369
## 10 (0.328,0.727] y1 3126
## cutyhat Var2 Freq_expected
## 1 [0.0385,0.163] yhat0 6931.040
## 2 (0.163,0.21] yhat0 6497.373
## 3 (0.21,0.263] yhat0 6108.523
## 4 (0.263,0.328] yhat0 5644.838
## 5 (0.328,0.727] yhat0 4849.534
## 6 [0.0385,0.163] yhat1 1057.960
## 7 (0.163,0.21] yhat1 1490.627
## 8 (0.21,0.263] yhat1 1880.477
## 9 (0.263,0.328] yhat1 2343.162
## 10 (0.328,0.727] yhat1 3139.466