A dataset designed for performing Decision Curve Analysis (DCA). It includes patient characteristics, an outcome variable (cardiac_event), and predicted probabilities from several different risk models.
Usage
data(dca_test)Format
A data frame with 50 rows and 17 variables:
- patient_id
Integer. Unique patient identifier.
- age
Integer. Patient's age in years.
- sex
Character. Patient's sex.
- diabetes
Character. Diabetes status (e.g., "Yes", "No").
- hypertension
Character. Hypertension status (e.g., "Yes", "No").
- smoking
Character. Smoking status (e.g., "Yes", "No").
- cholesterol
Integer. Cholesterol level.
- troponin
Numeric. Troponin level.
- creatinine
Numeric. Creatinine level.
- cardiac_event
Character. The outcome variable, indicating if a cardiac event occurred.
- basic_model
Numeric. Predicted probability of a cardiac event from a basic model.
- enhanced_model
Numeric. Predicted probability from an enhanced model.
- biomarker_model
Numeric. Predicted probability from a model including a biomarker.
- miscalibrated_model
Numeric. Predicted probability from a deliberately miscalibrated model.
- poor_model
Numeric. Predicted probability from a poorly performing model.
- risk_category
Character. A categorized risk based on some criteria.
- hospital
Character. Hospital identifier or group.
Examples
data(dca_test)
str(dca_test)
#> 'data.frame': 50 obs. of 17 variables:
#> $ patient_id : int 1 2 3 4 5 6 7 8 9 10 ...
#> $ age : int 67 59 72 61 74 55 68 70 63 76 ...
#> $ sex : chr "Male" "Female" "Male" "Female" ...
#> $ diabetes : chr "Yes" "No" "Yes" "No" ...
#> $ hypertension : chr "No" "Yes" "Yes" "No" ...
#> $ smoking : chr "Former" "Never" "Current" "Never" ...
#> $ cholesterol : int 245 198 267 201 234 278 189 298 212 256 ...
#> $ troponin : num 2.1 0.8 3.2 1.1 1.8 2.5 0.9 4.1 1.4 2.8 ...
#> $ creatinine : num 1.3 0.9 1.6 1 1.2 1.1 0.8 1.8 1.1 1.5 ...
#> $ cardiac_event : chr "Yes" "No" "Yes" "No" ...
#> $ basic_model : num 0.234 0.089 0.421 0.067 0.198 0.312 0.123 0.567 0.098 0.389 ...
#> $ enhanced_model : num 0.289 0.102 0.523 0.071 0.223 0.398 0.134 0.678 0.112 0.467 ...
#> $ biomarker_model : num 0.312 0.098 0.578 0.069 0.234 0.423 0.128 0.723 0.109 0.512 ...
#> $ miscalibrated_model: num 0.421 0.16 0.758 0.121 0.356 0.562 0.221 0.894 0.176 0.7 ...
#> $ poor_model : num 0.156 0.067 0.298 0.045 0.123 0.234 0.089 0.423 0.078 0.267 ...
#> $ risk_category : chr "Moderate" "Low" "High" "Low" ...
#> $ hospital : chr "Hospital A" "Hospital B" "Hospital C" "Hospital A" ...
head(dca_test)
#> patient_id age sex diabetes hypertension smoking cholesterol troponin
#> 1 1 67 Male Yes No Former 245 2.1
#> 2 2 59 Female No Yes Never 198 0.8
#> 3 3 72 Male Yes Yes Current 267 3.2
#> 4 4 61 Female No No Never 201 1.1
#> 5 5 74 Male No Yes Former 234 1.8
#> 6 6 55 Female Yes No Current 278 2.5
#> creatinine cardiac_event basic_model enhanced_model biomarker_model
#> 1 1.3 Yes 0.234 0.289 0.312
#> 2 0.9 No 0.089 0.102 0.098
#> 3 1.6 Yes 0.421 0.523 0.578
#> 4 1.0 No 0.067 0.071 0.069
#> 5 1.2 No 0.198 0.223 0.234
#> 6 1.1 Yes 0.312 0.398 0.423
#> miscalibrated_model poor_model risk_category hospital
#> 1 0.421 0.156 Moderate Hospital A
#> 2 0.160 0.067 Low Hospital B
#> 3 0.758 0.298 High Hospital C
#> 4 0.121 0.045 Low Hospital A
#> 5 0.356 0.123 Moderate Hospital B
#> 6 0.562 0.234 High Hospital C
summary(dca_test$basic_model)
#> Min. 1st Qu. Median Mean 3rd Qu. Max.
#> 0.0560 0.0980 0.2105 0.2600 0.4160 0.6230
table(dca_test$cardiac_event)
#>
#> No Yes
#> 30 20