Algorithm Type:
Case, Control
Institution:
Implementation Details:
We have implemented the algorithm utilizing allergy structured data for ACEi.
- For identifying cases. We reviewed 50 patients. We had no false positive at all. PPV of 100%. Total of 147 cases identified using the allergy section.
- For controls, we excluded all patients with an ARB ever. Total remaining number is 1471. Reviewed 50 patients. 48 patients were correct controls (PPV for identifying controls=96%).
NOTE: patients without a documented allergy to ACEi but with later use of ARBs, are a potential source for additional cases if reviewed manually. We reviewed 105 patients of that group and identified 50 additional cases for a total of ~200 cases (including the ones extracted using allergy data). We have 150 more patients in that group and we expect a yield between 40-50% of additional cases. This method can be used to enrich our cases and may be of significant importance to sites without structured allergy data.
Phenotype:
| Cases (0) | Actual Class (Expectation) | |
|
Predicted Class (Observation) |
True Positive: 50.00 |
False Positive: 0.00 |
|
Case Positive Predictive Value:
100%
|
||
| Control (0) | Actual Class (Expectation) | |
|
Predicted Class (Observation) |
True Positive: 48.00 |
False Positive: 2.00 |
|
Control Positive Predictive Value:
96%
|
||