II. Definitions
- Test Sensitivity
- Screening Test's ability to identify true disease
- A Test with high sensitivity has few False Negatives
- Independent of disease Prevalence in the community
- Sensitive Tests allow user to RULE OUT a condition (mnemonic "SNout")
- Recall
- Identical to the Test Sensitivity, but Recall term is used more in data science
- Reflects what percentage of relevant items are positive (not missed as False Negatives)
- See Test Precision (Positive Predictive Value)
III. Calculation
- Test Sensitivity
- True positive tests per total affected patients tested
- Test Sensitivity = P(positive test | disease)
- Where P (A | B) = Probability of A given B
- Expressed as a percentage
- False Negative Rate
- True cases missed by the Screening Test (test negative despite presence of condition)
- False Negative Rate = 1 - Test Sensitivity
IV. Example: A new Screening Test for Crohn's Disease
- Patients with known Crohn's Disease tested: 45
- Patients with known Crohn's Disease who have a positive test: 36
- Sensitivity = 36/45 or 80%
V. References
- Hennekens (1987) Epidemiology Medicine, p.327-47
- Gates (2001) Am Fam Physician 63(3):513-22 [PubMed]
- MacLean (1996) Med Clin North Am 80(1):1-14 [PubMed]
- Nielsen (1999) Med Clin North Am 83(6):1323-37 [PubMed]