Solution:
Let D = Person has disease
Let T+ = Test positive
Given:
- P(D) = 0.005 (0.5% prevalence)
- P(D') = 0.995
- P(T+|D) = 0.98 (sensitivity)
- P(T+|D') = 0.05 (1 - specificity)
Using Total Probability:
P(T+) = P(D)P(T+|D) + P(D')P(T+|D')
= 0.005 × 0.98 + 0.995 × 0.05
= 0.0049 + 0.04975
= 0.05465
Answer: 0.0547 or 5.47% (approximately)
Solution:
We need P(D|T+)
Using Bayes' Theorem:
P(D|T+) = [P(D) × P(T+|D)] / P(T+)
= (0.005 × 0.98) / 0.05465
= 0.0049 / 0.05465
= 0.0897 (approximately)
Answer: 0.090 or 9.0% (approximately)
Critical Insight: Even with a highly accurate test (98% sensitivity, 95% specificity), only about 9% of people who test positive actually have the disease. This is due to the very low prevalence (0.5%) of the disease.
Solution:
We need P(D'|T-)
First find P(T-):
P(T-) = 1 - P(T+) = 1 - 0.05465 = 0.94535
P(T-|D') = 0.95 (specificity)
Using Bayes' Theorem:
P(D'|T-) = [P(D') × P(T-|D')] / P(T-)
= (0.995 × 0.95) / 0.94535
= 0.94525 / 0.94535
= 0.9999 (approximately)
Answer: 0.9999 or 99.99% (approximately)
Interpretation: A negative test is very reliable - about 99.99% of people who test negative truly don't have the disease.
Solution:
False positive means: Test positive BUT doesn't have disease = P(D'|T+)
P(D'|T+) = 1 - P(D|T+)
= 1 - 0.0897
= 0.9103
Alternative using Bayes:
P(D'|T+) = [P(D') × P(T+|D')] / P(T+)
= (0.995 × 0.05) / 0.05465
= 0.04975 / 0.05465
= 0.9103 ✓
Answer: 0.910 or 91.0% (approximately)
Critical Point: About 91% of positive tests are false positives! This demonstrates why confirmatory testing is essential for rare diseases.
Solution:
Number of false positives = Total screened × P(T+ ∩ D')
P(T+ ∩ D') = P(D') × P(T+|D')
= 0.995 × 0.05
= 0.04975
Expected false positives = 10,000 × 0.04975
= 497.5
Answer: Approximately 498 people
Complete Screening Picture for 10,000 people:
- Have disease: 50 people (0.5%)
- Test positive: 49 (98% of 50)
- Test negative: 1 (2% of 50) - false negative
- Don't have disease: 9,950 people
- Test positive: 498 (5% of 9,950) - false positives
- Test negative: 9,452 (95% of 9,950)