Solution:
Let D = Person has disease
Let D' = Person does not have disease
Let T+ = Test result is positive
Given:
- P(D) = 0.02 (2% have disease)
- P(D') = 0.98 (98% don't have disease)
- P(T+|D) = 0.95 (sensitivity)
- P(T+|D') = 0.10 (1 - specificity = 1 - 0.90)
Using Theorem of Total Probability:
P(T+) = P(D) × P(T+|D) + P(D') × P(T+|D')
= 0.02 × 0.95 + 0.98 × 0.10
= 0.019 + 0.098
= 0.117
Answer: 0.117 or 11.7%
Solution:
Using the complement rule:
P(T-) = 1 - P(T+)
= 1 - 0.117
= 0.883
Alternative method:
P(T-) = P(D) × P(T-|D) + P(D') × P(T-|D')
= 0.02 × 0.05 + 0.98 × 0.90
= 0.001 + 0.882
= 0.883 ✓
Answer: 0.883 or 88.3%
Solution:
We need to find P(D|T+)
Using Bayes' Theorem:
P(D|T+) = [P(D) × P(T+|D)] / P(T+)
= (0.02 × 0.95) / 0.117
= 0.019 / 0.117
= 0.1624 (approximately)
Answer: 0.162 or 16.2% (approximately)
Interpretation: Even though the test is quite accurate, only about 16% of people who test positive actually have the disease. This is due to the low prevalence (2%) of the disease in the population.
Solution:
We need to find P(D'|T-)
Using Bayes' Theorem:
P(D'|T-) = [P(D') × P(T-|D')] / P(T-)
= (0.98 × 0.90) / 0.883
= 0.882 / 0.883
= 0.9989 (approximately)
Answer: 0.999 or 99.9% (approximately)
Interpretation: A negative test result is very reliable - about 99.9% of people who test negative truly don't have the disease.
Solution:
The test is correct when:
- Person has disease AND test is positive, OR
- Person doesn't have disease AND test is negative
P(Correct) = P(D ∩ T+) + P(D' ∩ T-)
= P(D) × P(T+|D) + P(D') × P(T-|D')
= 0.02 × 0.95 + 0.98 × 0.90
= 0.019 + 0.882
= 0.901
Answer: 0.901 or 90.1%
Interpretation: The test gives the correct result about 90% of the time overall.