16.1
%
83.9
%
99.95
%
95
%
950
people
4,950
people
94,050
people
50
people
5,900
people
94,100
people
5.21
16.1
%
83.9
%
99.95
%
95
%
950
people
4,950
people
94,050
people
50
people
5,900
people
94,100
people
5.21
The False Positive Paradox Calculator demonstrates a counterintuitive result of Bayes' theorem: even with a highly accurate test, most positive results can be false positives when the condition being tested for is rare. This phenomenon is crucial for understanding diagnostic testing, screening programs, and probabilistic reasoning.
For example, if a disease affects 1% of the population and a test is 95% sensitive and 95% specific, only about 16% of positive results will be true positives. The remaining 84% are false alarms. This paradox has profound implications for medical screening, criminal justice, and quality control.
Using Bayes' theorem, the Positive Predictive Value (PPV) — the probability that a person testing positive actually has the disease — is:
$$PPV = \frac{\text{Sensitivity} \times \text{Prevalence}}{\text{Sensitivity} \times \text{Prevalence} + (1 - \text{Specificity}) \times (1 - \text{Prevalence})}$$
The False Discovery Rate is the complement:
$$FDR = 1 - PPV = \frac{(1 - \text{Spec}) \times (1 - \text{Prev})}{\text{Sens} \times \text{Prev} + (1 - \text{Spec}) \times (1 - \text{Prev})}$$
The Negative Predictive Value — probability that a negative result is truly negative:
$$NPV = \frac{\text{Specificity} \times (1 - \text{Prevalence})}{\text{Specificity} \times (1 - \text{Prevalence}) + (1 - \text{Sensitivity}) \times \text{Prevalence}}$$
The Overall Accuracy is the proportion of all results (positive and negative) that are correct:
$$\text{Accuracy} = \text{Sens} \times \text{Prev} + \text{Spec} \times (1 - \text{Prev})$$
The paradox arises because when prevalence is low, the pool of healthy people is enormous relative to the pool of sick people. Even a small false positive rate applied to a huge healthy population generates more false positives than true positives from the small sick population.
A PPV of 0.16 means only 16% of positive test results are truly positive — 84% are false alarms. This does not mean the test is bad; it means the condition is rare.
The FDR directly shows the false alarm rate among positive results. High FDR is expected with low prevalence even when the test has high sensitivity and specificity.
NPV is typically very high when prevalence is low, meaning a negative result can be trusted confidently.
Accuracy can be misleadingly high with rare diseases because correctly identifying the vast majority of healthy people inflates the overall accuracy.
Inputs
Results
Despite 95% accuracy, only 16.1% of positive results are true positives! 83.9% are false alarms.
Inputs
Results
With 20% prevalence, PPV jumps to 69.2%. The false positive paradox is much less severe.
Because the number of healthy people vastly exceeds sick people. Even a 5% false positive rate applied to 990 healthy people (in a population of 1000 with 1% prevalence) produces ~50 false positives, while 95% sensitivity on 10 sick people produces only ~10 true positives. False positives outnumber true positives 5:1.
Three strategies: (1) Increase specificity — a more specific test produces fewer false positives. (2) Test only high-risk populations (higher effective prevalence). (3) Use confirmatory testing — a second positive test greatly increases PPV via serial Bayesian updating.
Exactly. Initial screening tests are designed to be highly sensitive (catch all cases) at the expense of specificity. Positive screens then undergo a confirmatory test with high specificity to weed out false positives. This two-stage approach dramatically improves final PPV.
Yes. When community prevalence was low, PCR tests (sensitivity ~98%, specificity ~99.5%) still produced notable false positive rates in mass screening. This is why testing asymptomatic populations requires careful interpretation and why confirmatory testing was recommended.
PPV increases with prevalence. At very low prevalence, PPV is low regardless of test accuracy. At 50% prevalence, PPV approximately equals sensitivity. This is why targeted testing (in symptomatic or high-risk groups) is more informative than universal screening.
Yes. With 1% prevalence, a test that always says 'negative' achieves 99% accuracy by correctly classifying all healthy people while missing every sick person. Accuracy is dominated by the majority class and should never be the sole measure of test quality.
Roboculator Team
The Roboculator Team explains calculations, planning tools, and practical formulas in clear language for real-life situations.
How helpful was this calculator?
Be the first to rate!
Lottery Odds Calculator
Probability Calculators
Probability Calculator
Probability Calculators
Conditional Probability Calculator
Probability Calculators
Monty Hall Problem Calculator
Probability Calculators
Combinations Calculator (nCr)
Probability Calculators
Permutations Calculator (nPr)
Probability Calculators
