Rapid Antibody Test — HIV
Test Characteristics
| Metric | Value |
|---|---|
| False-negative rate | 0.5% (sensitivity 99.5%) |
| False-positive rate | 0.5% (specificity 99.5%) |
| Bayes factor | 100× |
| Base rate | 0.5% overall US adult prevalence |
Interpreting Results
| Scenario | Prior | + Result | − Result |
|---|---|---|---|
| Average-risk US adult | 0.5% | 100 × 0.5% = 50%0.5%× 10050% | 1/100 × 0.5% = 0.005%0.5%÷ 1000.005% |
| IV drug use or MSM, high-prevalence area | 5% | 100 × 5% ≥ 100%5%× 100100% | 1/100 × 5% = 0.05%5%÷ 1000.05% |
100 × 5% ≥ 100%5%× 100100%: exact posterior is 84%. The key teaching example: even with BF = 100×, a positive test in the general population gives only 50% probability — about half of positives are false. Confirmatory testing always required. Negative is extraordinarily reassuring, except during window period (<4 weeks).
- + result: risk goes from identical twin (0.5%) to coin flip (50%)
- − result: risk drops to well below fatal car crash (0.005%)
Sources:
- Meta-analysis of RDT operational characteristics, 2012–2020. medRxiv.
- aidsmap. Sensitivity and specificity of HIV tests.