Good! I appreciate that these take a little bit of thinking to get to the answer. Here’s the three steps:
- True negative = (1 – prevalence) * specificity * population = 0.99 * 0.9 * 10,000 = 8,910
- False negative = prevalence * (1 – sensitivity) * population = 0.01 * 0.1 * 10,000 = 10
- NPV = True negative / (True negative + False negative) = 8,910 / (8,910 + 10) = 0.9989 or 99.89%
Now that we have looked at both sides, we should be able to answer this question:
What happens to the positive predictive value (PPV) and negative predictive value (NPV) of a test when incidence rates double.