Positive and negative diagnostic test results can be expressed as probabilities (of being correct). These are called the positive predictive value (PVP) and negative predictive value (NPV), respectively. To calculate PVP or NPV one must know the sensitivity and specificity of the test being used and know or estimate the prevalence of infection in the population of animals being tested. Specific formulas are given in the part of the website concerning serologic diagnosis of paratuberculosis in cattle.
