Of course, the actual numbers vary depending on the testing population. This is because interpreting of the results of any medical test (assuming no test is 100% accurate) depends upon the initial degree of belief, or the prior probability that an individual has, or does not have a disease. Generally the prior probability is estimated using the prevalence of a disease within a population or at a given testing location. The positive predictive value and negative predictive value of all tests, including HIV tests, take into account the prior probability of having a disease along with the accuracy of the testing method to determine a new degree of belief that an individual has or does not have a disease (also known as the posterior probability ). The chance that a positive test accurately indicates an HIV infection increases as the prevalence or rate of HIV infection increases in the population. Conversely, the negative predictive value will decrease as the HIV prevalence rises. Thus a positive test in a high-risk population, such as people who frequently engage in unprotected anal intercourse with unknown partners, is more likely to correctly represent HIV infection than a positive test in a very low-risk population, such as unpaid blood donors.