In the real world, there are many factors that can influence the
sequence of tests, such as the length of the delay and the probability
that the delayed test will be needed. When we ignore these many factors
and pay attention only to the simplified model presented here, it makes
sense to do all of the desired immediate tests before we do any of the
desired delayed tests. We do not know to what extent this actually
occurs in the real world. One complication is that medical doctors in
most industrialized countries are not directly affected by the cost of
the tests they select. In fact, fear of law suits gives them incentive
to order unnecessary tests.