The foregoing discussion emphasized the fact that pancreatography can document changes that are relatively specific for chronic pancreatitis but that similar changes can be seen in other clinical conditions and even as normal variants. In addition, the exact clinical implication of minor or equivocal changes is unclear and care should be taken not to overinterpret ERP findings. It also must be realized that ERP may miss a substantial number of patients with earlier or less advanced chronic pancreatitis. ERP also may document pancreas divisum, but is not helpful in explaining the patient's clinical condition in the absence of dorsal duct abnormalities. Finally, tests of pancreatic function--in particular, hormonal stimulation tests--are complementary to tests of pancreatic morphology and allow the diagnosis of less advanced or earlier chronic pancreatitis, as well as patients with divisum and normal dorsal ducts who nonetheless have obstruction to flow at the minor papilla. The evaluation of a patient with presumed chronic pancreatitis therefore should begin with simple, noninvasive tests that are able to detect advanced forms of chronic pancreatitis. These include plain abdominal radiograph and serum trypsin. If either of these is markedly abnormal, no further diagnostic testing is generally required. In patients in whom diagnostic uncertainty still exists, reasonable second-echelon tests include abdominal CT, bentiromide testing, or secretin stimulation testing. Of these, hormonal stimulation testing offers the most sensitivity but is not universally available. More invasive evaluations--in particular, ERP--should be reserved for patients in whom the diagnosis is still unclear or in whom therapeutic rather than diagnostic information is required (e.g., a patient classified a medical failure being considered for Peustow procedure).