One thousand patients admitted to a 575-bed general hospital during a 6-month period each underwent 20 chemical and hematologic tests. The potential clinical benefit was assessed. There were 2223 abnormal results found; 675 were predicted on clinical assessment, 1325 did not yield new diagnoses, and the remaining 223 led to 83 new diagnoses in 77 patients. On critical evaluation of the new diagnoses, none were unequivocally beneficial to the patient. Up to 30 patients might have benefited had these abnormal findings been followed up diligently, 39 others had findings or diagnoses of no lasting significance, and in 14 patients asymptomatic mild biochemical diabetes was discovered. Although screening may reveal many abnormal test results, the clinical benefits are not impressive.