Reassessment of preoperative laboratory testing has changed the test-ordering patterns of physicians

To test the hypothesis that physicians have substantially reduced the ordering of unwarranted preoperative tests, the authors reviewed 2,093 medical records of patients having four surgical procedures performed at three institutions in three cities in 1979, 1981, 1983, 1985 or 1987. Excluding hemoglobin measurements, the incidence of ordering preoperative laboratory tests unwarranted by findings on history or physical examination decreased from 32.2 to 25.9 percent during this decade, representing a 19.6 percent reduction. This decrease was irregular and varied from operation to operation, test to test and institution to institution. Overall, the percentage of preoperative tests ordered that were unwarranted decreased from 66.9 percent in 1979 to 60.1 percent in 1987. Extrapolating these results, the authors estimate that more than $320 million was saved annually by elimination of unwarranted tests and that the potential savings could exceed $1.35 billion a year. Unexpectedly, the preoperative ordering of medically indicated tests also decreased (from 92.9 to 80.9 percent, representing a 12.9 percent reduction). Because the benefit of performing justified tests is probably greater than the benefit of avoiding unwarranted tests, the net change has probably not been beneficial. A better system for obtaining justified tests and for eliminating the unwarranted tests may be necessary before a net benefit occurs. Punitive measures to reduce testing without prior establishment of such a system may save money, but impair health.