Through the present delta value check used in quality control programs is a powerful tool for detecting random errors in clinical chemistry analysis, it has some problems, such as missed true errors and delays in reporting time, because it also has the potential of showing erroneous positive results. Recently, new calculation methods for delta check with delta difference, delta percent change, rate difference, and rate percent change have been suggested by Lacher and Connelly (Clin Chem 34:1966-1970, 1988). Based on this new delta check method, we made the new criteria of which calculation method is applied to the clinical chemistry tests, i.e., the differential application of rate and delta check, and selectively applied the new method to 17 chemistry tests in order to solve the above problems. The applied criteria were the time dependence of the test item and the coefficient of variation of the absolute delta difference. Calcium, inorganic phosphorus, total protein, albumin, sodium, potassium, and chloride were classified as delta difference calculation method group; glucose and cholesterol as delta percent change group; creatinine, total and direct bilirubin as rate difference group; and urea nitrogen, uric acid, ALP, ALT, and AST as rate percent change group. With the previous criteria by Whitehurst et al. (Clin Chem 221:87-92) for 5045 specimens, the check-out rate was 47.8% (2,411 out of 5,045), and the positive predictive value was 0.41% (10 out of 2,411). For the new criteria, the check-out rate was 12.7% (621 out of 5,045), and the positive predictive value was 1.8% (nine out of 621).(ABSTRACT TRUNCATED AT 250 WORDS)