ABSTRACT
Objective:To conduct periodic revalidation of the 15 items and 43 terms autoverification rules of blood analysis after 1 year of application, analyze the application suitability and make the rules improved.Methods:Track the results of 528 010 blood analysis samples of our hospital from August 1, 2019 to January 31, 2020, and analyze the pass rate and interception rate of autoverification; 600 specimens in total were selected randomly for microscope examination, including 300 specimens which touched autoverification rules (1 012 items of autoverification rules) and were intercepted by autoverification and 300 specimens which untouched autoverification rules and were released by autoverification. The abnormal characteristics and unacceptable Delta check of the specimens also need to be concerned at the same time.The false negative rate and false positive rate, true negative rate, true positive rate and pass correct rate of autoverification were verified and compared with the rate of the second phase verification when the autoverification rule was established. The false negative rate, false positive rate, true negative rate and true positive rate of the Delta check rule which 54 716 specimens touched were calculated and compared with the second phase verification rate when the autoverification rule was established.The results of microscopic examination were used as the gold standard for the calculation of the rates, and P<0.05 was considered as a significant difference. The false positive and true positive of 1 012 autoverification rules were analyzed item by item.The false positive and true positive of 108 specimens which touched blast cell autoverification rule were analyzed terms by terms. The mean TAT and median TAT of 528 010 specimens and 193 750 outpatient specimens were calculated respectively, and the report percentages of 528 010 samples that TAT<30, 30-60 and>60 min were calculated respectively. Analyze and evaluate the application suitability of autoverification rules to juge whether they meet the needs of doctors and laboratory. The design process and the rules and application process of autoverification were optimized and improved.Results:The autoverification pass rate was 63.06% (332 971/528 010), the interception rate was 36.94% (195 039/528 010). The false negative rate was 1.00% (1/600), the false positive rate was 12.67% (76/600), the true negative rate was 49% (294/600), the true positive rate was 37.33% (224/600), and the correct rate was 98% (294/300). The pass rate, true negative rate, true positive rate and correct rate of the periodic reverification group were higher than the second phase verification group, the false negative rate and false positive rate were lower than that the second phase verification group. The false negative rate and true positive rate of the Delta check of periodic verification group were lower than that the second phase verification group, the false positive rate and true negative rate were higher than the second phase verification group, there were significant differences in the comparition results. The mean TAT of 528 010 specimens was25 min, and the median TAT was 22 min. The mean TAT of 193 750 outpatient specimens was 23 min, and the median TAT was 20 min. The report percentages of 528 010 samples that TAT<30 min, 30 min-60 min and>60 min were 83.30% (439 819/528 010), 8.00% (42 250/528 010) and 8.70% (45 941/528 010), respectively.Conclusion:The results of periodic revalidation of autoverification after 1 years application show that the 15 items and 43 terms autoverification rules of blood analysis could meet requirements about the accuracy and efficiency of the laboratory, and have a good suitability for application.