Abstract: ObjectiveTo distinguish and eliminate the interference of testing delays (time factors) on coagulation parameters in rat carcinogenic test using statistical method.MethodsIn four groups of low, middle and high dose, and control, 168 female Wistar rats were tested for prothrombin time (PT), activated partial thromboplastic time (APTT) and thrombin time (TT) by magnetic beads method after finishing the carcinogenic test (104 weeks). After blood samples were collected and tested within 3-24 h, the data were analyzed using factor analysis method combining with partial correlation and multiple regression analysis. ResultsThe factor analysis showed that APTT was independently affected by time and dose, TT was influenced by time and interacted with the dose of exposure, and PT was also influenced by time. Partial correlation and multiple regression analyses showed that APTT and TT were separately correlated with time and the time was selected into the regression model of APTT and TT, respectively; PT and time had no correlation and can’t establish a regression model. Based on the statistical results it is judged that testing delay had affect APTT and TT, and the time threshold was 5 h. After removing the problematic data, mean values of APTT in four groups of low, middle and high dose, and control decreased by 1.85%, 4.56%, 20.08%, and 30.45%，and those of TT decreased by 4.01%, 6.78%, 11.76%, and 16.64%, respectively.ConclusionThe combined use of factor analysis, partial correlation analysis and multiple regression analysis can distinguish and eliminate the interference of testing delays on coagulation parameters APTT and TT in rat carcinogenic test.