Objective: To validation and optimization the indicator system of risk assessment for mechanical cuts. Methods: The risk assessment index system of mechanical cutting injury established earlier was used to assess the risk of mechanical cutting injury in 40 cases of mechanical cutting injury registered from January 2015 to December 2017 and 40 similar positions without accidents in the same period. The multiple stepwise regression analysis was used to screen the indicator system, and to adjust the weight coefficient of each index. The total coincidence rate and Kappa value were compared between before and after optimization respectively. Results: The new index system has 3 first-class indicators, 10 second-class indicators and 14 three-class indicators, fewer than the old index system which has 3 first-class indicators, 10 second-class indicators, 34 three-class indicators. There three indicators have revamped in the first-class. The total of coincidence rates of the new and old indicator systems were 67.50% and 90.00%, the difference was statistically significant (P<0.01). The Kappa value were 0.35 and 0.80, respectively. Conclusion The evaluation results with new indicator systems is more consistent with the actual hazard detection the the old indicator systems, and scientific, reasonable and practical, and the indicator system of risk assessment for mechanical cuts can be used for the risk assessment of mechanical cutting injuries.

OBJECTIVE:To establish HPLC fingerprints of Miao medicine Ardisia crenata.METHODS:HPLC method was adopted.The determination was performed on Diamonsil C18 column with mobile phase consiste of methanol-water (gradient elution) at the flow rate of 1.0 mL/min.The detection wavelength was 220 nm,and column temperature was maintained at 30 ℃.The sample size was 10 μL.Using 11-O-(3',4',5'-three-o-galloylhyperin)-bergeninum as reference,HPLC fingerprints of 16 batches of samples were determined.Common identification and similarity evaluation were performed by using TCM Chromatographic Fingerprint Similarity Evaluation Software (2012 edition).Cluster analysis of fmgerprrints was conducted.RESULTS:There were 6 common peaks in HPLC fingerprints of 16 batches of samples.The similarity among 8 batches was more than 0.9.The 16 batches of samples could be clustered into 4 categories.CONCLUSIONS:Established fingerprints can provide reference for identification and quality evaluation ofA.crenata.