Abstract: Coal and gas outburst prediction is the key link of two “four-in-one” comprehensive outburst prevention measures for outburst coal seam. It is of great significance to determine the sensitive index and critical value of outburst prediction.In the past, the determination of sensitive indexes and critical values of outburst prediction, including in-situ stress index and gas index, were determined through repeated field tests and experiments. However, the current outburst prevention and control management model does not support this method, so laboratory experiments can only be used mainly for the study of gas-related indicators, combined with part of the field verification to determine.In this paper,six coal samples were selected from Dalong Mine No.12 coal seam, Luling No.8 coal seam, Qinan No.3 coal seam, Zhuxianzhuang No.10 coal seam, Tongting No.7 coal seam and Xinjing No.3 Coal Seam. The coal types range from medium metamorphic gas and fat coal to high metamorphism. For anthracite coal, the relationship between gas pressure P and gas content W, gas pressure Pand gas desorption index K1 and Δh2, as well as drilling cuttings gas desorption index K1 and Δh2 are studied experimentally.The results show that: there is a single value correspondce between the gas indicators for outburst prediction, but the corresponding relationship of each coal sample is changing. The proposed critical values of outburst prediction indexes given by relevant national standards do not correspond to each other, nor can they fully reflect the actual outburst risk and outburst severity of coal seams. It is difficult to have a unified critical value for coal seams with different metamorphic degrees. The critical value of the sensitive index of outburst prediction can be indirectly determined by the corresponding relationship between the gas related indexes of coal sample and the reality of outburst coal seam, such as the initial outburst depth, outburst dynamic phenomenon and drilling dynamic phenomenon.At the same time, the future refined outburst prevention and control will have higher requirements for the sensitive indicators of outburst prediction. The further research direction should include indicators that reflect the ground stress status, coal seam occurrence under the influence of small structures, prediction of coal seam potential outburst strength, and new outburst prediction methods and measures adapted to the differences of outburst disasters.