Objective To research the quantization diagnostic method for primary insomnia （PI） of liver depression pattern （LDP）. Methods Totally 1 173 PI patients were randomized into the training sample group of 924 cases including 685 cases of LDP and 239 cases of non-LDP and the testing sample group of 249 cases including 194 cases of LDP and 55 cases of non-LDP. On the basis of epidemiological survey

the typing standards for LDP and common clinical symptoms were taken as candidate correlated factors. The correlated factors were selected with one-factor analysis and binary Logistic stepwise regression. The correlated factors were scored with conditional probability conversion method. The threshold value of quantization diagnosis was determined with maximum likelihood method. The retrospective and prospective tests were performed on the established a quantization diagnostic standard for LDP. The sensitivity

specificity

accuracy rating and Youden index were calculated. Results The related factors of quantitative criteria for PI of LDP were: difficulty falling asleep

depression

irritability

frequent sigh

chest tightness

chest pain

dry mouth and throat

bitter taste in mouth

string pulse

vomiting or belching

worries

shortness of breath and little interest. These 13 factors were scored 3 to 9 points respectively. The threshold value of quantization diagnosis for PI of LDP was 25 points. The sensitivity

specificity

accuracy rating and Youden index were 83.5%

86.6%

84.3% and 0.701 in the retrospective test and 84.5%

85.5%

84.7% and 0.700 in the prospective test. Conclusion It is feasible to establish a quantization diagnostic standard for PI of LDP based on symptom frequency.