2.7. Statistics

SPSS statistical software v17.0 (IBM, USA) was used for the statistical analysis. Counting data and grade data are described with the composition ratio or frequency. Counting data were tested using a χ² test; rank data were analysed using a rank sum test. After a normality test, data that conformed to a normal distribution (approximate normal distribution) or symmetrical distribution were expressed by mean ± standard deviation (x ± s) and by median or quartile otherwise. ANOVA was used to detect the homogeneity of variance, t-test was used to compare the two groups, and Pearson's correlation test was used to analyse the correlations between the two groups. Analysis of covariance (ANCOVA) was performed on the improvement of traditional Chinese medicine syndrome scores, with HDRS scale scores as a covariate. P < 0.05 was considered as a significant difference; P < 0.001 was considered as an extremely significant difference.