Statistical method:
|
All statistical analysis of this project is conducted in R language. First, descriptive statistics of basic information and outcome indicators were conducted for the subjects included in each center, and Breslow Day test was used to evaluate the difference between the two centers. Then, Corchran Mental Haenszel method was used to eliminate the impact of the difference between the two centers on the efficacy, reduce the heterogeneity of the central effect, so as to combine the data of the two centers. Then, the symptom efficacy, efficacy index, TCM symptom score, and health survey summary of the combined data were counted. Descriptive statistics were made on the data before the intervention of the two groups, and baseline analysis was conducted using independent sample t-test, independent sample Mann Whitney test, and chi square test. The data after intervention before intervention, follow-up after intervention, and follow-up before intervention were tested by independent sample parameters or non parameters. Specifically, for continuous data, including efficacy index, TCM symptom score, and health survey summary, if the data conforms to normal distribution, covariance analysis is used to include age, gender, and previous stage scores to statistically correct the results; If the data do not conform to the normal distribution, Mann Whitney test is used. Mann Whitney test was used for the main outcome indicators, symptoms and efficacy. All tests were bilateral tests, with a significant difference of 0.05.
|