The inclusion criteria were discharged patients who took Chinese herbal medicine, and the exclusion criteria were patients who did not take Chinese herbal medicine or were diagnosed as dead.

By using the "data analysis" and "statistical report" of the Ancient and Modern Medical Records (AMMR) cloud platform, the medication rules of 340 patients' Chinese herbal medicine prescriptions were mined. According to the 2015 edition of Chinese Pharmacopoeia ^[6], the Chinese medicine names involved were unified, the above contents were input into Excel 2013 software one by one, and then checked by two persons.

The AMMR cloud platform was used to construct the database, and patient data were imported into the AMMR cloud platform in batch. The medical record data were standardized and added to the analysis pool. Data mining methods, such as frequency statistics and rule analysis, were used to analyze medical records, and the rules of core prescriptions and drug pair groups were found to facilitate summarizing the medication rule against COVID-19.

A total of 340 COVID-19 patients were discharged from the hospital, the youngest was 25 years old, and the oldest 90 years old; there are 100 cases of 60~69 years old, more than other age groups. See Table 1 and Figure 1. There are 143 males and 197 females, with the ratio of male to female of 0.73 ∶ 1, as shown in Figure 2.

The frequency statistics was conducted for the Chinese herbal medicine data of 340 discharged patients of COVID-19. Totally 204 herbs were administered for 5,423 times, and the frequency of drugs with high frequency (top 10 herbs) is arranged from high to low. Among them, the use frequency of Poria cocos was the highest, which was 225 times, as shown in Table 2 and Figure 3.