3.1. General characteristics of registered trials
80 registrations were retrieved, all of which met the inclusion criteria. By July 24, 2020, 27 (33.75%) registrations have reached the expected completion time. Among them, 25 (92.59%) are still under study, 1 (3.70%) registration has been published and 1 (3.70%) registration has been completed but not yet published. The number of daily registrations for COVID-19 related to traditional Chinese Medicine fluctuated significantly. The earliest registration date was February 5, 2020. The largest number of registrations (7, 8.75%) was on April 22, 2020. The registration time of included SRs is shown in Figure 1.
Among the 80 included SRs, 1 (1.25%) SR was performed by only one author, 56 (70.00%) SRs had 2-5 participants, 21 (26.25%) SRs had 6-10 participants, and 2 (2.50%) SRs had more than 10 participants. All 80 studies were systematic reviews, including meta-analysis (64, 80.00%), network meta-analysis (8, 10.00%), review of reviews (3, 3.75%), narrative synthesis (3, 3.75%), synthesis of qualified studies (1, 1.25%), individual patient data (IPD) meta-analysis (1, 1.25%). The types of studies included in these SRs are diverse. 3 (3.75%) SRs did not limit the types of included studies. RCTs (77, 96.25%) was the most involved type. The second to the fourth were observational studies (10, 12.50%), case-control studies (7, 8.75%), and quasi randomized studies (4, 5.00%).
Of the 80 SRs, 75 (93.75%) assessed the risk of bias of primary studies included and 5 (6.25%) did not. Cochrane risk of bias (73, 91.25%) was the most widely used bias risk assessment tool, followed by NOS (14, 17.50%), the modified downs and black instrument (2,2.50%), and the Hoy risk of bias tool (1, 1.25%). Data analysis software was reported in 59 (73.75%) SRs. Review manager software (49, 68.06%) was the most frequently used software, the remaining software included STATA (7, 9.72%), R (6, 8.33%), Excel (3, 4.17%), and SPSS (2, 2.78%). 42 (52.50%) SRs were funded and 38 (47.50%) were not. 14 (17.55%) were supported by the National Natural Science Foundation of China, Table 1.
3.2. Database
All SRs reported the retrieved databases. 77 (96.25%) SRs searched English and Chinese databases, and 3 (3.75%) only searched English databases. PubMed/Medline (78,97.5%) was the most frequently used English database and CNKI (77,96.25%) was the most frequently used Chinese database. Wanfang (61,76.25%), CBM (57,71.25%) and the Cochrane library (34,42.5%) were also commonly searched. As for the combination of databases, 75 (93.75%) SRs searched CNKI and PubMed/Medline, 61 (76.25%) SRs retrieved CNKI and Wanfang, and 75 (93.75%) SRs retrieved PubMed/Medline and EMBASE, more details are presented in Table 2.
3.3. Country and Province
Among the included SRs, 76 (95.00%) were completed by only one country, including 72 (94.73%) from China, 1 (1.31%) from Australia, 1 (1.31%) from Singapore, 1 (1.31%) from India and 1 (1.31%) from the UK. 4 (5.00%) SRs were completed by national cooperation, including one (25.00%) by China and Australia, two (50.00%) by China and Northern Ireland, and one (25.00%) by China and the United Kingdom.
21 provinces from China participated in the registration of 76 SRs. The top six productive provinces are Sichuan (25, 31.25%), Beijing (13, 16.25%), Shaanxi (7, 8.75%), Jilin (5, 6.25%), Jiangsu (4, 5.00%) and Guangdong (4, 5.00%). The 7th to 12th productive provinces are Henan (3, 3.75%), Tianjin (3, 3.75%), Jiangxi (3, 3.75%), Shandong (3, 3.75%), Hebei (2, 2.50%) and Gansu (2, 2.50%). Hainan, and Heilongjiang, Hong Kong, Chongqing, Hunan, Liaoning, Shanghai, Taiwan, and Zhejiang registered only 1 (1.25%) SR, Table 3. A social network analysis of provinces revealed that 7 provinces formed a cooperative relationship (Figure 2). However, the collaborations between provinces were sparse.
3.4. Institutions
A total of 81 institutions contributed to the registrations of COVID-19 SRs of TCM. These institutions are universities and hospitals. 38 (47.50%) SRs were completed by one institution, 26 (32.50%) by two organizations, 10 (12.50%) by three organizations, 5 (6.25%) by four organizations, and 1 (1.25%) by five organizations. The top six productive institutions are Chengdu University of traditional Chinese medicine (18, 24.66%), Shaanxi University of Traditional Chinese medicine (7, 9.59%), Beijing University of Traditional Chinese medicine (6, 8.22%), Chengdu University of traditional Chinese medicine hospital (5, 6.85%), Guang'anmen Hospital (5, 6.85%) and Changchun University of Traditional Chinese medicine (5, 6.85%), Table 4.
A social network analysis of institutions revealed that 10 institutions formed a cooperative relationship (Figure 3). Beijing Traditional Chinese medicine, located in the network center, had more cooperation with other institutions. There was less cooperation between institutions.
3.5. Interventions
78 (97.50%) registrations reported interventions and 2 (2.50%) did not. The reported interventions were classified into TCM (46, 58.97%), acupuncture (8, 10.25%), integrated Chinese and Western medicine (26, 33.33%), and Taijiquan (2, 2.56%). 17 registrations reported the specific names of traditional Chinese medicines used. They were Lianhua Qingwen capsule (6, 35.29%), Shufeng Jiedu capsule (3, 17.65%), Jinhua qinggan granule (2, 11.76%), Reduning injection (2, 11.76%), Xuebijing injection (2, 11.76%), ginseng (1, 5.88%), and Qingfei Paidu decoction (1, 5.88%). The administration methods included oral administration (14, 82.35%) and injection (3, 17.64%). The control measure was placebo or conventional medicine or no treatment (80, 100%).
3.6. Outcome measures
3.6.1. Primary outcome measures
All the included SRs used multiple outcome indicators. The primary outcome measures used in each study were related to clinical symptoms, signs, and laboratory tests. The most frequently used primary outcome measure was clinical efficiency (45, 56.25%), followed by the time of clinical improvement time (31, 38.75%), clinical symptoms improvement (31, 38.75%), antipyretic time (29, 36.25%), lung function (26, 32.50%), body temperature (22, 27.50%), mortality (21, 26.25%), rate of viral nucleic acid turning negative (14, 17.50%), time of viral nucleic acid turning negative (12, 15.00%), and incidence of adverse events (12, 15.00%), Table 5.
3.6.2. Secondary outcome measures
Of the 80 included SRs, 65 (81.25%) reported secondary outcome measures. The most frequently used secondary outcome measure was adverse events (50, 62.5%), followed by hospital stay (15, 18.75%), clinical symptoms improvement (15, 18.75%), clinical recovery time (14, 17.50%), fever disappearance time (13, 16.25%), safety measurements (13, 16.25%), TNF-a/IL-6 (13, 16.25%), WBC (11, 13.75%), chest/lung CT (10, 12.50%), and rate of progression to severe (9, 11.25%), Table 6.