Efficacy and Safety of “Three Chinese Patent Medicines and Three TCM Prescriptions” for COVID-19: A Systematic Review and Network Meta-Analysis

Objective To systematically evaluate the efficacy, safety, and precision of TMTP for COVID-19. Methods Randomized controlled trials and retrospective studies were searched in 11 electronic databases. This network meta-analysis included trials using TMTP to treat patients with COVID-19. The traditional pairwise meta-analysis was done by using Stata 15, and Bayesian network meta-analysis was done with WinBUGS. Results 18 trials were included with 2036 participants and 7 drugs. The results showed that LHQW had the most significant effects on improving expectoration, shortness of breath, sore throat, nausea, emesis, inappetence, muscle soreness, and headache, and it could produce the least adverse reactions. XBJ was the best drug for fever, fatigue, and diarrhea, which showed great advantages in lowering WBC levels. XFBD was the most effective drug for cough and chest distress, which had the least exacerbation rate. JHQG was the most effective for rhinobyon and rhinorrhea, while QFPD was the best drug in decreasing CRP levels. Conclusion This study was the first most large-scale and comprehensive research of TMTP for COVID-19. The results showed that LHQW had good efficacy without obvious adverse reactions. Therefore, we believe that it should be firstly recommended for COVID-19 treatment. In addition, XBJ is recommended for patients with a severe fever, fatigue, and diarrhea, and JHQG is recommended for patients with obvious rhinobyon and rhinorrhea; then, XFBD is recommended for patients with cough and chest tightness as the main manifestation. Our findings will help experts develop new COVID-19 treatment guidelines to better guide clinical medication for protecting the health of COVID-19 patients.

COVID-19, no therapeutic approach has been effective in its treatment up to now [3]. Moreover, during the sudden outbreak of COVID-19, there is a lack of effective chemical drugs (CDs) for prevention and control in clinical practice, and the interim screening and development of new drugs may take a long time, making it difficult to apply them in a timely way for clinical treatment. erefore, the unique advantages of traditional Chinese medicine (TCM) multitarget interventions had become an indispensable systemic approach for patients, and significant progress has been made in China's battle against COVID-19 in China [4].
TCM has been proven to be effective in treating patients with influenza and has very successful experience in the prevention and treatment of infectious diseases. ree Chinese patent medicines and three TCM prescriptions (TMTP) were the key recommended drugs in the Novel Coronavirus Infection for the Diagnosis and Treatment of Pneumonia published by the National Health Commission of the People's Republic of China. And the China Administration of Traditional Chinese Medicine pointed out that TMTP had a great clinical advantage in the prevention and control of COVID-19. In addition, three Chinese patent medicines included Lian-Hua-Qing-Wen (LHQW), Jin-Hua-Qing-Gan (JHQG), and Xue-Bi-Jing (XBJ), and three TCM prescriptions included Qing-Fei-Pai-Du (QFPD), Xuan-Fei-Bai-Du (XFBD), and Hua-Shi-Bai-Du (HSBD), which had played an important role in the fight against the epidemic by treating the disease based on syndrome differentiation according to the patient's individuality and seasonal and local conditions. Although TMTP was widely used in clinical practice, there were inconsistent research conclusions due to the different sample sizes, outcome indicators, and unified evaluation criteria in clinical reports, and accurate conclusions cannot be drawn, which made it difficult to grasp its therapeutic effect and clinical advantages. erefore, we conducted a comprehensive systematic review and meta-analysis to evaluate TMTP efficacy and safety of COVID-19 patients from a variety of clinical symptoms, in order to serve for COVID-19 prevention and control better. It is a challenging task to comprehensively analyze the present clinical evidence for the efficacy and safety of TMTP for COVID-19 by using traditional meta-analysis methods because there is a lack of head-to-head trials that directly compare certain treatments among existing trials. Bayesian network meta-analysis, also known as a mixed treatment comparison method, is a valuable tool in comparative effectiveness research, which enables the comparison of multiple interventions to incorporate clinical evidence from both direct and indirect treatment comparisons in a network of treatments and associated trials [5], and allows indirect comparison without head-to-head trials to simultaneously compare several treatments by using a common comparator and combined direct and indirect comparisons while retaining randomness in individual trials [6][7][8].
is study aimed to systematically evaluate the efficacy and safety of TMTP for COVID-19 using Bayesian network meta-analysis, and the competing drugs in each outcome were ranked to obtain the most effective one. e study focused on the diversity of symptoms in patients with COVID-19, which was the most large-scale and comprehensive research on this issue so far. We hope that our study will provide up-to-date information on the treatment of COVID-19.

Data and Methods
is systematic review and Bayesian network meta-analysis was conducted and reported following the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) with Cochrane methodology [9]. is study has been registered and the PROSPERO number is CRD42021240869.

Type of Research.
is network meta-analysis included clinical trials using TMTP to treat COVID-19. Trials were excluded if (a) no control group was used; (b) TMTP was not used in the experimental group; (c) there is a combination with other drugs; (d) trials on effective analysis data cannot be obtained; (e) they are reviews, conference paper, case reports, experience sharing, animal trials, etc.; (f ) they are repeatedly published articles and plagiarized trials.

Preventative
Measures. e intervention measures in the experimental group should be TMTP single drug combined with CD, and the control group should be CD.

Results.
e primary outcomes were defined as clinical effect and CT recovery rate. Clinical effect was evaluated based on the improvement of clinical symptoms of patients before and after treatment, which could better reflect the therapeutic effect of drugs. Symptoms were classified as significant, effective, or ineffective according to the degree of relief. Moreover, CT played an important role in early screening and disease surveillance for COVID-19, and its changes could significantly reflect the therapeutic effect. e secondary outcomes included nucleic acid negative rate, disappearance rate of primary symptoms (fever, cough, fatigue), disappearance rate of respiratory (expectoration, shortness of breath, chest distress, rhinobyon, rhinorrhea, sore throat), gastrointestinal (nausea, diarrhea, emesis, inappetence), and other symptoms (muscular soreness, headache), inflammatory biomarkers such as C-reactive protein (CRP) and white blood cell (WBC), as well as exacerbation rate and adverse reactions.

Study Selection and Data Extraction.
Before starting the screening processes, two researchers who participated in training and calibration exercises independently screened the titles and abstracts of potentially eligible trials that were in duplicate; then, they independently retrieved and reviewed the full text of the possible trials in duplicate based on the inclusion and exclusion criteria and compared their results. If there was disagreement, they agreed through discussion or submitted it to a third party for evaluation. And before the screening process, the third party used a standardized screening form and performed calibration exercises. e screening process was conducted in Endnote X9.
Before the data extraction process began, we conducted various forms of calibration exercises and pilots. en, according to the inclusion and exclusion criteria mentioned above, the two researchers used standardized tables to independently extract data in duplicate from all eligible trials. In case of disagreement, they agreed through discussion or submitted it to a third party. All the eligible trials were published in English.
For all eligible trials, the researchers extracted data on the following characteristics: (1) e basic information of the study (author's name, title of the study, year of publication, country/region, and publication status) e methodological quality of each included study was independently assessed by two reviewers based on 2 tools. e Cochrane collaboration tool has been used to assess the quality of randomized controlled trials, and it comprised the following 7 aspects: random sequence generation, allocation concealment, blind method, incomplete result data, selective reporting, and other biases. e quality assessment results of each item can be divided into three grades: "low risk," "high risk," and "unclear." e risk coefficient is lower because of the more rigorous design and higher methodological quality of each RCT. Newcastle Ottawa Scale (NOS) has been used to assess the quality of retrospective studies. is method includes 3 aspects of evaluation: the selection method, comparability, and contact exposure assessment method of case and control groups. e higher the score is, the greater the quality of the learning is. When necessary, the consensus on this issue was studied with the help of a third party.

Statistical Analysis.
e traditional pairwise metaanalysis was done by using Stata 15 (StataCorp, College Station, TX, USA), and Bayesian network meta-analysis was done with WinBUGS version 14.3 (MRC Biostatistics Unit, Cambridge, UK). Both the continuous and dichotomous outcomes were derived from the included trials without any conversion. e dichotomous outcomes were described by relative risk (RR) and 95% confidence interval (CI); in addition, mean difference (MD) and 95% CI were used to describe the effect value of the intergroup comparison. Heterogeneity was determined according to the results of the I 2 test. I 2 < 50% indicated the low heterogeneity of interstudy, and the fixed effect model was adopted. Furthermore, the random effect model was adopted when I 2 > 50% [11], which was also used to generate direct and mixed treatment comparison estimates. Direct estimates of any two interventions were obtained by pooling data from clinical trials that compared the same interventions face to face. And the mixed treatment comparison estimates of interventions were obtained by combining direct clinical trial data for comparative interventions with the indirect estimates between the interventions through a common comparator. Normal prior distributions, noninformative uniform, and 3 different sets of starting values were used to fit the model. At the same time, 4 chains, 2.5 initial values Evidence-Based Complementary and Alternative Medicine scaling, 20000 tuning iterations, 50000 simulation iterations, and 10 thinning intervals were used to obtain the posterior distributions of model parameters. Subgroup analysis was used to evaluate the therapeutic effects of different drugs. Inverted funnel plots and Egger's regression test were used to determine publication bias when the number of included studies exceeded 10 in the network meta-analysis [12].

Risk of Bias Assessment of the Literature Included in the Study.
e methodological quality of 10 randomized controlled trials was summarized in Table A.3, and the criteria in the Cochrane Handbook for Systematic Reviews of Interventions were used to assess the risk of bias in the study. Although randomization was announced in all 10 trials, 5 trials used random number table [15,18,21,23,27], 1 used the R 3.6.2 software [30], 1 used the method of flip a coin [24], and 3 did not report the method [13,19,29]. Moreover, only 1 trial reported allocation concealment [21], and all the trials did not report the blind method. e quality of 8 retrospective studies was assessed by NOS. Table A.4 summarized the NOS scores of each study, all of which were of fair quality. Since fewer than 10 trials were included in each subgroup, publication bias could not be adequately analyzed.

Clinical Effect.
Clinical effect was reported in 6 trials, in which 2 used LHQW vs. CD [18,27], 2 used XBJ vs. CD [13,29], and 2 used QFPD vs. CD [19,20]. Figure 3 provided the forest plots for the network meta-analysis of each relevant drug. e meta-analysis showed that using TMTP to treat COVID-19 could significantly increase the clinical effect (RR, 1.20; 95% CI, 1.10 to 1.31). Compared with CD, LHQW significantly improved the clinical efficacy, which was 1.22 times higher than that of CD (RR, 1.22; 95% CI, 1.10 to 1.35). However, there is no remarkable difference between XBJ and CD (RR, 1.37; 95% CI, 0.94 to 1.99) or QFPD and CD (RR, 1.10; 95% CI, 0.91 to 1.32) on clinical effect. It should be noted that the mixed treatment comparison and drug sequencing were not performed because there was no closed ring in the network plot.

CT Recovery Rate.
CT results are of great significance in the diagnosis of COVID-19, which is mainly characterized by ground-glass opacity. e improvement of COVID-19 was determined according to the CT changes before and after treatment in the clinic. A total of 7 trials reported the CT recovery of patients after treatment, in which 4 used LHQW vs. CD [14,18,25,27], 2 used XBJ vs. CD [17,29], and 1 used QFPD vs. CD [28]. Figure 4 provided the forest plots for the network meta-analysis of each relevant drug. e results exhibited that the CT recovery rate of TMTP was increased significantly more than that of CD alone (RR, 1.24; 95% CI, 1.15 to 1.35). e rates of LHQW, XBJ, and QFPD were 1.22 (RR, 1.22; 95% CI, 1.10 to 1.36), 1.36 (RR, 1.36; 95% CI, 1.06 to 1.75), and 1.26 (RR, 1.26; 95% CI, 1.10 to 1.43) times higher than those of CD. However, because there was no closed ring in the network plot, the mixed treatment comparison and drug sequencing were not performed.

Nucleic Acid Negative
Rate. 2 trials reported the nucleic acid of patients before and after treatment [23,29], all of which used XBJ vs. CD. Figure A.2 showed the forest plots. However, compared with CD, XBJ did not show an obvious increasing efficacy on it (RR, 0.97; 95% CI, 0.72 to 1.29).
Compared with CD, LHQW showed a significant effect on increasing the disappearance rate of primary symptoms, which were 1.42 (RR, 1.42; 95% CI, 1.22 to 1.65), 1.97 (RR, 1.97; 95% CI, 1.45 to 2.68), and 1.52 (RR, 1.52; 95% CI, 1.15 to 2.03) times higher than those of CD in fever, cough, and fatigue respectively. JHQG and XBJ could remarkably improve the fever symptoms of patients, which were 1.51 (RR, 1.51; 95% CI, 1.07 to 2.41) and 15.71 (RR, 15.71; 95% CI, 1.03 to 240.75) times higher than those of CD, but they did not show an obvious advantage in cough and fatigue. A significant improvement of cough was observed by XFBD, whose efficacy was 1.97 times higher than that of CD (RR, 1.97; 95% CI, 1.04 to 3.72). However, there was no remarkable difference in fever and fatigue.
e results of network meta-analysis showed that XBJ was the most effective drug for improving fever, followed by XFBD, LHQW, JHQG, and CD. XFBD was the best drug for cough, followed by LHWE, JHQG, CD, and XBJ. In addition, XBJ was best for fatigue, followed by XFBD, LHQW, JHQG, and CD, while LHQW and JHQG had similar effects on fatigue.
Using LHQW to treat COVID-19 could obviously increase the disappearance rate of expectoration, shortness of  between LHQW and CD on rhinobyon, rhinorrhea, and sore throat. JHQG could also improve expectoration, and its efficacy was 1.85 times higher than that of CD (RR, 1.85; 95% CI, 1.01 to 3.38), but it did not show good effects on rhinobyon, rhinorrhea, and sore throat. According to the results, there was no obvious difference between XFBD and CD in shortness of breath and chest distress, as well as QFBD and CD in sore throat. e network meta-analysis exhibited that LHQW and JHQG were the best drugs for expectoration, followed by CD and XBJ. And LHQW was the most effective drug for shortness of breath, followed by XFBD and CD. XFBD was the best one in improving chest distress, followed by LHQW and CD. JHQG was the best drug for rhinobyon and rhinorrhea, in which the former was secondary to LHQW and CD, and the latter was secondary to CD and LHQW.   Evidence-Based Complementary and Alternative Medicine Additionally, LHQW was the most effective for sore throat, followed by JHQG, CD, and XFBD.
e results of the network meta-analysis identified that LHQW was the most effective in nausea, followed by JHQG, CD, and XFBD. XBJ was the best drug in improving diarrhea, followed by LHQW, CD, XFBD, and JHQG. For emesis, the drug sorts were LHQW, JHQG, CD, and XFBD. And LHQW was also the best drug for inappetence, followed by XFBD and CD.

Disappearance
Rate of Other Symptoms. 5 trials reported muscle soreness before and after treatment [14-16, 22, 26], and 3 reported headache [22,24,26]. Figure A.5 provided the forest plots for the network metaanalysis of each relevant drug. Compared with CD, TMTP exhibited significant effects on increasing the disappearance rate of muscle soreness (RR, 1.51; 95% CI, 1.03 to 2.20), but there was no obvious difference between TMTP and CD on headache. 4 trials comparing LHQW with CD reported muscle soreness before and after the intervention.
e meta-analysis revealed that LHQW significantly improved its efficacy, which was 1.88 times higher than that of CD (RR, 1.88; 95% CI, 1.05 to 3.36); however, it did not show a remarkable advantage on headache. Either JHQG or XFBD was not more effective than CD in the treatment of muscle soreness and headache.
According to the results of network meta-analysis, LHQW was the best drug for improving muscle soreness and headache, followed by JHQG and CD for muscle soreness, and XFBD and CD for headache.

Inflammatory Biomarkers (CRP and WBC). 4 trials
reported CRP level before and after treatment [13,17,23,27] and 3 reported WBC [17,20,23]. Figure A.6 provided the forest plots for the network meta-analysis of each relevant drug. Meta-analysis showed that using TMTP could significantly decrease CRP level (MD, −0.94; 95% CI, −1.79 to −0.09); however, there was no obvious difference between TMTP and CD in WBC. e reduction in CRP was remarkably greater for XFBD than that of CD (MD, −0.46; 95% CI, −0.69 to −0.23). But XBJ had no obvious advantages in decreasing the levels of CRP and WBC, and there was no significant difference between QFPD and CD in WBC. e network meta-analysis results identified that QFPD was the best drug for decreasing CRP levels, followed by CD and XBJ, and XBJ was the best for WBC, followed by CD and QFPD.
e results of network meta-analysis showed that using LHQW to treat COVID-19 could produce the least adverse reaction, followed by QFPD, CD, XBJ, and JHQG.

Discussion
e efficacy and safety of TMTP for COVID-19 were evaluated by Bayesian network meta-analysis. 18 trials that contained 2036 participants were included. e traditional meta-analysis exhibited that using LHQW to treat COVID-19 could significantly increase the efficacy, and its clinical and CT effect was 1.22 times higher than that of CD, which could also improve most of the symptoms. Its effects on fever, cough, fatigue, expectoration, shortness of breath, chest distress, and muscle soreness were 1.42, 1.97, 1.52, 2.58, 2.79, 2.15, and 1.88 times those of CD, respectively, and the exacerbation rate was 0.52 times that of CD. e CT effects of XBJ and QFPD were 1.36 and 1.26 times those of CD, and the adverse reaction rate of QFPD was 0.72 times that of CD. e effects of JHQG on the improvement of fever and expectoration were 1.51 and 1.85 times those of CD. e cough effect and exacerbation rate of XFBD were 1.97 and 0.29 times those of CD, and it could also lower than the CRP level. e network meta-analysis identified that LHQW was the most effective drug in improving expectoration, shortness of breath, sore throat, nausea, emesis, inappetence, muscle soreness, and headache of COVID-19 patients, and it could produce the least adverse reactions. XBJ was the most effective in improving fever, fatigue, and diarrhea, and it showed great advantages in lowering WBC levels. XFBD was the best drug in improving cough and chest distress, and it had the least exacerbation rate. JHQG was the best one in improving rhinobyon and rhinorrhea, and QFPD was the most effective drug in decreasing CRP levels.
Some clinical studies have proved the efficacy and safety of TCM in the treatment of COVID-19. It was shown that TCM treatment of COVID-19 could significantly reduce the mortality of patients and delay the progression of the disease, especially in the treatment of severe/critical cases, which showed good advantages [31]. Furthermore, it could significantly improve clinical remission rates and shorten the nucleic acid conversion time and hospitalization time. e combination of HSBD and TCM injection showed obvious superiority in the treatment of COVID-19 [32]. Meanwhile, HSBD alone could also remarkably shorten the fever time of COVID-19 patients, relieve symptoms such as cough,   fatigue, and chest discomfort, and improve the CT recovery rate [33]. Based on the two indicators of exacerbation rate and adverse reactions, this study showed that TCM could effectively prevent the deterioration and progression of the disease and had good safety. It could not only obviously relieve the multisystem clinical symptoms of COVID-19 patients and reduce clinical indicators but also significantly improve the clinical efficacy, which is a powerful and effective measure for the treatment of COVID-19.
Each prescription in TMTP is a combination of classical and famous formulae, which can play a role in the treatment of COVID-19 through multitarget comprehensive intervention. ey can be used clinically in combination with the actual situation of patients and are suitable for the treatment of mild, common, and severe COVID-19 patients.
LHQW is a combination of Maxing Shigan Decoction and Yinqiao Powder, which is composed of 12 kinds of herbs including Lianqiao (fructus of Mahuang disperses lung qi and dissipates phlegm; Shigao clears heat; Xingren improves cough and asthma; Banlangen cools blood to relieve sore throat; Guanzhong and Yuxingxao remove toxicity for eliminating carbuncles; Bohe dispels wind pathogens and relieves sore throat; Huoxiang removes dampness for regulating stomach; Dahuang eliminates heat; Hongjingtian moistens lung for arresting cough; Gancao clears heat-toxicity. us, LHQW has the effects of dispelling disease and detoxification, as well as relieving heat in the lung. Studies have shown that LHQW could significantly inhibit the replication of SARS-CoV-2 in Vero E6 cells at the mRNA level and greatly reduce the production of proinflammatory cytokines TNF-α, IL-6, CCL-2/MCP-1, and CXCL-10/IP-10, thus playing a role in the resistance to the virus; furthermore, it has a broad-spectrum effect on a series of influenza viruses by inhibiting virus proliferation and regulating immune function [34], which can not only enhance the body's immunity and inhibit respiratory inflammation [35] but also affect the relevant cytokines and ameliorate lung injury associated with inflammatory cell infiltration [36]. erefore, LHQW has antibacterial, antipyretic, analgesic, anti-inflammatory, cough relieving, and   Mahuang disperses lung qi and dissipates phlegm; Shigao clears heat; Xingren improves cough and asthma; Huangqin dispels heat and removes dampness; Zhebeimu reduces phlegm; Zhimu nourishes yin and clears heat; Niubangzi dispels heat pathogens and relieves throat disorder; Qinghao and Gancao clear heat-toxicity; Bohe dispels wind pathogens and relieves sore throat. JHQG could inhibit the replication of the influenza virus, promote virus clearance [38], alleviate rhinobyon, rhinorrhea, and other symptoms through a variety of mechanisms, and shorten the fever time by PTGS2 possibly [39]. XBJ is developed on the basis of Xuefu Zhuyu Decoction, including Honghua (floral bud of Carthamus tinctorius L.), Chishao (radix of Paeonia lactiflora Pall.), Chuanxiong (rhizome of Ligusticum chuanxiong Hort.), Danshen (radix and rhizome of Salvia miltiorrhiza Bge.), and Danggui (radix of Angelica sinensis (Oliv.) Diels), which is used for the mutual syndrome of stasis and poison in warm and hot diseases. XBJ has the function of activating blood circulation and removing blood stasis, which can antagonize endotoxin in vitro. It can not only improve a variety of infectious symptoms such as fever and fatigue by anti-inflammatory, antiendotoxin, and the improvement of blood coagulation function and alleviate digestive symptoms such as abdominal pain and diarrhea by reducing the inflammatory response of the body but also shorten the recovery time of white blood cells [40]. Furthermore, it could effectively reduce acute lung injury by regulating the expression of pulmonary inflammatory factors p-p38 MAPK, NF-κB 65, HIF-1α, p-IκB-α, and TGF-β1 [41] and improve dyspnea and hypoxemia in patients with severe COVID-19.   ). QFPD has effects of clearing heat toxicity, dispersing lung qi, and dissipating phlegm. It could inhibit proinflammatory cytokines such as IL-6 and IL-1β, increase anti-inflammatory cytokines such as IL-4 and IL-10, or inhibit NF-κB and MAPK signaling pathways, exert an anti-inflammatory and antiviral role, and significantly reduce inflammatory indicators [42,43].
XFBD ). XFBD has the effects of clearing heat toxicity, dispersing lung qi, and dissipating phlegm. It may block inflammatory cytokine storm by inhibiting excessive cytokine production, immune cell activation, and oxidative damage in vivo, which is the intervention mechanism after virus invasion [44]. erefore, XFBD can significantly relieve cough, chest tightness, and other symptoms of patients and inhibit the progression of the disease. Fisch.). HSBD has the effects of dispelling lung heat, preventing asthma, drying damp, strengthening the spleen, and removing blood stasis. It could inhibit the infection, replication, and proliferation of the SARS-CoV-2 virus to some extent, regulate the balance of the RAS system and inflammatory response accordingly, and effectively block the formation of the inflammatory storm after the infection of the SARS-CoV-2 virus [45,46]. ere were some important advantages in this study. Methodologically, our study benefits from rigorous methods, extensive search, repeated and independent screening, meticulous data abstraction process, and comprehensiveness of analytical indicators. In addition, the Bayesian network meta-analysis was used to compare therapies indirectly when no head-to-head trial existed, and more accurate evaluation for efficacy was obtained by jointly assessing direct and indirect comparisons. Moreover, we mapped drug sequencing figures through single and mixed analysis, further ranked the competing drugs, and summed up the best one for that outcome. e following limitations should be considered in this study. Due to the insufficient sample size, short duration, and partial retrospective study of TMTP in the treatment of COVID-19, the methodological quality had a certain risk bias. In addition, TMTP is mainly used in China because of the limited application of TCM in other countries. e Chinese government has issued health packages to Chinese people all over the world, while the data cannot be calculated and summarized well. erefore, although TMTP is widely used, the data is limited, which will have a certain impact on the results.

Conclusion
In summary, the study focused on the diversity of symptoms in COVID-19 patients, which was the most large-scale and comprehensive research on this issue so far. In this study, multiple outcomes were used to systematically evaluate TMTP efficacy for COVID-19 so as to determine the precise efficacy and safety of each drug for COVID-19 treatment. Bayesian network meta-analysis was used to integrate clinical evidence from direct and indirect treatment comparisons into a network, using a common comparator for indirect comparisons in the absence of head-to-head trials, and the competing drugs in each outcome were ranked. LHQW could significantly reduce proinflammatory cytokines and enhance immunity with antiviral effects. JHQG could inhibit influenza virus replication and promote virus clearance. XBJ could control infectious symptoms and alleviate acute lung injury effectively. QFPD had anti-inflammatory and antiviral effects, which could significantly reduce inflammatory indicators. XFBD could block the inflammatory cytokine storm and intervene in virus invasion. And HSBD could inhibit the infection, replication, and proliferation of the SARS-CoV-2 virus to a certain extent and, obviously, block the formation of the inflammatory storm after infection. As a result, LHQW could improve most symptoms of COVID-19 patients with no obvious adverse reactions. erefore, we believe that it should be Evidence-Based Complementary and Alternative Medicine