Since the outbreak of the pandemic coronavirus disease 2019 (COVID-19), various articles on different fields of the disease published in medical and related journals increased in a way of explosion. A recent literature search at PubMed using the term “COVID-19” only displayed more than 314 000 articles; a search term “treatment of COVID-19” would reveal more than 130 000, and for “antiviral therapies for COVID-19” showed that there were 13 082 articles. Antiviral treatment is one of the most important components of the treatments for COVID-19. To understand the status of antiviral therapies against COVID-19, we performed literature search at PubMed about 2 months ago, and after reading the abstracts, full texts and further searching, we finally enrolled 36 original articles of randomized controlled clinical trials (RCTs) for a literature study. To do some comparisons of the clinical trials conducted at different stages of the pandemic, we divided the articles according to the dates of publication roughly into two stages: stage 1 articles were those published between November 2019 and June 2021 (13 RCTs), and stage 2 articles were those published after June 2021 (23 RCTs). Overall, these articles on RCTs for antiviral therapies against COVID-19 provide much important and valuable evidence and messages concerning the efficacy and safety of more than 10 different antivirals and some other drugs that were tested in the clinical trials, and the trials provided evidence for some drugs that were previously believed to be effective and safe in the treatment of COVID-19 actually turned out to be ineffective and not safe, that is, chloroquine and hydroxychloroquine and more. Findings from clinical trials provided evidence for the efficacy and safety of remdesivir1 and quite a few monoclonal antibodies2-7 and other antivirals, and some of these drugs have already been approved for clinical applications.8, 9 These and more antiviral drugs will contribute together with other therapeutic approaches to the final success of global fights against the disease. However, some phenomena seen in some of the RCTs seemed not to be easy to understand, which prompted us to attempt to explore if there were some underlying factors that possibly led to the emergence of those phenomena. For example, there were 4 RCTs for the efficacy and safety of remdesivir.1, 10-12 Why did only one of them obtain evidence to support the effectiveness of remdesivir, and the remaining 3 could not? Furthermore, in one of the three trials that failed to support the efficacy of the drug, patients were randomized into 3 groups (1:1:1), patients in groups 1 and 2 were treated with remdesivir for 5 days and 10 days at the same daily dose in addition to the standard of care and group 3 was given standards of care alone, was such design reasonable enough? Especially, is a 5-day antiviral treatment long enough for a disease that has a course of illness of about 4 weeks or longer?11 The results showed that the main outcome of group 1 was significantly better than that of the control group, while group 2 (10-day treatment group) did not lead to significant improvement as compared with the control group, and the actual median days the group 2 patients received the drug was actually 6 days! It is difficult to understand or explain why treatment with the same drug at the same doses for treating 5 and 6 days could lead to such a paradoxical result. Another example: it was a surprise that of the 36 RCTs on antiviral treatments for COVID-19, only 50% had virological outcome/measurements included in the primary or secondary outcomes, that is, one-half of the RCTs did not provide any information about if the antiviral agents had any effect on the virus and if any, how the effects were. To reveal the possible underlying factors/reasons for such phenomena, we tried to sort out which factors could potentially be related to the results by closely reading the RCT reports and other references. The preliminarily perceived factors include the following: 1) virological outcome; 2) sample size; 3) control group; 4) time to begin the treatment, 5) stratification of participants, 6) subgroup analysis; 7) blinding and masking; and 8) reasonable designing of test group intervention. We have mentioned these factors in our recent review,13 in the present commentary, and future articles, we will further explore the relevant factors’ potential relation and how to reduce or prevent the problems. In the following texts, we describe the importance of antiviral therapies, virological outcomes, possible reasons for low application, and suggestions for how to set the virological outcomes. The etiologic agent of COVID-19 is severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2), which is the target of the antiviral therapy against COVID-19. As one of the most important therapeutic approaches against COVID-19, when the investigators design their clinical trials, especially when they consider how to set their primary and secondary outcomes, they should pay special attention to virological outcomes or endpoints. Because the investigators, caregivers, patients, and even the public would very much like to know: 1) If the test drug for antiviral treatment has any significant antiviral effect; 2) What kind of an effect was seen; 3) When the effect started to appear? How long did the effect persist? 4) Was the treatment safe? The virological status of patients by the end of a trial may be closely related to the patients’ further management. Some of the patients may still be positive for the viral RNA by the end of a trial. How should such patients be treated next? The virological outcome will provide some important information. In addition, virological outcomes can be quantitative or qualitative. The quantitative outcome measures may provide very informative, continuous, and detailed data reflecting the trend of viral loads at different stages of the treatment in comparison with the control data, as well-depicted in the figures of Dougan et al.2 and Udwadia et al.14 In fact, most of the clinical trials targeting any kind of pathogens causing diseases in the human body would set one or more outcomes that reflect the effects of the treatment on the pathogen. There are innumerable examples indicating the usefulness and importance of virological, bacteriological, and parasitological outcomes. A clinical trial protocol for the antiviral treatment of influenza included a virological outcome, “the change from baseline in infectious influenza virus titer” as one of its secondary outcomes.15 The virus titer is usually expressed as a geometric mean titer, which belongs to semiquantitative measurement. Some of the RCTs even set two (e.g., virus titers and viral RNA load)16 or even three (e.g., i) change in viral load at Day 7 with respect to baseline as one of the primary outcomes, and ii) trend over time in quantitative viral load at Days 7, 14 and 30 and iii) proportion of participants with virological clearance at Days 14 and 30)17 virological outcomes to measure the efficacy in different ways that may provide more useful information. During stage 1 of the pandemic, five of 13 (38.5%) RCTs had virological outcomes, but during stage 2, of the 23 RCTs, 13 (56.5%) had virological outcomes, indicating that more clinical studies have paid more attention to virological outcome measures. Clinicians and clinical researchers emphasize the importance of antiviral therapies for COVID-19 based on their knowledge of the pathogenesis and pathophysiology. Shiraki et al.18 pointed out: “Optimally, antivirals should be administered immediately after COVID-19 diagnosis, similar to that after influenza diagnosis, to prevent COVID-19 pneumonia and complications resulting from microangiopathy.” The previous pandemic of influenza, esp. that of 2009, and epidemics have left plenty of valuable experience and lessons, including encouraging early use of antiviral agents represented by oseltamivir.19, 20 For clinical trials of antiviral therapies, virological outcomes are essential for evaluating the efficacy and safety of the antiviral agents. During the pandemic of COVID-19, the clinical investigators participating in the design of clinical trials as well as the daily care of COVID-19 patients face and experience very high working load, threats from the disease itself, and very limited time to in-depth consider all the relevant issues of the design, shortage of facilities for patients’ treatment, isolation, and many more work and difficulties. These difficulties and shortages may affect the design of a clinical trial, and it was in many cases impossible to make a clinical trial perfect. However, after many of the clinical trial papers have been published, it is important to review, analyze and re-consider the clinical trials may provide some help in finding out what factors were related to the incompleteness of some clinical trials and finding ways to reduce the factors or take certain countermeasures. For example, preparedness for health emergencies and crises should include “Ready to use well-designed clinical trial protocols” for predicted public health events/crises/emergencies should be made available and applicable. Different considerations, that is, in the treatment of moderate to severe cases with COVID-19, the antiviral treatment itself might not be as important as high-dose glucocorticoid and/or respiratory supportive therapies based on the complicated pathogenesis of the disease. In the earlier months of the pandemic, there was almost no effective and safe antiviral drug available for the treatment of COVID-19, therefore antiviral treatment itself and virological outcomes could not be regarded as important enough to be listed among the primary or secondary outcome measures. However, during stage 2 of the pandemic, the situation is entirely different. Trials for many antiviral drugs have already obtained evidences from well-designed RCTs for efficacy and safety in the treatment of COVID-19, and more than 10 drugs (including remdesivir, molnupiravir, nirmatrelvir, and four monoclonal antibody drugs, that is, imdevimab, bamlanivimab, etesevimab, and casirivimab) obtained approval from the authorities for clinical use.8 More than 40 therapeutic monoclonal antibodies for the treatment of COVID-19 are under different stages of clinical trials.9 With many more effective and safe antiviral agents against COVID-19 and the encouragement of early use of antivirals, and many other measures including vaccinations, this COVID-19 pandemic will soon be extinguished. Quantitative measurements. The most frequently used quantitative virological determination is quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) for the detection of viral nucleic acid. Nasopharyngeal or oropharyngeal specimens are collected once daily or every 2 or 3 days as seen in many RCTs.2, 4, 5, 10, 21, 22 The results can also be expressed as “time to viral clearance” or “rate of viral nucleic acid conversion to negative”. Quantitative determination of viral load may provide continuous data of daily values of viral load, which will be informative and helpful in the judgment of patients’ status of viral infection. In qualitative measurements of viral nucleic acid or antigen, the report of results is usually positive or negative. In many clinical trials, the investigators often predefine one or two days in the process of the trial, for example, the viral clearance rate on day 7 since the beginning of treatment, the median time to viral clearance, and so on. The following are some examples of virological outcomes or endpoints. Shinkai et al.18 used “time to … recovery to SARS-CoV-2-negative” as one of the composite primary outcomes in that trial; it was defined as two consecutive negative results at least 24 h apart by qualitative nucleic acid amplification testing. Udwadia et al.’s14 virological outcome was similar: “time to the cessation of viral shedding”, it was also a primary outcome which was specified as two consecutively negative results of RT-PCR within the period from days 2 to day 28 on both oropharyngeal and nasopharyngeal swabs. The trial of Doi et al.23 had two virological endpoints: the primary one was “viral clearance by day 6” (this can be quantitative or qualitative) and the secondary one was “change in viral load by day 6” (this one must be quantitative). How to choose quantitative or qualitative and particular virological outcomes? The designers of the trial need to consider the availability of resources for the virological tests, which may include manpower, materials including the testing kits, and related financial sources. Quantitative and high-frequency sampling (e.g., several times a day) and large sample size will consume higher cost; qualitative and low-frequency sampling, that is, once or twice during the whole trial process sampled on a prespecified trial day, such as “rate of viral clearance on day 7”, and so on. However, despite the low-frequency sampling and qualitative testing may reduce the cost, the obtained data are not as informative and detailed as the high-frequency sampling and quantitative testing. We hope that all the RCTs for antiviral therapies against COVID-19 as well as any other viral diseases would set at least one virological outcome for evaluating the efficacy of the antiviral agent. Getu Zhaori is the editor-in-chief of Pediatric Investigation. Lu Lu, Chunyan Liu, Shujing Han, and Yongli Guo are the editorial staff of Pediatric Investigation.