1 INTRODUCTION
The coronavirus disease 2019 (COVID‐19) is an epidemic disease caused by the Severe Acute Respiratory Syndrome Corona Virus 2 (SARS‐CoV‐2), a kind of enveloped coronavirus with positive‐sense, single‐stranded RNA.1 COVID‐19 spread quickly and more than 4.52 million people had been diagnosed worldwide by May 17, 2020; the all‐cause mortality of COVID‐19 reached 6.8%.2 The World Health Organization declared this epidemic as a global health emergency since it threatened the health of the people all over the world. The common symptoms of COVID‐19 include fever, cough, myalgia, headache, sore throat, nausea, vomiting, and diarrhea, rare but important symptoms include conjunctivitis and pink eye.3
The vaccine of SARS‐CoV‐2 is currently at the stage of clinical trials while there are no specific drugs for SARS‐CoV‐2.4 Meanwhile, studies have shown that general therapy, antiviral therapy, and coronavirus‐specific therapy may be effective in the treatment of COVID‐19.5 Famotidine, remdesivir, Chloroquine, nicotine, ivermectin, Lopinavir/Ritonavir (LPV/r), interferon (IFN), Novaferon, and Arbidol are important drugs in the COVID‐19 therapy.6–9 Among them, LPV/r, a combination of Lopinavir and Ritonavir, has the effect of preventing gag‐pol polymerisation and the production of immature, non‐infectious viral particles by inhibiting human immunodeficiency virus (HIV) proteases,10 and which has been shown to have anti‐SARS‐CoV activity in vitro and in vivo.11, 12 Moreover, the use of LPV/r combined with pneumonia‐associated adjuvant drugs was suggested in some research.13 Some studies have proven that LPV/r was effective in the treatment of COVID‐19, and the combination of LPV/r and Arbidol was more effective than LPV/r alone.14 However, research has also shown that the efficacy of LPV/r in the treatment of COVID‐19 remains to be verified.15
Arbidol is often used to treat influenza virus infection and was added into the sixth edition of the new coronavirus pneumonia protocol of the National Health Commission of China. It has the effect of inhibiting the binding of the virus to the host cell.14 Studies have shown that Arbidol had good anti‐SARS‐CoV‐2 activity in vitro by preventing the attachment and release of viruses.16, 17 IFN is a broad‐spectrum antiviral drug that can activate the protection mechanism of the body to control and eliminate different viruses when they invade, and IFN was applied in the treatment of COVID‐19.18 In a study of using IFN spray therapy as a control, it was found that adding LPV/r or Arbidol based on IFN did not improve the symptoms or improve the time of negative nucleic acid conversion.19 Meanwhile, Novaferon, a recombinant IFNα like protein, has higher anti‐tumour and antiviral activity than IFNα,20 and a research has proven that Novaferon had more anti‐SARS‐CoV‐2 activity than LPV/r in vitro and in vivo.21
LPV/r often needs to be applied in combination with other antiretroviral drugs, which has been proven to be effective in treating human immunodeficiency virus (HIV) infection in adults.22 However, the efficacy of LPV/r and its combination with other drugs in COVID‐19 needs to be clarified. In this research, we evaluated the efficacy and the influencing factors of LPV/r and its combination with other commonly used drugs including IFN, Arbidol and Novaferon, to improve the rational use of antiviral drugs in the COVID‐19 treatment.
2 RESULTS
2.1 Characteristics of patients
We collected 245 mild and moderate patients admitted to the Changsha Public Health Treatment Center, and after screening by the inclusion and exclusion criteria, 170 patients were retained with a median age of 42.0 (IQR, 34.8–55.0). Among them, 89 cases (52.4%) were female, the most common complication was hypertension (23, 13.5%), and the top three symptoms were cough (131, 77.0%), fever (125, 73.5%) and fatigue (66, 33.8%).
Among the 170 patients, 131 cases contained LPV/r while 39 cases did not. There were 21, 30, 12 cases in the LPV/r alone group, the LPV/r + IFN group and the LPV/r + Novaferon group, respectively, and 14, 20 cases in the LPV/r + IFN+Novaferon group and the LPV/r + IFN+Arbidol group respectively. Treatment regimens with cases less than 10, including LPV/r + Arbidol, LPV/r + Arbidol+Novaferon, LPV/r + Chloroquine and LPV/r + Ribavirin, were excluded in the comparison of different regimes based on LPV/r patients without LPV/r received other types of antiviral treatment, including Chloroquine, IFN, Ribavirin, Arbidol, Novaferon and their combinations.
2.2 LPV/r vs LPV/r + Novaferon vs LPV/r + IFN
As is shown in Table 1, patients characteristics were well matched among the three groups since there was no significant difference in demographic information, coexisting conditions, symptoms and other aspects. However, difference was found in the blood platelet (PLT) of laboratory index [254.00 (201.50–332.00) vs 208.00 (155.50–249.75) vs 188.00 (144.25–263.75), P = .016], and which in LPV/r alone group was higher (Supplementary material).
Demographic and clinical characteristics of patients
| Characteristics | Treatment regimes |
P 1 |
P 2 |
||||
|---|---|---|---|---|---|---|---|
| LPV/r alone (n = 21) | LPV/r + Novaferon (n = 12) | LPV/r + IFN (n = 30) | LPV/r + IFN+Novaferon (n = 14) | LPV/r + IFN+Arbidol (n = 20) | |||
| Age (years), median (IQR) | 45.00 (34.00–55.00) | 43.50 (40.00–57.00) | 37.50 (29.75–49.25) | 45.50 (36.50–53.50) | 45.00 (37.50–61.50) | .183 | .179 |
| Female sex, n (%) | 16 (76.2) | 5 (41.7) | 14 (46.7) | 8 (57.1) | 10 (50.0) | .063 | .811 |
| Smokers, n (%) | 2 (9.5) | 3 (25.0) | 0 (0.0) | 2 (14.3) | 2 (10.0) | .125 | .048 |
| Complications, n (%) | |||||||
| Cerebrovascular disease | 0 (0.0) | 1 (8.3) | 1 (3.3) | 0 (0.0) | 2 (10.0) | .610 | .973 |
| Diabetes | 2 (9.5) | 3 (25.0) | 2 (6.7) | 2 (14.3) | 1 (5.0) | .593 | .516 |
| Hypertension | 5 (23.8) | 2 (16.7) | 4 (13.3) | 1 (7.1) | 3 (15.0) | .345 | .677 |
| COPD | 1 (4.8) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | .210 | |
| Chronic liver disease | 0 (0.0) | 1 (8.3) | 2 (6.7) | 1 (7.1) | 3 (15.0) | .318 | .709 |
| Tumor | 0 (0.0) | 0 (0.0) | 1 (3.3) | 1 (7.1) | 0 (0.0) | .325 | .739 |
| Cerobrain disease | 1 (4.8) | 0 (0.0) | 1 (3.3) | 0 (0.0) | 0 (0.0) | .848 | .320 |
| Rheu | 1 (4.8) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | .210 | |
| Others | 0 (0.0) | 0 (0.0) | 2 (6.7) | 1 (7.1) | 1 (5.0) | .160 | .976 |
| Family clusters, n (%) | 17 (81.0) | 10 (83.3) | 27 (90.0) | 11 (78.6) | 16 (80.0) | .354 | .263 |
| Exposure history, n (%) | 11 (52.4) | 7 (58.3) | 11 (36.7) | 8 (57.1) | 10 (50.0) | .225 | .392 |
| Temperature, median (IQR) | 37.25 (36.80–38.18) | 37.00 (36.40–37.50) | 37.35 (36.80–38.00) | 36.95 (36.43–38.23) | 37.60 (37.00–38.93) | .445 | .264 |
| Symptoms, n (%) | |||||||
| Fever | 15 (71.4) | 7 (58.3) | 24 (80.0) | 9 (64.3) | 17 (85.0) | .353 | .409 |
| Cough | 17 (81.0) | 8 (66.7) | 26 (86.7) | 10 (71.4) | 16 (80.0) | .494 | .234 |
| Expectoration | 7 (33.3) | 2 (16.7) | 10 (33.3) | 5 (35.7) | 9 (45.0) | .527 | .697 |
| Dyspnea | 6 (28.6) | 4 (33.3) | 6 (20.0) | 2 (14.3) | 7 (35.0) | .440 | .945 |
| Chill | 4 (19.0) | 2 (16.7) | 3 (10.0) | 2 (14.3) | 3 (15.0) | .354 | .641 |
| Fatigue | 12 (57.1) | 5 (41.7) | 12 (40.0) | 6 (42.9) | 8 (40.0) | .252 | .982 |
| Muscle soreness | 3 (14.3) | 0 (0.0) | 3 (10.0) | 1 (7.1) | 1 (5.0) | .731 | .627 |
| Dizziness | 1 (4.8) | 1 (8.3) | 2 (6.7) | 1 (7.1) | 1 (5.0) | .818 | .976 |
| Headache | 1 (4.8) | 0 (0.0) | 2 (6.7) | 2 (14.3) | 3 (15.0) | .674 | .352 |
| Abdominal pain | 1 (4.8) | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | .210 | |
| Diarrhea | 6 (28.6) | 1 (8.3) | 6 (20.0) | 1 (7.1) | 3 (15.0) | .561 | .290 |
| Nausea | 2 (9.5) | 0 (0.0) | 3 (10.0) | 1 (7.1) | 4 (20.0) | .836 | .895 |
| Vomiting | 3 (14.3) | 0 (0.0) | 2 (6.7) | 1 (7.1) | 4 (20.0) | .408 | .593 |
| Chest pain | 0 (0.0) | 1 (8.3) | 3 (10.0) | 2 (14.3) | 2 (10.0) | .168 | .727 |
| Sore throat | 3 (14.3) | 0 (0.0) | 5 (16.7) | 0 (0.0) | 2 (10.0) | .662 | .109 |
Note
- Data with significant differences were marked by bold font. The values shown are based on available data.
- Abbreviations: COPD, chronic obstructive pulmonary disease; IQR, interquartile range; P1 was the result of comparison of LPV/r alone, LPV/r + IFN and LPV/r + Novaferon; P2 was the result of comparison of LPV/r + IFN, LPV/r + IFN+Novaferon and LPV/r + IFN+Arbidol; Rheu, rheumatism.
There were significant differences in the time of negative nucleic acid conversion [9.00 (5.00–12.00) vs 6.00 (4.00–8.00) vs 9.00 (7.25–11.00) days, P = .036] and length of hospitalization [12.00 (11.00–15.00) vs 7.50 (5.00–10.00) vs 12.00 (10.00–13.50) days, P < .001] among LPV/r, LPV/r + IFN and LPV/r + Novaferon groups (Table 2). The LPV/r + Novaferon group had shorter time of negative nucleic acid conversion compared with the LPV/r + IFN group (P = .005), and the length of hospitalization in the LPV/r + Novaferon group was both shorter than LPV/r alone (P < .001) and the LPV/r + IFN (P < .001) groups (Figure 1).
Comparison of clinical efficacy in LPV/r, LPV/r + IFN and LPV/r + Novaferon group
| Outcomes | Treatment regimes |
P 1 |
P 2 |
||||
|---|---|---|---|---|---|---|---|
| LPV/r alone (n = 21) | LPV/r + Novaferon (n = 12) | LPV/r + IFN (n = 30) | LPV/r + IFN+Novaferon (n = 14) | LPV/r + IFN+Arbidol (n = 20) | |||
| Time of negative nucleic acid conversion (days), median (IQR) | 9.00 (5.00–12.00) | 6.00 (4.00–8.00) | 9.00 (7.25–11.00) | 10.00 (8.00–11.25) | 14.00 (9.75–19.00) | .036 | .031 |
| Length of hospitalization (days), median (IQR) | 12.00 (11.00–15.00) | 7.50 (5.00–10.00) | 12.00 (10.00–13.50) | 13.50 (11.50–17.00) | 19.50 (13.25–24.00) | <.001 | <.001 |
| ADR, (%) | |||||||
| Renal dysfunction | 1 (4.8） | 0 (0.0) | 1 (3.3) | 0 (0.0) | 2 (10.0) | .749 | .365 |
| Liver dysfunction | 0 (0.0) | 0 (0.0) | 0 (0.0) | 2 (14.3) | 2 (10.0) | .149 | |
| Secondary bacterial infection | 1 (4.8) | 0 (0.0) | 1 (3.3) | 1 (7.1) | 1 (5.0) | .749 | .878 |
| ICU, (%) | 4 (19.0) | 0 (0.0) | 3 (10.0) | 1 (7.1) | 2 (10.0) | .281 | .893 |
| ICU (days), median (IQR) | 0.00 (0.00–0.00) | 0.00 (0.00–0.00) | 0.00 (0.00–0.00) | 0.00 (0.00–0.00) | 0.00 (0.00–0.00) | .283 | .954 |
Note
- Data with significant differences were marked by Bold font. The values shown are based on available data.
- Abbreviations: ADR, adverse drug reaction; ICU, Intensive Care Unit; IQR, interquartile range; P1 was the comparison result of LPV/r alone, LPV/r + IFN and LPV/r + Novaferon; P2 was the comparison result of LPV/r + IFN, LPV/r + IFN+Novaferon and LPV/r + IFN+Arbidol.
The comparison of different antiviral medications on the time of negative nucleic acid conversion and the length of hospitalization. *.010 ≤ P ≤ .0167; **.001 ≤ P < .010; ***P < .001
2.3 LPV/r + IFN vs LPV/r + IFN+Novaferon vs LPV/r + IFN+Arbidol
The baselines among these three groups were well matched except that the distribution of current smokers [0 (0.0%) vs 2 (14.3%) vs 2 (10.0%), P = .048] was uneven (Table 1). Meanwhile, ALT [18.53 (12.78–27.32) vs 20.25 (17.40–30.67) vs 29.78（22.82–83.27）, P = .011], AST [23.16 (18.94–29.42) vs 21.51 (18.25–25.43) vs 31.28 (24.80–46.65), P = .009] and LDH [176.60 (149.58–215.98) vs 148.25 (121.08–180.90) vs 192.80 (156.40–260.30), P = .011] of laboratory index at the beginning of therapy among these groups were unequal (Supplementary material). The LDH of these three groups were within the normal range, but the ALT and AST of the LPV/r + IFN+Arbidol group were partially higher than the normal values.
Contrary to the second part, the combination of three kinds of drugs (LPV/r + IFN+Novaferon, LPV/r + IFN+Arbidol) had longer time of negative nucleic acid conversion [9.00 (7.25–11.00) vs 10.00 (8.00–11.25) vs 14.00 (9.75–19.00) days, P = .031] and length of hospitalization [12.00 (10.00–13.50) vs 13.50 (11.50–17.00) vs 19.50 (13.25–24.00) days, P < .001] (Table 2). Specifically, the LPV/r + IFN+Arbidol group had longer time of negative nucleic acid conversion and length of hospitalization compared with LPV/r + IFN (P = .009 & P < .001), and longer hospital stay compared with the LPV/r + IFN+Novaferon groups (P = .011) (Figure 1). Meanwhile, adding IFN based on LPV/r and Novaferon would not shorten the time of negative nucleic acid conversion and the length of hospitalization, and even prolonged them (P = .004 and P = .001).
Finally, we performed multiple linear regression to analyze variables that may affect the time of negative nucleic acid conversion and the length of hospitalization, and found that the medication regimen as an independent factor can affect the time of negative nucleic acid conversion (P = .018; 95% CI 0.184–1.857) and the length of hospitalization (P = .006; 95% CI 0.324–1.895) simultaneously (Table 3).
Multiple linear regression for the time of negative nucleic acid conversion and the length of hospitalization
| Variables | B | SD | P | 95% CI |
|---|---|---|---|---|
| The time of negative nucleic acid conversion | ||||
| Age | 0.023 | 0.050 | .649 | −0.077–0.123 |
| Sex | 0.699 | 1.195 | .561 | −1.690–3.088 |
| High blood pressure | 2.704 | 2.017 | .185 | −1.328–6.735 |
| Diabetes | −3.238 | 2.158 | .138 | −7.551–1.075 |
| Smokers | −1.422 | 2.116 | .504 | −5.651–2.808 |
| Medications | 1.020 | 0.419 | .018 | 0.184–1.857 |
| ALT | −0.011 | 0.030 | .727 | −0.071–0.050 |
| AST | 0.011 | 0.063 | .867 | −0.115–0.136 |
| LDH | 0.000 | 0.011 | .968 | −0.023–0.022 |
| PLT | −0.004 | 0.007 | .612 | −0.018–0.011 |
| The length of hospitalization | ||||
| Age | −0.023 | 0.041 | .571 | −0.104–0.058 |
| Sex | 0.475 | 1.110 | .670 | −1.732–2.682 |
| High blood pressure | 2.971 | 1.913 | .124 | −0.832–6.775 |
| Diabetes | −4.247 | 2.017 | .038 | −8.256–(−0.238) |
| Smokers | −2.078 | 1.811 | .254 | −5.679–1.523 |
| Medications | 1.110 | 0.395 | .006 | 0.324–1.895 |
| ALT | 0.071 | 0.028 | .012 | 0.016–0.126 |
| AST | −0.010 | 0.063 | .871 | −0.135–0.114 |
| LDH | 0.005 | 0.011 | .660 | −0.017–0.026 |
| PLT | 0.003 | 0.006 | .603 | −0.009–0.015 |
3 DISCUSSION
With the spread of COVID‐19 in the world, how to choose effective drugs and methods to treat COVID‐19 has attracted people’s attention. Due to the lack of special drugs for the treatment of COVID‐19, conventional antiviral therapy and coronavirus‐specific therapy have become an important way to control this epidemic.5 LPV/r is a commonly used drug for the treatment of COVID‐19, which was first recommended in the second edition of the new coronavirus pneumonia protocol of the National Health Commission of China.23 However, the efficacy evaluation of LPV/r for COVID‐19 was insufficient. Only one study of five patients with mild pneumonia indicated that LPV/r did not shorten the duration of SARS CoV‐2 shedding.24 Considering the factors of combination medication and sample size, we focused on evaluating the efficacy of LPV/r in combination with Arbidol, IFN and Novaferon in this study. The results of our research demonstrated that LPV/r + Novaferon had shorter time of negative nucleic acid conversion and hospitalization compared with LPV/r (P = .103 & P < .001) and LPV/r + IFN (P = .005 & P < .001). However, the combination of LPV/r with more drugs would not improve the primary outcomes or even make it worse, since adding Novaferon based on LPV/r + IFN had not shorten the time of negative nucleic acid conversion (P = .491) and the length of hospitalization (P = .155), and adding Arbidol prolonged them (P = .005 & P < .001) (Figure 1). In addition, the secondary endpoints of different medication regimens had no statistical differences.
As was shown in the results, the combination of LPV/r and Novaferon had better efficacy for the treatment of COVID‐19 compared with LPV/r alone and LPV/r + IFN, and there was no statistical difference between LPV/r alone and LPV/r + IFN. These phenomena are somewhat different from the existing research. Studies have shown that LPV/r was effective in the treatment of COVID‐19 and good responses were achieved in the remission of fever and inflammation.13 Therapeutic regimens of IFN‐α + LPV/r might also be beneficial for COVID‐19 treatment.25 Since Novaferon has a stronger effect than IFN no matter in antiviral or antitumour, it is speculated that the combination of LPV/r + Novaferon is more effective than LPV/r + IFN, and a study also indicated that regardless of whether it is combined with LPV/r, Novaferon is more effective against COVID‐19.21 However, Novaferon is currently only available in China, and the clinical research for it is still rare.
The combination of the three drugs has not achieved the desired effect. On the one hand, adding IFN based on LPV/r + Novaferon did not improve the primary outcomes, which might be due to the competition of the same combination target and different efficacy since IFN and Novaferon have a similar antiviral mechanism. Moreover, some studies indicated that Novaferon could inhibit the production of IFN‐α.26, 27 On the other hand, Arbidol reduced the efficacy of LPV/r + IFN. The mechanism of Arbidol is to block viral replication by inhibiting the fusion of influenza virus lipid membranes with host cells.14, 17 So, these three drugs have no obvious antagonism. Meanwhile, the efficacy of LPV/r + Arbidol was controversial. Studies have shown that the combination of LPV/r and Arbidol had antiviral effect for COVID‐19,28 and the efficacy of which surpassed LPV/r alone.14 However, a study also proved that their combination in the treatment of COVID‐19 was not beneficial.29 Furthermore, drug instructions of Arbidol indicate clearly that it can induce IFN production too. The cause of this phenomenon is not clear.
In addition, there were some limitations in our study. First, the sample size of the research was insufficient, which reduced the practicality of statistical results; second, the collection of laboratory index data before and after the treatment was incomplete; finally, uncontrollable bias existed in this study because it was retrospective. In the future research, the sample size needs to be expanded, and the antiviral mechanism of the drug needs further exploration.
In conclusion, the combination of LPV/r and Novaferon may have better efficacy against COVID‐19 in shortening the time of negative nucleic acid conversion and the length of hospitalization. However, adding IFN based on LPV/r + Novaferon or adding Arbidol based on LPV/r + IFN may not improve the efficacy.
4 PATIENTS AND METHODS
4.1 Ethics
The study was approved by Ethics Committee of the First Hospital of Changsha and written informed consent was waived because of the retrospective nature of the study. The implementation of this research conformed to the Helsinki Declaration published in 1964 and its later amendments.
4.2 Patients
This retrospective case‐control study involved patients admitted to the Changsha Public Health Treatment Center as of March 1, 2020. The inclusion criteria of this study were as follows: (a) Confirmed COVID‐19 patients who were tested positive for novel coronavirus nucleic acid for two respiratory specimens; (b) Mild and moderate patients in line with the diagnostic criteria in the ‘novel coronavirus infected pneumonia treatment scheme (trial seventh edition)’ 30; (c) Patients who have experienced antiviral treatment due to novel coronavirus nucleic acid infection. The medications complied with the ‘novel coronavirus infected pneumonia treatment scheme (trial seventh edition)’,30 and the choices for different patients under specific circumstances were based on the experience of clinicians. Exclusion criteria included: (a) Cases with incomplete clinical data; (b) Medications with patients less than 10 since too few cases would reduce the reliability of statistical results; (c) Medications with the treatment duration less than 72 hours. To protect the privacy of selected patients, all patients are anonymous, and all records are deleted after data coding and analysis.
4.3 Collection of clinical data
We collected the demographic characteristics, complications, epidemiological data, medications, laboratory indexes, adverse reactions and efficacy indicators of COVID‐19 patients in this study. The epidemiological data involved the history of family clusters and exposure and the common symptoms of COVID‐19 infection, like fever, cough and expectoration etc. These indicators were collected at the beginning of hospitalization. The efficacy indexes included the time of negative nucleic acid conversion that was tested at least twice, the length of hospitalization, and intense care unit (ICU) experiences.
4.4 Design for case‐control study
The primary objective was to assess the efficacy of LPV/r‐based regimens in COVID‐19. The primary outcomes in our study were the time of negative nucleic acid conversion. The secondary indicators included the length of hospitalization, the rate of adverse reaction, transferring to ICU and clinical mechanical therapy.
We compared the efficacy of the antiviral regimens including LPV/r alone, LPV/r + IFN, LPV/r + Novaferon, LPV/r + IFN+Arbidol andLPV/r + IFN+Novaferon. The detailed screening steps are shown in Figure 2. LPV/r alone meant that no other antiviral drugs were used except for LPV/r. Similarly, LPV/r + IFN, LPV/r + Novaferon, LPV/r + IFN+Arbidol and LPV/r + IFN+Novaferon represented that no other antiviral drugs were used except drugs in the respective combinations. The medication regimens of these drugs followed the instructions [LPV/r: P.O. 500 mg (400 mg Lopinavir + 100 mg Ritonavir) bid; Novaferon: Aerosol 20 μg bid; Arbidol: P.O. 0.2 g tid; IFN: Aerosol 500 × 104 IU bid.]. Moreover, we compared the basic information of the medication regimens in each group to ensure their baseline match, so as to meet the standard of random and control.
Detailed screening and grouping steps of this research
4.5 Statistical analysis
Data were processed using SPSS 25.0 statistical software (IBM, Armonk, NY, USA). The measurement data of non‐normal distribution were represented as the median (interquartile range [IQR]). Continuous variables were evaluated by the nonparametric test, including Kruskal–Wallis H test for three groups and Mann–Whitney U test for two groups. The number of dichotomous variables was expressed as the number of cases (percentage), and the Chi‐square test or Fisher’s exact test was used for the comparison among groups. Bonferroni correction was needed when comparing three groups of data pair by pair. P < .0167 was considered statistically significant. Multiple linear regression analysis was used to detect variables that affected the outcomes. The method for variable screening was entry regression and the confidence interval was 95%.
ACKNOWLEDGEMENTS
We thank the supporting grants of the Novel coronavirus pneumonia major project of Hunan (2020SK3014). We pay high respects to the medical staff at the front line of the COVID‐19 epidemic.
CONFLICT OF INTEREST
There is no conflict of interest in this work.
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
