Vitamin C reduces COVID-19 risk: real-time meta-analysis of 72 studies

@CovidAnalysis, October 2026, Version 88V88 (V1 2021)
c19early.org
October 2026

Media coverage of vitamin C for COVID‑19

Media covered none of the 32 studies showing significantly lower risk

No positive vitamin C studies were covered

0%

1 negative or inconclusive study was covered

2%
Abstract
Significantly lower risk is seen for mortality, ICU admission, hospitalization, and recovery. 25 studies from 25 independent teams in 12 countries show significant benefit.
Meta-analysis using the most serious outcome reported shows 21% [14‑27%] lower risk. Results are similar for Randomized Controlled Trials, higher quality studies, and peer-reviewed studies. Clinical outcomes suggest benefit while viral and case outcomes do not, consistent with an intervention that aids the immune system or recovery but may have limited antiviral effects.
Results are robust—in worst case exclusion sensitivity analysis 29 of 72 studies must be excluded before statistical significance is lost.
6 RCTs with 1,420 patients have not reported results (up to 5 years late).
The European Food Safety Authority has found evidence for a causal relationship between the intake of vitamin C and optimal immune system function1,2.
Early cessation of high-dose IV treatment may result in a detrimental rebound effect3. Ongoing treatment is more effective than early cessation: 33% [22‑42%] vs. 16% [-31‑46%].
No treatment is 100% effective. Protocols combine safe and effective options with individual risk/benefit analysis and monitoring. Other treatments are more effective. Dietary sources may be preferred. The quality of non-prescription supplements varies widely and the quantity of the active ingredient may be significantly lower than stated4-6. High doses may increase the risk of kidney stones7, with risk depending on formulation, predisposition, diet, and hydration8. Choi et al. show the highest bioavailability from fruit and vegetable juice. All data and sources to reproduce this analysis are in the appendix.
7 other meta-analyses show significant improvements with vitamin C for mortality10-14, progression15, severity10,14, and cases16.
Download PDF
Evolution of COVID-19 clinical evidence Meta-analysis results over time Vitamin C p=0.000000069 Acetaminophen p=0.00000021 2020 2021 2022 2023 2024 2025 2026 Lowerrisk Higherrisk c19early.org October 2026 50% 0% -50%

Vitamin C for COVID-19

Highlights
A

Vitamin C COVID-19 studies

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Su−135%2.35 · 0.67–8.27progressionn/an/a Thomas (RCT)−204%3.04 · 0.13–72.89death1/480/50COVIDAtoZ Zhao (PSM)72%0.28 · 0.08–0.93progression4/5512/55 Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75 Usanma Koban33%0.67 · 0.07–5.38viral+31 (n)95 (n) Rahman40%0.60 · 0.47–0.76hosp.128/47656/124
Early treatment34%0.66 · 0.48–0.90202/772114/39934% lower risk
Tau² = 0.04, I² = 38.2%, p = 0.0082
Krishnan31%0.69 · 0.47–0.92death40/7952/73 Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Yüksel (ICU)19%0.81 · 0.66–0.99death31/4240/44ICU patients Patel29%0.71 · 0.43–1.14death22/9626/80 Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 Jang51%0.49 · 0.23–1.01no recov.5/126/7ECMO patients JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Gao86%0.14 · 0.03–0.72death1/465/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT2 Al Sulaiman (PSM)15%0.85 · 0.61–1.12death46/14259/142 Mulhem−32%1.32 · 1.07–1.62death157/794359/2,425 Gadhiya−1%1.01 · 0.48–1.91death19/5536/226 Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT2 Elhadi (ICU)−12%1.12 · 0.96–1.31death175/277106/188ICU patients Suna21%0.79 · 0.44–1.41death17/15324/170 Pourhoseingholi13%0.87 · 0.63–1.19death54/199285/2,269 Li (ICU)−11%1.11 · 0.79–1.54death7/819/24ICU patients Vishnuram54%0.46 · 0.24–0.86death164/8,63410/241 Özgünay (ICU)9%0.91 · 0.63–1.30death17/3275/128ICU patients Tan25%0.75 · 0.10–2.98death/int.1/4614/115CT2 Zheng (PSM)−157%2.57 · 0.39–16.79death12/707/327 Simsek44%0.56 · 0.23–1.35death6/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Majidi (DB RCT)14%0.86 · 0.74–1.01death26/3167/69ICU patients Baguma−48%1.48 · 0.41–4.70death385 (n)96 (n) Tu83%0.17 · 0.08–0.35death8/11626/64 Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT2 Gavrielatou (ICU)58%0.42 · 0.12–1.48death2/1049/103ICU patients Salehi (ICU)10%0.90 · 0.65–1.25death22/4052/85ICU patients Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Hess (PSW)20%0.80 · 0.40–1.60death10/2537/75 Izzo41%0.59 · 0.50–0.69recovery869 (n)521 (n)LINCOLNLONG COVID OT1 CT2 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Kumar (DB RCT)23%0.77 · 0.40–1.47death10/3013/30ICU patients Özgülteki̇n (ICU)−5%1.05 · 0.81–1.36death18/2118/22ICU patients Doocy63%0.37 · 0.08–1.82death2/6422/80 Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Coskun (ICU)25%0.75 · 0.48–1.15death17/3824/40ICU patients Kyagambiddwa50%0.50 · 0.24–1.04death246 (all patients) Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200 Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT2 Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients Corrao39%0.61 · 0.23–1.60death9/1046/42 Uz84%0.16 · 0.02–0.97death41 (n)46 (n) Dinoi−32%1.32 · 0.79–2.20deathcase-control study Bepouka76%0.24 · 0.08–0.72death185 (n)225 (n)
Late treatment21%0.79 · 0.70–0.88950/14,9501,535/9,91021% lower risk
Tau² = 0.07, I² = 66.7%, p < 0.0001
Behera18%0.82 · 0.45–1.57casescase-control study Louca0%1.00 · 0.97–1.04casespopulation-based cohort Mahto−26%1.26 · 0.63–2.28IgG+34/14059/549 Bejan34%0.66 · 0.29–1.53death569 (n)8,637 (n) Holt−3%1.03 · 0.77–1.39cases49/1,580397/13,647COVIDENCE UK Abdulateef19%0.81 · 0.37–1.78hosp.8/13222/295 Aldwihi36%0.64 · 0.45–0.86hosp.142/50595/233 Mohseni−44%1.44 · 1.22–1.71cases34/43307/560 Nimer25%0.75 · 0.54–1.04hosp.52/651167/1,497 Shehab4%0.96 · 0.46–1.99severe case14/13912/114 Loucera28%0.72 · 0.58–0.88death840 (n)15,128 (n) Guldemir31%0.69 · 0.48–0.99hosp.33/17384/304 Sharif46%0.54 · 0.01–0.92severe casen/an/a Asoudeh69%0.31 · 0.14–0.65severe case250 (all patients) Vaisi38%0.62 · 0.31–1.23hosp.2,818 (n)1,137 (n) Akbar14%0.86 · 0.65–1.14cases665 (n)9,335 (n) Guan31%0.69 · 0.50–0.86symp. case28/462,017/2,454
Prophylaxis19%0.81 · 0.70–0.94394/8,3013,160/53,89019% lower risk
Tau² = 0.06, I² = 81.5%, p = 0.0058
All studies21%0.79 · 0.73–0.861,546/24,0234,809/64,19921% lower risk
Tau² = 0.06, I² = 74.0%, p < 0.000100.511.52+
1 OT: comparison with other treatment
2 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
B
Loading..
Fig. 1. A. Random-effects meta-analysis. This plot shows pooled effects, see the specific outcome analyses for individual outcomes. Analysis validating pooled outcomes for COVID-19 can be found below. Effect extraction is pre-specified, using the most serious outcome reported. For details see the appendix. B. Timeline of results in vitamin C studies. The marked dates indicate the time when efficacy was known with a statistically significant improvement of ≥10% from ≥3 studies for pooled outcomes, one or more specific outcome, and pooled outcomes in RCTs. Efficacy based on RCTs only was delayed by 6.8 months, compared to using all studies.
Introduction
Fig. 2. SARS-CoV-2 spike protein fibrin binding leads to thromboinflammation and neuropathology, from17.
Immediate treatment recommended
SARS-CoV-2 infection primarily begins in the upper respiratory tract and may progress to the lower respiratory tract, other tissues, and the nervous and cardiovascular systems, which may lead to cytokine storm, pneumonia, ARDS, neurological injury18-34 and cognitive deficits21,26, cardiovascular complications35-41, DNA damage42-45, organ failure, and death. Even mild untreated infections may result in persistent cognitive deficits46—the spike protein binds to fibrin leading to fibrinolysis-resistant blood clots, thromboinflammation, and neuropathology. Minimizing replication as early as possible is recommended.
Many treatments are expected to modulate infection
SARS-CoV-2 infection and replication involves the complex interplay of 500+ host and viral proteins and other factorsA,47-54, providing many therapeutic targets for which many existing compounds have known activity. Scientists have predicted that over 12,000 compounds may reduce COVID-19 risk55, either by directly minimizing infection or replication, by supporting immune system function, or by minimizing secondary complications.
Extensive supporting research
Vitamin C has been identified by the European Food Safety Authority (EFSA) as having sufficient evidence for a causal relationship between intake and optimal immune system function1,2,56. Vitamin C plays a key role in the immune system, supporting the production and function of leukocytes, or white blood cells, which defend against infection and disease, including the production of lymphocytes, which make antibodies, and enhancing phagocytosis, the process by which immune system cells ingest and destroy viruses and infected cells. Vitamin C is an antioxidant, protecting cells from damage caused by free radicals. Vitamin C inhibits SARS-CoV-2 3CLpro57,58, inhibits SARS-CoV-2 infection by reducing ACE2 levels in a dose-dependent manner59, and may limit COVID-19 induced cardiac damage by acting as an antioxidant and potentially reducing the reactive oxygen species (ROS) production induced by the spike protein that contributes to the activation of profibrotic pathways36. Vitamin C reduces inflammation, oxidative stress, and NETosis, supporting immune function and vascular protection60. Intracellular levels of vitamin C decline during COVID-19 hospitalization suggesting ongoing utilization and depletion of vitamin C61. Threonic acid, a metabolite of vitamin C, is lower in mild and severe cases, consistent with increased need for and metabolization of vitamin C with moderate infection, but more limited ability to produce threonic acid in severe infection due to depletion or existing lower levels of vitamin C62. Symptomatic COVID-19 is associated with a lower frequency of natural killer (NK) cells, and vitamin C has been shown to improve NK cell numbers and functioning63,64.
Other infections
Studies have shown efficacy with vitamin C for the common cold65,66 and acute respiratory tract infections67.
Analysis
We analyze all significant controlled studies of vitamin C for COVID-19. Search methods, inclusion criteria, effect extraction criteria (more serious outcomes have priority), all individual study data, PRISMA answers, and statistical methods are detailed in Appendix 1. We present random-effects meta-analysis results for all studies, studies within each treatment stage, individual outcomes, peer-reviewed studies, Randomized Controlled Trials (RCTs), and higher quality studies.
Treatment timing
Fig. 3 shows stages of possible treatment for COVID-19. Prophylaxis refers to regularly taking medication before becoming sick, in order to prevent or minimize infection. Early treatment refers to treatment immediately or soon after symptoms appear, while late treatment refers to more delayed treatment.
Prophylaxis pre-exposure post-exposure Early treatment Late treatment Treatment delay viral load disease severity exposed symptom onset hospitalized ICU Expected benefit of antiviral therapy −2 0 2 4 6 8 10 Time relative to symptom onset (days) trajectories vary widely immediately or soon after symptoms after disease progression
Fig. 3. Treatment stages.
Evaluation Notes
Preclinical Research
Vitamin C inhibits SARS-CoV-2 3CLpro57,58, inhibits SARS-CoV-2 infection by reducing ACE2 levels in a dose-dependent manner59, and may limit COVID-19 induced cardiac damage by acting as an antioxidant and potentially reducing the reactive oxygen species (ROS) production induced by the spike protein that contributes to the activation of profibrotic pathways36.
8 in silico studies support the efficacy of vitamin C58,75-81.
8 in vitro studies support the efficacy of vitamin C36,57-59,75,82-84.
An in vivo animal study supports the efficacy of vitamin C59.
Preclinical research is an important part of the development of treatments, however results may be very different in clinical trials. Preclinical results are not used in this paper.
Results
Table 1 summarizes the results for all stages combined, for Randomized Controlled Trials, for peer-reviewed studies, after exclusions, and for specific outcomes. Table 2 shows results by treatment stage. Fig. 4 plots individual results by treatment stage. Fig. 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, and 19 show forest plots for random-effects meta-analysis of all studies with pooled effects, mortality results, ventilation, ICU admission, hospitalization, progression, recovery, cases, viral clearance, high-dose IV studies, high-dose IV RCTs with early cessation vs. ongoing treatment, sufficiency studies, peer reviewed studies, non-symptomatic vs. symptomatic results, and long COVID.
Table 1. Random-effects meta-analysis for all stages combined, for Randomized Controlled Trials, for peer-reviewed studies, after exclusions, and for specific outcomes. Results show the relative risk with treatment and the 95% confidence interval. ** p<0.01  *** p<0.001  **** p<0.0001.
Relative Risk Studies Patients
All studies0.79 [0.73‑0.86]****7280K
After exclusions0.75 [0.67‑0.85]****4250K
Peer-reviewedPeer-reviewed0.78 [0.71‑0.85]****6580K
RCTsRCTs0.80 [0.69‑0.92]**204,585
Mortality0.81 [0.73‑0.91]***4440K
VentilationVent.0.84 [0.70‑1.01]910K
ICU admissionICU0.77 [0.66‑0.90]**710K
HospitalizationHosp.0.81 [0.70‑0.93]**1610K
Recovery0.71 [0.65‑0.78]****102,182
Cases0.97 [0.81‑1.16]730K
Viral1.05 [0.64‑1.73]2236
RCT mortality0.86 [0.72‑1.02]144,038
RCT hospitalizationRCT hosp.0.91 [0.76‑1.09]8856
Table 2. Random-effects meta-analysis results by treatment stage. Results show the relative risk with treatment and the 95% confidence interval.treatment and the 95% confidence interval. ** p<0.01  *** p<0.001  **** p<0.0001.
Early treatment Late treatment Prophylaxis
All studies0.66 [0.48‑0.90]**0.66**
[0.48‑0.90]		0.79 [0.70‑0.88]****0.79****
[0.70‑0.88]		0.81 [0.70‑0.94]**0.81**
[0.70‑0.94]
After exclusions0.94 [0.43‑2.08]0.94
[0.43‑2.08]		0.75 [0.63‑0.89]**0.75**
[0.63‑0.89]		0.74 [0.62‑0.89]**0.74**
[0.62‑0.89]
Peer-reviewedPeer-reviewed0.66 [0.48‑0.90]**0.66**
[0.48‑0.90]		0.76 [0.66‑0.87]****0.76****
[0.66‑0.87]		0.81 [0.70‑0.94]**0.81**
[0.70‑0.94]
RCTsRCTs0.70 [0.54‑0.90]**0.70**
[0.54‑0.90]		0.81 [0.70‑0.95]**0.81**
[0.70‑0.95]
Mortality3.04 [0.13‑72.89]3.04
[0.13‑72.89]		0.82 [0.73‑0.92]***0.82***
[0.73‑0.92]		0.71 [0.58‑0.87]**0.71**
[0.58‑0.87]
VentilationVent.0.84 [0.69‑1.03]0.84
[0.69‑1.03]		0.75 [0.35‑1.62]0.75
[0.35‑1.62]
ICU admissionICU0.77 [0.66‑0.90]**0.77**
[0.66‑0.90]		0.85 [0.43‑1.69]0.85
[0.43‑1.69]
HospitalizationHosp.0.60 [0.47‑0.76]****0.60****
[0.47‑0.76]		0.90 [0.76‑1.06]0.90
[0.76‑1.06]		0.71 [0.62‑0.82]****0.71****
[0.62‑0.82]
Recovery0.75 [0.63‑0.90]**0.75**
[0.63‑0.90]		0.71 [0.64‑0.79]****0.71****
[0.64‑0.79]
Cases0.97 [0.81‑1.16]0.97
[0.81‑1.16]
Viral1.05 [0.64‑1.73]1.05
[0.64‑1.73]
RCT mortality3.04 [0.13‑72.89]3.04
[0.13‑72.89]		0.85 [0.71‑1.02]0.85
[0.71‑1.02]
RCT hospitalizationRCT hosp.0.69 [0.12‑3.98]0.69
[0.12‑3.98]		0.91 [0.76‑1.10]0.91
[0.76‑1.10]
Loading..
Fig. 4. Scatter plot showing the most serious outcome in all studies, and for studies within each stage. Diamonds shows the results of random-effects meta-analysis.

Vitamin C COVID-19 studies

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Su−135%2.35 · 0.67–8.27progressionn/an/a Thomas (RCT)−204%3.04 · 0.13–72.89death1/480/50COVIDAtoZ Zhao (PSM)72%0.28 · 0.08–0.93progression4/5512/55 Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75 Usanma Koban33%0.67 · 0.07–5.38viral+31 (n)95 (n) Rahman40%0.60 · 0.47–0.76hosp.128/47656/124
Early treatment34%0.66 · 0.48–0.90202/772114/39934% lower risk
Tau² = 0.04, I² = 38.2%, p = 0.0082
Krishnan31%0.69 · 0.47–0.92death40/7952/73 Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Yüksel (ICU)19%0.81 · 0.66–0.99death31/4240/44ICU patients Patel29%0.71 · 0.43–1.14death22/9626/80 Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 Jang51%0.49 · 0.23–1.01no recov.5/126/7ECMO patients JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Gao86%0.14 · 0.03–0.72death1/465/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT2 Al Sulaiman (PSM)15%0.85 · 0.61–1.12death46/14259/142 Mulhem−32%1.32 · 1.07–1.62death157/794359/2,425 Gadhiya−1%1.01 · 0.48–1.91death19/5536/226 Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT2 Elhadi (ICU)−12%1.12 · 0.96–1.31death175/277106/188ICU patients Suna21%0.79 · 0.44–1.41death17/15324/170 Pourhoseingholi13%0.87 · 0.63–1.19death54/199285/2,269 Li (ICU)−11%1.11 · 0.79–1.54death7/819/24ICU patients Vishnuram54%0.46 · 0.24–0.86death164/8,63410/241 Özgünay (ICU)9%0.91 · 0.63–1.30death17/3275/128ICU patients Tan25%0.75 · 0.10–2.98death/int.1/4614/115CT2 Zheng (PSM)−157%2.57 · 0.39–16.79death12/707/327 Simsek44%0.56 · 0.23–1.35death6/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Majidi (DB RCT)14%0.86 · 0.74–1.01death26/3167/69ICU patients Baguma−48%1.48 · 0.41–4.70death385 (n)96 (n) Tu83%0.17 · 0.08–0.35death8/11626/64 Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT2 Gavrielatou (ICU)58%0.42 · 0.12–1.48death2/1049/103ICU patients Salehi (ICU)10%0.90 · 0.65–1.25death22/4052/85ICU patients Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Hess (PSW)20%0.80 · 0.40–1.60death10/2537/75 Izzo41%0.59 · 0.50–0.69recovery869 (n)521 (n)LINCOLNLONG COVID OT1 CT2 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Kumar (DB RCT)23%0.77 · 0.40–1.47death10/3013/30ICU patients Özgülteki̇n (ICU)−5%1.05 · 0.81–1.36death18/2118/22ICU patients Doocy63%0.37 · 0.08–1.82death2/6422/80 Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Coskun (ICU)25%0.75 · 0.48–1.15death17/3824/40ICU patients Kyagambiddwa50%0.50 · 0.24–1.04death246 (all patients) Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200 Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT2 Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients Corrao39%0.61 · 0.23–1.60death9/1046/42 Uz84%0.16 · 0.02–0.97death41 (n)46 (n) Dinoi−32%1.32 · 0.79–2.20deathcase-control study Bepouka76%0.24 · 0.08–0.72death185 (n)225 (n)
Late treatment21%0.79 · 0.70–0.88950/14,9501,535/9,91021% lower risk
Tau² = 0.07, I² = 66.7%, p < 0.0001
Behera18%0.82 · 0.45–1.57casescase-control study Louca0%1.00 · 0.97–1.04casespopulation-based cohort Mahto−26%1.26 · 0.63–2.28IgG+34/14059/549 Bejan34%0.66 · 0.29–1.53death569 (n)8,637 (n) Holt−3%1.03 · 0.77–1.39cases49/1,580397/13,647COVIDENCE UK Abdulateef19%0.81 · 0.37–1.78hosp.8/13222/295 Aldwihi36%0.64 · 0.45–0.86hosp.142/50595/233 Mohseni−44%1.44 · 1.22–1.71cases34/43307/560 Nimer25%0.75 · 0.54–1.04hosp.52/651167/1,497 Shehab4%0.96 · 0.46–1.99severe case14/13912/114 Loucera28%0.72 · 0.58–0.88death840 (n)15,128 (n) Guldemir31%0.69 · 0.48–0.99hosp.33/17384/304 Sharif46%0.54 · 0.01–0.92severe casen/an/a Asoudeh69%0.31 · 0.14–0.65severe case250 (all patients) Vaisi38%0.62 · 0.31–1.23hosp.2,818 (n)1,137 (n) Akbar14%0.86 · 0.65–1.14cases665 (n)9,335 (n) Guan31%0.69 · 0.50–0.86symp. case28/462,017/2,454
Prophylaxis19%0.81 · 0.70–0.94394/8,3013,160/53,89019% lower risk
Tau² = 0.06, I² = 81.5%, p = 0.0058
All studies21%0.79 · 0.73–0.861,546/24,0234,809/64,19921% lower risk
Tau² = 0.06, I² = 74.0%, p < 0.000100.511.52+
1 OT: comparison with other treatment
2 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 5. Random-effects meta-analysis for all studies. This plot shows pooled effects, see the specific outcome analyses for individual outcomes. Analysis validating pooled outcomes for COVID-19 can be found below. Effect extraction is pre-specified, using the most serious outcome reported. For details see the appendix.

Vitamin C COVID-19 mortality results

c19early.org
October 2026
StudyImprovementRR · 95% CITreatmentControlRelative Risk
Thomas (RCT)−204%3.04 · 0.13–72.891/480/50COVIDAtoZ
Early treatment−204%3.04 · 0.13–72.891/480/50204% higher risk
Tau² = 0.00, I² = 0.0%, p = 0.5
Krishnan31%0.69 · 0.47–0.9240/7952/73 Zhang (RCT)50%0.50 · 0.20–1.506/2711/29ICU patients Yüksel (ICU)19%0.81 · 0.66–0.9931/4240/44ICU patients Patel29%0.71 · 0.43–1.1422/9626/80 Kumari (RCT)36%0.64 · 0.26–1.557/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.563/303/30 Gao86%0.14 · 0.03–0.721/465/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.515/409/40CT1 Al Sulaiman (PSM)15%0.85 · 0.61–1.1246/14259/142 Mulhem−32%1.32 · 1.07–1.62157/794359/2,425 Gadhiya−1%1.01 · 0.48–1.9119/5536/226 Elhadi (ICU)−12%1.12 · 0.96–1.31175/277106/188ICU patients Suna21%0.79 · 0.44–1.4117/15324/170 Pourhoseingholi13%0.87 · 0.63–1.1954/199285/2,269 Li (ICU)−11%1.11 · 0.79–1.547/819/24ICU patients Vishnuram54%0.46 · 0.24–0.86164/8,63410/241 Özgünay (ICU)9%0.91 · 0.63–1.3017/3275/128ICU patients Zheng (PSM)−157%2.57 · 0.39–16.7912/707/327 Simsek44%0.56 · 0.23–1.356/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.250/184/26 Majidi (DB RCT)14%0.86 · 0.74–1.0126/3167/69ICU patients Baguma−48%1.48 · 0.41–4.70385 (n)96 (n) Tu83%0.17 · 0.08–0.358/11626/64 Gavrielatou (ICU)58%0.42 · 0.12–1.482/1049/103ICU patients Salehi (ICU)10%0.90 · 0.65–1.2522/4052/85ICU patients Hess (PSW)20%0.80 · 0.40–1.6010/2537/75 Kumar (DB RCT)23%0.77 · 0.40–1.4710/3013/30ICU patients Özgülteki̇n (ICU)−5%1.05 · 0.81–1.3618/2118/22ICU patients Doocy63%0.37 · 0.08–1.822/6422/80 Labbani-.. (DB RCT)33%0.67 · 0.20–2.174/376/37 Coskun (ICU)25%0.75 · 0.48–1.1517/3824/40ICU patients Kyagambiddwa50%0.50 · 0.24–1.04246 (all patients) Rana (DB RCT)55%0.45 · 0.16–1.275/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.988/20110/200 Adhikari (RCT)−19%1.19 · 0.98–1.461,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.195/227/25SAFE EVICT CORONA-ALIICU patients Corrao39%0.61 · 0.23–1.609/1046/42 Uz84%0.16 · 0.02–0.9741 (n)46 (n) Dinoi−32%1.32 · 0.79–2.20case-control study Bepouka76%0.24 · 0.08–0.72185 (n)225 (n)
Late treatment18%0.82 · 0.73–0.92935/13,8571,504/9,12318% lower risk
Tau² = 0.06, I² = 61.5%, p = 0.00082
Bejan34%0.66 · 0.29–1.53569 (n)8,637 (n) Loucera28%0.72 · 0.58–0.88840 (n)15,128 (n)
Prophylaxis29%0.71 · 0.58–0.871,409 (n)23,765 (n)29% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.0011
All studies19%0.81 · 0.73–0.91936/15,3141,504/32,93819% lower risk
Tau² = 0.06, I² = 61.2%, p = 0.0003400.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 6. Random-effects meta-analysis for mortality results.

Vitamin C COVID-19 mechanical ventilation results

c19early.org
October 2026
StudyImprovementRR · 95% CITreatmentControlRelative Risk
Kumari (RCT)20%0.80 · 0.40–1.5912/7515/75 JamaliMo.. (RCT)−25%1.25 · 0.37–4.215/304/30 Özgünay (ICU)−1%1.01 · 0.79–1.2923/3291/128ICU patients Hess (PSW)40%0.60 · 0.28–1.0018/2554/75 Kumar (DB RCT)21%0.79 · 0.43–1.4411/3014/30ICU patients Coskun (ICU)2%0.98 · 0.76–1.2828/3830/40ICU patients Rana (DB RCT)44%0.56 · 0.19–1.625/1399/139ICU patients Mousaviasl (DB RCT)−200%3.00 · 0.12–73.091/2010/200
Late treatment16%0.84 · 0.69–1.03103/570217/71716% lower risk
Tau² = 0.02, I² = 32.6%, p = 0.1
Bejan25%0.75 · 0.35–1.62572 (n)8,657 (n)
Prophylaxis25%0.75 · 0.35–1.61572 (n)8,657 (n)25% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.47
All studies16%0.84 · 0.70–1.01103/1,142217/9,37416% lower risk
Tau² = 0.02, I² = 23.8%, p = 0.06200.511.52+
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 7. Random-effects meta-analysis for ventilation.

Vitamin C COVID-19 ICU results

c19early.org
October 2026
StudyImprovementRR · 95% CITreatmentControlRelative Risk
Hakamifard (RCT)46%0.54 · 0.14–2.083/385/34CT1 Suna−2%1.02 · 0.46–2.2411/15312/170 Simsek10%0.90 · 0.55–1.4618/5828/81 Hess (PSW)27%0.73 · 0.41–1.0422/2563/75 Mousaviasl (DB RCT)−33%1.33 · 0.47–3.758/2016/200 Corrao−102%2.02 · 0.46–8.8310/1042/42
Late treatment23%0.77 · 0.66–0.9072/579116/60223% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.0012
Bejan15%0.85 · 0.43–1.69577 (n)8,690 (n)
Prophylaxis15%0.85 · 0.43–1.69577 (n)8,690 (n)15% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.65
All studies23%0.77 · 0.66–0.9072/1,156116/9,29223% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.001100.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 8. Random-effects meta-analysis for ICU admission.

Vitamin C COVID-19 hospitalization results

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Thomas (RCT)31%0.69 · 0.12–3.98hosp.2/483/50COVIDAtoZ Rahman40%0.60 · 0.47–0.76hosp.128/47656/124
Early treatment40%0.60 · 0.47–0.76130/52459/17440% lower risk
Tau² = 0.00, I² = 0.0%, p < 0.0001
Kumari (RCT)24%0.76 · 0.66–0.87hosp. time75 (n)75 (n) JamaliMo.. (RCT)−31%1.31 · 1.03–1.66hosp. time30 (n)30 (n) Hamidi-A.. (RCT)38%0.62 · 0.45–0.86hosp. time40 (n)40 (n)CT1 Hakamifard (RCT)1%0.99 · 0.91–1.07hosp. time38 (n)34 (n)CT1 Tehrani (RCT)18%0.82 · 0.60–1.13hosp. time18 (n)26 (n) Labbani-.. (DB RCT)−13%1.13 · 0.81–1.58hosp. time37 (n)37 (n) Rana (DB RCT)37%0.63 · 0.00–1358hosp. time139 (n)139 (n)ICU patients Corrao25%0.75 · 0.50–1.12hosp. time104 (n)42 (n)
Late treatment10%0.90 · 0.76–1.06481 (n)423 (n)10% lower risk
Tau² = 0.03, I² = 74.4%, p = 0.21
Bejan0%1.00 · 0.67–1.49hosp.626 (n)9,122 (n) Abdulateef19%0.81 · 0.37–1.78hosp.8/13222/295 Aldwihi36%0.64 · 0.45–0.86hosp.142/50595/233 Nimer25%0.75 · 0.54–1.04hosp.52/651167/1,497 Guldemir31%0.69 · 0.48–0.99hosp.33/17384/304 Vaisi38%0.62 · 0.31–1.23hosp.2,818 (n)1,137 (n)
Prophylaxis29%0.71 · 0.62–0.82235/4,905368/12,58829% lower risk
Tau² = 0.00, I² = 0.0%, p < 0.0001
All studies19%0.81 · 0.70–0.93365/5,910427/13,18519% lower risk
Tau² = 0.04, I² = 72.0%, p = 0.002100.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 9. Random-effects meta-analysis for hospitalization.

Vitamin C COVID-19 progression results

c19early.org
October 2026
StudyImprovementRR · 95% CITreatmentControlRelative Risk
Su−135%2.35 · 0.67–8.27n/an/a Zhao (PSM)72%0.28 · 0.08–0.934/5512/55
Early treatment21%0.79 · 0.10–6.364/5512/5521% lower risk
Tau² = 1.91, I² = 84.4%, p = 0.84
Tan73%0.27 · 0.09–0.617/4641/115CT1 Coppock (RCT)5%0.95 · 0.16–7.844/442/22 Labbani-.. (DB RCT)16%0.84 · 0.68–1.0537 (n)37 (n) Seely (DB RCT)48%0.52 · 0.10–2.712/424/44CT1
Late treatment44%0.56 · 0.27–1.1813/16947/21844% lower risk
Tau² = 0.33, I² = 66.9%, p = 0.13
All studies39%0.61 · 0.32–1.1717/22459/27339% lower risk
Tau² = 0.37, I² = 67.8%, p = 0.1400.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 10. Random-effects meta-analysis for progression.

Vitamin C COVID-19 recovery results

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Thomas (RCT)18%0.82 · 0.63–1.07recov. time48 (n)50 (n)COVIDAtoZ Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75
Early treatment25%0.75 · 0.63–0.9069/21046/12525% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.0024
Kumari (RCT)26%0.74 · 0.64–0.86recov. time75 (n)75 (n) Jang51%0.49 · 0.23–1.01no recov.5/126/7ECMO patients Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT2 Coppock (RCT)22%0.78 · 0.61–0.98no disch.31/4420/22 Izzo41%0.59 · 0.50–0.69recovery869 (n)521 (n)LINCOLNLONG COVID OT1 CT2 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Coskun (ICU)28%0.72 · 0.57–0.90sofa38 (n)40 (n)ICU patients Seely (DB RCT)14%0.86 · 0.61–1.22no recov.34 (n)24 (n)CT2
Late treatment29%0.71 · 0.64–0.7936/1,11426/73329% lower risk
Tau² = 0.01, I² = 37.3%, p < 0.0001
All studies29%0.71 · 0.65–0.78105/1,32472/85829% lower risk
Tau² = 0.01, I² = 27.8%, p < 0.000100.511.52+
1 OT: comparison with other treatment
2 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 11. Random-effects meta-analysis for recovery.

Vitamin C COVID-19 case results

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Behera18%0.82 · 0.45–1.57casescase-control study Louca0%1.00 · 0.97–1.04casespopulation-based cohort Holt−3%1.03 · 0.77–1.39cases49/1,580397/13,647COVIDENCE UK Mohseni−44%1.44 · 1.22–1.71cases34/43307/560 Vaisi10%0.90 · 0.69–1.29symp. case2,818 (n)1,137 (n) Akbar14%0.86 · 0.65–1.14cases665 (n)9,335 (n) Guan31%0.69 · 0.50–0.86symp. case28/462,017/2,454
Prophylaxis3%0.97 · 0.81–1.16111/5,1522,721/27,1333% lower risk
Tau² = 0.04, I² = 79.6%, p = 0.74
All studies3%0.97 · 0.81–1.16111/5,1522,721/27,1333% lower risk
Tau² = 0.04, I² = 79.6%, p = 0.7400.511.52+
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 12. Random-effects meta-analysis for cases.

Vitamin C COVID-19 viral clearance results

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Zhao (PSM)−8%1.08 · 0.64–1.80viral time55 (n)55 (n) Usanma Koban33%0.67 · 0.07–5.38viral+31 (n)95 (n)
Early treatment−5%1.05 · 0.64–1.7386 (n)150 (n)5% higher risk
Tau² = 0.00, I² = 0.0%, p = 0.86
All studies−5%1.05 · 0.64–1.7386 (n)150 (n)5% higher risk
Tau² = 0.00, I² = 0.0%, p = 0.8600.511.52+
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 13. Random-effects meta-analysis for viral clearance.

Vitamin C COVID-19 high-dose IV studies

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Zhao (PSM)72%0.28 · 0.08–0.93progression4/5512/55 Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75
Early treatment49%0.51 · 0.22–1.1973/21758/13049% lower risk
Tau² = 0.26, I² = 62.0%, p = 0.12
Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Yüksel (ICU)19%0.81 · 0.66–0.99death31/4240/44ICU patients Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Gao86%0.14 · 0.03–0.72death1/465/30 Suna21%0.79 · 0.44–1.41death17/15324/170 Li (ICU)−11%1.11 · 0.79–1.54death7/819/24ICU patients Özgünay (ICU)9%0.91 · 0.63–1.30death17/3275/128ICU patients Simsek44%0.56 · 0.23–1.35death6/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT1 Gavrielatou (ICU)58%0.42 · 0.12–1.48death2/1049/103ICU patients Hess (PSW)20%0.80 · 0.40–1.60death10/2537/75 Özgülteki̇n (ICU)−5%1.05 · 0.81–1.36death18/2118/22ICU patients Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients Corrao39%0.61 · 0.23–1.60death9/1046/42 Uz84%0.16 · 0.02–0.97death41 (n)46 (n)
Late treatment19%0.81 · 0.70–0.94148/2,391341/2,23319% lower risk
Tau² = 0.04, I² = 45.1%, p = 0.0068
All studies22%0.78 · 0.67–0.91221/2,608399/2,36322% lower risk
Tau² = 0.04, I² = 48.8%, p = 0.001100.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 14. Random-effects meta-analysis for high-dose IV studies. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details. Analysis validating pooled outcomes for COVID-19 can be found below.

Vitamin C COVID-19 early cessation vs. ongoing HDIV RCTs

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID
Early cessation16%0.84 · 0.54–1.319/1,66917/1,27316% lower risk
Tau² = 0.10, I² = 57.2%, p = 0.46
Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75 Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT1 Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients
Ongoing33%0.67 · 0.58–0.7890/34482/27033% lower risk
Tau² = 0.00, I² = 0.0%, p < 0.0001
All studies26%0.74 · 0.58–0.9599/2,01399/1,54326% lower risk
00.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 15. Random-effects meta-analysis for high-dose IV RCTs with early cessation vs. ongoing treatment. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details. Analysis validating pooled outcomes for COVID-19 can be found below.
Fig. 16. Random-effects meta-analysis for sufficiency studies. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details. Analysis validating pooled outcomes for COVID-19 can be found below.

Vitamin C COVID-19 peer reviewed studies

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Su−135%2.35 · 0.67–8.27progressionn/an/a Thomas (RCT)−204%3.04 · 0.13–72.89death1/480/50COVIDAtoZ Zhao (PSM)72%0.28 · 0.08–0.93progression4/5512/55 Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75 Usanma Koban33%0.67 · 0.07–5.38viral+31 (n)95 (n) Rahman40%0.60 · 0.47–0.76hosp.128/47656/124
Early treatment34%0.66 · 0.48–0.90202/772114/39934% lower risk
Tau² = 0.04, I² = 38.2%, p = 0.0082
Krishnan31%0.69 · 0.47–0.92death40/7952/73 Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Patel29%0.71 · 0.43–1.14death22/9626/80 Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 Jang51%0.49 · 0.23–1.01no recov.5/126/7ECMO patients Gao86%0.14 · 0.03–0.72death1/465/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT2 Mulhem−32%1.32 · 1.07–1.62death157/794359/2,425 Gadhiya−1%1.01 · 0.48–1.91death19/5536/226 Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT2 Elhadi (ICU)−12%1.12 · 0.96–1.31death175/277106/188ICU patients Suna21%0.79 · 0.44–1.41death17/15324/170 Li (ICU)−11%1.11 · 0.79–1.54death7/819/24ICU patients Vishnuram54%0.46 · 0.24–0.86death164/8,63410/241 Özgünay (ICU)9%0.91 · 0.63–1.30death17/3275/128ICU patients Tan25%0.75 · 0.10–2.98death/int.1/4614/115CT2 Zheng (PSM)−157%2.57 · 0.39–16.79death12/707/327 Simsek44%0.56 · 0.23–1.35death6/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Majidi (DB RCT)14%0.86 · 0.74–1.01death26/3167/69ICU patients Tu83%0.17 · 0.08–0.35death8/11626/64 Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT2 Gavrielatou (ICU)58%0.42 · 0.12–1.48death2/1049/103ICU patients Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Hess (PSW)20%0.80 · 0.40–1.60death10/2537/75 Izzo41%0.59 · 0.50–0.69recovery869 (n)521 (n)LINCOLNLONG COVID OT1 CT2 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Kumar (DB RCT)23%0.77 · 0.40–1.47death10/3013/30ICU patients Özgülteki̇n (ICU)−5%1.05 · 0.81–1.36death18/2118/22ICU patients Doocy63%0.37 · 0.08–1.82death2/6422/80 Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Coskun (ICU)25%0.75 · 0.48–1.15death17/3824/40ICU patients Kyagambiddwa50%0.50 · 0.24–1.04death246 (all patients) Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200 Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT2 Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Corrao39%0.61 · 0.23–1.60death9/1046/42 Uz84%0.16 · 0.02–0.97death41 (n)46 (n) Dinoi−32%1.32 · 0.79–2.20deathcase-control study Bepouka76%0.24 · 0.08–0.72death185 (n)225 (n)
Late treatment24%0.76 · 0.66–0.87789/14,0901,089/7,21924% lower risk
Tau² = 0.09, I² = 71.3%, p < 0.0001
Behera18%0.82 · 0.45–1.57casescase-control study Louca0%1.00 · 0.97–1.04casespopulation-based cohort Mahto−26%1.26 · 0.63–2.28IgG+34/14059/549 Bejan34%0.66 · 0.29–1.53death569 (n)8,637 (n) Holt−3%1.03 · 0.77–1.39cases49/1,580397/13,647COVIDENCE UK Abdulateef19%0.81 · 0.37–1.78hosp.8/13222/295 Aldwihi36%0.64 · 0.45–0.86hosp.142/50595/233 Mohseni−44%1.44 · 1.22–1.71cases34/43307/560 Nimer25%0.75 · 0.54–1.04hosp.52/651167/1,497 Shehab4%0.96 · 0.46–1.99severe case14/13912/114 Loucera28%0.72 · 0.58–0.88death840 (n)15,128 (n) Guldemir31%0.69 · 0.48–0.99hosp.33/17384/304 Sharif46%0.54 · 0.01–0.92severe casen/an/a Asoudeh69%0.31 · 0.14–0.65severe case250 (all patients) Vaisi38%0.62 · 0.31–1.23hosp.2,818 (n)1,137 (n) Akbar14%0.86 · 0.65–1.14cases665 (n)9,335 (n) Guan31%0.69 · 0.50–0.86symp. case28/462,017/2,454
Prophylaxis19%0.81 · 0.70–0.94394/8,3013,160/53,89019% lower risk
Tau² = 0.06, I² = 81.5%, p = 0.0058
All studies22%0.78 · 0.71–0.851,385/23,1634,363/61,50822% lower risk
Tau² = 0.06, I² = 76.3%, p < 0.000100.511.52+
1 OT: comparison with other treatment
2 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 17. Random-effects meta-analysis for peer reviewed studies. Zeraatkar et al. analyze 356 COVID-19 trials, finding no significant evidence that preprint results are inconsistent with peer-reviewed studies. They also show extremely long peer-review delays, with a median of 6 months to journal publication. A six month delay was equivalent to around 1.5 million deaths during the first two years of the pandemic. Authors recommend using preprint evidence, with appropriate checks for potential falsified data, which provides higher certainty much earlier. Davidson et al. also showed no important difference between meta-analysis results of preprints and peer-reviewed publications for COVID-19, based on 37 meta-analyses including 114 trials. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details. Analysis validating pooled outcomes for COVID-19 can be found below.

Vitamin C COVID-19 symptomatic vs. case outcomes

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Krishnan31%0.69 · 0.47–0.92death40/7952/73 Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Yüksel (ICU)19%0.81 · 0.66–0.99death31/4240/44ICU patients Patel29%0.71 · 0.43–1.14death22/9626/80 Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 Jang51%0.49 · 0.23–1.01no recov.5/126/7ECMO patients Su−135%2.35 · 0.67–8.27progressionn/an/a JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Thomas (RCT)−204%3.04 · 0.13–72.89death1/480/50COVIDAtoZ Mahto−26%1.26 · 0.63–2.28IgG+34/14059/549 Gao86%0.14 · 0.03–0.72death1/465/30 Bejan34%0.66 · 0.29–1.53death569 (n)8,637 (n) Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT2 Al Sulaiman (PSM)15%0.85 · 0.61–1.12death46/14259/142 Mulhem−32%1.32 · 1.07–1.62death157/794359/2,425 Gadhiya−1%1.01 · 0.48–1.91death19/5536/226 Abdulateef19%0.81 · 0.37–1.78hosp.8/13222/295 Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT2 Zhao (PSM)72%0.28 · 0.08–0.93progression4/5512/55 Elhadi (ICU)−12%1.12 · 0.96–1.31death175/277106/188ICU patients Suna21%0.79 · 0.44–1.41death17/15324/170 Aldwihi36%0.64 · 0.45–0.86hosp.142/50595/233 Pourhoseingholi13%0.87 · 0.63–1.19death54/199285/2,269 Li (ICU)−11%1.11 · 0.79–1.54death7/819/24ICU patients Vishnuram54%0.46 · 0.24–0.86death164/8,63410/241 Özgünay (ICU)9%0.91 · 0.63–1.30death17/3275/128ICU patients Tan25%0.75 · 0.10–2.98death/int.1/4614/115CT2 Zheng (PSM)−157%2.57 · 0.39–16.79death12/707/327 Simsek44%0.56 · 0.23–1.35death6/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75 Majidi (DB RCT)14%0.86 · 0.74–1.01death26/3167/69ICU patients Baguma−48%1.48 · 0.41–4.70death385 (n)96 (n) Tu83%0.17 · 0.08–0.35death8/11626/64 Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT2 Gavrielatou (ICU)58%0.42 · 0.12–1.48death2/1049/103ICU patients Nimer25%0.75 · 0.54–1.04hosp.52/651167/1,497 Shehab4%0.96 · 0.46–1.99severe case14/13912/114 Salehi (ICU)10%0.90 · 0.65–1.25death22/4052/85ICU patients Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Hess (PSW)20%0.80 · 0.40–1.60death10/2537/75 Izzo41%0.59 · 0.50–0.69recovery869 (n)521 (n)LINCOLNLONG COVID OT1 CT2 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Sinnberg42%0.58 · 0.18–1.91death0/159/59 Loucera28%0.72 · 0.58–0.88death840 (n)15,128 (n) Kumar (DB RCT)23%0.77 · 0.40–1.47death10/3013/30ICU patients Özgülteki̇n (ICU)−5%1.05 · 0.81–1.36death18/2118/22ICU patients Doocy63%0.37 · 0.08–1.82death2/6422/80 Guldemir31%0.69 · 0.48–0.99hosp.33/17384/304 Sharif46%0.54 · 0.01–0.92severe casen/an/a Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Asoudeh69%0.31 · 0.14–0.65severe case250 (all patients) Coskun (ICU)25%0.75 · 0.48–1.15death17/3824/40ICU patients Vaisi38%0.62 · 0.31–1.23hosp.2,818 (n)1,137 (n) Kyagambiddwa50%0.50 · 0.24–1.04death246 (all patients) Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200 Boerenkamp56%0.44 · 0.17–1.12severe case5/3412/36 Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT2 Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Rahman40%0.60 · 0.47–0.76hosp.128/47656/124 Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients Guan31%0.69 · 0.50–0.86symp. case28/462,017/2,454 Corrao39%0.61 · 0.23–1.60death9/1046/42 Uz84%0.16 · 0.02–0.97death41 (n)46 (n) Dinoi−32%1.32 · 0.79–2.20deathcase-control study Bepouka76%0.24 · 0.08–0.72death185 (n)225 (n)
Symptomatic24%0.76 · 0.69–0.831,468/21,7534,126/40,65724% lower risk
Tau² = 0.07, I² = 65.3%, p < 0.0001
Behera18%0.82 · 0.45–1.57casescase-control study Louca0%1.00 · 0.97–1.04casespopulation-based cohort Holt−3%1.03 · 0.77–1.39cases49/1,580397/13,647COVIDENCE UK Zhao (PSM)−8%1.08 · 0.64–1.80viral time55 (n)55 (n) Mohseni−44%1.44 · 1.22–1.71cases34/43307/560 Usanma Koban33%0.67 · 0.07–5.38viral+31 (n)95 (n) Akbar14%0.86 · 0.65–1.14cases665 (n)9,335 (n)
Cases−6%1.06 · 0.89–1.2683/2,374704/23,6926% higher risk
Tau² = 0.03, I² = 68.4%, p = 0.56
All studies21%0.79 · 0.73–0.861,551/24,1274,830/64,34921% lower risk
Tau² = 0.06, I² = 73.2%, p < 0.000100.511.52+
1 OT: comparison with other treatment
2 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 18. Random-effects meta-analysis for non-symptomatic vs. symptomatic results. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details. Analysis validating pooled outcomes for COVID-19 can be found below.

Vitamin C COVID-19 long COVID results

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Seely (DB RCT)12%0.88 · 0.19–4.02PASC3/333/29CT1
Late treatment12%0.88 · 0.19–4.023/333/2912% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.88
All studies12%0.88 · 0.19–4.023/333/2912% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.8800.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 19. Random-effects meta-analysis for long COVID. Effect extraction is pre-specified, using the most serious outcome reported, see the appendix for details. Analysis validating pooled outcomes for COVID-19 can be found below.
Randomized Controlled Trials (RCTs)
Fig. 20 shows a comparison of results for RCTs and observational studies. Fig. 21, 22, and 23 show forest plots for random-effects meta-analysis of all Randomized Controlled Trials, RCT mortality results, and RCT hospitalization results. RCT results are included in Table 1 and Table 2.
Loading..
Fig. 20. Results for RCTs and observational studies.
RCTs have many potential biases
RCTs help to make study groups more similar and can provide a higher level of evidence, however they are subject to many biases87, and analysis of double-blind RCTs has identified extreme levels of bias88. For COVID-19, the overhead may delay treatment, dramatically compromising efficacy; they may encourage monotherapy for simplicity at the cost of efficacy which may rely on combined or synergistic effects; the participants that sign up may not reflect real world usage or the population that benefits most in terms of age, comorbidities, severity of illness, or other factors; standard of care may be compromised and unable to evolve quickly based on emerging research for new diseases; errors may be made in randomization and medication delivery; and investigators may have hidden agendas or vested interests influencing design, operation, analysis, reporting, and the potential for fraud. All of these biases have been observed with COVID-19 RCTs. There is no guarantee that a specific RCT provides a higher level of evidence.
Conflicts of interest for COVID-19 RCTs
RCTs are expensive and many RCTs are funded by pharmaceutical companies or other organizations with conflicts of interest, for example governments that previously denied treatment with the study drug. For COVID-19, this creates an incentive to show efficacy for patented commercial products, and an incentive to show a lack of efficacy for inexpensive treatments. The bias is expected to be significant, for example Als-Nielsen et al. analyzed 370 RCTs from Cochrane reviews, showing that trials funded by for-profit organizations were 5 times more likely to recommend the experimental drug compared with those funded by nonprofit organizations. Bekelman et al. and Lundh et al. show that industry-sponsored studies are more likely to be favorable. For COVID-19, some major philanthropic organizations are largely funded by investments with extreme conflicts of interest for and against specific COVID-19 interventions.
RCTs for novel acute diseases requiring rapid treatment
High quality RCTs for novel acute diseases are more challenging, with increased ethical issues due to the urgency of treatment, increased risk due to enrollment delays, and more difficult design with a rapidly evolving evidence base. For COVID-19, the most common site of initial infection is the upper respiratory tract. Immediate treatment is likely to be most successful and may prevent or slow progression to other parts of the body. For a non-prophylaxis RCT, it makes sense to provide treatment in advance and instruct patients to use it immediately on symptoms, just as some governments have done by providing medication kits in advance. Unfortunately, no RCTs have been done in this way. Every treatment RCT to date involves delayed treatment. Among the 227 treatments we have analyzed, 67% of RCTs involve very late treatment 5+ days after onset. No non-prophylaxis COVID-19 RCTs match the potential real-world use of early treatments. They may more accurately represent results for treatments that require visiting a medical facility, e.g., those requiring intravenous administration.
RCT bias for widely available treatments
RCTs have a bias against finding an effect for interventions that are widely available—patients that believe they need the intervention are more likely to decline participation and take the intervention. RCTs for vitamin C are more likely to enroll low-risk participants that do not need treatment to recover, making the results less applicable to clinical practice. This bias is likely to be greater for widely known treatments, and may be greater when the risk of a serious outcome is overstated. This bias does not apply to the typical pharmaceutical trial of a new drug that is otherwise unavailable.
Fig. 24. For COVID-19, observational study results do not systematically differ from RCTs, RR 0.97 [0.91‑1.03] across 227 treatments92.
Observational studies have been shown to be reliable
Evidence shows that observational studies can also provide reliable results. Concato et al. found that well-designed observational studies do not systematically overestimate the magnitude of the effects of treatment compared to RCTs. Anglemyer et al. analyzed reviews comparing RCTs to observational studies and found little evidence for significant differences in effect estimates.
We performed a similar analysis across the 227 treatments we cover, showing no significant difference in the results of RCTs compared to observational studies, RR 0.97 [0.91‑1.03]72. Similar results are found for all low-cost treatments, RR 0.98 [0.90‑1.06]. High-cost treatments show a non-significant trend towards RCTs showing greater efficacy, RR 0.93 [0.85‑1.02]. Details can be found in the supplementary data.
Lee et al. showed that only 14% of the guidelines of the Infectious Diseases Society of America were based on RCTs. Evaluation of studies relies on an understanding of the study and potential biases. Limitations in an RCT can outweigh the benefits, for example excessive dosages, excessive treatment delays, or remote survey bias may have a greater effect on results. Ethical issues may also prevent running RCTs for known effective treatments. For more on issues with RCTs see96,97.
RCTs may be less reliable
Concato et al. report a paradoxical finding—RCT results had higher variability, and only RCTs were found to sometimes report significant results the opposite of the overall result. The same trend is seen for the most popular (most politicized) COVID-19 treatments—considering all statistically significant results reported in studies, RCTs are slightly more likely to report a result in the opposite direction. In other words, for these COVID-19 treatments and for the topics covered by Concato et al., assuming causality from a single study is more likely to result in an incorrect conclusion for RCTs.
Increased risk of inconsistent results for RCTs suggests higher prevalence of bias, which may arise due to many issues including design bias, conflicts of interest, treatment differences by physicians aware of allocation, attrition bias, ascertainment bias, randomization failures, errors, or fraud.
Using all studies identifies efficacy 8+ months faster (9+ months for low-cost treatments)
Currently, 59 of the treatments we analyze show statistically significant efficacy or harm, defined as ≥10% decreased risk or >0% increased risk from ≥3 studies. Of these, 54% have been confirmed in RCTs, with a mean delay of 7.8 months (62% with 8.7 months delay for low-cost treatments). The remaining treatments either have no RCTs, or the point estimate is consistent.
All studies must be carefully analyzed
Neither observational studies nor RCTs prove causation—any study can be flawed or fraudulent. We need much more, for example a combination of results from many independent teams, detailed understanding of each study, knowledge of conflicts/team reliability, dose-response relationships, delay-response relationships, logical results across outcomes, or details consistent with preclinical expectations.
All studies must be evaluated individually. RCTs for a given medication and disease may be more reliable, however they may also be less reliable. For off-patent medications, very high conflict of interest trials may be more likely to be RCTs, and more likely to be large trials that dominate meta-analyses.

Vitamin C COVID-19 Randomized Controlled Trials

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Thomas (RCT)−204%3.04 · 0.13–72.89death1/480/50COVIDAtoZ Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75
Early treatment30%0.70 · 0.54–0.9070/21046/12530% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.0059
Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT1 Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT1 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Majidi (DB RCT)14%0.86 · 0.74–1.01death26/3167/69ICU patients Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT1 Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Kumar (DB RCT)23%0.77 · 0.40–1.47death10/3013/30ICU patients Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200 Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT1 Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients
Late treatment19%0.81 · 0.70–0.9588/2,309163/1,94119% lower risk
Tau² = 0.02, I² = 31.2%, p = 0.0073
All studies20%0.80 · 0.69–0.92158/2,519209/2,06620% lower risk
Tau² = 0.02, I² = 30.9%, p = 0.001600.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 21. Random-effects meta-analysis for all Randomized Controlled Trials. This plot shows pooled effects, see the specific outcome analyses for individual outcomes. Analysis validating pooled outcomes for COVID-19 can be found below. Effect extraction is pre-specified, using the most serious outcome reported. For details see the appendix.

Vitamin C COVID-19 RCT mortality results

c19early.org
October 2026
StudyImprovementRR · 95% CITreatmentControlRelative Risk
Thomas (RCT)−204%3.04 · 0.13–72.891/480/50COVIDAtoZ
Early treatment−204%3.04 · 0.13–72.891/480/50204% higher risk
Tau² = 0.00, I² = 0.0%, p = 0.5
Zhang (RCT)50%0.50 · 0.20–1.506/2711/29ICU patients Kumari (RCT)36%0.64 · 0.26–1.557/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.563/303/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.515/409/40CT1 Tehrani (RCT)87%0.13 · 0.01–2.250/184/26 Majidi (DB RCT)14%0.86 · 0.74–1.0126/3167/69ICU patients Kumar (DB RCT)23%0.77 · 0.40–1.4710/3013/30ICU patients Labbani-.. (DB RCT)33%0.67 · 0.20–2.174/376/37 Rana (DB RCT)55%0.45 · 0.16–1.275/13911/139ICU patients Mousaviasl (DB RCT)20%0.80 · 0.32–1.988/20110/200 Adhikari (RCT)−19%1.19 · 0.98–1.461,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.195/227/25SAFE EVICT CORONA-ALIICU patients
Late treatment15%0.85 · 0.71–1.0279/2,143152/1,79715% lower risk
Tau² = 0.02, I² = 26.0%, p = 0.072
All studies14%0.86 · 0.72–1.0280/2,191152/1,84714% lower risk
Tau² = 0.02, I² = 22.4%, p = 0.08300.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 22. Random-effects meta-analysis for RCT mortality results.

Vitamin C COVID-19 RCT hospitalization results

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Thomas (RCT)31%0.69 · 0.12–3.98hosp.2/483/50COVIDAtoZ
Early treatment31%0.69 · 0.12–3.982/483/5031% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.7
Kumari (RCT)24%0.76 · 0.66–0.87hosp. time75 (n)75 (n) JamaliMo.. (RCT)−31%1.31 · 1.03–1.66hosp. time30 (n)30 (n) Hamidi-A.. (RCT)38%0.62 · 0.45–0.86hosp. time40 (n)40 (n)CT1 Hakamifard (RCT)1%0.99 · 0.91–1.07hosp. time38 (n)34 (n)CT1 Tehrani (RCT)18%0.82 · 0.60–1.13hosp. time18 (n)26 (n) Labbani-.. (DB RCT)−13%1.13 · 0.81–1.58hosp. time37 (n)37 (n) Rana (DB RCT)37%0.63 · 0.00–1358hosp. time139 (n)139 (n)ICU patients
Late treatment9%0.91 · 0.76–1.10377 (n)381 (n)9% lower risk
Tau² = 0.04, I² = 77.1%, p = 0.34
All studies9%0.91 · 0.76–1.092/4253/4319% lower risk
Tau² = 0.03, I² = 73.4%, p = 0.3100.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 23. Random-effects meta-analysis for RCT hospitalization results.
NIH
NIH provides an analysis of vitamin C for COVID-1998, concluding that there is insufficient evidence to recommend for or against use. However, they appear to have not examined the majority of the evidence. For example, considering RCTs providing clinical results for COVID-19 and vitamin C, they reference only99-104, and appear not to know about 14 other RCTs105-118 as shown in Fig. 25. Notably, the NIH selection does not correspond to the most relevant and highest quality studies, for example including Zhang et al., with very late treatment of ICU patients. Authors do not reference any of the 52 observational studies. For COVID-19, observational study results do not systematically differ from RCTs, RR 0.97 [0.91‑1.03] across 227 treatments92.

Vitamin C RCTs missing in NIH analysis

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Thomas (RCT)−204%3.04 · 0.13–72.89death1/480/50COVIDAtoZ Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75< STUDY MISSING >
Early treatment30%0.70 · 0.54–0.9070/21046/12530% lower risk
Tau² = 0.00, I² = 0.0%, p = 0.0059
Zhang (RCT)50%0.50 · 0.20–1.50death6/2711/29ICU patients Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30< STUDY MISSING > Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT1< STUDY MISSING > Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT1< STUDY MISSING > Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26< STUDY MISSING > Majidi (DB RCT)14%0.86 · 0.74–1.01death26/3167/69ICU patients< STUDY MISSING > Yang (RCT)33%0.67 · 0.55–0.81recov. time10 (n)10 (n)CT1< STUDY MISSING > Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n)< STUDY MISSING > Kumar (DB RCT)23%0.77 · 0.40–1.47death10/3013/30ICU patients< STUDY MISSING > Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37< STUDY MISSING > Rana (DB RCT)55%0.45 · 0.16–1.27death5/13911/139ICU patients< STUDY MISSING > Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200< STUDY MISSING > Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT1< STUDY MISSING > Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Fowler (DB RCT)19%0.81 · 0.30–2.19death5/227/25SAFE EVICT CORONA-ALIICU patients< STUDY MISSING >
Late treatment19%0.81 · 0.70–0.9588/2,309163/1,94119% lower risk
Tau² = 0.02, I² = 31.2%, p = 0.0073
All studies20%0.80 · 0.69–0.92158/2,519209/2,06620% lower risk
Tau² = 0.02, I² = 30.9%, p = 0.001600.511.52+
1 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
NIH ANALYSIS:
70% OF RCTs MISSING
Fig. 25. Analysis by NIH is missing 14 RCTs.
Unreported RCTs
6 vitamin C RCTs have not reported results119-124. The trials report a total of 1,420 patients, with 3 trials having actual enrollment of 602, and the remainder estimated. The results are delayed from 3 years to over 5 years.
Exclusions
To avoid bias in the selection of studies, we analyze all non-retracted studies. Here we show the results after excluding studies with major issues likely to alter results, non-standard studies, and studies where very minimal detail is currently available. Our bias evaluation is based on analysis of each study and identifying when there is a significant chance that limitations will substantially change the outcome of the study. We believe this can be more valuable than checklist-based approaches such as Cochrane GRADE, which can be easily influenced by potential bias, may ignore or underemphasize serious issues not captured in the checklists, and may overemphasize issues unlikely to alter outcomes in specific cases (for example certain specifics of randomization with a very large effect size and well-matched baseline characteristics).
The studies excluded are as below. Fig. 26 shows a forest plot for random-effects meta-analysis of all studies after exclusions.
Abdulateef, unadjusted results with no group details.
Coskun, very late stage, ICU patients.
Elhadi, unadjusted results with no group details; very late stage, ICU patients.
Fowler, very late stage, ICU patients.
Gadhiya, substantial unadjusted confounding by indication likely.
Gavrielatou, very late stage, ICU patients.
Guldemir, unadjusted results with no group details.
Holt, significant unadjusted confounding possible.
Jang, very late stage, ECMO patients.
Krishnan, unadjusted results with no group details.
Kumar (B), very late stage, ICU patients.
Li, very late stage, ICU patients.
Majidi, very late stage, ICU patients.
Mohseni, unadjusted results with no group details.
Mulhem, substantial unadjusted confounding by indication likely; substantial confounding by time likely due to declining usage over the early stages of the pandemic when overall treatment protocols improved dramatically.
Rahman, unadjusted results with no group details; significant unadjusted confounding possible.
Rana, very late stage, ICU patients.
Salehi, unadjusted results with no group details; very late stage, ICU patients.
Shehab, unadjusted results with no group details.
Suna, substantial confounding by time likely due to declining usage over the early stages of the pandemic when overall treatment protocols improved dramatically.
Thomas (B), unadjusted differences between groups.
Tu, unadjusted results with no group details.
Vishnuram, unadjusted results with no group details; minimal details of groups provided.
Yang (B), combined treatments may contribute significantly to the effect seen.
Yüksel, very late stage, ICU patients.
Zhang, very late stage, ICU patients.
Zhao, substantial confounding by time likely due to declining usage over the early stages of the pandemic when overall treatment protocols improved dramatically.
Zheng, substantial unadjusted confounding by indication likely; immortal-time bias may significantly affect results; treatment start times unknown, treatment may not have started at baseline.
Özgülteki̇n, very late stage, ICU patients.
Özgünay, substantial unadjusted confounding by indication likely; very late stage, ICU patients.

Vitamin C COVID-19 studies after exclusions

c19early.org
October 2026
StudyImprovementRR · 95% CIOutcomeTreatmentControlRelative Risk
Su−135%2.35 · 0.67–8.27progressionn/an/a Ried (RCT)31%0.69 · 0.54–0.89no recov.69/16246/75 Usanma Koban33%0.67 · 0.07–5.38viral+31 (n)95 (n)
Early treatment6%0.94 · 0.43–2.0869/19346/1706% lower risk
Tau² = 0.24, I² = 42.9%, p = 0.89
Patel29%0.71 · 0.43–1.14death22/9626/80 Kumari (RCT)36%0.64 · 0.26–1.55death7/7511/75 JamaliMo.. (RCT)0%1.00 · 0.22–4.56death3/303/30 Gao86%0.14 · 0.03–0.72death1/465/30 Hamidi-A.. (RCT)44%0.56 · 0.20–1.51death5/409/40CT2 Al Sulaiman (PSM)15%0.85 · 0.61–1.12death46/14259/142 Hakamifard (RCT)46%0.54 · 0.14–2.08ICU3/385/34CT2 Pourhoseingholi13%0.87 · 0.63–1.19death54/199285/2,269 Tan25%0.75 · 0.10–2.98death/int.1/4614/115CT2 Simsek44%0.56 · 0.23–1.35death6/5815/81 Tehrani (RCT)87%0.13 · 0.01–2.25death0/184/26 Baguma−48%1.48 · 0.41–4.70death385 (n)96 (n) Coppock (RCT)5%0.95 · 0.16–7.84progression4/442/22 Hess (PSW)20%0.80 · 0.40–1.60death10/2537/75 Izzo41%0.59 · 0.50–0.69recovery869 (n)521 (n)LINCOLNLONG COVID OT1 CT2 Fogleman (DB RCT)4%0.96 · 0.65–1.40recovery32 (n)34 (n) Doocy63%0.37 · 0.08–1.82death2/6422/80 Labbani-.. (DB RCT)33%0.67 · 0.20–2.17death4/376/37 Kyagambiddwa50%0.50 · 0.24–1.04death246 (all patients) Mousaviasl (DB RCT)20%0.80 · 0.32–1.98death8/20110/200 Seely (DB RCT)48%0.52 · 0.10–2.71progression2/424/44CT2 Adhikari (RCT)−19%1.19 · 0.98–1.46death1,303 (n)903 (n)REMAP-CAP Adhikari (DB RCT)28%0.72 · 0.45–1.17death190 (n)194 (n)LOVIT-COVID Corrao39%0.61 · 0.23–1.60death9/1046/42 Uz84%0.16 · 0.02–0.97death41 (n)46 (n) Dinoi−32%1.32 · 0.79–2.20deathcase-control study Bepouka76%0.24 · 0.08–0.72death185 (n)225 (n)
Late treatment25%0.75 · 0.63–0.89187/4,310523/5,44125% lower risk
Tau² = 0.07, I² = 51.7%, p = 0.0011
Behera18%0.82 · 0.45–1.57casescase-control study Louca0%1.00 · 0.97–1.04casespopulation-based cohort Mahto−26%1.26 · 0.63–2.28IgG+34/14059/549 Bejan34%0.66 · 0.29–1.53death569 (n)8,637 (n) Aldwihi36%0.64 · 0.45–0.86hosp.142/50595/233 Nimer25%0.75 · 0.54–1.04hosp.52/651167/1,497 Loucera28%0.72 · 0.58–0.88death840 (n)15,128 (n) Sharif46%0.54 · 0.01–0.92severe casen/an/a Asoudeh69%0.31 · 0.14–0.65severe case250 (all patients) Vaisi38%0.62 · 0.31–1.23hosp.2,818 (n)1,137 (n) Akbar14%0.86 · 0.65–1.14cases665 (n)9,335 (n) Guan31%0.69 · 0.50–0.86symp. case28/462,017/2,454
Prophylaxis26%0.74 · 0.62–0.89256/6,2342,338/38,97026% lower risk
Tau² = 0.06, I² = 82.3%, p = 0.0013
All studies25%0.75 · 0.67–0.85512/10,7372,907/44,58125% lower risk
Tau² = 0.06, I² = 71.3%, p < 0.000100.511.52+
1 OT: comparison with other treatment
2 CT: study uses combined treatment
Rotate screen for more detailsIncrease width for more details
← Vitamin C
reduces riskVitamin C
increases risk →
Fig. 26. Random-effects meta-analysis for all studies after exclusions. This plot shows pooled effects, see the specific outcome analyses for individual outcomes. Analysis validating pooled outcomes for COVID-19 can be found below. Effect extraction is pre-specified, using the most serious outcome reported. For details see the appendix.
Media Bias
Low-cost treatments were subject to bias and censorship during the pandemic. Scientific bias is seen in the design, analysis, presentation, and selective reporting of studies, which often favored negative results. A similar bias is seen in the media coverage for low-cost treatments. While broadly seen, bias was particularly notable for ivermectin and hydroxychloroquine, e.g., Scott Alexander noted that "if you say anything in favor of ivermectin you will be cast out of civilization and thrown into the circle of social hell reserved for Klan members and 1/6 insurrectionists. All the health officials in the world will shout 'horse dewormer!' at you and compare you to Josef Mengele."68.
We analyze media coverage for the 227 treatments we cover using Altmetric148 and PlumX Metrics149. Altmetric covers ~14,000 news outlets150 and PlumX covers ~55,000 news sources151. The two services are complementary with coverage of different news sources and different methods for identifying articles not containing a DOI. We consider studies to have received significant media coverage if they were covered by at least 100 articles in either Altmetric of PlumX151,152. The most widespread study was covered by 1,047 outlets. Studies covered by major syndicated sources shows hundreds of news articles, while studies covered by few outlets typically appear in niche medical news sources or press releases. Fig. 27 and 28 show bias towards negative results for low-cost treatments, in contrast to the opposite bias for high-profit treatments.
Fig. 29 shows that the mainstream media did not cover any of the positive studies for vitamin C.
The media bias may result in widespread incorrect perceptions on the relative efficacy of high-profit and low-cost treatments. The impact is significant—increased cost limits the use of high-profit treatments and treatment equity. High-profit treatments were also more difficult to access, especially for earlier treatment which improves efficacy and minimizes community transmission.
c19early.org
October 2026

Media coverage of low-cost COVID‑19 treatments

Media covered only 0.8% of studies showing significantly lower COVID-19 risk

Only 5 positive studies were covered

0.8%

one each: budesonide, cannabidiol, fluvoxamine, HCQ, vitamin D

34 negative studies were covered

4%

HCQ 12, ivermectin 6, lopinavir/ritonavir 4, azithromycin 3, vitamin D 3, fluvoxamine 2, one each: ibuprofen, metformin, vitamin C, zinc

5.4×
Negative bias for low-cost treatments
negative studies were 5.4 times more likely to be covered

Data from Altmetric and PlumX: studies receiving significant mainstream media coverage from 6,000+ studies for 227 treatments, defined as ≥100 news outlets in either Altmetric or PlumX, which cover ~14,000 and ~55,000 outlets respectively. Media coverage was significantly higher for negative studies, p < 0.001 (Fisher's exact test). Studies during the pandemic (until May 5, 2023).

Fig. 27. Mainstream media coverage was biased against positive results for low-cost treatments. When restricting to RCTs, the bias was even larger, with negative RCTs 6.9× more likely to be covered.
c19early.org
October 2026

Media coverage of high-profit COVID‑19 treatments

Media was 3.8× more likely to cover positive studies

19 positive studies were covered

11%

paxlovid 5, remdesivir 3, casirivimab/imdevimab 2, convalescent plasma 2, molnupiravir 2, one each: antiandrogens, peginterferon lambda, sargramostim, tixagevimab/cilgavimab, tocilizumab

Only 6 negative studies were covered

3%

remdesivir 3, one each: bamlanivimab/etesevimab, bebtelovimab, sotrovimab

3.8×
Positive bias for high-profit treatments
positive studies were 3.8 times more likely to be covered

Data from Altmetric and PlumX: studies receiving significant mainstream media coverage from 6,000+ studies for 227 treatments, defined as ≥100 news outlets in either Altmetric or PlumX, which cover ~14,000 and ~55,000 outlets respectively. Media coverage was significantly higher for positive studies, p = 0.002 (Fisher's exact test). Studies during the pandemic (until May 5, 2023).

Fig. 28. For high-profit treatments, mainstream media coverage was biased towards positive results. When restricting to RCTs, the bias was even larger, with positive RCTs 12.0× more likely to be covered.
c19early.org
October 2026

Media coverage of vitamin C for COVID‑19

Media covered none of the 32 studies showing significantly lower risk

No positive vitamin C studies were covered

0%

1 negative or inconclusive study was covered

2%

Data from Altmetric and PlumX: studies receiving significant mainstream media coverage from 72 studies for vitamin C treatment, defined as ≥100 news outlets in either Altmetric or PlumX, which cover ~14,000 and ~55,000 outlets respectively.

Fig. 29. Mainstream media did not cover any of the positive studies for vitamin C.
A combination of factors may have led to the very low media coverage of low-cost treatments:
•
Politicization
led to a media environment where coverage was often framed to support a political narrative rather than to provide objective scientific information. As Scott Alexander said:
"if you say anything in favor of ivermectin you will be cast out of civilization and thrown into the circle of social hell reserved for Klan members and 1/6 insurrectionists. All the health officials in the world will shout 'horse dewormer!' at you and compare you to Josef Mengele."
There was strong social pressure to discredit low-cost treatments.
•
Censorship
of information conflicting with selected authorities. For example, individuals and organizations presenting conflicting science were often shadowbanned or completely banned on Twitter and YouTube if the science they were reporting conflicted with the opinions of specific authorities selected by those companies, including when the authorities had not even reviewed the scientific results in question.
•
FDA requires "no adequate, approved, and available alternatives"
in order to grant an EUA for novel high-profit interventions, creating an incentive for authorities to ignore or downplay existing low-cost treatments.
•
Regulatory capture
biases authorities towards high-profit interventions.
•
Authorities ignored most evidence for low-cost treatments
, for example the NIH references only 2% of studies in delayed, rarely-updated, biased commentaries with no quantitive analysis.
•
Media coverage of science is often not very accurate
, e.g., misunderstanding confounding issues. For example the media widely considered the RECOVERY HCQ RCT to be conclusive on efficacy, but very late treatment of late-stage patients (mostly on oxygen already) with an excessive toxic dose (shown dangerous in a dose comparison RCT) provides no information on the recommended early/prophylactic treatment. With difficulting in understanding basic confounders like treatment delay and dose, the media may favor deferring to authorities. Many studies for low-cost treatments require greater expertise to analyze.
•
Substantial funding from pharmaceutical advertising
biases editorial decisions towards high-profit interventions.
•
PR power
- companies/teams with strong PR presence are favored in the media, which correlates with high-profit and high conflict of interest studies.
•
The media was very negative in general
, inflating risk, and elevating fear and anxiety. A negative bias may improve ratings and revenue, increasing motivation to continue watching coverage.
Authority Delay
26 low-cost treatments were approved in one or more countries, yet many countries approved no low-cost treatments. The countries that did adopt low-cost treatments analyzed the evidence early and made timely approvals. With few exceptions, authorities did not change their initial views, regardless of how much evidence accumulated showing either efficacy or harm. Why?
The harms of smoking here hidden for 25 yearsC. Authorities did not analyze the data in real-time, failing to act when harm was known. Widespread acknowledgement of harm came only after attempts by two new surgeon generals, along with pressure from health advocates and a new president, and a review of 7,000 studies.
Similarly for COVID-19, most authorities and experts did not proactively analyze data in real-time. This guarantees delayed recognition of efficacy or harm, by which time moral, legal, career, and reputational liabilities strongly disincentivize any admission of error. Claims of no efficacy (for effective treatments) or safety (for harmful treatments) were often made prior to strong data being available. Correction would require admitting to errors that increased mortality, which is unlikely with the same generation of officials.
Analysis of potential treatments was rarely done, and when done these were typically minimal efforts. For example, NIH reviews were highly delayed, cover only a tiny fraction of treatments, reference only 2% of studies for the treatments covered, and include no quantitative analysis. They appear as rarely updated side projects from external panels implicitly tasked with justifying prior failures. As with smoking, the thousands of studies could (and should) have been analyzed and acted on in real-time.
A key structural improvement, applicable to all current and future diseases, is for authorities to implement real-time proactive analysis of clinical evidence. This does not remove all bias, but does make it possible to act on evidence, whereas delayed action may be unlikely due to moral, legal, career, and reputational liabilities.
c19early.org
Delayed public health acknowledgments
Official acknowledgment of efficacy or harm is often delayed—legal, career, and status risks disincentivize admission of error.
Evidence Official Acknowledgment Approx. Delay
Citrus Fruit (vitamin C) for Scurvy (effectiveness) 1747: James Lind conducted one of the first-ever controlled clinical trials, proving that oranges and lemons cured scurvy in sailors. 1795: The British Royal Navy finally made a daily ration of lemon juice a standard issue for all its sailors, effectively eliminating the disease. 48 years
Handwashing (lower mortality) 1847: Dr. Ignaz Semmelweis provided conclusive proof that having doctors wash their hands with a chlorine solution before delivering babies reduced maternal mortality rates from over 18% to around 1%. ~1870s: Semmelweis's findings were rejected and he was ridiculed. His work was only validated decades later (after his death). ~20+ years
Helicobacter pylori (bacteria causes ulcers) 1982-1984: Marshall and Warren discovered that Helicobacter pylori bacteria causes ulcers, confirmed via direct exposure. Officials maintained that ulcers were caused by stress and spicy food. 1994: The US NIH released a consensus statement officially recommending antibiotics as the standard treatment for peptic ulcers, overturning decades of acid-suppression therapy. ~12 years
Asbestos (causes asbestosis & cancer) 1924: The British Medical Journal published the first case study of a death from "asbestosis." By 1918, U.S. insurance companies had stopped selling life insurance to asbestos workers. 1971 (US): The Occupational Safety and Health Administration (OSHA) was formed and began regulating asbestos as a carcinogen, setting the first federal workplace safety standards for it. ~47 years
Leaded Gasoline (neurotoxicity) ~1924: Dangers of low-level lead exposure were known. Experts like Alice Hamilton warned the U.S. Surgeon General that adding lead to gasoline would cause widespread public poisoning. 1973 (US): The Environmental Protection Agency (EPA) ordered the first phasedown of lead in gasoline, following the Clean Air Act of 1970. A full ban for on-road vehicles took effect in 1996. ~49 years
Harms of Smoking (causes lung cancer) 1939: Franz Müller (Germany) published the first case-control epidemiological study strongly linking tobacco smoking to lung cancer. This was followed by major U.S. & U.K. studies in the 1950s. 1964 (US): The U.S. Surgeon General's report, "Smoking and Health," was released. It was the first U.S. government report to definitively link smoking to lung cancer and heart disease. 25 years
Heterogeneity
Heterogeneity in COVID-19 studies arises from many factors including:
Treatment delay
The time between infection or the onset of symptoms and treatment may critically affect how well a treatment works. For example an antiviral may be very effective when used early but may not be effective in late stage disease, and may even be harmful. Oseltamivir, for example, is generally only considered effective for influenza when used within 0-36 or 0-48 hours153,154. Baloxavir marboxil studies for influenza also show that treatment delay is critical — Ikematsu et al. report an 86% reduction in cases for post-exposure prophylaxis, Hayden et al. show a 33 hour reduction in the time to alleviation of symptoms for treatment within 24 hours and a reduction of 13 hours for treatment within 24-48 hours, and Kumar (C) et al. report only 2.5 hours improvement for inpatient treatment.
Table 3. Studies of baloxavir marboxil for influenza show that early treatment is more effective.
Treatment delayResult
Post-exposure prophylaxis86% fewer cases155
<24 hours-33 hours symptoms156
24-48 hours-13 hours symptoms156
Inpatients-2.5 hours to improvement157
Fig. 30 shows a mixed-effects meta-regression for efficacy as a function of treatment delay in COVID-19 studies from 227 treatments, showing that efficacy declines rapidly with treatment delay. Early treatment is critical for COVID-19.
Loading..
Fig. 30. Early treatment is more effective. Meta-regression showing efficacy as a function of treatment delay in COVID-19 studies from 227 treatments.
Patient demographics
Details of the patient population including age and comorbidities may critically affect how well a treatment works. For example, many COVID-19 studies with relatively young low-comorbidity patients show all patients recovering quickly with or without treatment. In such cases, there is little room for an effective treatment to improve results, for example as in López-Medina et al.
SARS-CoV-2 variants
Efficacy may depend critically on the distribution of SARS-CoV-2 variants encountered by patients. Risk varies significantly across variants159, for example the Gamma variant shows significantly different characteristics160-163. Different mechanisms of action may be more or less effective depending on variants, for example the degree to which TMPRSS2 contributes to viral entry can differ across variants164,165.
Treatment regimen
Effectiveness may depend strongly on the dosage and treatment regimen.
Medication quality
The quality of medications may vary significantly between manufacturers and production batches, which may significantly affect efficacy and safety. Williams et al. analyze ivermectin from 11 different sources, showing highly variable antiparasitic efficacy across different manufacturers. Xu (B) et al. analyze a treatment from two different manufacturers, showing 9 different impurities, with significantly different concentrations for each manufacturer. Non-prescription supplements may show very wide variations in quality4,5.
Other treatments
The use of other treatments may significantly affect outcomes, including supplements, other medications, or other interventions such as prone positioning. Treatments may be synergistic168-192, therefore efficacy may depend strongly on combined treatments.
Effect measured
Across all studies there is a strong association between different outcomes, for example improved recovery is strongly associated with lower mortality. However, efficacy may differ depending on the effect measured, for example a treatment may be more effective against secondary complications and have minimal effect on viral clearance.
Meta-analysis
The distribution of studies will alter the outcome of a meta-analysis. Consider a simplified example where everything is equal except for the treatment delay, and effectiveness decreases to zero or below with increasing delay. If there are many studies using very late treatment, the outcome may be negative, even though early treatment is very effective. All meta-analyses combine heterogeneous studies, varying in population, variants, and potentially all factors above, and therefore may obscure efficacy by including studies where treatment is less effective. Generally, we expect the estimated effect size from meta-analysis to be less than that for the optimal case. Looking at all studies is valuable for providing an overview of all research, important to avoid cherry-picking, and informative when a positive result is found despite combining less-optimal situations. However, the resulting estimate does not apply to specific cases such as early treatment in high-risk populations. While we present results for all studies, we also present treatment time and individual outcome analyses, which may be more informative for specific use cases.
Pooled Effects
Pooled effects are no longer required to show efficacy as of September 2020
This section validates the use of pooled effects for COVID-19, which enables earlier detection of efficacy, however pooled effects are no longer required for vitamin C as of September 2020. Efficacy is now known based on specific outcomes.
Combining studies is required
For COVID-19, delay in clinical results translates into additional death and morbidity, as well as additional economic and societal damage. Combining the results of studies reporting different outcomes is required. There may be no mortality in a trial with low-risk patients, however a reduction in severity or improved viral clearance may translate into lower mortality in a high-risk population. Different studies may report lower severity, improved recovery, and lower mortality, and the significance may be very high when combining the results. "The studies reported different outcomes" is not a good reason for disregarding results. Pooling the results of studies reporting different outcomes allows us to use more of the available information. Logically we should, and do, use additional information when evaluating treatments—for example dose-response and treatment delay-response relationships provide additional evidence of efficacy that is considered when reviewing the evidence for a treatment.
Specific outcome and pooled analyses
We present both specific outcome and pooled analyses. In order to combine the results of studies reporting different outcomes we use the most serious outcome reported in each study, based on the thesis that improvement in the most serious outcome provides comparable measures of efficacy for a treatment. A critical advantage of this approach is simplicity and transparency. There are many other ways to combine evidence for different outcomes, along with additional evidence such as dose-response relationships, however these increase complexity.
Ethical and practical issues limit high-risk trials
Trials with high-risk patients may be restricted due to ethics for treatments that are known or expected to be effective, and they increase difficulty for recruiting. Using less severe outcomes as a proxy for more serious outcomes allows faster and safer collection of evidence.
Validating pooled outcome analysis for COVID-19
For many COVID-19 treatments, a reduction in mortality logically follows from a reduction in hospitalization, which follows from a reduction in symptomatic cases, which follows from a reduction in PCR positivity. We can directly test this for COVID-19.
Analysis of the the association between different outcomes across studies from all 227 treatments we cover confirms the validity of pooled outcome analysis for COVID-19. Fig. 31 shows that lower hospitalization is very strongly associated with lower mortality (p < 0.0000000001). Similarly, Fig. 32 shows that improved recovery is very strongly associated with lower mortality (p < 0.0000000001). Considering the extremes, Singh et al. show an association between viral clearance and hospitalization or death, with p = 0.003 after excluding one large outlier from a mutagenic treatment, and based on 44 RCTs including 52,384 patients. Fig. 33 shows that improved viral clearance is strongly associated with fewer serious outcomes. The association is very similar to Singh et al., with higher confidence due to the larger number of studies. As with Singh et al., the confidence increases when excluding the outlier treatment, from p = 0.0000000074 to p = 0.00000000015.
Loading..
Fig. 31. Lower hospitalization is associated with lower mortality, supporting pooled outcome analysis.
Loading..
Fig. 32. Improved recovery is associated with lower mortality, supporting pooled outcome analysis.
Loading..
Fig. 31. Improved viral clearance is associated with fewer serious outcomes, supporting pooled outcome analysis.
Pooled outcomes identify efficacy 5 months faster (7 months for RCTs)
Currently, 59 of the treatments we analyze show statistically significant efficacy or harm, defined as ≥10% decreased risk or >0% increased risk from ≥3 studies. 85% of these have been confirmed with one or more specific outcomes, with a mean delay of 4.6 months. When restricting to RCTs only, 53% of treatments showing statistically significant efficacy/harm with pooled effects have been confirmed with one or more specific outcomes, with a mean delay of 7.5 months. Fig. 34 shows when treatments were found effective during the pandemic. Pooled outcomes often resulted in earlier detection of efficacy.
Loading..
Loading..
Fig. 34. The time when studies showed that treatments were effective, defined as statistically significant improvement of ≥10% from ≥3 studies. Pooled results typically show efficacy earlier than specific outcome results. Results from all studies often shows efficacy much earlier than when restricting to RCTs. Results reflect conditions as used in trials to date, these depend on the population treated, treatment delay, and treatment regimen.
Limitations
Pooled analysis could hide efficacy, for example a treatment that is beneficial for late stage patients but has no effect on viral clearance may show no efficacy if most studies only examine viral clearance. In practice, it is rare for a non-antiviral treatment to report viral clearance and to not report clinical outcomes; and in practice other sources of heterogeneity such as differences in treatment delay are more likely to hide efficacy.
Summary
Analysis validates the use of pooled effects and shows significantly faster detection of efficacy on average. However, as with all meta-analyses, it is important to review the different studies included. We also present individual outcome analyses, which may be more informative for specific use cases.
Kidney Stones
High doses of vitamin C may increase the risk of kidney stones7, with doses exceeding around 1,000 mg per day elevating risk, and a clearer association with over 2,000 mg per day. Vitamin C metabolization produces oxalate, a key component of calcium oxalate kidney stones, which are the most common type. The risk depends on individual predisposition (e.g., history of kidney stones, dehydration, or metabolic conditions), total oxalate intake from the diet, and hydration levels.
Risk depends on the formulation. Liposomal vitamin C enhances absorption and may allow lower effective doses, potentially reducing oxalate production compared to equivalent doses of ascorbic acid. Time-release formulations spread absorption over time, which may lower peak oxalate levels. Natural sources (e.g., fruit) also contain other compounds (like citrate in citrus fruits) that may inhibit stone formation. Dietary intake is less commonly associated with increased risk. Risk may also be reduced with hydration—adequate water intake dilutes urinary oxalate and calcium, reducing stone formation.
Polytherapy
There are many complementary mechanisms of action for COVID-19 treatments, with over 500 therapeutic targets73. Studies show complementary and synergistic effects with polytherapy168-192. For example, Jitobaom et al.169 shows >10x reduction in IC50 with ivermectin and niclosamide, an RCT by Said et al.176 showed the combination of nigella sativa and vitamin D was more effective than either alone, and an RCT by Wannigama et al.194 showed improved results with fluvoxamine combined with additional treatments, compared to fluvoxamine alone.
SARS-CoV-2 can rapidly acquire mutations altering infectivity, disease severity, and drug resistance even without selective pressure195-202. Antigenic drift can undermine more variant-specific treatments like monoclonal antibodies and more specific antivirals. Treatment with targeted antivirals may select for escape mutations203. The efficacy of treatments varies depending on cell type204 due to differences in viral receptor expression, drug distribution and metabolism, and cell-specific mechanisms. Efficacy may also vary based on genetic variants205-215.
Variable efficacy across variants, cell types, tissues, and host genetics, along with the complementary and synergistic actions of different treatments, all point to greater efficacy with polytherapy. In many studies, the standard of care given to all patients includes other treatments—efficacy seen in these trials may rely in part on synergistic effects. Less variant specific treatments and polytherapy targeting multiple viral and host proteins may be more effective. Meta-analysis of all early treatment trials shows 71% [60‑79%] lower risk for studies using combined treatments, compared to 27% [24‑30%] for single treatments.
Polytherapy vs. monotherapy for all treatments
Monotherapy Polytherapy
Early treatment
27% [24–30%]
71% [60–79%]
Late treatment
18% [16–20%]
49% [41–55%]
Prophylaxis
28% [25–30%]
90% [48–98%]
Polytherapy (combined treatment) is more effective than monotherapy. Analysis of results across 227 treatments. p = 0.000000057, p < 0.0000000001, and p = 0.02.
Studies where combined treatments may significantly contribute to efficacy are excluded from this meta-analysis216-219. While not reaching statistical significance for all stages, vitamin C studies using polytherapy also show greater efficacy than monotherapy studies. This comparison includes combined treatment studies excluded from the rest of this paper216-219.
Polytherapy vs. monotherapy for vitamin C
Monotherapy Polytherapy
Early treatment
34% [10–52%]
52% [-123–91%]
Late treatment
18% [8–27%]
47% [32–58%]
Prophylaxis
19% [6–30%]
36% [4–57%]
Polytherapy (combined treatment) is more effective than monotherapy. Analysis of results for vitamin C studies. p = 0.71, p = 0.0013, and p = 0.28.
c19early.org
Combined treatment is more effective
SARS-CoV-2 involves the complex interplay of 500+ host and viral proteins and factors, providing many therapeutic targets. Many compounds have antiviral activity for SARS-CoV-2, with many different mechanisms including blocking attachment, entry, and replication. Combinations of treatments, with careful attention to potential side effects, allows improved efficacy via complementary and synergistic mechanisms.
Tissue and cell coverage Different compounds have different tissue penetration profiles and efficacy across cell types
Variant coverage Compounds with different viral resistance profiles
Intra/extracellular Disabling and removing intracellular and extracellular viral particles
Resistance Minimizing persistence and emergence of resistant variants
Genetics Robustness against individual variations in efficacy based on genetics
Lower doses Potentially lower doses of individual agents, reducing toxicity
Viral and host-directed Antivirals targeting viral and host proteins
Immune system function Supporting or enhancing natural immune system function
Disease phases Addressing multiple disease phases (viral replication, inflammation, secondary complications)
Discussion
Results for other infections
Studies have also shown efficacy with vitamin C for the common cold65,66 and acute respiratory tract infections67.
Publication bias
Publishing is often biased towards positive results, however evidence suggests that there may be a negative bias for inexpensive treatments for COVID-19. Both negative and positive results are very important for COVID-19, media in many countries prioritizes negative results for inexpensive treatments (inverting the typical incentive for scientists that value media recognition), and there are many reports of difficulty publishing positive results220-223.
One method to evaluate bias is to compare prospective vs. retrospective studies. Prospective studies are more likely to be published regardless of the result, while retrospective studies are more likely to exhibit bias. For example, researchers may perform preliminary analysis with minimal effort and the results may influence their decision to continue. Retrospective studies also provide more opportunities for the specifics of data extraction and adjustments to influence results.
Fig. 35 shows a scatter plot of results for prospective and retrospective treatment studies. Prospective studies show 20% [8‑31%] improvement in meta-analysis, compared to 21% [13‑29%] for retrospective studies, showing no significant difference.
Loading..
Fig. 35. Prospective vs. retrospective studies. The diamonds show the results of random-effects meta-analysis.
0% 25% 50% 75% 100% Studies usingearly treatment c19early.org October 2026 92% Regdanvimab 89% Sotrovimab 83% Molnupiravir 79% Casirivimab/im.. 73% Paxlovid 11% HCQ 11% Vitamin C 13% Melatonin 15% Alkalinization 16% Vitamin D 21% Azvudine 23% Nitric Oxide 24% Zinc 27% Budesonide ↑ Mostly early treatment ↓ Mostly late treatment
Fig. 36. Early treatment was more common for high-profit drugs.
Late treatment bias
Studies for vitamin C were mostly late treatment studies, in contrast with typical high-profit drugs that were more likely to be tested with early treatment.
Funnel plots have traditionally been used for analyzing publication bias. This is invalid for COVID-19 acute treatment trials — the underlying assumptions are invalid, which we can demonstrate with a simple example. Consider a set of hypothetical perfect trials with no bias. Fig. 37 plot A shows a funnel plot for a simulation of 80 perfect trials, with random group sizes, and each patient's outcome randomly sampled (10% control event probability, and a 30% effect size for treatment). Analysis shows no asymmetry (p > 0.05). In plot B, we add a single typical variation in COVID-19 treatment trials — treatment delay. Consider that efficacy varies from 90% for treatment within 24 hours, reducing to 10% when treatment is delayed 3 days. In plot B, each trial's treatment delay is randomly selected. Analysis now shows highly significant asymmetry, p < 0.0001, with six variants of Egger's test all showing p < 0.05224-231. Note that these tests fail even though treatment delay is uniformly distributed. In reality treatment delay is more complex — each trial has a different distribution of delays across patients, and the distribution across trials may be biased (e.g., late treatment trials may be more common). Similarly, many other variations in trials may produce asymmetry, including dose, administration, duration of treatment, differences in SOC, comorbidities, age, variants, and bias in design, implementation, analysis, and reporting.
Log Risk Ratio Standard Error 1.406 1.055 0.703 0.352 0 -3 -2 -1 0 1 2 A: Simulated perfect trials p > 0.05 Log Risk Ratio Standard Error 1.433 1.074 0.716 0.358 0 -4 -3 -2 -1 0 1 2 B: Simulated perfect trials with varying treatment delay p < 0.0001
Fig. 37. Example funnel plot analysis for simulated perfect trials.
Conflicts of interest
Pharmaceutical drug trials often have conflicts of interest whereby sponsors or trial staff have a financial interest in the outcome being positive. Vitamin C for COVID-19 lacks this because it is an inexpensive and widely available supplement. In contrast, most COVID-19 vitamin C trials have been run by physicians on the front lines with the primary goal of finding the best methods to save human lives and minimize the collateral damage caused by COVID-19. While pharmaceutical companies are careful to run trials under optimal conditions (for example, restricting patients to those most likely to benefit, only including patients that can be treated soon after onset when necessary, and ensuring accurate dosing), not all vitamin C trials represent the optimal conditions for efficacy.
High use by healthcare professionals, and with higher education
Treatments like vitamin C, D, and zinc were often not recommended by health authorities in contradiction to the clinical evidence. However use was common, with increased prevalence for healthcare professionals232-234 (and specifically those with more experience235), more highly educated individuals233,236-243, and with higher income235,238,241,242. Healthcare professionals also had lower risk than expected for COVID-19244,245, consistent with greater use of effective treatments.
Physician case series results
Table 4 shows the reported results of physicians that used combined early treatment protocols for COVID-19, compared to results for physicians following guidelines focusing on late treatment (Dr. Uip reportedly prescribed early treatment for himself, but not for patients246). The protocols vary, but include ivermectin, HCQ, vitamin D, vitamin C, zinc, melatonin, quercetin, budesonide, and other treatments. These results are subject to selection and ascertainment bias and more accurate analysis requires details of the patient populations and follow-up, however results are consistently better across many teams, and consistent with the extensive controlled trial evidence that shows a significant reduction in risk with many early treatments, and improved results with the use of multiple treatments in combination.

Physician-reported outcomes: early vs. late treatment of COVID-19

c19early.org
October 2026

0.1% mortality with early treatment vs. 6.2% with late treatment (49× lower)

Early treatment0.1%average mortality2.2% hospitalized
Late treatment6.2%average mortality39% hospitalized
PhysicianCountryPatientsHospitalizedDied
Early treatmenttreatment on presentation
Dr. Roberto Alfonso AccinelliPeru1,26570.6%0/360 deaths for treatment within 3 days
Dr. Mohammed Tarek AlamBangladesh10000%patients up to 84 years old
Dr. Oluwagbenga AlongeNigeria31000%
Dr. Raja BhattacharyaIndia14821.4%up to 88yo, 81% comorbidities
Dr. Flavio CadegianiBrazil3,45040.1%00%
Dr. Alessandro CapucciItaly350164.6%
Dr. Shankara ChettySouth Africa8,00000%
Dr. Deborah ChisholmUSA10000%
Dr. Ryan ColeUSA40000%00%
Dr. Marco CosentinoItaly392256.4%10.3%earlier treatment results were better
Dr. Jeff DavisUSA6,00000%
Dr. DhanajayIndia50000%
Dr. Bryan Tyson & Dr. George FareedUSA20,00060.0%40.0%
Dr. Raphael FurtadoBrazil17010.6%00%
Rabbi Yehoshua GerziIsrael86010.1%00%
Dr. Heather GesslingUSA1,50010.1%
Dr. Ellen GuimarãesBrazil50081.6%20.4%
Dr. Syed HaiderUSA4,00050.1%00%
Dr. Mark HancockUSA2400%
Dr. Sabine HazanUSA1,00000%
Dr. Mollie JamesUSA3,500401.1%10.0%
Dr. Roberta LacerdaBrazil55081.5%20.4%
Dr. Katarina LindleyUSA10055.0%00%
Dr. Ben MarbleUSA150,00040.0%
Dr. Edimilson MigowskiBrazil2,00070.3%20.1%
Dr. Abdulrahman MohanaSaudi Arabia2,73300%
Dr. Carlos NigroBrazil5,000450.9%230.5%
Dr. Benoit OchsLuxembourg80000%
Dr. OrtoreItaly24031.2%00%
Dr. Valerio PascuaHonduras415266.3%10.2%one patient already on oxygen died
Dr. Sebastian PopRomania30000%
Dr. Brian ProctorUSA869202.3%20.2%
Dr. Anastacio QueirozBrazil70000%
Dr. Didier RaoultFrance8,3152142.6%50.1%
Dr. Karin RiedTurkey23710.4%up to 99yo, 73% comorbidities
Dr. Roman RozencwaigCanada8000%patients up to 86 years old
Dr. Vipul ShahIndia8,00050.1%
Dr. Silvestre SobrinhoBrazil116108.6%00%
Dr. UnknownBrazil957161.7%20.2%
Dr. Vladimir ZelenkoUSA2,200120.5%20.1%
Early treatment236,1812.2%0.1%
Late treatmentno treatment until hospitalization
Dr. Jake Scott (**)USA1,00010010%
Dr. David Uip (*)Brazil2,20085039%542.5%
Late treatment3,20039%6.2%

Early-treatment physicians treated patients on presentation, while late-treatment physicians followed official guidance with treatment only on hospitalization. Groups are not randomized. Results are subject to selection and ascertainment bias and accurate analysis requires details of the patient populations and followup, however the results are consistent across many teams, and consistent with the extensive controlled clinical evidence showing a significant reduction in risk with many early treatments, and complementary/synergistic benefits with combined treatment.

Table 4. Physicians using early combined treatment protocols had much lower hospitalization and mortality rates compared with those following guidelines focusing on late treatment. Results are subject to selection and ascertainment bias and accurate analysis requires details of the patient populations and followup, however the results are consistent across many teams, and consistent with the extensive controlled clinical evidence showing a significant reduction in risk with many early treatments, and complementary/synergistic benefits with combined treatments. (*) Dr. Uip reportedly prescribed early treatment for himself, but not for patients246. (**) Dr. Scott reports treating hundreds of patients and losing over a hundred, but has not provided specific numbers247. Dr. Scott reports following (and helping create) US guidelines.
Limitations
Summary statistics from meta-analysis necessarily lose information. As with all meta-analyses, studies are heterogeneous, with differences in treatment delay, treatment regimen, patient demographics, variants, conflicts of interest, standard of care, and other factors. We provide analyses for specific outcomes and by treatment delay, and we aim to identify key characteristics in the forest plots and summaries. Results should be viewed in the context of study characteristics.
Some analyses classify treatment based on early or late administration, as done here, while others distinguish between mild, moderate, and severe cases. Viral load does not indicate degree of symptoms — for example patients may have a high viral load while being asymptomatic. With regard to treatments that have antiviral properties, timing of treatment is critical — late administration may be less helpful regardless of severity.
Details of treatment delay per patient is often not available. For example, a study may treat 90% of patients relatively early, but the events driving the outcome may come from 10% of patients treated very late. Our 5 day cutoff for early treatment may be too conservative, 5 days may be too late in many cases.
Comparison across treatments is confounded by differences in the studies performed, for example dose, variants, and conflicts of interest. Trials with conflicts of interest may use designs better suited to the preferred outcome.
In some cases, the most serious outcome has very few events, resulting in lower confidence results being used in pooled analysis, however the method is simpler and more transparent. This is less critical as the number of studies increases. Restriction to outcomes with sufficient power may be beneficial in pooled analysis and improve accuracy when there are few studies, however we maintain our pre-specified method to avoid any retrospective changes.
Studies show that combinations of treatments can be highly synergistic and may result in many times greater efficacy than individual treatments alone168-192. Therefore standard of care may be critical and benefits may diminish or disappear if standard of care does not include certain treatments.
This real-time analysis is constantly updated based on submissions. Accuracy benefits from widespread review and submission of updates and corrections from reviewers. Less popular treatments may receive fewer reviews.
No treatment or intervention is 100% available and effective for all current and future variants. Efficacy may vary significantly with different variants and within different populations. All treatments have potential side effects. Propensity to experience side effects may be predicted in advance by qualified physicians. We do not provide medical advice. Before taking any medication, consult a qualified physician who can compare all options, provide personalized advice, and provide details of risks and benefits based on individual medical history and situations.
Notes
1 of the 72 studies compare against other treatments, which may reduce the effect seen. 6 of 72 studies combine treatments. The results of vitamin C alone may differ. 4 of 20 RCTs use combined treatment. 7 other meta-analyses show significant improvements with vitamin C for mortality10-14, progression15, severity10,14, and cases16. Choi et al. show the highest bioavailability from fruit and vegetable juice compared with raw fruit and vegetables and vitamin C treatment, with the lowest bioavailability from vitamin C treatment.
Reviews
Many reviews cover vitamin C for COVID-19, presenting additional background on mechanisms and related results, including3,60,236,248-262.
Other studies
Additional preclinical or review papers suggesting potential benefits of vitamin C for COVID-19 include292-308. We have not reviewed these studies in detail.
Perspective
Results compared with other treatments
SARS-CoV-2 infection and replication involves a complex interplay of 500+ host and viral proteins and other factors47-54, providing many therapeutic targets. Over 12,000 compounds have been predicted to reduce COVID-19 risk55, either by directly minimizing infection or replication, by supporting immune system function, or by minimizing secondary complications. Fig. 38 shows an overview of the results for vitamin C in the context of multiple COVID-19 treatments, and Fig. 39 shows a plot of efficacy vs. cost for COVID-19 treatments.
Loading..
Fig. 38. Scatter plot showing results within the context of multiple COVID-19 treatments. Diamonds shows the results of random-effects meta-analysis. 0.5% of 12,000+ proposed treatments show efficacy309.
Loading..
Loading..
Fig. 39. Efficacy vs. cost for COVID-19 treatments.
Conclusion
Vitamin C is an effective treatment for COVID-19. Significantly lower risk is seen for mortality, ICU admission, hospitalization, and recovery. 25 studies from 25 independent teams in 12 countries show significant benefit. Meta-analysis using the most serious outcome reported shows 21% [14‑27%] lower risk. Results are similar for Randomized Controlled Trials, higher quality studies, and peer-reviewed studies. Clinical outcomes suggest benefit while viral and case outcomes do not, consistent with an intervention that aids the immune system or recovery but may have limited antiviral effects. 2 sufficiency studies analyze outcomes based on serum levels, showing 55% [-10‑81%] lower risk for patients with higher vitamin C levels.
Results are robust—in worst case exclusion sensitivity analysis 29 of 72 studies must be excluded before statistical significance is lost.
The European Food Safety Authority has found evidence for a causal relationship between the intake of vitamin C and optimal immune system function1,2.
Early cessation of high-dose IV treatment may result in a detrimental rebound effect3. Ongoing treatment is more effective than early cessation: 33% [22‑42%] vs. 16% [-31‑46%].
7 other meta-analyses show significant improvements with vitamin C for mortality10-14, progression15, severity10,14, and cases16.
Studies have also shown efficacy with vitamin C for the common cold65,66 and acute respiratory tract infections67.
High doses may increase the risk of kidney stones7, with risk depending on formulation, predisposition, diet, and hydration8. Choi et al. show the highest bioavailability from fruit and vegetable juice compared with raw fruit and vegetables and vitamin C treatment, with the lowest bioavailability from vitamin C treatment.
 
Contact.
Contact us on X at @CovidAnalysis.
Funding.
We have received no funding or compensation in any form, and do not accept donations. This is entirely volunteer work.
Conflicts of interest.
We have no conflicts of interest. We have no affiliation with any pharmaceutical companies, supplement companies, governments, political parties, or advocacy organizations.
AI.
We use AI models (Gemini, Grok, Claude, and ChatGPT) tasked with functioning as additional peer-reviewers to check for errors, suggest improvements, and review spelling and grammar. Any corrections are manually verified. Our preference for em dashes is independent of AI.
Updates.
Our COVID-19 meta-analyses involve the extraction of over 229,000 datapoints from thousands of papers for 227 treatments. We thank the thousands of scientists, physicians, and other contributors that have provided updates, suggestions, feedback, and corrections. These are all welcome and can be submitted at https://c19early.org/cmeta.html.
Dedication.
This work is dedicated to top evidence-based physicians that worked tirelessly to analyze evidence and greatly reduce mortality and morbidity during the pandemic. In alphabetical order: Dr. Thomas J. Borody, Dr. Mary Talley Bowden, Dr. Flavio Cadegiani, Dr. Shankara Chetty, Dr. Ryan Cole, Dr. George Fareed, Dr. Sabine Hazan, Dr. Pierre Kory, Dr. Tess Lawrie, Dr. Robert Malone, Dr. Paul Marik, Dr. Peter McCullough, Dr. Didier Raoult, Dr. Harvey Risch, Dr. Jackie Stone, Dr. Brian Tyson, Dr. Joseph Varon, and Dr. Vladimir Zelenko.
Public domain.
This is a public domain work distributed in accordance with the Creative Commons CC0 1.0 Universal license, which dedicates the work to the public domain by waiving all rights worldwide under copyright law. You can distribute, remix, adapt, and build upon this work in any medium or format, including for commercial purposes, without asking permission. Referenced material and third-party images retain any original copyrights or restrictions. See: https://creativecommons.org/publicdomain/zero/1.0/.
Study Notes
Survey of 428 recovered COVID-19 patients in Iraq, showing fewer hospital visits for patients on prophylactic vitamin C or D. Hospitalization was lower for those on vitamin C, D, or zinc, without statistical significance.

The study observes that patients taking prophylactic vitamins C and D for less than 1 week had a higher proportion of hospital visits compared to those taking them for more than 2 weeks. This fails to account for protopathic bias: patients feeling the early onset of COVID-19 symptoms likely initiated supplement use right before diagnosis, placing them in the '<1 week' category with a higher baseline severity trajectory. Submit Corrections or Updates.
Late treatmentRCT · NCT04401150 · 384 patients · multinational
Vitamin C for COVID-19
LOVIT-COVID Adhikari et al. JAMA · October 2023
Outcome
Improvement
Relative Risk · 95% CI
Mortality, all LOVIT-COVID patients
28%0.72 · 0.45–1.17
Mortality...
28%
Mortality...
28%
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality with vitamin C
Not statistically significant · p = 0.19
REMAP-CAP results listed separately
Very late stage (APACHE II 8 and 12 for non-critical and critical) RCT with publication delayed over a year showing higher ventilation and no significant difference in mortality with vitamin C.

Authors have combined what was to be two separate trials into one trial, however there are very large differences between the trials. The results for each source trial are shown separately here103,104.

eTable 15 shows that results in LOVIT-COVID were substantially better than those in REMAP-CAP. eTable 13 shows improved survival for LOVIT-COVID and worse survival for REMAP-CAP (authors provide mortality breakdown only for hospital survival):

LOVIT-COVID shows 85% and 82% probability of superiority of vitamin C (critical and non-critical).
REMAP-CAP shows 12% and 7% probability of superiority of vitamin C.

Notably, LOVIT-COVID patients were blinded, while REMAP-CAP was open-label, introducing additional opportunity for bias for this highly politicized treatment. REMAP-CAP had more patients and dominates the combined results.

eFigure 8b also shows that the REMAP-CAP results were initially positive, switching to negative around September 2021. Authors note that they were unable to explain this reversal. The overall negative result is only due to the larger number of patients in the REMAP-CAP later time period.

Results for intubation are much worse than mortality, with statistically significant higher intubation for the treatment group. Hypothetically, if the actual risk matched other trials (~20% lower risk in meta analysis of 18 RCTs at the time), and there was something causing biased intubation of treatment patients in this mostly open-label trial, we may get the observed results whereby intubation is significantly worse due to the bias, but this has a muted effect on mortality which may reflect the change in risk due to intubation combined with that due to treatment.

Results varied dramatically over time. For example, during 22 Jan - 21 Apr 2021, the probability of superiority for vitamin C was 1.0 for critical and 0.97 for non-critical (eTable 17).

There were dramatic changes in randomization proportions and in overall clinical outcomes over time, leading to potential issues and inaccuracies in the attempted adjustment for confounding by time.

The very long delay between the end of the trial and publication also raises questions.

See also Hemilä et al. which shows that the poor results may be explained by a rebound effect due to the abrupt termination of treatment after 4 days.

NCT04401150 (LOVIT-COVID) and NCT02735707 (REMAP-CAP). 50mg/kg vitamin C administered intravenously over 30-60 minutes every 6 hours for 4 days. Submit Corrections or Updates.
Late treatmentRCT · NCT04401150 · 2,206 patients · multinational
Vitamin C for COVID-19
REMAP-CAP Adhikari et al. JAMA · October 2023
Outcome
Improvement
Relative Risk · 95% CI
Mortality, all REMAP-CAP patients
-19%1.19 · 0.98–1.46
Mortality...
-16%
Mortality...
-27%
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Higher mortality with vitamin C
Not statistically significant · p = 0.078
LOVIT-COVID results listed separately
Very late stage (APACHE II 8 and 12 for non-critical and critical) RCT with publication delayed over a year showing higher ventilation and no significant difference in mortality with vitamin C.

Authors have combined what was to be two separate trials into one trial, however there are very large differences between the trials. The results for each source trial are shown separately here103,104.

eTable 15 shows that results in LOVIT-COVID were substantially better than those in REMAP-CAP. eTable 13 shows improved survival for LOVIT-COVID and worse survival for REMAP-CAP (authors provide mortality breakdown only for hospital survival):

LOVIT-COVID shows 85% and 82% probability of superiority of vitamin C (critical and non-critical).
REMAP-CAP shows 12% and 7% probability of superiority of vitamin C.

Notably, LOVIT-COVID patients were blinded, while REMAP-CAP was open-label, introducing additional opportunity for bias for this highly politicized treatment. REMAP-CAP had more patients and dominates the combined results.

eFigure 8b also shows that the REMAP-CAP results were initially positive, switching to negative around September 2021. Authors note that they were unable to explain this reversal. The overall negative result is only due to the larger number of patients in the REMAP-CAP later time period.

Results for intubation are much worse than mortality, with statistically significant higher intubation for the treatment group. Hypothetically, if the actual risk matched other trials (~20% lower risk in meta analysis of 18 RCTs at the time), and there was something causing biased intubation of treatment patients in this mostly open-label trial, we may get the observed results whereby intubation is significantly worse due to the bias, but this has a muted effect on mortality which may reflect the change in risk due to intubation combined with that due to treatment.

Results varied dramatically over time. For example, during 22 Jan - 21 Apr 2021, the probability of superiority for vitamin C was 1.0 for critical and 0.97 for non-critical (eTable 17).

There were dramatic changes in randomization proportions and in overall clinical outcomes over time, leading to potential issues and inaccuracies in the attempted adjustment for confounding by time.

The very long delay between the end of the trial and publication also raises questions.

See also Hemilä et al. which shows that the poor results may be explained by a rebound effect due to the abrupt termination of treatment after 4 days.

NCT04401150 (LOVIT-COVID) and NCT02735707 (REMAP-CAP). 50mg/kg vitamin C administered intravenously over 30-60 minutes every 6 hours for 4 days. Submit Corrections or Updates.
Retrospective 10,000 adults in Qatar, showing lower risk of COVID-19 cases with vitamin C supplementation, without statistical significance. Authors do not analyze COVID-19 severity. Submit Corrections or Updates.
Retrospective 158 critically ill patients receiving vitamin C and propensity matched controls, showing mortality OR 0.77 [0.48-1.23], and statistically significantly lower thrombosis, OR 0.42 [0.18-0.94]. 1000mg of vitamin C was given daily. Submit Corrections or Updates.
Retrospective survey-based analysis of 738 COVID-19 patients in Saudi Arabia, showing lower hospitalization with vitamin C, turmeric, zinc, and nigella sativa, and higher hospitalization with vitamin D. For vitamin D, most patients continued prophylactic use. For vitamin C, the majority of patients continued prophylactic use. For nigella sativa, the majority of patients started use during infection. Authors do not specify the fraction of prophylactic use for turmeric and zinc. Submit Corrections or Updates.
Retrospective 250 recovered COVID-19 patients, showing lower risk of severe cases with higher vitamin C intake. Submit Corrections or Updates.
Retrospective COVID+ hospitalized patients in Uganda, 385 patients receiving vitamin C treatment, showing higher mortality with treatment, without statistical significance. Submit Corrections or Updates.
Retrospective matched case-control prophylaxis study for HCQ, ivermectin, and vitamin C with 372 healthcare workers, showing lower COVID-19 incidence for all treatments, with statistical significance reached for ivermectin.

HCQ OR 0.56, p = 0.29
Ivermectin OR 0.27, p < 0.001
Vitamin C OR 0.82, p = 0.58 Submit Corrections or Updates.
ProphylaxisRetrospective · 9,748 patients · USA
Vitamin C for COVID-19
Bejan et al. Clinical Pharmacology & Therapeutics · Feb 2021
Outcome
Improvement
Relative Risk · 95% CI
Mortality
34%0.66 · 0.29–1.53
Ventilation
25%0.75 · 0.35–1.62
ICU admission
15%0.85 · 0.43–1.69
Hospitalization
0%1.00 · 0.67–1.49
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality and ventilation
Not statistically significant · p = 0.33 (mortality) and p = 0.47 (ventilation)
Retrospective 9,748 COVID-19 patients in the USA showing lower risk of mortality, ventilation, and ICU admission with vitamin C prophylaxis, without statistical significance. Submit Corrections or Updates.
Retrospective 410 hospitalized COVID-19 patients in the Democratic Republic of Congo showing significantly lower mortality with vitamin C treatment. Submit Corrections or Updates.
Analysis of serum and intracellular vitamin C levels in hospitalized COVID-19 patients. Low vitamin C levels were common with 36% having serum levels <26 μmol/L and 15% <11 μmol/L.

Intracellular vitamin C levels in peripheral blood mononuclear cells (PBMCs) were low at admission and declined during hospitalization, suggesting ongoing utilization and depletion of vitamin C stores.

Critical patients had higher odds of low serum vitamin C levels. There was a weak negative correlation between serum vitamin C levels and severity, without statistical significance. Submit Corrections or Updates.
150 patient vitamin C early treatment RCT with results not reported over 3 years after completion. Submit Corrections or Updates.
Late treatmentRCT · 66 patients · USA
Vitamin C for COVID-19
Coppock et al. Life · March 2022
Outcome
Improvement
Relative Risk · 95% CI
Progression
5%0.95 · 0.16–7.84
Improvement
50%0.50 · 0.16–1.33
Discharge
22%0.78 · 0.59–1.02
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Greater improvement and higher discharge
Not statistically significant · p = 0.16 (improvement) and p = 0.071 (discharge)
RCT with 66 very late stage (8 days from symptom onset) hospitalized patients, 44 treated with vitamin C and 22 control patients, showing no significant differences with treatment. Submit Corrections or Updates.
Late treatmentProspective · NCT04323514 · 146 patients · Italy
Vitamin C for COVID-19
Corrao et al. J. Clinical Medicine · July 2024
Outcome
Improvement
Relative Risk · 95% CI
Mortality
39%0.61 · 0.23–1.60
Death/ICU
19%0.81 · 0.56–1.12
ICU admission
-102%2.02 · 0.46–8.83
Hospitalization time
25%0.75 · 0.50–1.12
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality and death/ICU
Not statistically significant · p = 0.37 (mortality) and p = 0.24 (death/ICU)
Prospective study of 146 hospitalized COVID-19 patients showing shorter hospitalization with high-dose intravenous vitamin C. 104 patients received 10g of vitamin C intravenously daily for 3 days and 42 patients received only standard care. Mortality was lower with treatment (8.7% vs 14.3%) without statistical significance. Treatment was associated with significantly shorter hospitalization in multivariable analysis (-4.95, p=0.041). No adverse events were reported in the vitamin C group. Submit Corrections or Updates.
ICU patientsRetrospective · 78 patients · Turkey · Mar – Jun 2020
Vitamin C for COVID-19
Coskun et al. SiSli Etfal Hastanesi Tip Bulteni / The Medical Bulle.. · Mar 2023
Outcome
Improvement
Relative Risk · 95% CI
Mortality
25%0.75 · 0.48–1.15
Ventilation
2%0.98 · 0.76–1.28
ICU time
0%1.00 · 1.00–1.00
SOFA score, @96 hours
28%0.72 · 0.57–0.90
no CI
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Improved recovery with vitamin C
p = 0.005
Retrospective 78 ICU patients in Turkey, showing lower mortality with high-dose vitamin C treatment, without statistical significance. The SOFA score was significantly better with treatment at day 4.

Authors incorrectly state that "HDVC treatment did not reduce the short-term mortality...". Mortality was lower with treatment, although not statistically significant given the sample size.

6g of vitamin C daily in 4 equal doses every 6h, for a total of 96h. Submit Corrections or Updates.
Retrospective 247 non-survivors and 247 matched survivors in hospitalized COVID-19 patients in Italy showing results for several treatments. Submit Corrections or Updates.
Prospective study of 144 hospitalized COVID-19 patients in the DRC and South Sudan, showing lower mortality with vitamin C treatment. Submit Corrections or Updates.
Prospective study of 465 COVID-19 ICU patients in Libya showing no significant differences with treatment. Submit Corrections or Updates.
Early terminated low-risk patient RCT with 32 low-dose vitamin C, 32 melatonin, and 34 placebo patients, showing faster resolution of symptoms with melatonin in spline regression analysis, and no significant difference for vitamin C. All patients recovered with no serious outcomes reported. Baseline symptoms scores were higher in the melatonin and vitamin C arms (median 27 and 24 vs. 18 for placebo). Submit Corrections or Updates.
RCT 47 ICU patients showing no significant differences with vitamin C treatment. Submit Corrections or Updates.
Retrospective 283 patients in the USA showing higher mortality with all treatments (not statistically significant). Confounding by indication is likely. In the supplementary appendix, authors note that the treatments were usually given for patients that required oxygen therapy. Oxygen therapy and ICU admission (possibly, the paper includes ICU admission for model 2 in some places but not others) were the only variables indicating severity used in adjustments. Submit Corrections or Updates.
Estimated 160 patient vitamin C late treatment RCT with results not reported over 4 years after estimated completion. Submit Corrections or Updates.
Retrospective 76 COVID-19 patients, 46 treated with intravenous high-dose vitamin C, showing lower mortality and improved oxygen requirements with treatment. Dosage was 6g intravenous infusion per 12hr on the first day, and 6g once for the following 4 days. Submit Corrections or Updates.
Retrospective 113 consecutive mechanically ventilated COVID+ ICU patients in Greece, 10 receiving high-dose IV vitamin C, showing lower mortality with treatment, without statistical significance (p=0.11).

The associated meta analysis includes only 11 studies, while there are currently 72 studies, 44 with mortality results. Authors only include critical patients, however not all studies with critical patients are included, for example108,112,143,147. The meta analysis also uses unadjusted results, while PSM, Cox proportional hazards, or KM results are reported by99,145,277,278. For99 authors use 28 day mortality, while the study reports longer term in-hospital mortality. Submit Corrections or Updates.
ProphylaxisRetrospective · 2,621 patients · China · Dec 2022 – Jan 2023
Vitamin C for COVID-19
Guan et al. American J. Medical Sciences · May 2024
Outcome
Improvement
Relative Risk · 95% CI
Symp. case, high dose
31%0.69 · 0.50–0.86
Symp. case, medium dose
18%0.82 · 0.68–0.94
Symp. case, low dose
7%0.93 · 0.84–1.01
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Fewer symptomatic cases with vitamin C
p = 0.0073
Retrospective 2,746 individuals in China showing significantly lower incidence of COVID-19 symptoms and fever with higher vitamin C intake, with a dose response relationship. Submit Corrections or Updates.
Retrospective 477 COVID+ public transportation workers in Turkey, showing lower risk of hospitalization with vitamin C use in unadjusted results. Submit Corrections or Updates.
Late treatmentRCT · 72 patients · Iran · Mar – Apr 2020
Vitamin C for COVID-19
Hakamifard et al. Immunopathologia Persa · April 2021
Outcome
Improvement
Relative Risk · 95% CI
ICU admission
46%0.54 · 0.14–2.08
Hospitalization time
1%0.99 · 0.91–1.07
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower ICU admission with treatment
Not statistically significant · p = 0.46
RCT with 38 patients treated with vitamin C and vitamin E, and 34 control patients, showing lower ICU admission with treatment, but not statistically significant. Submit Corrections or Updates.
Late treatmentRCT · NCT04370288 · 80 patients · Iran · Apr – Sep 2020
Vitamin C for COVID-19
Hamidi-Alamdari et al. Clinical and Translational Investigation · Mar 2021
Outcome
Improvement
Relative Risk · 95% CI
Mortality
44%0.56 · 0.20–1.51
Hospitalization time
38%0.62 · 0.45–0.86
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Shorter hospitalization with treatment
p = 0.004
RCT 80 hospitalized patients with severe COVID-19, 40 treated with methylene blue + vitamin C + N-acetylcysteine, showing lower mortality, shorter hospitalization, and significantly improved SpO2 and respiratory distress with treatment. Submit Corrections or Updates.
60 patient vitamin C late treatment RCT with results not reported over 5 years after completion. Submit Corrections or Updates.
Late treatmentRetrospective · 100 patients · USA · Mar – Jul 2020
Vitamin C for COVID-19
Hess et al. Internal and Emergency Medicine · March 2022
Lower mortality and ICU admission
Not statistically significant · p = 0.54 (mortality) and p = 0.11 (ICU admission)
Retrospective 100 severe condition hospitalized patients in the USA, 25 treated with high-dose IV vitamin C, showing lower mechanical ventilation and cardiac arrest, and increased length of survival with treatment. 3g IV vitamin C every 6h for 7 days. Submit Corrections or Updates.
Prospective survey-based study with 15,227 people in the UK, showing lower risk of COVID-19 cases with vitamin A, vitamin D, zinc, selenium, probiotics, and inhaled corticosteroids; and higher risk with metformin and vitamin C. Statistical significance was not reached for any of these. Except for vitamin D, the results for treatments we follow were only adjusted for age, sex, duration of participation, and test frequency. Submit Corrections or Updates.
Long COVIDProspective · 1,390 patients · Italy
Vitamin C for COVID-19
LINCOLN Izzo et al. Pharmacological Research · July 2022
Outcome
Improvement
Relative Risk · 95% CI
Recovery
41%0.59 · 0.50–0.69
Recovery (b)
68%0.33 · 0.19–0.57
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Improved recovery with vitamin C + L-arginine
p < 0.000001 Study compares with another combination of treatments
Long COVID trial comparing L-arginine + vitamin C with multivitamin treatment (vitamin B1, B2, B6, B12, nicotinamide, folic acid, pantothenic acid), showing significant improvement in symptoms with L-arginine + vitamin C treatment. Submit Corrections or Updates.
Late treatmentRCT · 60 patients · Iran · Apr – May 2020
Vitamin C for COVID-19
JamaliMoghadamSiahkali et al. Research Square · January 2021
Outcome
Improvement
Relative Risk · 95% CI
Mortality
0%1.00 · 0.22–4.56
Ventilation
-25%1.25 · 0.37–4.21
Hospitalization time
-31%1.31 · 1.03–1.66
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Longer hospitalization with vitamin C
p = 0.028
Small late stage RCT for the addition of vitamin C to HCQ and lopinavir/ritonavir, with 30 treatment and 30 control patients, finding a significant reduction in temperature and a significant improvement in oxygenation after 3 days in the vitamin C group. However, hospitalization time was longer and there was no significant difference in mortality. Submit Corrections or Updates.
Retrospective 19 COVID-19 ECMO patients in South Korea, showing a higher rate of weaning from ECMO with vitamin C treatment, without statistical significance. Authors perform multivariate analysis but do not provide full results, only reporting p > 0.05. Submit Corrections or Updates.
Retrospective 152 mechanically ventilated patients in the USA showing unadjusted lower mortality with vitamin C, vitamin D, HCQ, and zinc treatment, statistically significant only for vitamin C. Submit Corrections or Updates.
ICU patientsDouble-blind RCT · CTRI/2020/11/029230 · 60 patients · India
Vitamin C for COVID-19
Kumar et al. J. Family Medicine and Primary Care · Aug 2022
Outcome
Improvement
Relative Risk · 95% CI
Mortality
23%0.77 · 0.40–1.47
Ventilation
21%0.79 · 0.43–1.44
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality with vitamin C
Not statistically significant · p = 0.6
RCT 60 ICU patients in India, showing no significant difference in outcomes with vitamin C. Mortality was lower in the vitamin C arm despite having more severe cases at baseline (87% vs. 67%). 1 gram intravenous vitamin C 8 hourly for four days. Submit Corrections or Updates.
Late treatmentRCT · 150 patients · Pakistan · Mar – Jul 2020
Vitamin C for COVID-19
Kumari et al. Cureus · November 2020
Outcome
Improvement
Relative Risk · 95% CI
Mortality
36%0.64 · 0.26–1.55
Ventilation
20%0.80 · 0.40–1.59
Recovery time
26%0.74 · 0.64–0.86
Hospitalization time
24%0.76 · 0.66–0.87
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Faster recovery and shorter hospitalization
p = 0.0001 (recovery) and p = 0.0001 (hospitalization)
RCT 150 hospitalized patients in Pakistan showing 26% faster recovery, p < 0.0001. 36% lower mortality, not statistically significant due to the small number of events. Dosage was 50 mg/kg/day of intravenous vitamin C. Submit Corrections or Updates.
Retrospective 246 severe COVID-19 patients in Uganda, showing lower mortality with vitamin C treatment, without statistical significance (p = 0.06). Submit Corrections or Updates.
Late treatmentRCT · 74 patients · Iran · Apr – Nov 2020
Vitamin C for COVID-19
Labbani-Motlagh et al. J. Research in Pharmacy Practice · December 2022
Lower progression with vitamin C
Not statistically significant · p = 0.12
RCT 74 patients in Iran, showing no significant differences in outcomes with high-dose vitamin C treatment. Tables 1b and 2a show conflicting baseline SOFA scores. The percentages of patients receiving antiviral treatments and corticosteroids are switched between the text and Table 1b. Authors indicate ICU admission was an outcome, but the result is not provided. AKI was lower with treatment, though not reaching statistical significance. Submit Corrections or Updates.
392 patient vitamin C late treatment RCT with results not reported over 3 years after completion. The companion non-COVID trial NCT03680274 has reported results. Submit Corrections or Updates.
PSM retrospective 8 ICU patients treated with vitamin C and 24 matched controls, showing no significant difference. Authors note that "it is possible for the delayed timing of IV vitamin C to have blunted the beneficial effects as these patients may have already progressed to the late fibroproliferative phase or ARDS". IV vitamin C 1.5 grams every 6 hours. Submit Corrections or Updates.
Estimated 608 patient vitamin C late treatment RCT with results not reported over 3 years after estimated completion. Submit Corrections or Updates.
Survey analysis of dietary supplements showing no significant difference in PCR+ cases with vitamin C usage in the UK, however significant reductions were found in the US and Sweden. These results are for PCR+ cases only, they do not reflect potential benefits for reducing the severity of cases. A number of biases could affect the results, for example users of the app may not be representative of the general population, and people experiencing symptoms may be more likely to install and use the app. Submit Corrections or Updates.
Retrospective 15,968 COVID-19 hospitalized patients in Spain, showing lower mortality with existing use of several medications including metformin, HCQ, azithromycin, aspirin, vitamin D, vitamin C, and budesonide. Since only hospitalized patients are included, results do not reflect different probabilities of hospitalization across treatments. Submit Corrections or Updates.
Retrospective 689 healthcare workers in India, showing no significant difference in IgG positivity with vitamin C prophylaxis. Submit Corrections or Updates.
RCT 100 ICU patients in Iran, 31 treated with vitamin C, showing lower mortality with treatment. Submit Corrections or Updates.
Retrospective 603 patients in Iran, 34 taking vitamin C supplements, showing increased risk of COVID-19 cases in unadjusted results. IR.SHOUSHTAR.REC.1399.015. Submit Corrections or Updates.
Late treatmentDouble-blind RCT · 401 patients · Iran · Nov 2020 – May 2021
Vitamin C for COVID-19
Mousaviasl et al. Disease and Diagnosis · July 2023
Outcome
Improvement
Relative Risk · 95% CI
Mortality
20%0.80 · 0.32–1.98
Mortality, ICU
-99%1.99 · 0.18 to >9
Ventilation
-200%3.00 · 0.12 to >9
ICU admission
-33%1.33 · 0.47–3.75
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Trial underpowered to detect differences
RCT 401 hospitalized COVID-19 patients showing no significant differences with low-dose oral vitamin C (1000mg daily for 5 days). Submit Corrections or Updates.
Retrospective database analysis of 3,219 hospitalized patients in the USA. Very different results in the time period analysis (Table S2), and results significantly different to other studies for the same medications (e.g., heparin OR 3.06 [2.44-3.83]) suggest significant confounding by indication and confounding by time. Submit Corrections or Updates.
ProphylaxisRetrospective · 2,148 patients · Jordan · Mar – Jul 2021
Vitamin C for COVID-19
Nimer et al. Bosnian J. Basic Medical Sciences · Feb 2022
Outcome
Improvement
Relative Risk · 95% CI
Hospitalization
25%0.75 · 0.54–1.04
Severe case
17%0.83 · 0.62–1.09
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower hospitalization and severe cases
Not statistically significant · p = 0.08 (hospitalization) and p = 0.18 (severe cases)
Retrospective 2,148 COVID-19 recovered patients in Jordan, showing lower risk of severity and hospitalization with vitamin C prophylaxis, without statistical significance. Submit Corrections or Updates.
Retrospective 176 hospitalized patients, 96 treated with oral vitamin C (from 500mg to 1500mg daily), showing lower mortality with treatment. Submit Corrections or Updates.
Prospective study of 2,468 hospitalized COVID-19 patients in Iran, showing no significant difference with vitamin C treatment. IR.MUQ.REC.1399.013. Submit Corrections or Updates.
Retrospective 416 non-hospitalized and 184 hospitalized COVID-19 patients in Bangladesh, showing higher acetaminophen and lower vitamin C usage for hospitalized patients. Confounding may be significant and baseline details per treatment group are not provided, however fever and symptomatic patients were more common in the non-hospitalized group. Note there is an alignment mismatch in Table 1. Submit Corrections or Updates.
ICU patientsRCT · 278 patients · Pakistan · Dec 2020 – Apr 2022
Vitamin C for COVID-19
Rana et al. Biological and Clinical Sciences Research J. · Jun 2023
Outcome
Improvement
Relative Risk · 95% CI
Mortality
55%0.45 · 0.16–1.27
Ventilation
44%0.56 · 0.19–1.62
Hospitalization time
37%0.63 · 0.00 to >9
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality and ventilation
Not statistically significant · p = 0.2 (mortality) and p = 0.41 (ventilation)
RCT 278 COVID-19 ICU patients in Pakistan, showing lower mortality and ventilation, and shorter length of stay with high-dose vitamin C treatment, without statistical significance. 30 grams IV vitamin C for four days. Submit Corrections or Updates.
RCT 237 patients in Turkey, 162 treated with IV vitamin C in addition to HCQ/AZ/zinc/vitamin D used for all patients, showing significantly faster recovery with the addition of IV vitamin C.

97% of patients were vitamin D deficient, and lower vitamin D levels were associated with ICU admission and longer hospital stay.

Only 1 of 237 hospitalized patients died (average age 63, range 22-99) - a 70-year-old patient with heart and lung disease and severely deficient vitamin D levels (6 nmol/L). IV vitamin C (sodium ascorbate) was given as 50 mg/kg every six hours on day 1, followed by 100 mg/kg every six hours (four times daily, 400 mg/kg/day) for seven days. NCT04395768. Submit Corrections or Updates.
Retrospective 125 mechanically ventilated ICU patients in Iran, showing no significant difference with vitamin C treatment in unadjusted results. Submit Corrections or Updates.
Late treatmentRCT · 90 patients · Canada · Sep 2021 – Apr 2022
Vitamin C for COVID-19
Seely et al. BMJ Open · September 2023 · NCT04780061
Patients may have mostly recovered before treatment
Early terminated low-risk population (no hospitalization) very late treatment (mean 8 days) RCT with 44 patients treated with vitamin C, D, K, and zinc, and 46 control patients, showing no significant differences.

Authors acknowledge that the very late treatment is a major limitation, noting that in an ideal setting, "patients would begin taking therapeutic interventions immediately after noticing symptoms". Authors note that patients already had a low symptom burden at baseline and that "it is likely that the majority of the participants had almost fully recovered before starting treatment."

Authors note that most participants were young, had few comorbidities and had excellent self-rated health at baseline, leaving less room for improvement.

There was low compliance with completing surveys. Data from only 64% of patients was in the main analysis.

Authors claim "high internal validity", but the loss of data was statistically significantly different between arms, without analysis or mention. Since the study involves widely available treatments, one possibility is that patients in the control arm who feel sick may be more likely to independently take the treatments (via supplementation or food/sun exposure), believing that they are in the control arm or that additional dosing is safe, and they may then feel it's inappropriate to continue submitting the surveys.

Discussion is biased, stating that "evidence for the use of these products in people with COVID-19 is limited", however there were 219 controlled studies at the time, including 8, 27, and 16 RCTs for vitamin C, D, and zinc. Authors claim high similarity between arms however there was 60% vs. 41% male patients, and 88% vs. 68% of patients that received a third dose.

Authors claim that treatment "showed no beneficial effects for overall health or symptom burden". However 48% lower ER visits is beneficial, and most outcomes show a benefit. The only statistically significant effect was the loss of data, however significant clinical effects are not expected based on the small sample, very late treatment, event rates, and outcomes. Submit Corrections or Updates.
Retrospective 962 COVID-19 patients in Bangladesh, showing significantly lower severity with vitamin C, vitamin D, and zinc supplementation, and improved results from the combination of all three.

Table 2 has a typographical error where the columns for Vitamin C and Vitamin D are duplicates. The OR for "Taking vitamin C, D and zinc as medication" in Table 4 is likely a typo, with the point estimate outside the CI.

The study aims to evaluate the risk of infection by comparing COVID-19 PCR-positive patients to a control group of suspected patients who had close contact and symptoms but were PCR-negative, however this introduces selection bias. These estimates are likely to be less reliable and are excluded in our analysis. Submit Corrections or Updates.
Estimated 50 patient vitamin C late treatment RCT with results not reported over 5 years after estimated completion. Submit Corrections or Updates.
Retrospective survey-based analysis of 349 COVID-19 patients, showing no significant difference with vitamin C prophylaxis in unadjusted analysis. REC/UG/2020/03. Submit Corrections or Updates.
Late treatmentRetrospective · 139 patients · Turkey
Vitamin C for COVID-19
Simsek et al. Annals of Medical Research · September 2021
Outcome
Improvement
Relative Risk · 95% CI
Mortality
44%0.56 · 0.23–1.35
ICU admission
10%0.90 · 0.55–1.46
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality with vitamin C
Not statistically significant · p = 0.19
Retrospective 139 hospitalized patients in Turkey, 58 treated with high-dose vitamin C, showing improved kidney functioning with treatment. Mortality was lower with treatment, but not reaching statistical significance with the small sample size. Submit Corrections or Updates.
SufficiencyRetrospective · 74 patients · Germany · Feb – Nov 2020
Vitamin C for COVID-19
Sinnberg et al. Antioxidants · August 2022
Outcome
Improvement
Relative Risk · 95% CI
Mortality
42%0.58 · 0.18–1.91
Ventilation
41%0.59 · 0.28–1.25
Hospitalization
61%0.39 · 0.15–0.96
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower hospitalization with higher vitamin C levels
p = 0.05
Analysis of 74 COVID-19 patients and 8 controls in Germany, showing low vitamin C levels associated with mortality. There was no significant difference for vitamin A, D, or E levels. Very few group details are provided, for example the age of patients in the control group and each severity group is not provided. Submit Corrections or Updates.
Retrospective 616 patients in China showing increased risk of disease progression with vitamin C treatment. Submit Corrections or Updates.
Retrospective 323 hospitalized patients, 153 treated with vitamin C, showing no significant differences. Patients in each group were in different time periods, with the vitamin C group first. Time based confounding is possible due to improvements in SOC. Submit Corrections or Updates.
PSM retrospective 207 hospitalized patients in China, 46 treated with diammonium glycyrrhizinate and vitamin C, showing lower risk of ARDS with treatment. Submit Corrections or Updates.
Late treatmentRCT · 44 patients · Iran · Mar – May 2020
Vitamin C for COVID-19
Tehrani et al. Urology J. · November 2021
Outcome
Improvement
Relative Risk · 95% CI
Mortality
87%0.13 · 0.01–2.25
Hospitalization time
18%0.82 · 0.60–1.13
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality and shorter hospitalization
Not statistically significant · p = 0.13 (mortality) and p = 0.23 (hospitalization)
RCT 54 late stage patients, 18 treated with IV vitamin C (2g every 6h for 5 days), showing significant relative improvements in oxygen saturation and respiratory rate. Submit Corrections or Updates.
Early treatmentRCT · 98 patients · USA · Apr 2020 – Feb 2021
Vitamin C for COVID-19
COVIDAtoZ Thomas et al. JAMA Network Open · February 2021
Outcome
Improvement
Relative Risk · 95% CI
Mortality
-204%3.04 · 0.13 to >9
Hospitalization
31%0.69 · 0.12–3.98
Recovery time
18%0.82 · 0.63–1.07
primary
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Faster recovery with vitamin C
Not statistically significant · p = 0.15
Small 214 low-risk outpatient RCT showing non-statistically significant faster recovery with zinc and with vitamin C. A secondary analysis concludes that vitamin C increases recovery rate by 71% (p = 0.036)310. See also311.

The 'both' (combination) treatment group had much higher rates of diabetes (29.3% vs 6.0% in SOC), hypertension (46.6% vs 28.0%), and dyslipidemia (41.4% vs 18.0%), yet the statistical analysis did not adjust for these major confounders. This likely biased the results against the active treatment arm and undermined the futility stopping criteria. The treatment groups were also older, with the combination group mean age of 48.7 vs. 42 for the control group.

The reported mean 'Time to 50% reduction' in Table 2 appears to only include patients who successfully achieved the reduction, effectively excluding the worst outcomes (treatment failures) from the primary continuous measure. Using ANOVA on this truncated time-to-event data is statistically inappropriate and introduces selection bias.

Table 2 reports the mean (SD) for 'no shortness of breath' in the standard of care group as 3.6 (45) days. An SD of 45 days is not possible in a 28-day study and may be a typographical error for 4.5.

Table 1 (footnote c) indicates the 12-component score was collected for 78 patients (20 SOC, 14 AA, 21 zinc, 23 both). However, Table 2 reports outcomes for only 75 patients (19 SOC, 14 AA, 21 zinc, 21 both) with no explanation provided for the 3 missing patients.

An author reports receiving consulting fees from Gilead, the maker of remdesivir. Submit Corrections or Updates.
Retrospective 180 hospitalized COVID-19 patients in Sierra Leone, showing lower mortality with vitamin C treatment in unadjusted results. Submit Corrections or Updates.
Retrospective 126 patients in Turkey, showing no significant difference in PCR+ at day 14 with vitamin C treatment. Submit Corrections or Updates.
Retrospective 270 moderate/severe hospitalized COVID-19 patients, showing lower mortality with high (25 g/day) or low-dose (2 g/day) intraveneous vitamin C. Submit Corrections or Updates.
Analysis of nutrient intake and COVID-19 outcomes for 3,996 people in Iran, showing lower risk of COVID-19 hospitalization with sufficient vitamin A, vitamin C, and selenium intake, with statistical significance for vitamin A and selenium. Submit Corrections or Updates.
Retrospective 8,634 hospitalized patients in India, showing lower mortality with high-dose vitamin C in unadjusted results. No group details are provided, the text and table appear to show different results, and some numbers do not match. Submit Corrections or Updates.
Late treatmentRCT · 20 patients · China · February 2020
Vitamin C for COVID-19
Yang et al. American J. Translational Research · Jan 2022 · ChiCTR2000032717
Faster recovery and viral clearance
p < 0.0001 (recovery) and p < 0.0001 (viral clearance)
Is late treatment with vitamin C + TCM beneficial for COVID-19?
c19early.org
Prospective study of 60 patients in China with three arms: SOC, SOC+TCM, and SOC+TCM+high-dose vitamin C, showing successively faster recovery with the addition of TCM and the addition of high-dose vitamin C. TCM included inhaled vitamin C 10g, 3-7 times per day. IV vitamin C 10g/60kg twice a day, and oral vitamin C 3g three times a day. Group C vs. group A includes combined treatment with TCM, while group C vs. group B both include vitamin C (high vs. low dose). Submit Corrections or Updates.
PSM retrospective 86 ICU patients on mechanical ventilation in Turkey, showing lower mortality with high-dose vitamin C treatment (≥200mg/kg for 4 days). Submit Corrections or Updates.
ICU patientsRCT · 56 patients · China · Feb – Mar 2020
Vitamin C for COVID-19
Zhang et al. Annals of Intensive Care · August 2020
Outcome
Improvement
Relative Risk · 95% CI
Mortality
50%0.50 · 0.20–1.50
Mortality (b)
80%0.20 · 0.10–0.90
Mortality (c)
50%0.50 · 0.20–1.80
Mortality (d)
70%0.30 · 0.21–1.10
00.511.52+
← Vitamin C
reduces riskVitamin C
increases risk →
Lower mortality with vitamin C
Not statistically significant · p = 0.2
Small RCT for high-dose vitamin C for ICU patients showing reduced (but not statistically significant) mortality. Dosage was 12g of vitamin C/50ml every 12 hours for 7 days at a rate of 12ml/hour. Submit Corrections or Updates.
PSM retrospective 110 patients, 55 treated with high-dose IV vitamin C, showing lower progression to severe disease with treatment. Patients in each group were in different time periods, time based confounding is likely due to SOC improving over time. ChiCTR2000033050. Submit Corrections or Updates.
Late treatmentRetrospective · 397 patients · China · February 2020
Vitamin C for COVID-19
Zheng et al. Open Medicine · September 2021
Higher mortality and worse improvement
Not statistically significant · p = 0.33 (mortality) and p = 0.17 (improvement)
Retrospective 397 severe COVID-19 patients in China, showing worse outcomes with vitamin C treatment, without statistical significance. IV vitamin C 2-4g/day. Subject to confounding by indication and immortal time bias. Exclusion criteria were (a) the duration of hospitalization was less than 3 days; (b) vitamin C treatment started before admission; and (c) the length of vitamin C use was less than 3 days. Includes vitamin C use started at any time during hospitalization, for many patients this was >15 days later (Figure A2). Duration of treatment varied widely (Figure A1). Treatment was determined by clinicians according to the condition of each patient. Submit Corrections or Updates.
Retrospective 43 ICU patients in Turkey, 21 treated with vitamin C, showing no significant difference in mortality and increased renal failure. Treatment included stage 1 AKI patients. Vitamin C 45-50 g/day for 5 days. Submit Corrections or Updates.
Retrospective 160 ICU patients, 32 with raised neutrophil/lymphocyte ratio treated with vitamin C, showing no significant differences. Submit Corrections or Updates.
Appendix 1. Methods and Data
Search methods
We perform ongoing searches of PubMed, medRxiv, Europe PMC, ClinicalTrials.gov, The Cochrane Library, Google Scholar, Research Square, ScienceDirect, Oxford University Press, the reference lists of other studies and meta-analyses, and submissions to the site c19early.org, which regularly receives notification of studies upon publication. Search terms are "vitamin C", "ascorbic acid" and COVID-19 or SARS-CoV-2. Automated searches are performed twice daily, with all matches reviewed for inclusion. All studies regarding the use of vitamin C for COVID-19 that report a comparison with a control group are included in the main analysis. Sensitivity analysis is performed, excluding studies with major issues, epidemiological studies, and studies with minimal available information. Studies with major unexplained data issues, for example major outcome data that is impossible to be correct with no response from the authors, are excluded.
Fig. 40. Mid-recovery results can more accurately reflect efficacy when almost all patients recover. Mateja et al. confirm that intermediate viral load results more accurately reflect hospitalization/death.
Effect extraction
We extracted effect sizes and associated data from all studies. If studies report multiple kinds of effects then the most serious outcome is used in pooled analysis, while other outcomes are included in the outcome-specific analyses. For example, if effects for mortality and cases are reported then they are both used in specific outcome analyses, while mortality is used for pooled analysis. If symptomatic results are reported at multiple times, we use the latest time, for example if mortality results are provided at 14 days and 28 days, the results at 28 days have preference. Mortality alone is preferred over combined outcomes. Outcomes with zero events in both arms are not used, the next most serious outcome with one or more events is used. For example, in low-risk populations with no mortality, a reduction in mortality with treatment is not possible, however a reduction in hospitalization, for example, is still valuable. Clinical outcomes are considered more important than viral outcomes. When basically all patients recover in both treatment and control groups, preference for viral clearance and recovery is given to results mid-recovery where available. After most or all patients have recovered there is little or no room for an effective treatment to do better, however faster recovery is valuable. An IPD meta-analysis confirms that intermediate viral load reduction is more closely associated with hospitalization/death than later viral load reduction312. If only individual symptom data is available, the most serious symptom has priority, for example difficulty breathing or low SpO2 is more important than cough.
Statistical methods
Forest plots are computed using PythonMeta313 with the DerSimonian and Laird random-effects model (the fixed effect assumption is not plausible in this case) and inverse variance weighting. Results are presented with 95% confidence intervals. Heterogeneity among studies was assessed using the I2 statistic. When results provide an odds ratio, we compute the relative risk when possible, or convert to a relative risk according to Zhang (C) et al. Reported confidence intervals and p-values are used when available, and adjusted values are used when provided. If multiple types of adjustments are reported propensity score matching and multivariable regression has preference over propensity score matching or weighting, which has preference over multivariable regression. Adjusted results have preference over unadjusted results for a more serious outcome when the adjustments significantly alter results. When needed, conversion between reported p-values and confidence intervals followed Altman, Altman (B), and Fisher's exact test was used to calculate p-values for event data. If continuity correction for zero values is required, we use the reciprocal of the opposite arm with the sum of the correction factors equal to 1317. Results are expressed with RR < 1.0 favoring treatment, and using the risk of a negative outcome when applicable (for example, the risk of death rather than the risk of survival). If studies only report relative continuous values such as relative times, the ratio of the time for the treatment group versus the time for the control group is used. Calculations are done in Python (3.15.1) with scipy (1.18.1), pythonmeta (1.26), numpy (2.5.4), statsmodels (0.15.0), and plotly (6.9.0). Mixed-effects meta-regression results are computed with R (4.4.0) using the metafor (4.6-0) and rms (6.8-0) packages, and using the most serious sufficiently powered outcome. For all statistical tests, a p-value less than 0.05 was considered statistically significant. Grobid 0.8.2 is used to parse PDF documents.
When evaluating potential effect modification across groups, we use an interaction test as described by Altman (C) et al. We compared the log-transformed relative risks using a z-test, deriving the standard error of the difference from the 95% confidence intervals. A two-sided interaction p-value of < 0.05 was considered a statistically significant difference in treatment effect between the groups.
Quality evaluation
Cochrane RoB 2/ROBINS-I are often used to evaluate studies, and have the advantage of providing standardized rules that can be applied with minimal understanding of the domain and study. However, the rules do not account for many real-world issues, often overemphasize or underemphasize others, and studies show low inter-rater reliability325. Certain domains are more applicable for these tools, however the time-sensitive nature of a pandemic, with significant mortality for every day of delay in evidence assessment, and the characteristics of COVID-19 make them inappropriate for this domain. This can be demonstrated with examples where expert RoB 2/ROBINS-I ratings do not match reality for COVID-19. Popp et al. use RoB 2 to classify Reis et al. as low risk of bias, however this is the opposite of reality—the trial not only has very high risk of bias, but has very high actual known bias, refusing to release data despite pledging to, reporting multiple impossible numbers, having blinding and randomization failure, and many other issues327. Axfors et al. use RoB 2 to classify Horby et al. as low risk of bias, however this is the opposite of reality—the very late treatment and excessive dosage used produces results with no relevance to recommended usage. HCQ shows poor results with late treatment and excessive dosage, and the combination shows harmD. Hempenius et al. use ROBINS-I to classify 33 studies for HCQ. The two rated as having the lowest risk of bias323,324 are far from the most informative. Both involve very late treatment, providing no information on recommended usage, and ROBINS-I does a very poor job of accounting for the impact of confounding factorsE.
Our quality evaluation focuses on known issues and bias, and the potential impact on outcomes, rather than just the risk of bias. The estimated potential impact of each confounding factor, and the direction of the impact is considered. For example, consider a study that shows significantly lower risk, the value of the study varies significantly if confounding points to an underestimate or an overestimate of efficacy. In one case, the real effect may be null, while the other case provides stronger evidence of efficacy (which may be greater than the study shows). Analysis focusing on the risk of bias, while simpler, may penalize studies for theoretical or technical issues that have no or minimal impact on outcomes. Analysis also depends on the outcome, for example certain issues are less relevant for objective outcomes such as mortality. Inaccurate penalization, and inaccurate high-quality evaluation in the face of known major issues affecting outcomes, increases in significance during a pandemic when immediate recognition of new evidence is critical, and when considering all global studies, as required during a pandemic. Investigators in other countries may have different customs for design, analysis, and reporting, and different English language skills, however they may not be less diligent or have greater bias. Investigators in lower-pharmaceutical-profit countries may have lower bias towards profitable interventions.
Treatment time
We have classified studies as early treatment if most patients are not already at a severe stage at the time of treatment (for example based on oxygen status or lung involvement), and treatment started within 5 days of the onset of symptoms. If studies contain a mix of early treatment and late treatment patients, we consider the treatment time of patients contributing most to the events (for example, consider a study where most patients are treated early but late treatment patients are included, and all mortality events were observed with late treatment patients). We note that a shorter time may be preferable. Antivirals are typically only considered effective when used within a shorter timeframe, for example 0-36 or 0-48 hours for oseltamivir, with longer delays not being effective153,154.
Living analysis
This is a living analysis and is updated regularly. Submit updates or corrections with the form below. We received no funding, this research is done in our spare time. We have no affiliation with any pharmaceutical companies, supplement companies, governments, political parties, or advocacy organizations.
A summary of study results is below. Please submit updates and corrections at the bottom of this page.
A summary of study results is below. Please submit updates and corrections at https://c19early.org/cmeta.html.
Effect extraction follows pre-specified rules as detailed above and gives priority to more serious outcomes. For pooled analyses, the first (most serious) outcome is used, which may differ from the effect a paper focuses on. Other outcomes are used in outcome specific analyses.
Boukef, 2/28/2023, Double Blind Randomized Controlled Trial, placebo-controlled, Tunisia, trial NCT05670444 (history).
Submit Corrections or Updates.
 150 patient RCT with results unknown and over 3 years late.
Rahman, 11/8/2023, retrospective, Bangladesh, peer-reviewed, 5 authors, dosage not specified, excluded in exclusion analyses: unadjusted results with no group details; significant unadjusted confounding possible.
Submit Corrections or Updates.
 risk of hospitalization, 40.5% lower, RR 0.60, p < 0.001, treatment 128 of 476 (26.9%), control 56 of 124 (45.2%), NNT 5.5.
Ried, 11/25/2021, Randomized Controlled Trial, Turkey, peer-reviewed, 3 authors, study period January 2021 - June 2021, average treatment delay 4.0 days, dosage 50mg/kg qid day 1, 100mg/kg qid days 2-7, trial ACTRN12620000557932.
Submit Corrections or Updates.
 risk of no recovery, 30.6% lower, RR 0.69, p = 0.008, treatment 69 of 162 (42.6%), control 46 of 75 (61.3%), NNT 5.3, mid-recovery, day 15.
Su, 12/23/2020, retrospective, China, peer-reviewed, 9 authors, study period 20 January, 2020 - 30 April, 2020, dosage 10000mg days 1-3, 5-15g per day for at least 3 days.
Submit Corrections or Updates.
 risk of progression, 135.3% higher, HR 2.35, p = 0.18, adjusted per study, binary logistic regression.
improvement time, 34.2% worse, relative time 1.34, p = 0.04, adjusted per study, inverted to make RR<1 favor treatment, Cox proportional hazards.
Thomas (B), 2/12/2021, Randomized Controlled Trial, USA, peer-reviewed, 11 authors, study period 8 April, 2020 - 11 February, 2021, dosage 8000mg days 1-10, trial NCT04342728 (history) (COVIDAtoZ), excluded in exclusion analyses: unadjusted differences between groups.
Submit Corrections or Updates.
 risk of death, 204.2% higher, RR 3.04, p = 0.49, treatment 1 of 48 (2.1%), control 0 of 50 (0.0%), continuity correction due to zero event (with reciprocal of the contrasting arm).
risk of hospitalization, 30.6% lower, RR 0.69, p = 1.00, treatment 2 of 48 (4.2%), control 3 of 50 (6.0%), NNT 55.
recovery time, 17.9% lower, relative time 0.82, p = 0.15, treatment mean 5.5 (±3.7) n=48, control mean 6.7 (±4.4) n=50, mean time to a 50% reduction in symptoms, primary outcome.
Usanma Koban, 6/7/2022, retrospective, Turkey, peer-reviewed, 3 authors, study period 1 March, 2020 - 30 September, 2020, dosage not specified.
Submit Corrections or Updates.
 risk of no viral clearance, 33.0% lower, OR 0.67, p = 0.73, treatment 31, control 95, adjusted per study, multivariable, day 14, RR approximated with OR.
Zhao, 4/22/2021, retrospective, propensity score matching, China, peer-reviewed, 15 authors, average treatment delay 4.0 days, dosage 100mg/kg days 1-7, excluded in exclusion analyses: substantial confounding by time likely due to declining usage over the early stages of the pandemic when overall treatment protocols improved dramatically.
Submit Corrections or Updates.
 risk of progression, 72.0% lower, RR 0.28, p = 0.03, treatment 4 of 55 (7.3%), control 12 of 55 (21.8%), NNT 6.9, adjusted per study, PSM.
time to viral-, 7.7% higher, relative time 1.08, p = 0.79, treatment 55, control 55, PSM.
Effect extraction follows pre-specified rules as detailed above and gives priority to more serious outcomes. For pooled analyses, the first (most serious) outcome is used, which may differ from the effect a paper focuses on. Other outcomes are used in outcome specific analyses.
Adhikari (B), 10/25/2023, Double Blind Randomized Controlled Trial, multiple countries, peer-reviewed, 82 authors, dosage 50mg/kg qid days 1-4, trial NCT04401150 (history) (LOVIT-COVID).
Submit Corrections or Updates.
 risk of death, 27.8% lower, HR 0.72, p = 0.19, treatment 190, control 194, all LOVIT-COVID patients.
risk of death, 27.5% lower, HR 0.72, p = 0.34, treatment 84, control 97, inverted to make HR<1 favor treatment, LOVIT-COVID critical.
risk of death, 28.1% lower, HR 0.72, p = 0.37, treatment 106, control 97, inverted to make HR<1 favor treatment, LOVIT-COVID non-critical.
Adhikari, 10/25/2023, Randomized Controlled Trial, multiple countries, peer-reviewed, 82 authors, dosage 50mg/kg qid days 1-4, trial NCT04401150 (history) (REMAP-CAP).
Submit Corrections or Updates.
 risk of death, 19.5% higher, HR 1.19, p = 0.08, treatment 1,303, control 903, all REMAP-CAP patients.
risk of death, 16.3% higher, HR 1.16, p = 0.22, treatment 953, control 434, inverted to make HR<1 favor treatment, REMAP-CAP critical.
risk of death, 26.6% higher, HR 1.27, p = 0.19, treatment 350, control 469, inverted to make HR<1 favor treatment, REMAP-CAP non-critical.
Al Sulaiman, 4/2/2021, retrospective, propensity score matching, Saudi Arabia, preprint, 12 authors, dosage 1000mg days 1-11.
Submit Corrections or Updates.
 risk of death, 14.9% lower, RR 0.85, p = 0.27, treatment 46 of 142 (32.4%), control 59 of 142 (41.5%), NNT 11, odds ratio converted to relative risk, PSM.
Baguma, 12/28/2021, retrospective, Uganda, preprint, 16 authors, study period March 2020 - October 2021, dosage not specified.
Submit Corrections or Updates.
 risk of death, 48.5% higher, RR 1.48, p = 0.54, treatment 385, control 96, adjusted per study, inverted to make RR<1 favor treatment, odds ratio converted to relative risk, multivariable, control prevalance approximated with overall prevalence.
Bepouka, 5/14/2025, retrospective, DR Congo, peer-reviewed, 14 authors, study period 20 March, 2020 - 2 January, 2022.
Submit Corrections or Updates.
 risk of death, 76.0% lower, OR 0.24, p = 0.01, treatment 185, control 225, adjusted per study, multivariable, RR approximated with OR.
Coppock, 3/19/2022, Randomized Controlled Trial, USA, peer-reviewed, 14 authors, dosage 300mg/kg day 1, 600mg/kg day 2, 900mg/kg days 3-6.
Submit Corrections or Updates.
 risk of progression, 5.0% lower, HR 0.95, p = 0.64, treatment 4 of 44 (9.1%), control 2 of 22 (9.1%), adjusted per study, within 36 hours.
risk of no improvement, 49.7% better, RR 0.50, p = 0.16, treatment 6 of 44 (13.6%), control 6 of 22 (27.3%), NNT 7.3, adjusted per study, inverted to make RR<1 favor treatment, odds ratio converted to relative risk, within 36 hours.
risk of no hospital discharge, 22.5% lower, RR 0.78, p = 0.07, treatment 31 of 44 (70.5%), control 20 of 22 (90.9%), NNT 4.9, within 36 hours.
Corrao, 7/8/2024, prospective, Italy, peer-reviewed, 7 authors, trial NCT04323514 (history).
Submit Corrections or Updates.
 risk of death, 39.4% lower, RR 0.61, p = 0.37, treatment 9 of 104 (8.7%), control 6 of 42 (14.3%), NNT 18.
risk of death/ICU, 19.0% lower, OR 0.81, p = 0.24, treatment 104, control 42, adjusted per study, multivariable, RR approximated with OR.
risk of ICU admission, 101.9% higher, RR 2.02, p = 0.51, treatment 10 of 104 (9.6%), control 2 of 42 (4.8%).
hospitalization time, 25.0% lower, relative time 0.75, p = 0.16, treatment 104, control 42.
Coskun, 3/21/2023, retrospective, Turkey, peer-reviewed, 7 authors, study period March 2020 - June 2020, trial NCT04710329 (history), excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 25.4% lower, RR 0.75, p = 0.26, treatment 17 of 38 (44.7%), control 24 of 40 (60.0%), NNT 6.6.
risk of mechanical ventilation, 1.8% lower, RR 0.98, p = 1.00, treatment 28 of 38 (73.7%), control 30 of 40 (75.0%), NNT 76.
relative SOFA score, 28.4% better, RR 0.72, p = 0.005, treatment 38, control 40, mean SOFA score, day 4.
Dinoi, 2/20/2025, retrospective, Italy, peer-reviewed, 11 authors, study period 17 March, 2020 - 15 June, 2021, dosage not specified.
Submit Corrections or Updates.
 risk of death, 31.5% higher, OR 1.32, p = 0.30, treatment 38 of 247 (15.4%) cases, 30 of 247 (12.1%) controls, case control OR.
Doocy, 10/19/2022, prospective, multiple countries, peer-reviewed, 6 authors, study period December 2020 - June 2021, dosage not specified, trial NCT04568499 (history).
Submit Corrections or Updates.
 risk of death, 62.8% lower, RR 0.37, p = 0.22, treatment 2 of 64 (3.1%), control 22 of 80 (27.5%), NNT 4.1, adjusted per study, inverted to make RR<1 favor treatment, multivariable.
Elhadi, 4/30/2021, prospective, Libya, peer-reviewed, 21 authors, study period 29 May, 2020 - 30 December, 2020, dosage not specified, excluded in exclusion analyses: unadjusted results with no group details; very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 12.0% higher, RR 1.12, p = 0.15, treatment 175 of 277 (63.2%), control 106 of 188 (56.4%).
Fogleman, 7/27/2022, Double Blind Randomized Controlled Trial, placebo-controlled, USA, peer-reviewed, mean age 52.0, 7 authors, study period 5 October, 2020 - 21 June, 2021, average treatment delay 6.0 days, dosage 1000mg days 1-14, trial NCT04530539 (history).
Submit Corrections or Updates.
 relative recovery, 4.4% better, RR 0.96, p = 0.83, treatment mean 17.59 (±13.1) n=32, control mean 16.82 (±15.7) n=34, mid-recovery, relative symptom improvement, day 9.
Fowler, 4/4/2024, Double Blind Randomized Controlled Trial, placebo-controlled, USA, preprint, 1 author, trial NCT04344184 (history) (SAFE EVICT CORONA-ALI), excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 18.8% lower, RR 0.81, p = 0.75, treatment 5 of 22 (22.7%), control 7 of 25 (28.0%), NNT 19.
relative WHO status, 1.7% worse, RR 1.02, p = 0.28, treatment mean 3.05 (±0.22) n=21, control mean 3.0 (±0.0) n=23, day 27.
Gadhiya, 4/8/2021, retrospective, USA, peer-reviewed, 4 authors, dosage not specified, excluded in exclusion analyses: substantial unadjusted confounding by indication likely.
Submit Corrections or Updates.
 risk of death, 0.7% higher, RR 1.01, p = 0.98, treatment 19 of 55 (34.5%), control 36 of 226 (15.9%), adjusted per study, odds ratio converted to relative risk, multivariate logistic regression.
Galindo, 5/15/2022, Double Blind Randomized Controlled Trial, placebo-controlled, Colombia, trial NCT05029037 (history).
Submit Corrections or Updates.
 Estimated 160 patient RCT with results unknown and over 4 years late.
Gao, 2/26/2021, retrospective, China, peer-reviewed, 14 authors, dosage 12000mg day 1, 6,000mg days 2-5.
Submit Corrections or Updates.
 risk of death, 86.0% lower, HR 0.14, p = 0.04, treatment 1 of 46 (2.2%), control 5 of 30 (16.7%), NNT 6.9, adjusted per study, KM.
Gavrielatou, 2/11/2022, retrospective, Greece, peer-reviewed, 10 authors, study period 21 October, 2020 - 8 March, 2021, average treatment delay 5.5 days, excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 58.0% lower, RR 0.42, p = 0.11, treatment 2 of 10 (20.0%), control 49 of 103 (47.6%), NNT 3.6.
Hakamifard, 4/14/2021, Randomized Controlled Trial, Iran, peer-reviewed, 8 authors, study period March 2020 - April 2020, dosage 1000mg daily, this trial uses multiple treatments in the treatment arm (combined with vitamin E) - results of individual treatments may vary.
Submit Corrections or Updates.
 risk of ICU admission, 46.3% lower, RR 0.54, p = 0.46, treatment 3 of 38 (7.9%), control 5 of 34 (14.7%), NNT 15.
hospitalization time, 1.0% lower, relative time 0.99, p = 0.82, treatment 38, control 34.
Hamidi-Alamdari, 3/8/2021, Randomized Controlled Trial, Iran, peer-reviewed, 23 authors, study period 19 April, 2020 - 21 September, 2020, this trial uses multiple treatments in the treatment arm (combined with methylene blue and N-acetyl cysteine) - results of individual treatments may vary, trial NCT04370288 (history).
Submit Corrections or Updates.
 risk of death, 44.4% lower, RR 0.56, p = 0.38, treatment 5 of 40 (12.5%), control 9 of 40 (22.5%), NNT 10.0.
hospitalization time, 37.6% lower, relative time 0.62, p = 0.004, treatment 40, control 40.
He, 1/31/2021, Single Blind Randomized Controlled Trial, China, trial NCT04664010 (history).
Submit Corrections or Updates.
 60 patient RCT with results unknown and over 5 years late.
Hess, 3/29/2022, retrospective, USA, peer-reviewed, 9 authors, study period March 2020 - July 2020.
Submit Corrections or Updates.
 risk of death, 20.0% lower, HR 0.80, p = 0.54, treatment 10 of 25 (40.0%), control 37 of 75 (49.3%), NNT 11, time to event analysis, propensity score weighting.
risk of mechanical ventilation, 39.5% lower, RR 0.60, p = 0.05, treatment 18 of 25 (72.0%), control 54 of 75 (72.0%), odds ratio converted to relative risk, propensity score weighting.
risk of mechanical ventilation, 50.0% lower, HR 0.50, p = 0.03, treatment 18 of 25 (72.0%), control 54 of 75 (72.0%), time to event analysis, propensity score weighting.
risk of ICU admission, 27.2% lower, RR 0.73, p = 0.10, treatment 22 of 25 (88.0%), control 63 of 75 (84.0%), odds ratio converted to relative risk, propensity score weighting.
risk of ICU admission, 30.0% lower, HR 0.70, p = 0.19, treatment 22 of 25 (88.0%), control 63 of 75 (84.0%), time to event analysis, propensity score weighting.
Izzo, 7/19/2022, prospective, Italy, peer-reviewed, 21 authors, this trial compares with another treatment - results may be better when compared to placebo, this trial uses multiple treatments in the treatment arm (combined with L-arginine) - results of individual treatments may vary, LINCOLN trial.
Submit Corrections or Updates.
 relative recovery, 41.4% better, RR 0.59, p < 0.001, treatment mean 8.15 (±1.3) n=869, control mean 13.9 (±2.3) n=521, relative symptom score.
relative recovery, 67.5% better, RR 0.33, p < 0.001, treatment 869, control 521, relative Borg score.
JamaliMoghadamSiahkali, 1/9/2021, Randomized Controlled Trial, Iran, preprint, 17 authors, study period April 2020 - May 2020, dosage 1500mg qid days 1-5.
Submit Corrections or Updates.
 risk of death, no change, RR 1.00, p = 1.00, treatment 3 of 30 (10.0%), control 3 of 30 (10.0%).
risk of mechanical ventilation, 25.0% higher, RR 1.25, p = 1.00, treatment 5 of 30 (16.7%), control 4 of 30 (13.3%).
hospitalization time, 30.8% higher, relative time 1.31, p = 0.03, treatment 30, control 30.
Jang, 12/16/2020, retrospective, South Korea, peer-reviewed, median age 63.0, 10 authors, study period February 2020 - April 2020, dosage not specified, excluded in exclusion analyses: very late stage, ECMO patients.
Submit Corrections or Updates.
 risk of no recovery, 51.4% lower, RR 0.49, p = 0.15, treatment 5 of 12 (41.7%), control 6 of 7 (85.7%), NNT 2.3, weaning from ECMO.
Krishnan, 7/20/2020, retrospective, USA, peer-reviewed, 13 authors, dosage not specified, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of death, 30.7% lower, RR 0.69, p = 0.04, treatment 40 of 79 (50.6%), control 52 of 73 (71.2%), NNT 4.9, odds ratio converted to relative risk.
Kumar (B), 8/30/2022, Double Blind Randomized Controlled Trial, placebo-controlled, India, peer-reviewed, mean age 57.0, 11 authors, average treatment delay 7.5 days, dosage 1000mg tid days 1-4, trial CTRI/2020/11/029230, excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 23.1% lower, RR 0.77, p = 0.60, treatment 10 of 30 (33.3%), control 13 of 30 (43.3%), NNT 10.0.
risk of mechanical ventilation, 21.4% lower, RR 0.79, p = 0.60, treatment 11 of 30 (36.7%), control 14 of 30 (46.7%), NNT 10.0.
Kumari, 11/30/2020, Randomized Controlled Trial, Pakistan, peer-reviewed, 10 authors, study period March 2020 - July 2020, dosage 50mg/kg daily.
Submit Corrections or Updates.
 risk of death, 36.4% lower, RR 0.64, p = 0.45, treatment 7 of 75 (9.3%), control 11 of 75 (14.7%), NNT 19.
risk of mechanical ventilation, 20.0% lower, RR 0.80, p = 0.67, treatment 12 of 75 (16.0%), control 15 of 75 (20.0%), NNT 25.
recovery time, 26.0% lower, relative time 0.74, p < 0.001, treatment 75, control 75, days to symptom-free.
hospitalization time, 24.3% lower, relative time 0.76, p < 0.001, treatment 75, control 75, days spent in hospital.
Kyagambiddwa, 5/11/2023, retrospective, Uganda, peer-reviewed, mean age 39.0, 15 authors, study period May 2020 - August 2022, dosage not specified.
Submit Corrections or Updates.
 risk of death, 50.0% lower, HR 0.50, p = 0.06, adjusted per study, multivariable, Cox proportional hazards.
Labbani-Motlagh, 12/14/2022, Double Blind Randomized Controlled Trial, placebo-controlled, Iran, peer-reviewed, 12 authors, study period 5 April, 2020 - 19 November, 2020, dosage 12000mg days 1-4, trial IRCT20190917044805N2.
Submit Corrections or Updates.
 risk of death, 33.3% lower, RR 0.67, p = 0.74, treatment 4 of 37 (10.8%), control 6 of 37 (16.2%), NNT 18, day 28.
hospitalization time, 12.8% higher, relative time 1.13, p = 0.49, treatment mean 9.24 (±7.5) n=37, control mean 8.19 (±5.34) n=37.
risk of progression, 15.9% lower, RR 0.84, p = 0.12, treatment 37, control 37, SOFA, day 5.
risk of progression, 9.3% higher, RR 1.09, p = 0.47, treatment 37, control 37, NEWS, day 5.
risk of progression, 5.8% higher, RR 1.06, p = 0.38, treatment 37, control 37, WHO, day 5.
risk of progression, 60.0% lower, RR 0.40, p = 0.14, treatment 4 of 37 (10.8%), control 10 of 37 (27.0%), NNT 6.2, AKI.
Lamontagne, 12/6/2022, Double Blind Randomized Controlled Trial, placebo-controlled, Canada, trial NCT04401150 (history) (LOVIT-COVID).
Submit Corrections or Updates.
 392 patient RCT with results unknown and over 3 years late.
Li, 6/8/2021, retrospective, propensity score matching, USA, peer-reviewed, 6 authors, excluded in exclusion analyses: very late stage, ICU patients; very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 10.5% higher, RR 1.11, p = 1.00, treatment 7 of 8 (87.5%), control 19 of 24 (79.2%), PSM.
Liu, 6/1/2023, Single Blind Randomized Controlled Trial, China, trial NCT05694975 (history) (CEMVISCC).
Submit Corrections or Updates.
 Estimated 608 patient RCT with results unknown and over 3 years late.
Majidi, 12/15/2021, Double Blind Randomized Controlled Trial, Iran, peer-reviewed, 16 authors, study period May 2020 - July 2020, dosage 500mg days 1-14, trial IRCT20151226025699N5, excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 13.6% lower, RR 0.86, p = 0.03, treatment 26 of 31 (83.9%), control 67 of 69 (97.1%), NNT 7.6, day 28.
Mousaviasl, 7/22/2023, Double Blind Randomized Controlled Trial, placebo-controlled, Iran, peer-reviewed, 13 authors, study period November 2020 - May 2021, dosage 500mg bid days 1-5.
Submit Corrections or Updates.
 risk of death, 20.4% lower, RR 0.80, p = 0.64, treatment 8 of 201 (4.0%), control 10 of 200 (5.0%), NNT 98, day 28.
risk of death, 99.0% higher, RR 1.99, p = 1.00, treatment 2 of 201 (1.0%), control 1 of 200 (0.5%), ICU mortality.
risk of mechanical ventilation, 199.5% higher, RR 3.00, p = 1.00, treatment 1 of 201 (0.5%), control 0 of 200 (0.0%), continuity correction due to zero event (with reciprocal of the contrasting arm).
risk of ICU admission, 32.7% higher, RR 1.33, p = 0.79, treatment 8 of 201 (4.0%), control 6 of 200 (3.0%).
Mulhem, 4/7/2021, retrospective, database analysis, USA, peer-reviewed, 3 authors, dosage not specified, excluded in exclusion analyses: substantial unadjusted confounding by indication likely; substantial confounding by time likely due to declining usage over the early stages of the pandemic when overall treatment protocols improved dramatically.
Submit Corrections or Updates.
 risk of death, 32.2% higher, RR 1.32, p = 0.01, treatment 157 of 794 (19.8%), control 359 of 2,425 (14.8%), adjusted per study, odds ratio converted to relative risk, logistic regression.
Patel, 10/1/2020, retrospective, USA, peer-reviewed, 8 authors, dosage 1000mg days 1-4, 500mg to 1500mg daily.
Submit Corrections or Updates.
 risk of death, 29.5% lower, RR 0.71, p = 0.18, treatment 22 of 96 (22.9%), control 26 of 80 (32.5%), NNT 10.
risk of death, 15.6% lower, RR 0.84, p = 0.60, treatment 15 of 30 (50.0%), control 16 of 27 (59.3%), NNT 11, ICU patients.
Pourhoseingholi, 5/26/2021, prospective, Iran, preprint, mean age 57.9, 11 authors, study period 2 February, 2020 - 20 July, 2020, average treatment delay 7.4 days, dosage not specified.
Submit Corrections or Updates.
 risk of death, 13.0% lower, HR 0.87, p = 0.38, treatment 54 of 199 (27.1%), control 285 of 2,269 (12.6%), adjusted per study, multivariable, Cox proportional hazards.
Rana, 6/28/2023, Double Blind Randomized Controlled Trial, placebo-controlled, Pakistan, peer-reviewed, 10 authors, study period 28 December, 2020 - 10 April, 2022, dosage 30000mg days 1-4, trial NCT04682574 (history), excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 54.5% lower, RR 0.45, p = 0.20, treatment 5 of 139 (3.6%), control 11 of 139 (7.9%), NNT 23.
risk of mechanical ventilation, 44.4% lower, RR 0.56, p = 0.41, treatment 5 of 139 (3.6%), control 9 of 139 (6.5%), NNT 35.
hospitalization time, 36.8% lower, relative time 0.63, p = 0.91, treatment 139, control 139.
Salehi, 3/11/2022, retrospective, Iran, preprint, mean age 62.0, 11 authors, study period April 2021 - September 2021, excluded in exclusion analyses: unadjusted results with no group details; very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 10.1% lower, RR 0.90, p = 0.56, treatment 22 of 40 (55.0%), control 52 of 85 (61.2%), NNT 16.
Seely, 9/22/2023, Double Blind Randomized Controlled Trial, placebo-controlled, Canada, peer-reviewed, mean age 39.9, 10 authors, study period September 2021 - April 2022, this trial uses multiple treatments in the treatment arm (combined with vitamin C, D, K2, and zinc) - results of individual treatments may vary, trial NCT04780061 (history).
Submit Corrections or Updates.
 ER visit, 47.6% lower, RR 0.52, p = 0.68, treatment 2 of 42 (4.8%), control 4 of 44 (9.1%), NNT 23.
relative mean cumulative symptom score, 13.8% better, RR 0.86, p = 0.41, treatment mean 166.3 (±92.3) n=34, control mean 192.9 (±153.6) n=24.
EQ-VAS average score <80, 29.4% lower, RR 0.71, p = 0.54, treatment 7 of 34 (20.6%), control 7 of 24 (29.2%), NNT 12, average daily EQ-VAS score <80.
relative EQ5D improvement, 28.6% better, RR 0.71, p = 0.44, treatment 32, control 31, relative improvement in EQ5D, week 1.
relative EQ5D improvement, 14.3% better, RR 0.86, p = 0.73, treatment 33, control 30, relative improvement in EQ5D, week 2.
relative EQ5D improvement, 50.0% better, RR 0.50, p = 0.17, treatment 32, control 33, relative improvement in EQ5D, week 3.
relative EQ5D improvement, 12.5% worse, RR 1.12, p = 0.47, treatment 30, control 25, relative improvement in EQ5D, week 4.
recovery time, 4.0% higher, relative time 1.04, p = 0.81, treatment 34, control 24.
risk of long COVID, 12.1% lower, RR 0.88, p = 1.00, treatment 3 of 33 (9.1%), control 3 of 29 (10.3%), NNT 80, 12 weeks.
risk of long COVID, 35.7% lower, RR 0.64, p = 0.69, treatment 3 of 35 (8.6%), control 4 of 30 (13.3%), NNT 21, 8 weeks.
risk of long COVID, 0.6% lower, RR 0.99, p = 1.00, treatment 6 of 35 (17.1%), control 5 of 29 (17.2%), NNT 1015, 4 weeks.
Sharmin, 9/1/2021, Double Blind Randomized Controlled Trial, placebo-controlled, Bangladesh, trial NCT04558424 (history).
Submit Corrections or Updates.
 Estimated 50 patient RCT with results unknown and over 5 years late.
Simsek, 9/27/2021, retrospective, Turkey, peer-reviewed, 16 authors, dosage 25000mg days 1-7.
Submit Corrections or Updates.
 risk of death, 44.1% lower, RR 0.56, p = 0.18, treatment 6 of 58 (10.3%), control 15 of 81 (18.5%), NNT 12.
risk of ICU admission, 10.2% lower, RR 0.90, p = 0.66, treatment 18 of 58 (31.0%), control 28 of 81 (34.6%), NNT 28.
Suna, 5/11/2021, retrospective, Turkey, peer-reviewed, 5 authors, dosage 2000mg daily, excluded in exclusion analyses: substantial confounding by time likely due to declining usage over the early stages of the pandemic when overall treatment protocols improved dramatically.
Submit Corrections or Updates.
 risk of death, 21.3% lower, RR 0.79, p = 0.52, treatment 17 of 153 (11.1%), control 24 of 170 (14.1%), NNT 33.
risk of ICU admission, 1.9% higher, RR 1.02, p = 1.00, treatment 11 of 153 (7.2%), control 12 of 170 (7.1%).
Tan (B), 7/26/2021, retrospective, China, peer-reviewed, 7 authors, dosage 500mg tid days 1-7, this trial uses multiple treatments in the treatment arm (combined with diammonium glycyrrhizinate) - results of individual treatments may vary.
Submit Corrections or Updates.
 risk of death/intubation, 24.5% lower, RR 0.75, p = 0.74, treatment 1 of 46 (2.2%), control 14 of 115 (12.2%), NNT 10.0, odds ratio converted to relative risk, primary outcome.
risk of ARDS, 73.3% lower, RR 0.27, p = 0.002, treatment 7 of 46 (15.2%), control 41 of 115 (35.7%), NNT 4.9, odds ratio converted to relative risk.
Tehrani, 11/8/2021, Randomized Controlled Trial, Iran, peer-reviewed, 10 authors, study period March 2020 - May 2020, average treatment delay 9.0 days, dosage 2000mg qid days 1-5.
Submit Corrections or Updates.
 risk of death, 87.1% lower, RR 0.13, p = 0.13, treatment 0 of 18 (0.0%), control 4 of 26 (15.4%), NNT 6.5, relative risk is not 0 because of continuity correction due to zero events (with reciprocal of the contrasting arm).
hospitalization time, 17.6% lower, relative time 0.82, p = 0.23, treatment 18, control 26.
Tu, 1/13/2022, retrospective, Sierra Leone, peer-reviewed, 11 authors, study period 31 March, 2020 - 11 August, 2020, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of death, 83.0% lower, RR 0.17, p < 0.001, treatment 8 of 116 (6.9%), control 26 of 64 (40.6%), NNT 3.0.
Uz, 11/13/2024, retrospective, Turkey, peer-reviewed, 6 authors.
Submit Corrections or Updates.
 risk of death, 84.2% lower, OR 0.16, p = 0.04998, treatment 41, control 46, adjusted per study, high dose, multivariable, RR approximated with OR.
risk of death, 90.2% lower, OR 0.10, p = 0.004, treatment 183, control 46, adjusted per study, low dose, multivariable, RR approximated with OR.
Vishnuram, 6/30/2021, retrospective, India, peer-reviewed, 5 authors, dosage 4000mg days 1-10, excluded in exclusion analyses: unadjusted results with no group details; minimal details of groups provided.
Submit Corrections or Updates.
 risk of death, 54.2% lower, RR 0.46, p = 0.03, treatment 164 of 8,634 (1.9%), control 10 of 241 (4.1%), NNT 44.
Yang (B), 1/15/2022, Randomized Controlled Trial, China, peer-reviewed, 11 authors, study period 1 February, 2020 - 29 February, 2020, this trial uses multiple treatments in the treatment arm (combined with TCM) - results of individual treatments may vary, trial ChiCTR2000032717, excluded in exclusion analyses: combined treatments may contribute significantly to the effect seen.
Submit Corrections or Updates.
 recovery time, 32.9% lower, relative time 0.67, p < 0.001, treatment mean 10.2 (±1.75) n=10, control mean 15.2 (±2.49) n=10, symptom disappearance, severe patients, group C vs. group A.
recovery time, 44.6% lower, relative time 0.55, p < 0.001, treatment mean 4.1 (±0.88) n=10, control mean 7.4 (±1.26) n=10, symptom disappearance, non-severe patients, group C vs. group A.
recovery time, 23.9% lower, relative time 0.76, p = 0.006, treatment mean 10.2 (±1.75) n=10, control mean 13.4 (±2.76) n=10, symptom disappearance, severe patients, group C vs. group B (high vs. low dose).
recovery time, 28.1% lower, relative time 0.72, p = 0.003, treatment mean 4.1 (±0.88) n=10, control mean 5.7 (±1.16) n=10, symptom disappearance, non-severe patients, group C vs. group B (high vs. low dose).
recovery time, 27.1% lower, relative time 0.73, p = 0.002, treatment mean 13.45 (±3.11) n=10, control mean 18.45 (±3.12) n=10, disease recovery, severe patients, group C vs. group A.
recovery time, 23.2% lower, relative time 0.77, p < 0.001, treatment mean 7.0 (±0.94) n=10, control mean 9.11 (±1.25) n=10, disease recovery, non-severe patients, group C vs. group A.
recovery time, 15.4% lower, relative time 0.85, p = 0.15, treatment mean 13.45 (±3.11) n=10, control mean 15.89 (±4.06) n=10, disease recovery, severe patients, group C vs. group B (high vs. low dose).
recovery time, 14.6% lower, relative time 0.85, p = 0.02, treatment mean 7.0 (±0.94) n=10, control mean 8.2 (±1.14) n=10, disease recovery, non-severe patients, group C vs. group B (high vs. low dose).
Yüksel, 9/20/2020, retrospective, Turkey, preprint, 13 authors, dosage 200mg/kg days 1-4, excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 18.8% lower, RR 0.81, p = 0.04, treatment 31 of 42 (73.8%), control 40 of 44 (90.9%), NNT 5.8, propensity score matching.
Zhang, 8/10/2020, Randomized Controlled Trial, China, peer-reviewed, 11 authors, study period 14 February, 2020 - 29 March, 2020, dosage 12000mg bid days 1-7, excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 50.0% lower, RR 0.50, p = 0.20, treatment 6 of 27 (22.2%), control 11 of 29 (37.9%), NNT 6.4, adjusted per study, ICU mortality.
risk of death, 80.0% lower, RR 0.20, p = 0.04, treatment 5 of 27 (18.5%), control 11 of 29 (37.9%), NNT 5.2, adjusted per study, ICU mortality for SOFA ≥3.
risk of death, 50.0% lower, RR 0.50, p = 0.31, treatment 6 of 27 (22.2%), control 10 of 29 (34.5%), NNT 8.2, adjusted per study, 28 day mortality.
risk of death, 70.0% lower, RR 0.30, p = 0.07, treatment 5 of 27 (18.5%), control 10 of 29 (34.5%), NNT 6.3, adjusted per study, 28 day mortality for SOFA ≥3.
Zheng, 9/22/2021, retrospective, China, peer-reviewed, 10 authors, study period 13 February, 2020 - 29 February, 2020, dosage 3000mg days 1-5, excluded in exclusion analyses: substantial unadjusted confounding by indication likely; immortal-time bias may significantly affect results; treatment start times unknown, treatment may not have started at baseline.
Submit Corrections or Updates.
 risk of death, 157.0% higher, HR 2.57, p = 0.33, treatment 12 of 70 (17.1%), control 7 of 327 (2.1%), adjusted per study, propensity score matching.
risk of death, 169.0% higher, HR 2.69, p = 0.07, treatment 12 of 70 (17.1%), control 7 of 327 (2.1%), adjusted per study, IPTW.
clinical improvement ≥ 2 points, 35.1% worse, HR 1.35, p = 0.17, treatment 18 of 70 (25.7%), control 16 of 327 (4.9%), adjusted per study, inverted to make HR<1 favor treatment, propensity score matching.
clinical improvement ≥ 2 points, 31.6% worse, HR 1.32, p = 0.11, treatment 18 of 70 (25.7%), control 16 of 327 (4.9%), adjusted per study, inverted to make HR<1 favor treatment, IPTW.
Özgülteki̇n, 9/22/2022, retrospective, Turkey, peer-reviewed, 4 authors, study period March 2020 - June 2020, excluded in exclusion analyses: very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 4.8% higher, RR 1.05, p = 1.00, treatment 18 of 21 (85.7%), control 18 of 22 (81.8%).
Özgünay, 7/4/2021, retrospective, Turkey, peer-reviewed, 7 authors, dosage 2000mg tid days 1-10, excluded in exclusion analyses: substantial unadjusted confounding by indication likely; very late stage, ICU patients.
Submit Corrections or Updates.
 risk of death, 9.3% lower, RR 0.91, p = 0.69, treatment 17 of 32 (53.1%), control 75 of 128 (58.6%), NNT 18.
risk of mechanical ventilation, 1.1% higher, RR 1.01, p = 1.00, treatment 23 of 32 (71.9%), control 91 of 128 (71.1%).
Effect extraction follows pre-specified rules as detailed above and gives priority to more serious outcomes. For pooled analyses, the first (most serious) outcome is used, which may differ from the effect a paper focuses on. Other outcomes are used in outcome specific analyses.
Abdulateef, 4/8/2021, retrospective, Iraq, peer-reviewed, 7 authors, study period July 2020 - August 2020, dosage varies, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of hospitalization, 18.7% lower, RR 0.81, p = 0.69, treatment 8 of 132 (6.1%), control 22 of 295 (7.5%), NNT 72, unadjusted.
Akbar, 11/7/2023, retrospective, Qatar, peer-reviewed, mean age 40.3, 9 authors, study period March 2020 - September 2020, dosage not specified.
Submit Corrections or Updates.
 risk of case, 14.0% lower, OR 0.86, p = 0.29, treatment 665, control 9,335, adjusted per study, multivariable, model 2, RR approximated with OR.
Aldwihi, 5/11/2021, retrospective, Saudi Arabia, peer-reviewed, survey, mean age 36.5, 8 authors, study period August 2020 - October 2020, dosage not specified.
Submit Corrections or Updates.
 risk of hospitalization, 36.3% lower, RR 0.64, p = 0.006, treatment 142 of 505 (28.1%), control 95 of 233 (40.8%), NNT 7.9, adjusted per study, odds ratio converted to relative risk, multivariable.
Asoudeh, 3/21/2023, retrospective, Iran, peer-reviewed, 10 authors, study period June 2021 - September 2021.
Submit Corrections or Updates.
 risk of severe case, 69.0% lower, OR 0.31, p = 0.003, adjusted per study, T3 vs. T1, multivariable, model 3, RR approximated with OR.
Behera, 11/3/2020, retrospective, India, peer-reviewed, 13 authors.
Submit Corrections or Updates.
 risk of case, 18.0% lower, OR 0.82, p = 0.58, treatment 29 of 67 (43.3%) cases, 38 of 148 (25.7%) controls, adjusted per study, case control OR, model 2 conditional logistic regression.
risk of case, 29.0% lower, OR 0.71, p = 0.24, treatment 29 of 67 (43.3%) cases, 38 of 148 (25.7%) controls, adjusted per study, case control OR, matched pair analysis.
Bejan, 2/28/2021, retrospective, USA, peer-reviewed, mean age 42.0, 6 authors.
Submit Corrections or Updates.
 risk of death, 34.0% lower, OR 0.66, p = 0.33, treatment 569, control 8,637, adjusted per study, RR approximated with OR.
risk of mechanical ventilation, 25.0% lower, OR 0.75, p = 0.47, treatment 572, control 8,657, adjusted per study, RR approximated with OR.
risk of ICU admission, 15.0% lower, OR 0.85, p = 0.65, treatment 577, control 8,690, adjusted per study, RR approximated with OR.
risk of hospitalization, no change, OR 1.00, p = 1.00, treatment 626, control 9,122, adjusted per study, RR approximated with OR.
Guan, 5/22/2024, retrospective, China, peer-reviewed, 5 authors, study period December 2022 - January 2023.
Submit Corrections or Updates.
 risk of symptomatic case, 31.4% lower, RR 0.69, p = 0.007, treatment 28 of 46 (60.9%), control 2,017 of 2,454 (82.2%), NNT 4.7, adjusted per study, odds ratio converted to relative risk, high dose, multivariable.
risk of symptomatic case, 18.5% lower, RR 0.82, p = 0.02, treatment 55 of 79 (69.6%), control 2,017 of 2,454 (82.2%), NNT 8.0, adjusted per study, odds ratio converted to relative risk, medium dose, multivariable.
risk of symptomatic case, 6.8% lower, RR 0.93, p = 0.13, treatment 129 of 167 (77.2%), control 2,017 of 2,454 (82.2%), NNT 20, adjusted per study, odds ratio converted to relative risk, low dose, multivariable.
Guldemir, 11/16/2022, retrospective, Turkey, peer-reviewed, 3 authors, study period 30 March, 2020 - 23 September, 2020, dosage not specified, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of hospitalization, 31.0% lower, RR 0.69, p = 0.046 (Fisher's exact test), treatment 33 of 173 (19.1%), control 84 of 304 (27.6%), NNT 12.
Holt, 3/30/2021, prospective, United Kingdom, peer-reviewed, 34 authors, study period 1 May, 2020 - 5 February, 2021, trial NCT04330599 (history) (COVIDENCE UK), excluded in exclusion analyses: significant unadjusted confounding possible.
Submit Corrections or Updates.
 risk of case, 2.9% higher, RR 1.03, p = 0.86, treatment 49 of 1,580 (3.1%), control 397 of 13,647 (2.9%), adjusted per study, odds ratio converted to relative risk, minimally adjusted, group sizes approximated.
Louca, 11/30/2020, retrospective, United Kingdom, peer-reviewed, 26 authors, dosage not specified.
Submit Corrections or Updates.
 risk of case, no change, RR 1.00, p = 1.00, odds ratio converted to relative risk, United Kingdom, all adjustment model.
Loucera, 8/16/2022, retrospective, Spain, peer-reviewed, 8 authors, study period January 2020 - November 2020.
Submit Corrections or Updates.
 risk of death, 28.3% lower, HR 0.72, p = 0.002, treatment 840, control 15,128, Cox proportional hazards, day 30.
Mahto, 2/15/2021, retrospective, India, peer-reviewed, 6 authors.
Submit Corrections or Updates.
 risk of IgG positive, 25.9% higher, RR 1.26, p = 0.49, treatment 34 of 140 (24.3%), control 59 of 549 (10.7%), adjusted per study, odds ratio converted to relative risk, multivariable.
Mohseni, 8/4/2021, retrospective, Iran, peer-reviewed, 4 authors, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of case, 44.2% higher, RR 1.44, p = 0.002, treatment 34 of 43 (79.1%), control 307 of 560 (54.8%).
Nimer, 2/28/2022, retrospective, Jordan, peer-reviewed, survey, 4 authors, study period March 2021 - July 2021, dosage not specified.
Submit Corrections or Updates.
 risk of hospitalization, 24.7% lower, RR 0.75, p = 0.08, treatment 52 of 651 (8.0%), control 167 of 1,497 (11.2%), NNT 32, adjusted per study, odds ratio converted to relative risk, multivariable.
risk of severe case, 17.0% lower, RR 0.83, p = 0.18, treatment 66 of 651 (10.1%), control 194 of 1,497 (13.0%), NNT 35, adjusted per study, odds ratio converted to relative risk, multivariable.
Sharif, 11/26/2022, retrospective, Bangladesh, peer-reviewed, 14 authors, study period 13 December, 2020 - 4 February, 2021.
Submit Corrections or Updates.
 risk of severe case, 46.0% lower, OR 0.54, p = 0.001, adjusted per study, multivariable, RR approximated with OR.
Shehab, 2/28/2022, retrospective, multiple countries, peer-reviewed, survey, 7 authors, study period September 2020 - March 2021, excluded in exclusion analyses: unadjusted results with no group details.
Submit Corrections or Updates.
 risk of severe case, 4.3% lower, RR 0.96, p = 1.00, treatment 14 of 139 (10.1%), control 12 of 114 (10.5%), NNT 220, unadjusted, severe vs. mild cases.
Vaisi, 5/11/2023, retrospective, Iran, peer-reviewed, 5 authors.
Submit Corrections or Updates.
 risk of hospitalization, 37.9% lower, HR 0.62, p = 0.17, treatment 2,818, control 1,137, adjusted per study, inverted to make HR<1 favor treatment, sufficient vs. insufficient intake, multivariable, Cox proportional hazards.
risk of symptomatic case, 9.6% lower, HR 0.90, p = 0.71, treatment 2,818, control 1,137, adjusted per study, inverted to make HR<1 favor treatment, sufficient vs. insufficient intake, multivariable, Cox proportional hazards.
Footnotes
a.
Viral infection and replication involves attachment, entry, uncoating and release, genome replication and transcription, translation and protein processing, assembly and budding, and release. Each step can be disrupted by therapeutics.
b.
Monoclonal antibodies were previously included. Other treatments such as dexamethasone, tocilizumab, and baricitinib were recommended for late stage hospitalized patients.
c.
Smoking was known to cause lung cancer since at least 1939, but this was not widely recognized in the US until 1964, 25 years later. Surgeon general Leroy Burney tried publicizing the danger starting in 1957, with limited success. Surgeon general Luther Terry, appointed in 1961, prompted by President Kennedy in 1962 amid pressure from health advocates, finally got recognition in 1964. The 1964 report reviewed 7,000+ studies, but these could (and should) have been reviewed and acted upon in real-time as they were published. Historians attribute the 25 year delay to an industry campaign to manufacture doubt and controversy, through tactics like funding biased and fraudulent research from "independent" organizations, attacking scientists, and political lobbying. The success of the industry campaign is only possible because officials did not analyze the research in detail. Fraudulent industry research supported prior failures, but would have been called out as fraudulent by officials that analyzed and understood the research in real-time.
d.
When administered late in infection, HCQ may enhance viral egress by further increasing lysosomal pH beyond the effect of ORF3a's water channel activity, thereby promoting lysosomal exocytosis, inactivating degradative enzymes, and facilitating the release of SARS-CoV-2 particles into the extracellular environment319,320. Research also suggests potential cardioprotective effects at lower doses, but cardiotoxicity with excessive dosage321. Bobrowski et al. also indicate negative effects if HCQ and remdesivir are combined.
e.
Peters (B) et al. is subject to confounding by calendar-time (SOC evolved rapidly early in the pandemic, the linear covariate does not reflect non-linear SOC changes and hospital specific effects), hospital type (non-treatment hospitals were tertiary university centers), confounding by indication (4/7 hospitals initiated treatment on deterioration), immortal-time bias for as-treated (exposure assigned after baseline), significant differences for other experimental treatments, potential overadjustment from collider bias (steroid use and indication bias), limited baseline severity information, differences in hospice referral propensity across hospitals, unadjusted difference in time from onset to admission, difference in PCR positivity, and other factors. Mahévas et al. is subject to confounding by hospital (treatment highly dependent on the hospital, different SOC/ICU transfer practices, not included in PS), immortal time (only partly addressed in sensitivity analysis), co-treatment differences, calendar-time (SOC evolved rapidly early in the pandemic), binary coding for age (age ≥65 despite steep age-risk gradient), residual imbalance (variables dropped from PS), a composite outcome dependent on hospital triage/capacity, and other factors.
References