Treatments

Wikipedia has a list of repurposed drugs, a bit outdated. 

These are now part of standard care for vulnerable (e.g. elderly) non-hospitalized patients:

These are now part of standard care for hospitalized patients:

  • Dexamethasone, 17% mortality reduction.
  • Remdesivir (AKA Veklury), injected. There was some controversy about its effectiveness, especially given its cost, but a more recent study found it cut the chance of going on a ventilator by half.

These are good:

  • Monoclonal antibodies tend to work for a while and then become ineffective, so I stopped monitoring them for a while. This is what I currently know:
    • This press release (2024-01-10) says that pemivibart neutralizes about 75% of the past and current COVID-19 strains.
  • Oral therapies:
    • Metformin:
      • This preprint from USA (2025-01-15) reports that metformin didn’t reduce hospital stays significantly.
      • This paper from USA (2024-09-17) reports that, compared to people taking other diabetes medications, people taking metformin had ~15% lower risk of Long COVID or death.
      • This paper from USA (2024-05-25) suggests that metformin might be even better for women than for men.
      • This paper from USA (2024-05-01) found that people who took metformin were 28% less likely to detectable viral load than placebo at day 5 or 10, and 32% less likely to have viral rebound than people who took a placebo.
      • This paper from USA (2024-06-08) reports that COVID-positive overweight or obese (but not diabetic) patients were 41% less likely to get Long COVID than people who got a placebo.
      • This preprint from the USA (2023-06-07) says that metformin reduced the COVID-19 viral load by 3.6x over placebo.
      • This paper from the USA (2022-12-24) says that metformin reduces Long COVID by almost half.
      • This paper from USA (2022-11-17) reports that people with Type 2 diabetes were 18% less likely to die if they took metformin compared to those who took dipeptidyl-peptidase-4-inhibitors and 64% less likely to die compared to those who took sulfonylureas.
      • This paper from USA (2022-11-14) reports that
        • prediabetics who took metformin had a 37% lower risk of mild COVID-19 and a 82% lower risk of severe COVID-19 than those who took levothyroxine;
        • prediabetics who took metformin had a 96% lower risk of mild COVID-19 and a 81% lower risk of severe COVID-19 than those who took ondansetron;
        • people with polycystic ovary syndrome (POS) who took metformin had the same COVID-19 risks as those who took levothyroxine;
        • people with POS who took metformin had a 90% lower risk of mild COVID-19 and a 99% lower risk of severe COVID-19 than those who took ondansetron.
      • This paper from the USA (2022-08-18) says that metformin reduces the chance of at least one of death, hospitalization, emergency visits, and hypoxemia by 16%.
      • This paper from USA (2021-01-12) reports that diabetics who were taking metformin before getting COVID-19 infections were 67% less likely to die than those who were not.
      • This paper from the USA (2020-12-03) found that obese or diabetic women hospitalized for COVID-19 were 22% less likely to die if they took metformin than if they did not. (Curiously, there was no benefit for men.)
    • Ensitrelvir (brand name Xocova, approved in Japan for emergency use on 2022-11-22, full approval on 2024-03-05):
      • This paper (2024-02-09) found that patients who took ensitrelvir recovered about a day faster than those who did not.
      • This 2023-02-22 press release reports that patients who took Ensitrelvir Fumaric Acid took 80% less time to test negative.
      • 2023-07-13 preprint: Patients who took ensitrelvir got better about 24 hrs faster than those who didn’t.
    • There are quite a few papers now that say that the gut biome is messed up by COVID-19. Some explicitly say that probiotics help. This Twitter thread has more (if Twitter is still alive), but here’s a list:
      • 2022-09-28 paper: hospitalized patients who took a probiotic containing the K12 strain of S. salivarius were 3.4 times less likely to die than controls.
      • 2021-08-15 Letter to Editor: None of 33 Italian schoolchildren taking S. salivarius K12 supplements got COVID-19, while 24 of 46 (52%) of the controls did.
      • S. Salivarius is available online in some countries, e.g. BioNAZE in the US.
      • See this Twitter thread for someone else’s summary of probiotics.
    • antihistamines:
      • H1 antagonists. NOTE: First-generation antihistamines DO cross the blood-brain barrier and second-/third-generation antihistamines DO NOT.
        • This paper using data from Honduras (2024-11-26) reports that patients given intranasal chlorpheniramine (first-gen) immediately after a positive COVID-19 test had WAY fewer Long COVID symptoms than people who got placebo.
        • This paper from China (2024-07-02) reports that a variety of H1 antagonists did a good job of blocking COVID-19 from entering cells in test tubes and mice with human immune systems. (They tested it as a prophylactic, but if they are correct about the mechanism, it also should be a good treatment.)
        • This paper from Spain (2023-04-25) found that COVID-positive people given H1-antagonist antihistamines (mostly cetirizine (second-gen) or dexchlorpheniramine (first-gen)) had about half the risk of hospitalization as controls. They did not break out first-gen antihistamine results vs. second-gen. (NB: Patients over 40 were also given azithromycin, an antibiotic.)
        • This case study (2022-02-07) reports on two women with bad Long COVID who got ~90% of function back after starting to take daily H1 antagonist antihistamines (diphenhydramine and hydroxyzine (both first-gen for one woman and diphenhydramine (first-gen) for the other) and . Interestingly, the second woman only had improvement when she switched from fexofenadine (a second-gen) to diphenhydramine (a first-gen).
        • 2021-01-25 paper from Spain: 0% — absolutely none — of nursing home residents at two specific nursing homes who got COVID-19 (which was all of them!) treated with H1 antagonists (mostly dexchlorpheniramine (first-gen), with some cetirizine or loratadine (both second-gen)) died or even went to the hospital, while peer nursing homes had a 28% fatality rate.
        • This paper from USA (2020-09-07) reported a 15.5% decrease in inpatient mortality risk in hospitalized COVID patients who were given both cetirizine (H1, second-gen) and famotidine (H2, second-generation).
      • There was some excitement early in the pandemic about H2 antagonists — electronic record data seemed to indicate that famotidine helped against COVID — but later research seemed to say that H2 agonists weren’t helpful. (Sorry, I appear to have deleted the H2 section when I thought H2 antagonist antihistamines weren’t useful.)
        • This paper from USA (2023-07-10) reports that famotidine (an H2 antagonist) helped hospitalized patients a lot. Famotidine users had 35% lower risk of all-cause mortality.
      • Glycyrrhizin (licorice root extract, which is commonly used in Chinese herbal medicine) has some research supporting it BUT please be careful. It apparently can cause problems in pregnancy and might be risky if you have kidney issues or high blood pressure.
        • This article (2025-01-02) reports that a study in Iran (which I could not find) found that patients in critical care who took licorice root extract spent 13.1 days in hospital, while the placebo group spent 25 days in hospital.
        • This paper (2024-10-09) reports that glycyrrhizin blocks cell attachment in test tubes.
        • This paper (2023-10) reports that glycyrrhizin reduced time in the ICU, but didn’t change mortality rates or duration of ventilation time.
        • This paper (2021-07-27) reports that patients hospitalized with COVID-19 who took diammonium glycyrrhizinate plus Vitamin C has a 86% lower risk of acute respiratory distress syndrome (ARDS).
        • This paper (2021-04-02) reports that glycyrrhizin blocks Mpro (which is important in viral replication of SARS-CoV-2) test tubes.
        • This paper (2021-11-26) reports that screening identified licorice-saponin A3 and glycyrrhizin as compounds which might bind to the SARS-CoV-2 spike’s receptor binding domain (and hence prevent the virus from docking with the human cell).
        • This paper (2020-06-24), this paper (2024-09-09), this paper (2021-02-17), this paper (2021-06-08), and this paper (2022-11-30) give theoretical arguments on why glycyrrhizin might work against SARS-CoV-2.
  • Nasal therapies (also useful as a prophylactic sometimes, but note this Reddit post (2025-02-22) or this blog post (2025-02-22) (the same content, different formats) which takes issue with many of these studies):
    • Sinus irrigation (aka “neti pot”)
      • 20231108 This paper from USA says that people who gargled and irrigated their nose with saline had ~1/3 the hospitalization rate of controls.
      • 20220825 This preprint says that nasal irrigation of good old saline is quite effective at reducing the severity of COVID-19. (The effectiveness that the article gives says that it’s more effective than Paxlovid!)
      • This paper says symptoms resolved 4-7 days earlier for people who irrigated with saline.
      • This paper from right before the pandemic says that people with the common cold (note: many colds are from non-COVID-19 coronaviruses) who irrigated with saline resolved symptoms about two days earlier than those who did not, and their household transmission was 35% lower than for controls. Looking at the graphs of symptoms over time for each group, it looks to me like the difference is mostly in eliminating long-duration outliers.
  • Mouthwashes: I was sceptical that mouthwashes might help reduce transmission or disease severity because killing viruses in the mouth doesn’t get rid of virus in the nose, lungs, gut, or other organs. HOWEVER, there is some evidence that it does help:
    • This paper from Ghana (2022-06-25) reported that way fewer health care workers and patients who rinsed their mouth with a hydrogen peroxide solution got COVID-19 than those who did not use the mouthwash.
    • 2022-06-21 This paper from Spain reports that the saliva of people who gargled with a 0.07% cetylpyridinium chloride (CPC) mouthwash showed a significant increase in proteins from the SARS-CoV-2 nucleocapsid compared to placebo. (Translation: the CPC ripped the SARS-CoV-2 to shreds, leaving lots of pieces of the interior of the SARS-CoV-2 virus in people’s saliva.)
    • 2022-03-05: This paper says that in a clinic where patients rinsed with dilute hydrogen peroxide once they entered a treatment room, they found virus in the waiting room’s filter but not in the treatment rooms’ filters.
    • 2002-04-27: This prepandemic paper found that Japanese schoolchildren who gargled with Povidone-Iodine had an absentee rate of 19.8% for upper respiratory infections compared to 25.5% in schools where the students did not gargle.
    • 2005-10-19: This prepandemic paper from Japan found that people who gargled with povidone/iodine had an 11% lower risk of getting an upper respiratory disease than controls. Interestingly, people who gargled with tap water had a 36% lower risk than controls! (Maybe this is due to the tap water in the study’s city being heavily chlorinated? I saw something about that in a different paper from Japan.)
    • 2002-04-27: This prepandemic paper from Japan found that adults who had histories of ongoing problems with repeated upper respiratory infections decreased their infection frequency dramatically after gargling with povidone/iodine regularly (4x/day).
    • For which solution to rinse with:
      • 20210301: This paper measured the effects of various mouthwashes and concluded that hydrogen peroxide (Colgate Peroxyl) was best at killing COVID-19 in the mouth, followed by povidone-iodine, chlorhexidine gluconate (CHG), and Listerine (essential oils and alcohol).
      • This paper (20230703) and this paper (20201210) say povidone iodine gargle reduces viral load in the mouth.
      • 20200815 This review article discusses mouthwashes and viral loads in the mouth, but was kind of wishy-washy about relative merits of different mouthwashes. Same with this review article (20200905).
      • 20210707: This paper found that there wasn’t much difference between Chlorhexidine 0.2% and 1% Povidone-iodine solutions.
  • Eye drops
    • 20201124 This article says that eye drops Lumify and Besivance got rid of COVID-19 in the eye. That probably isn’t going to help you avoid getting COVID-19 any better than a pair of glasses, but maybe? it will help COVID-related conjunctivitis?

There are quite a few papers now that say that the gut biome is messed up by COVID-19. Some explicitly say that probiotics help. See this Twitter thread for a summary.

In the pipeline:

This paper says that glycine supplements help defend against viruses in general.

Potentially effective?

  • sarilumab
  • inhaled budesonide — one study says that it reduces hospitalization, fever, recovery period
  • amodiaquine (an antimalarial) shows good results in hamsters
  • ADG20, a monoclonal antibody
  • Ciclesonide
  • Vitamin D looked good, then it didn’t, and now here’s a press release saying it is useful.
  • 20231025 This paper found that people who got simvastatin — a statin — were 12% more likely to survive a serious COVID-19 infection. However, this was not enough to meet the prespecified criteria for superiority to controls.

Also see https://www.canada.ca/en/health-canada/services/drugs-health-products/covid19-industry/drugs-vaccines-treatments/authorization/applications.html

These used to be good:

Long COVID

Prevention

These treatments have been shown to reduce the incidence of Long COVID:

Long COVID Treatments

In Galen’s Long Covid Survival Guide, a guy tells what worked (partially) for him with Long COVID. tl;dr: antihistamine, low-dose Naltrexone, nicotine patches, antibiotic Amox-Clav, probiotic, better nutrition, liposomal Glutathione supplement, Melatonin.

Advice

This page has a big section on what to do if you get COVID-19.

Not Effective

  • This paper (20221215) found that gabapentin was effective at restoring smell, but this random controlled trial (20230921) found that it was not. The first paper was smaller and did not have controls, so I trust the second.
  • @@@TBD@@@ list all the monoclonal antibodies which don’t work any more against acute COVID-19