It is very unfortunate that many people are spreading misinformation about the crucial issue of which medicines have already been shown to help COVID19 and which have not. This is not made any easier by the tendencies for even the health authorities in some countries to leap in with both feet if there is any hint that something might help.

UPDATE – Read this post on Adrian’s other blog: Why are Christians, who are meant to love truth, promoting lies and crazy conspiracy theories about COVID19?

Recently the UK medical scientists who are responsible for a number of breakthroughs that are being shared with the rest of the world have been critical of the US medical system for using treatments for COVID19 without testing them.

Their specific point was that convalescent plasma, which is felt likely to be a promising option for the treatment of people with COVID19, ought not to have been given to 10,000 patients without using a randomised control trial to simultaneously demonstrate whether it was working or not. After all these patients we still do not know for sure if it is helping, making patients worse or making no difference. You could almost hear them sigh as they explained that yes, we in the UK are currently also in the middle of yet another RCT to help inform the whole world whether or not this intervention actually works.

At the core of the huge difference between the medical attitudes of the UK and the USA is a difference in philosophy. UK scientists are sceptical and well aware that for example even now they know that the Oxford vaccine prompts the creation of COVID-19 adapted antibodies and T-cells, we should not dare to presume that it will definitely work in terms of preventing infections or at least significantly reducing the severity of infections. Of course the hope and expectation is that it will work and we are planning on manufacturing over a billion doses of just that vaccine. But the phrase do not count your chickens before they have hatched springs to mind. And so the randomised control trials continue. We are not rushing in to assuming this vaccine will definitely work.

Some American scientists might reply “but you are going too slowly. People are dying. We have to do something. WE think this intervention works, and it is unethical to not treat someone when they might die in the meantime of this dreadful disease.” It is very emotive perhaps, but some of us have been involved in clincial research long enough to know that many a promising treatment has been shown not to work in a proper clinical trial.

There is also a more individualistic mind set in the USA I think. There is a concern that some patients might be denied a medicine that would work for them. But we would argue firstly that we do not yet know if they are instead being denied a drug that was going to harm them not help them. And we have a collectivist desire to find out as quickly as we possibly can what is good for everyone.

A lot of people, including it seems many doctors seem to think it is easy to tell if a medicine is working. Take a sick person, give them the medicine, and see if they get better. If they got better the medicine worked. If they didn’t then the medicine failed.

But this reminds me of an old medical joke about when someone came to their GP with a bad dose of a cold, maybe even a man flu. “Take yourself to bed for a rest young man, and if you like take this medicine. If you take it you will be better in a week, if you do not you will be better in seven days”

The nature of COVID19 is that many people are now being tested positive who have very mild, or in some cases non-existent, symptoms. If you take a group of such patients and give them a medicine to stop them getting worse then you could find that none of your patients did get worse and proclaim you had a wonder drug on your hands. But you have proved nothing as they might have remained well, or even if they were sick got better whatever treatment they were given.

It is nothing short of shameful that some doctors have even been claiming to have discovered a cure for COVID19. There is no such thing. And when people are dying it is foolish to tell the world that we do not need masks any more because they claim a cure has been found! This will further encourage people not to socially distance and further increase the infection and death rate.

Today I want to not only give you a heads up of where we are as far as medicines that have been shown to work for some patients with COVID19 but also help you understand how to assess the evidence base, and appreciate why some of the wild conspiracy claims that are circulating are just that conspiracies.

What I want us to learn to focus on with a razor sharp edge is randomised control trial. Yes, we can look at cases series and other studies some of them quite large and feel impressed by the reported results enough to want to proceed to a trial. But anything except a RCT proves nothing. It is too possible that any observational trial no matter how large is too biassed to be able to confidently come to any firm conclusions.

Imagine if you will you believe your drug will work. The temptation as a doctor is to only give it to the least unwell patients. And maybe you have to refer on the sicker patients anyway so cannot discuss starting your study medicine with them. And perhaps the sickest patients of all cannot take your medicine because it needs to be swallowed and they cant swallow. If then, say 80% of the patients who started your drug were not even on oxygen but 80% of the patients who did not start your drug were on oxygen or even a ventilator, then it doesn’t take a genius to predict that when you look at the results the people who take your drug are going to be more likely to survive and to have left the hospital more quickly.

Bias is why we need randomised control trials. They work in a remarkably simple way. You take a group of patients who are all willing to take the drug or not, and you effectively toss a coin to decide who gets the treatment and who doesn’t. You have to think carefully about what to give to the person not on treatment. Maybe they will get another drug to compare with. Or maybe they will get nothing if there is no treatment yet. And you have to think if it is ethical to allow the patients in the no treatment arm to be rescued by the treatment later in the study if they have not yet got better.

These days of course we use a computer to randomise rather than a coin toss but it is the same idea. We have no bias now and the patients should be evenly spread between the treatment groups. You can then look at what proportion of people taking your new treatment got better compared to the other group. Then you know for SURE if your medicine is working or not, unless your study was badly conducted, or was not big enough to give a confident “no”. If you do ten small studies even if randomised they might all say “no difference found” but if you had done one study ten times as big then you might have found the difference.

So how does this apply to our current situation. Lets look at which medicines may and may not work.

Does Hydroxychloroquine  work against COVID-19?

First up Hydroxychloroquine. No the government and big pharma is not hiding evidence this works! In fact the evidence is very weak indeed. The story all seems to have started with with an enterprising French doctor who reported giving the drug to a few patients who got better. This was all very exciting. But it was not randomised. There has been some suggestion since then that the medicine might be making a difference but this has only come from non-randomised case series. One of which was large but this may still be affected by bias. Randomised controlled trials are the gold standard for a reason.

Several large randomised control trials (RCTs) have shown no benefit whatsoever for hydroyxchloruqine either in hospitalised patients, nor mild patients, nor as a preventive measure.

The UK Recovery study has recruited over 12,000 patients and has clearly demonstrated the futility of the use of hydroxychloroquine in hospitalised patients. 26.8% of patients randomised to  hydroxychloroquine arm vs. 25.0% of those on usual care died. This difference was not statistically significant, but clearly given the large numbers involved, these results showed the medicine was not working.

This all really should have been enough to kill the idea of using a malaria drug to treat COVID19 altogether. What is the point of doing these trials if we do not listen to the results? But there remains a lot of interest in Hydroxychloroquine, and it continues to be studied in over one hundred and fifty clinical trials, which seems a little excessive to me. It does seem highly unlikely that any new randomised controlled trial will contra-indicate the results of previous studies. How many failed studies will it take before we give up?

It seems we can say with a high degree of confidence that Hydroxychloroquine categorically does not work against COVID19. Or at the very least if there is any real effect it must be very small and we must have been very unlucky not to detect it in any RCT studies so far.

There are a handful of other RCTs I haven’t mentioned, and if you are interested in understanding the full picture then I highly recommend this video and article by Dr Perry Wilson, a USA medic who goes through each of them carefully clearly and simply.

Dexamethasone – the first treatment shown to save lives in COVID-19.

We mentioned the Recovery trial earlier. That impressive study has clearly demonstrated that a simple old cheap medicine, dexamethasone saves many lives if used in hospitalised patients with COVID19.

In the overall group we see a small but statistically significant difference in mortality, 482 patients (22.9%) in the dexamethasone group and 1110 patients (25.7%) in the usual care group died within 28 days after randomisation (age-adjusted rate ratio, 0.83; 95% confidence interval [CI], 0.75 to 0.93; P<0.001).

But the main effect of this medicine seems to be on the most severely unwell patients. There was no difference seen between patients who were not receiving any respiratory support at randomisation. But in those receiving mechanical ventilation only 29.3% died when treated with dexamethasone vs. 41.4% who were not treated. This is a dramatic difference and this study has changed practice in many countries.

The idea peddled by some that somehow big Pharma is suppressing the notion that older drugs can work against COVID19 is surely contra-indicated by this data being clearly reported throughout the medical literature. There is no reason why every hospitalised patient should not now be considered for this medicine by their doctors, particularly those needing respiratory support.

Lopinavir-Ritonavir

This has also been shown to be ineffective by the Recovery trial.

Azithromycin

Currently the Recovery trial continues to look at data for this medicine which implies that the data so far has neither convincingly demonstrated that the medicine works nor that it definitely does not. We await further results with interest. The recovery trial continues to also study Tocilizumab (an anti-inflammatory treatment given by injection) and Convalescent plasma (collected from donors who have recovered from COVID-19 and contains antibodies against the SARS-CoV-2 virus).

Remdesivir

A study of this medicine including patients clearly demonstrated that there was a reduced time to recovery in patients taking Remdesivir. There was a numerical difference suggesting lower mortality also, but this was not statistically significant perhaps suggesting the study was a bit underpowered for that end point.

What about Oleandrin?

Despite recent reports that the White House is enthusiastic, this plant poison has not passed clinical trials for COVID19 and is at this stage just another bit of over-enthusiasm. To quote CNN

“No studies involving oleandrin and Covid-19 have been published in peer-reviewed medical journals, and there’s no public information to show oleandrin has been tested in Covid-19 patients. One pre-print study, which hasn’t been peer-reviewed or published, found the extract had antiviral effects against Covid-19 in cells in the lab.”

CNN

Monoclonal Antibodies

Currently several monoclonal antibodies are in clinical trials are . The idea here is a form of passive immunisation, similar to the use of convalescent plasma. Antibodies have been manufactured and purified and the hope is that these will both prevent infection with COVID19 and treat cases. We await the results of these trials with interest as the hope would be that such antibodies might be of use even to patients with severe immune suppression such as is often seen in blood cancer. Pharma companies involved include Lilly, Regeneron and Astra Zeneca.

There are many other treatments either being looked at pre-clinically or in early clinical trials. If you are unfortunate enough to catch COVID19 and have the opportunity to enter a clinical trial, do consider participating. The whole world will be in debt to the many thousands of volunteers who are helping discover which medicines actually work for COVID-19.

Learn More

• I highly recommend this video and article by Dr Perry Wilson, a USA medic who goes through each of the Hydroxychloroquine RCTs carefully clearly and simply.
• CNN Update on treatments for COVID19
• Covid19 Treatment Tracker
• Which drugs and therapies are proven to work