I see many problems with all the caution around that study:
1/ chloroquine is a very well known cheap drug. There is no new risk associated to using it that doctors arent already aware of.
2/ we do have a lot of statistics now on the outcome of patient not treated with that drug. There's no need for establishing a benchmark. Just proper categorization of existing patient should be enough to compare.
3/ people are oversaturating ICU right now. And as such, it is an emergency situation. We should take the reverse reasoning we usually take : if there's no suscipicion this drug could cause new problems, we should have people massively use it whenever possible and look at the result after 6 days (since that's the time the original study says it takes for the first results to show).
> 3/ people are oversaturating ICU right now. And as such, it is an emergency situation.
From what I can gather chloroquine is already used as a first line treatment in Italian hospitals (at least it is indicated as such by the COVID treatment guidelines published by the society of Italian infectious disease specialists [0], and has been for quite some time - this document is from 10 days ago, but an earlier revision was not different regarding chloroquine). If this is the case, it's pretty clear that this is no miracle cure - it has at best a marginal effect.
The number of patients already overwhelmed the healthcare system.
Per some interview with some Chinese doctors who battled COVID19, they said that for many critical patients, their viral load is actually low, but the virus had caused damages to all parts of their bodies thus antiviral isn't going to do anything significant at that stage.
This might suggest chloroquine could be helpful at earlier stage of COVID19, to help prevent patients developing into server/critical stage, which would still be a very good thing if hold true.
It might be, we don't know how the Italian system would cope without it. Due to increasing but low test numbers, we don't know how many infected patients there are in Italy. The situation in some hospitals is severe, but it has been for ~2 weeks.
I believe with the number of patients they have, they would've updated the guideline if it wasn't effective at all. But how effective it is we don't know. If a large percentage of some areas is already infected but the health system is still in the state it's in now, it could actually be a good sign. If the current numbers are realistic (which I don't believe at a positive test rate of >30%), it's probably not. We simply cannot know.
Thanks for this reference, it seems they are proposing 2 potential treatments (p. 13) for the more severe patients
> Il trattamento deve essere accompagnato da trattamento antivirale (lopinavir/ritonavir o remdesivir + clorochina/idrossiclorochina) e/o steorideo (desametasone)
This is from the 13th of March so I guess there will be still some days until we see results of this (either positive or negative)
For the record since that decision day, the growth rate in the number of deaths in Italy has fallen in half[1]. In the 9 days from March 13 to March 22 the daily number of deaths grew 160% from 250 to 651. Prior to that, the same level of growth occurred in 4 days between March 9 (96 deaths) to March 13.
That seems like pretty strong evidence that the new treatment protocol is drastically decelerating the pandemic in Italy.
This does not make much sense to me. If it was effective, it should decrease the lethality rate, not the growth in the number of deaths which is much more affected by the actual number of cases. In the page you link you can see that lethality rate has not fallen. Also, as I wrote I'm pretty sure that I saw an earlier version of that document, dating IIRC to the beginning of the month, that prescribed more or less the same combination of chloroquine and antivirals as a first line treatment.
I have no clue how much it is actually used, couldn't find any info and fortunately I don't have any direct experience yet. If it is used, the lethality rate in Italian hospital definitely show it's not incredibly effective, at least in the way they are using it.
> we do have a lot of statistics now on the outcome of patient not treated with that drug. There's no need for establishing a benchmark. Just proper categorization of existing patient should be enough to compare.
This is very wrong; we have good aggregate statistics across lots of different hospitals, but for a study with a small number of sites where the patients are treated with the drug, there's enough inter-site variability that may obscure any effect, or create a false effect.
This is doubly true as health care varies as sites get overwhelmed, and healthcare workers get increasingly strained from long shifts. There will be far more variability as time goes on.
And for how small we expect the effect to be, based on the data in the paper, we really should include controls.
All that said, I think that the publication of this flawed study is very good, and why pop-science takes that excoriate science as "most research finding are false" really communicate the wrong way to think about science, and limit its applicability. Flawed datasets like this still give us clues, and publishing flawed data is still useful to others to accelerate the pace of science. Which is why we need to take the attitude that publications are point in time guesses at what's going on, and only rarely does one come out that can be considered on its own as definitive, and that it's good to have mostly "here's some data and analysis but we don't have a complete theory yet"-type-papers. Without those intermediate publications, science would grind to a halt.
> but for a study with a small number of sites where the patients are treated with the drug, there's enough inter-site variability that may obscure any effect, or create a false effect.
The simple solution is to spread it over enough sites so that the law of large numbers quantifies that variability into the dataset. For example if we run trials over 10 hospitals in 20 different states, 200 sites in the dataset would almost certainly capture inter-site variance.
That only normalizes natural variances in your patients, not artificial biases introduced by the healthcare system. Pump and other hardware measuring out dosages might come from different manufacturers or be calibrated and serviced by different technicians. Maybe the study includes one too many hospitals in a region dominated by one manufacturer whose hardware tends to error more in one direction than other manufacturers. The generics used by pharmacies and hospitals might be slightly different because of local relationships. Exercise, ability to have visitation in quarantine, and other amenities in hospitals will have differing impacts on confounding factors like exercise and morale. And on and on.
There are a bunch of "contract research organizations" all over the world, quite a few of them making billions of dollars a year, who specialize only in running clinical trials because these things are so complicated. The documentation for clinical trials are often thousands of pages long describing tiny details like device calibration across hospitals trying to eliminate as many risks as possible. These organizations tend to have long standing relationships with hospitals and host many different trials in each one with a bunch more subcontractors who do patient selection.
As I said, drugs just have effects. Humans then look at the results, decide which are good and bad, and then label the bad ones "side effect". So I think the real question is: why do you think 100% of the novel effects will be later put in the "good" bucket?
Opinions like this seem to be popular these days, but they're dangerous. Because this is an emergency we want reliable information.
Think about it like this: If these researchers had done a proper trial - and there's absolutely no reason to think that they couldn't have done that in the same amount of time with the same amount of effort - then we'd actually know something useful now. With that trial - we don't.
Agreed, this kind of thinking comes from fear and desperation (and another kind is denial of reality - this can't be happening so it isn't. My neighbours still don't realise it's actually happening).
If we want to deal with covid we should have started years ago. As ever, we didn't.
They are in a war zone in Italy and presumably in the US very soon. If hydroxychloroquine is safe (except for some people with a known predisposed genetic condition), then we should be giving it widely and early. We can worry about the science and studies later on, but right now we need to save lives and decrease ICU overload.
> chloroquine is a very well known cheap drug. There is no new risk associated to using it that doctors arent already aware of.
I mean, you could say the same of practically any older drug. It's a fairly well-understood, cheap, somewhat dangerous drug.
> we do have a lot of statistics now on the outcome of patient not treated with that drug.
Sure. We also have a lot of statistics on the outcome of patients not treated with thalidomide, and morphine, and basically any other cheap somewhat dangerous drug you care to mention.
> if there's no suscipicion this drug could cause new problems
While I don't think there are, there's the certainty that it will cause the usual old problems that this drug causes. Which you probably don't want when you also have covid-19.
We know the side effects for a healthy individual or those with malaria, but how do we know for certain this won't cause new problems in those with COVID-19?
This is irresponsible. This drug has severe side effects, it makes absolutely no sense to prescribe it for a completely different disease without first confirming in clinical trials that it has positive effects.
> if there's no suscipicion this drug could cause new problems, we should have people massively use it
That's not how evidence-based medicine works. The drug is known to cause new problems because of the side effects. Clinical trials are needed to determine whether these are outweighed in patients by positive health effects.
You do not prescribe medicine on the basis of hunches.
> This is irresponsible. This drug has severe side effects
What are you talking about? The severe side effects are known and are usually associated with chronic usage of that drug. Chloroquine is taken for prophylaxis against malaria (daily) in many places in the southern hemisphere. Usual regimen is at least for 8 continuous weeks!
The suggestion on the table is to give this drug to extremely ill patients who suffer from a new viral disease on the basis of anecdotal evidence and without clinical trials. The suggestion was to massively prescribe it.
What health authorities want to do is to allow physicians to use the drug if the patient agrees, but only in a controlled setting in which data is collected and there is a control group. So at least afterwards we could draw conclusions about whether it killed more people or helped saving people. Does that not make sense to you?
There are actually several suggestions, which were repeatedly discussed a decade ago in multiple SARS studies (see my post below).
First is prophylaxis for front line health care workers, who are suffering grave infections likely exacerbated by long hours and low morale. Long term use of chloroquine or alternatives is a concern for this audience. There were fatalities last time, just like this time.
The second is for the patients themselves, who will otherwise endure uncontrolled viral replication for up to two weeks until the adaptive immune response recognizes the pathogen. Reducing transmission requires an earlier end to this stage.
Is the data from the previous studies worthless for these goals?
I feel like this thread and the arguments contained in it are a microcosm of what's going on in the world at large right now, and also explains why people were so slow to respond. Perfect is the enemy of good, after all.
In medicine, bad is the enemy of good. Improperly studied medicine rushed into a clinical setting can make things worse and are perfectly capable of making things worse on a grand scale, regardless of how terrible people's understanding of statistics and variance are.
Doctors have the authority in many countries to prescribe it off label based on their own review of the evidence and their patient's consent. That's the best we can do right now without risking something much worse.
As noted in the original post you replied to we don't really need a control group because we already have a lot of experience with people who didn't get any drugs. Quoting that post: "all that is needed is a proper categorization".
The goal here is to get enough evidence the treatment works or doesn't work. Conducting double blind studies is not the most efficient and fastest way to do that when you already have a lot of preexisting knowledge about both the patients not receiving any drugs and the safety profile of the drug in question.
This is again a situation where statisticians should be asked for help as medical professionals very often operate on very simplistic model (double blind or gtfo) which we just don't have time for this time around.
It does, but why can't one can have the control groups at the same as blanket prescribtion? In times like this when the health care apparatus is overwhelmed, is it a good alternative to do that, since side effects are well-known and given the fact that the drug might be ineffective if used late in the infection cycle [0]?
i absolutely don't see the need for a control group. We just need to monitor which patient had the treament and which didn't, globally. But if a doctor wants to prescribe all its patient with it, he should be allowed to, as long as it doesn't go against the traditional counter-indication for this very well known and very widely used on a global scale drug.
> We just need to monitor which patient had the treament and which didn't, globally
And you just described the reason for a control group. We simply do not have that information. Things are moving too fast and treatments have not been standardized yet, not to mention the built in variability between locations. If you want to know if a treatment is effective, you must have to comparable groups. The point of the control group is to gather that information. Just because there are many, many, global cases doesn’t mean you actually have been able to capture the data for a control. Things are moving that fast.
> But if a doctor wants to prescribe all its patient with it, he should be allowed to
And what about the supply of these drugs? There are people who do need these drugs for other conditions (I’ve read Lupus patients sometimes need it). These patients need the drug and the supply is getting scooped up by people who don’t know (a) if they have the virus and (b) if the treatment is effective. And if it is effective, is it effective for treatment of infection or prophylactically? Just because a drug is widely used does not mean that there is enough supply to distribute it to anyone who wants to try it.
These are major issues that can be answered with proper clinical trials. We know how to do this and they can (and are) being done as fast as possible.
I saw a retweet the other day by a Lupus patient who couldn't get their prescription refilled. That's going to make a crappy situation much worse for a whole class of people.
That is a fallacy right there, discredit the source of the argument instead of the argument itself. People should have the right to discuss things openly, even without credentials. They should not, however, actively act upon it causing actual harm. This is the realm of the specialists. As long as we have these distinctions, let people discuss.
Trouble is opinions can take on a life of their own. Not everyone is sensible enough to read opinions as just that. A lot of these articles are written with a tone of authority; espessially the ones from data scienctists. Discussion is fine, but bear in mind that its adding to a lot of unecessary noise.
You imply there is a necessary noise in the conversation. Who are we to define what credentials should be used to define what is noisy and what is not? I think the best way is to attack the argument itself, not where it is coming from.
If the discourse become authoritative and biased, we shut it, disregard it pointing out the flaws. But we have to be careful not to be authoritative ourselves when we don't even give the chance for the writer to be wrong because he lacks the right credentials.
Easy, in a pandemic like this only listen to medical professionals. Like I said, the time for amateur to examine and draw conclusions is after the pandemic is over.
I guess I should have articulated an argument instead of venting and using sarcasm.
Basically, I've seen my share of unqualified people (not that I'm qualified myself) building ridiculous models of exponential growth and predicting catastrophes, commenting on drug research protocols without necessarily understanding the details and so on.
What's the point of pollution the conversation with useless ill informed fake-authoritative opinions that no one should act upon? This isn't "I tried Python and had a lot of problems with dynamic types", it's uninformed specualtion on an arbitrary sample of information the they haven't trained to understand.
then use that against the argument. Break it for what it is not worth, but don't attack the person itself and the right to free speech. Where would you draw the line for defining what is polluting the conversation and what is not? Can you see the danger there? This trend of checking someones credentials before hearing the argument can become elitist real fast. Careful to not become the censorship we condemn
The problem is even for hydroxychloroquine the most at risk patients for serious COVID-19 illness (People with Diabetes and Heart issues) are under the well established precautions in using the hydroxychloroquine. To build publics faith in the idea that one medication is going to really come through is haphazard at best.
> You do not prescribe medicine on the basis of hunches.
Nor on the basis of panic; we should be able to manage this pandemic as long as people stick to the measures in place. Stay at home, limit your shopping, don't take the piss. I mean panic right now is causing more problems - e.g. hoarding and people trying to make a quick buck.
This thing is mainly a big problem because people can't keep calm and rational. Or they think they're rational, as in, "I should buy enough toilet paper because I should stay indoors for a while".
How long is a while? What happens if our supply chain is impacted? Grocery stores are open now, but isn't it possible for there to be outbreaks in the stores? What will become of the economy if this continues?
These are very real concerns for which no one has answers. People are stockpiling because of the uncertainty, which could be seen as a very rational stance in hindsight if this goes on for quite some time.
Hydroxycholoquine is very safe and is prescribed to lupus patients daily. The Korean recommended dose (400mg/day) is the same as what is prescribed to lupus patients. Chloroquine is a bit more dangerous, but only in high doses and over long periods of time. Patients would be on either for 10-20 days maximum which is much less than what it takes to be dangerous.
I'm not sure if any of what you wrote is accurate.
The drug is 90 years old, has been used by millions, has relatively modest side effects, is relatively well-understood, has shown efficacy against coronavirus in vitro.
Yes--in plain English, the trade-offs are different in different places. It's not a recreational drug so there is (usually) little risk it will get taken or abused unless it is really needed. That does not mean the drug has no risks and can be taken as a free-for-all.
And it's not just for drugs. It's also why DDT continues used in some countries but is illegal in others.
Exactly. Since the risks and side-effects are well-established, it’s more ethical at this point to experiment on humans with an “off label” usage. Doctors do that all the time with other meds.
Yes, doctors should experiment with the drug and I am certain, a lot of those, who have the necessary conditions to perform a proper experiment, will do.
What should not happen now is, that the general public starts taking the drug as a precaution, just because it might work. The known side-effects are too severe (e.g. blindness) and also, there are already shortages of the drug for those who already require it as part of their medical treatment (e.g. immune-supressed).
"Risk for toxicity is less with <5.0 mg/kg real weight/day for hydroxychloroquine and <2.3 mg/kg real weight/day for chloroquine[2]. Patients are at low risk during the first 5 years of treatment."
And management is as simple as stopping the drug if it starts affecting your vision:
"At the first signs of retinal toxicity, hydroxychloroquine should be stopped to prevent further retinal damage and visual loss[2]. "
I'd note that the phrase "further retinal damage and visual loss" implies that retinal damage and visual loss has already occurred... how much of that you are willing to tolerate is up to you, I suppose: I'd prefer 0 unless necessary. If the retinal damage is saving me from a death at the hands of COVID-19 maybe it's worth it, but if it's not...
Did you just read the list of side-effects without checking how often they happen? With that mindset any over the counter medicine will read like description of Zyklon B.
Being sick with a disease like Covid-19 could absolutely alter how a medication works on a person. The drug might be safe to a person with a certain profile and it has been tested against those, but each underlying issue is a new profile which it should be tested against as separately as possible.
Based on that reasoning, we shouldn't give any pharmaceuticals to coronavirus patients. Even simple fever relievers, like acetaminophen, or oral rehydration fluids should be prohibited because we just don't have enough data to know how they'll interact with COVID-19 specifically.
I think this is the fairest counter-argument I've heard so far about safety of HCQ in Covid-19. I still think the background of "you're in a house burning down, some tests tentatively indicate this may be effective..." weigh more against the risk of side-effects or worse progression due to CQ/HCQ. We ought to be able to devise some experiments to add certainty, using mainly retrospective data now available.
Most SLE/Lupus patients take HCQ, and reportedly none in the cohort of the doctor I know have sufficiently severe symptoms to qualify for pCR testing to even establish if they have Covid19. This is Spain with (currently) 708 cases per MM. Of course, caveats like absence of evidence is not (necessarily sufficient) evidence of absence. It may be that SLE or other immuno-modulation treatments have some effect - someone should do analysis with matched non-SLE HCQ and non-HCQ prophylaxis cohorts. But back to my main point with the house burning metaphor for exponential infection - the risk landscape demands we take more risk, give HCQ widely instead of waiting for the conclusion of prospective studies. We can withdraw it if it is shown to be no better than placebo or worsens outcomes when enough data is available to conclude.
2) It complicates the treatment of the patients who will experience the worst cases of COVID-19 causing panic... people above 65 with diabetes and heart conditions.
3) Again this drug has well known precautions that reads like the all the people who will experience the worst of COVID-19
Having Grandpa John walk in for COVID-19 and roll out for death by QT elongation is not something the public will have to litigate.
> Again this drug has well known precautions that reads like the all the people who will experience the worst of COVID-19
Except those precautions are at least two orders of magnitude less than the case fatality rate for coronavirus infection.
Equivalently, there are scenarios where a seatbelt can decapitate you or trap you in a drowning car. But you'd still be a fool to drive without one, because buckling is a thousand times more likely to save you than kill you.
The most serious cases of COVID-19 affect all the people who have the underlying conditions that have precautions listed using H-Chloroquine.
That makes the normal ICU case of COVID-19 more difficult (time consuming) to manage because now you have to spend time managing / monitoring: The condition, COVID-19, the side effects of H-Chloroquine on vital body systems already impaired, already being attacked by COVID.
Throw the magnitude two out the window like a lucky car crash victim while these patients suffer the crash in the hospital. Not even Bernie was going to pay for the specialty concierge care needed to manage such a patient and yet you some how assume that is how it works.
It isn't vitamin C. It has a lot of side effects, including serious psychiatric ones in limited cases. If you give it to 100 people you're going to get new health issues that you didn't have before in that population.
You cannot pour a drug into the populace that will cause side effects on a mass scale that might also overwhelm the healthcare system. It is nice to imagine we have the medicine already, that we can simply wave our hands distribute the pills and everything can go on as it was before this. But we do not and can not go back.
> There's no need for establishing a benchmark. Just proper categorization of existing patient should be enough to compare.
This is very difficult to do. In this study all the patients in the treatment group are from a hospital and most of the patients in the other hospital are from other hospital. Do the population of both hospitals have the same median income? Usually a low income cause a bad diet that may cause bad health that has a worse prognosis.
Are the patients in the other countries classified by income? Does the difference in income cause more/less consumption of milk? Bottled milk has an additional amount of vitamin D, and some unconfirmed reports say that the lack of vitamin D is bad for covid19. Does this difference depend on the country? For historical/genetic reason, French drink more milk that Chinese. What about wine, is the consumption of wine important?
What is the effect of temperature and humidity in the no-hydroxychloroquine group? The average climate conditions change from country to country and from place to place and from month to month.
To be included in the testing group the patient must give consent. Probably they have not to be too wasted to be able to read it. Probably not been to wasted is a good sign.
There are a lot of factors, and not all of them are known, not all of them are easy to measure, not all of them are measured. So you need a control group.
Many doctors are already actively prescrbign this. I heard that it is basically de facto standard of care in France. Doctors recognize that when eveidence based medicine doesn't provide an adequate treatment, they need to improvise. Off-label prescribing is very common
The point of these studies is to help the evidence catch up to improvised clinical practice. If the studies are poorly designed, that goal is not achieved
It is being used in several countries already (US and Europe, not sure about Asia). If you are interested in details, here is a video of a conference call at UCSF:
This is not some crazy experimental new drug that was unknown until past week. It's an experimental treatment, it's not an experimental drug, with unknown side effects. For a condition where time is critical.
Of course, we should also be quick to discard unpromising drugs, and not only focus on one promising drug.
In particular, if it turns out to reduce infection rate, taking it prophylactically will save the lives of everyone who would otherwise die for lack of a ventilator, or who would die even with one. If it does not turn out to reduce infection rate, no harm done.
If it turns out to shorten the illness, everyone sick, and also everyone taking it prophylactically who is actually infected but asymptomatic, will reduce their contagion rate by joining the ranks of the non-contagious sooner, reducing everyone's risk. If it does not turn out to shorten the illness, no harm done.
Chloroquine is well-known. If it caused an unacceptable rate of liver failure we would already know.
I see suggestions that it takes a very high doses to work. But that might be so only for the already severe infections they are using it on. It should be easy enough to discover whether low doses work prophylactically, and safe enough to try them while waiting on results.
Yeah, it's important to have it available if it's needed and it's important to conduct at least some sort of trial. Because the drug, combined with azithromycin is so fast acting, we should see the first results in a few days. Worst thing that can happen is it doesn't work. Best thing that can happen is we can save 1.8 trillion dollars and restart the economy. Under these circumstances one would be an idiot not to give it a try.
Coughing your lungs out but feel like CNN is not lying to you? Easy! Refuse the drug. Nobody can force you to take it.
Worst thing that can happen with a chloroquine and azithromycin combination is that it causes heart arrhythmia and sudden death in a patient that would not have otherwise died.
Studies of patients taking this combination of meds include daily EKGs, which is something the general population at home isn't going to have access to.
That's why they are starting with a trial, and not a full scale roll out. Other countries are using this combination of drug, and it's not because they're huge fans of Trump, I can assure you. We also have literally millions of devices out there capable of detecting arrhythmia - they're called Apple Watch.
That's like saying construction N95 masks can't be used in the medical setting just because they lack appropriate certification. That old shibboleth has been defenestrated weeks ago.
I bet it can, if FDA is beat over the head with a shovel, which it will be if the need arises. It's a real ECG verifiably capable of detecting irregular hearth rhythm, and it costs a tiny fraction of what a big-ass ECG machine would, and therefore can be deployed at a much larger scale. Just because it could only get approval to detect atrial fibrillation duing "peacetime" doesn't mean it can't be pressed into service for other things.
1/ chloroquine is a very well known cheap drug. There is no new risk associated to using it that doctors arent already aware of.
2/ we do have a lot of statistics now on the outcome of patient not treated with that drug. There's no need for establishing a benchmark. Just proper categorization of existing patient should be enough to compare.
3/ people are oversaturating ICU right now. And as such, it is an emergency situation. We should take the reverse reasoning we usually take : if there's no suscipicion this drug could cause new problems, we should have people massively use it whenever possible and look at the result after 6 days (since that's the time the original study says it takes for the first results to show).