The interesting part is in the comments section. A lot of ideas and personalities showing!
(Snip)
Here is a comment by one of the folks in the thread California posted. It sure makes sense to me. "Jim Andrews says:
30 March, 2020 at 11:54 am
As a surgeon and physician, in view of no other drugs have shown any effect, while Hydroxychloroquine and AZT has shown in vitro and in vivo results, I don't know any doctor wouldn't try it on severely symptomatic patients! Criticism of no large controlled trial has some merit but not when tens of thousands are dying with conventional supportive treatment, the patients has to be given an option of using the HCQ and AZT as last ditch resort. Having been a principal investigator for over 250 clinical trials I am not sure if a randomized control trial is ethical here! I believe patients who were not treated with such combinations in other hospitals should be used as control! To hear doctors that are critical of using HCQ AZT off label, they are either academic idiots who are not actively treating patients, or they suffer from ｡t's not a good idea because it's� not my idea syndrome".
The problem with just treating in the absence of a controlled trial is that we don’t learn whether a treatment is effective or not, and we don’t learn what the relevant harms of treatment are.
Yes, if you have a disease that is typically fatal then you can argue that a controlled trial is not necessary as it will be easy to tell if the treatment is effective. Even in that situation there are study designs that allow a more rational approach though. Give a patient the study drug, if he survives give the next patient the same drug. If the first patient fails then use a different (control) therapy on patient number 2. You switch back and forth depending on how the most recent patient did. This actually allows statistical analysis when you see that one treatment is better or worse than the other simply by noting that one treatment had fewer failures than the other. I’m blanking on the name of this study design but it overcomes concerns about the ethics of randomization and is pretty simple to do so long as investigators are honest.
If you ignore the power of controlled trials you end up with a situation where only anecdotal data is available and we know from long experience that this situation simply evolves into shouting matches between advocates of different theories. I worked in Russia for about five years studying their hospitals where the ranking professor, not controlled trials, determined what therapy was used in a given hospital. Mayhem. Hospitals literally across the street from each other used bizarre different therapies because the ranking professors subscribed to different theories and there was no way to adjudicate between them as no controlled trials were ever done. Sepsis was treated in one hospital by implanting fiberoptic catheters in the bloodstream and then delivering UV light through the catheter “because UV light kills bacteria”. Across the street they’d implant pig spleens to supplement the patient’s own spleen “because the spleen is important in immunity”. Truly nutty theories can be put into practice without any data on effectiveness or safety.
In vitro evidence of effectiveness is of limited help as in vitro conditions do not accurately reflect in vivo reality. I cannot even count the number of antibacterial or antiviral compounds that worked in the test tube but fail in the patient.
And, piling on untested therapies because it is “unethical” to withhold potentially helpful therapy (as identified by something like the Marseille trial) can make it much harder to do legitimate studies. If everyone is on chloroquine and azithromycin, how do you know that those drugs don’t affect the metabolism or safety or efficacy of a third experimental drug?
It’s helpful to look at how cancer treatment has evolved in the past fifty years. We learned, in pediatrics at least, that running rigidly controlled trials allowed steady progress and improved survival, step by step. We went from less than 10% survival in pediatric acute lymphocytic leukemia to over 90% survival. It wasn’t because cowboys physicians just treated any way they thought reasonable in a life threatening situation. It was because solid data allowed truly effective approaches to be identified and then improved upon with additional strictly controlled trials.
It may be hard to accept that carefully designed and executed studies take more time than a shotgun approach but in the long run it is the way to go. Whining about “unethical randomized trials” is BS. What is unethical is failing to do appropriate reliable trials so that the next 100 million patients actually benefit from the experience gained treating the first million.
