I think the numbers in the Pfizer trial were 170 out of 43,000 (168 from placebo and 8 from vaccine arm with positive PCR and ‘at least one symptom). Hence the famous 95%. This hurts my brain.
@@ritcha02 I see your point ... without the vaccine 168 people would have caught th virus ... with th vaccine only 8 people would have caught the virus. Now here's the hard part for me at what point do you shut down the economy mask all the citizens and keep children at home ... using the same numbers insisting on a vaccine disrupted 43000 lives ... potentially saved 160 lives ... still cost 8 lives though there's nothing to go home to .. but that's ok we'll just keeping printing money and pretending it's worth something just remember ... you get what you pay for. Eventually the quality of the underlying asset matches the value of what it is being exchanged for
This part was missing in my earlier comment : The FDA requires "Provide absolute risks, not just relative risks. Patients are unduly influenced when risk information is presented using a relative risk approach ; this can result in suboptimal decisions. Thus, an absolute risk format should be used"
Personally I agree with you... also relative risk reduction(which is vaccine efficacy) is just that- a relative risk reduction of contracting Covid. It is not the same as Protection against contracting Covid, which unfortunately many get confused by the figures given and their significance.
Absolute risk reduction is essential to doctors and patients alike to decide on risk-reward. Relative risk is normally used to compare between trials. But this being misrepresented in the media, and the wrong figures are shared to make an informed decision.
What is the significance of absolute risk? For example, if the of a particular negative outcome is 90% that would mean 90% of a population is going to get that negative outcome. An intervention with 50% efficacy would benefit 45% of the whole population. If the absolute risk for same negative outcome is only .09% then an intervention with 100% efficacy would only be benefiting.9% of the population. So absolute risk matters a lot. I don't think your talk it besides that very well.
The problem is the AR was not revealed to the public for any Covid vaccine even though the FDA requires full disclosure of both ARR and RRR so as to not mislead people.
@@tibitaylor according to the data they're getting from the global campaign, they're NEVER gonna get an approval unless they manipulate heavily the numbers.
Nice. Except if the vaccine had no side effects 95% relative risk reduction would be great. When the side effects of an intervention is significant the number needed to vaccinate minus 1 says how many people get the side effects with no benefit. 10000÷84 . More than 110 people vaccinated will experience the side effects without any benefit.
I understood that relative risk reduction was devised to help researchers in distinguishing between studies.Absolute risk reduction is the risk factor that should be told to the patients...
Thanks for clearing that up, doc! Wouldn't it be more transparent if they also included the risk of having a negative outcome of the treatment, medication or vaccine, like hosptalisation due to complications or even death?
think absolute reduction is important. Example: many big pharma companies claim their drug reduces in a 50% the chances to get disease X. But then the actual people getting the disease is 2 in 100000. So reducing it to 50% means now it's ONE person getting the disease, not TWO. That is a 50% reduction. See what they do there? Absolutes are important. Maybe even more. A 50% of a very small number is a small number.
Dr. Mo been - you did a phenomenal job with this explanation. I am still trying to understand why they only report one number - That being relative risk reduction - instead of following the FDA guideline which states they should report both numbers together. Thank you very much!!
Thank you very much for your excellent review. Engineers would consider such infinitesimal reductions in disease to be nothing more than statistical noise. We tend to work with larger deviations. 🙂 A slight bias in subject selection, for either group, would render such numbers meaningless. This is absolutely astounding.
My understanding is that typically the control group receives standard of care. By having corrupt medical leadership not treating outpatients with available interventions--vitamin D, HCQ, IVM, etc--the control group does not do as well as people should when being treated with the best combination of interventions available. So the vaccines appear to be more efficacious than they really are in comparison to alternative interventions.
In my view, I would rather stick to absolute risk reduction as the data demonstrates from the get go (without intervention) the risk is very low 0.88%. Vaccine intervention is not required. Relative risk reduction is a red herring that distracts one from that fact.
Absolutely, I have also posted a similar comment, it is unbelievable that Drbeen despite explaining correctly the difference between RRS and ARR still makes the wrong conclusion
I'm 75. My risk is higher than 0.88%. In other words, other factors need to be taken into account when talking about individual risk. Beyond that, I also care about the plight of society as a whole. Whenever others don't care about getting sick (because their risk is relatively low), then the number of people who can infect the more vulnerable like me, either directly or through a younger member of the same family, shoots up. That's a concern in a number of ways. For example, in whole societies the risk percentages get applied to very large numbers. Hospital capacities can be overwhelmed, causing even more preventable deaths. Vaccines can reduce the extent to which hospitals are overwhelmed, and relative risk reduction would shed some light on that problem. Finally, although the absolute risk reduction of 0.88% may seem low, every single infected person raises the possibility that a random mutation will lead to a truly nasty variant of the virus that is much more deadly and/or transmissible than the ones we already know about. That risk is too often neglected, in my opinion. I believe the situation is more complex than you imply.
@@RussCR5187 In that case you take the vaccine however let's not force the other 99% of us in that direction. Besides, have you looked into Ivermectin? You should.
@@motumer But what about the mutation concern I brought up in connection with the 99% who don't get the vaccine and don't care whether they get infected or not? PS -- Yes, for a number of months now I have had a supply of ivermectin large enough to cover my entire family, and I successfully used it myself as an early treatment after coming down with a fever and nasty headache. Don't actually know if the symptoms were from covid or influenza or common cold. But truth be told I don't really care what it was in this time of pandemic. Better safe than sorry.
I think you forgot an important part of the equation. The benefit of Absolute Risk Reduction is that it takes into account the size of the study’s participants. If you have a relatively small study group showing good efficacy vs a large group, this is very different. The good efficacy in the small group may be more greatly effected by the make up of the individuals. This effect is more balanced out in a large study. So ARR is important in fully evaluating the benefits of an intervention.
Plus. It can be easily manipulated. If the participants were excluded in the two groups in a proportion that ARR keeps the same but the CER becomes always lower, the RRR can be as high as they want, because every number divided by a denominator less than 1 becomes always higher, by has no relevance in clinical means, it only matters for scientists to evaluate experimental protocols. so my opinion is that the RRR shouldn’t even be cited because is affected by the experimental modeling , and the ARR is not, that why it’s called absolute.
So if the ARR has a large number, the results are more reliable. In the case of Pfizer then, is 18000 participants a good size? Which in turn would provide a reliable / reasonable RRR? So how is the Pfizer efficacy (RRR - 95%) misleading?
So the selection of the experimental and control groups becomes critically important when numbers are so small. They must be carefully age matched, matched for comorbidities, matched for asymptomatic previous Covid disease/ potential immunity. What am I forgetting ?
Very good Dr. Been. I wonder why other ppl are making the absolute risk reduction look bad. Thank you for making this video, it makes sense. I'm still not getting the jab lol for other obvious reasons.
@@DrBeenMedicalLectures Can you do a video to cover the role of the endocrine system in immune regulation. So patients can evaluate for themselves if endocrine disruptors "may" be playing a role? www.hormone.org/your-health-and-hormones/endocrine-disrupting-chemicals-edcs
You could have 1 million people in a trial, if 2 get sick in the control group and only 1 in the intervention group, you have a 50% RRR. Contrary to what is said in this video, relative risk reduction does not show the quality of the data but it is a statistical shenanigan to amplify noise and make it appear as real!
Thank you so much Dr Been for explaining what I know that most people aren't even aware of when it comes to statistical reporting let alone understand how it can be completely misunderstood and mis interpreted even by intellectuals alike. This however highlights my questioning of why we are trying to vaccinate everyone when only a fraction are at risk of getting seriously sick from the virus. To me it would be like giving everyone blood pressure medication when only those with chronic high blood pressure would need it. Why not just vaccinate the sick and elderly?
If we tell people that one out of 100 vaccinated people can be prevented to have the disease , I really doubt a lot of people will jump in to get vaccinated, not to mention the side effects. And the vaccine companies didn’t provide people the ARR for a good reason.
@@andreasmelkersson3592 It's actually less. 142 needed to be vaccinated by Pfizers vaccine to prevent 1 case. 88 for Modernas. However if it's being efficient for preventing severe disease as it logically should then it's still beneficial. However things like this from the vaccinecompanies aren't building much trust.
Something to consider: the number used to represent the risk of contracting the virus in the control group may be seriously flawed if they are using the PCR test. This value is easily manipulated by altering the cycle threshold. The question is: Are the researchers being transparent about the CT used for each group? Could there be any financial motivation to use different parameters in each arm of the study? The pharmaceutical industry does not have a good track record of integrity in self-reported data when it comes to justifying their bottom line.
Loved this discussion.. hoping to see more of these epi/biostat related videos... Kindly make more videos on Epi/Biostat related videos esp/for USMLE ...Thank You
There's another factor that's not taken into account. If the treatment itself causes the same or similar harms as the disease in a significant size, that needs to be taken into account. It also confounds the analysis with regards to causation.
And what if you found out the major difference in the cases between the placebo group and the test group is a couple groups in the placebo group went to a party but no one in the placebo group did anything? When the cases are that small it is definitely possible.
So In the Covid trials the did they measure who had positive PCR together with at least one symptom from each arm and that’s the relative risk? So the number takes into account that some would be expected to have got Covid anyway? Did they look at severity of symptoms?
Wouldn't it be if you gave it to 100 people that 20 %of those people would have gotten the events or but now 10 %. Also that is only tracking 2 weeks after second shot right? What are the relative risk of the side effects?
What people talking about covid in the comments are misunderstanding is that the control group is affected by non-pharmaceutical interventions (NPIs). Therefore it’s not a “true” control group. At the moment we cannot have a true control group, because govs have regulation in place to stop the spread. Hence, when talking about ARR for covid vaccines it’s in relation to the measures. So this means that the vaccines are a little more effective than current measures, allowing countries to reduce their NPIs for more societal and economic benefits again. Which is why they’re being rolled out.
I feel that we would have a perfectly good way of looking at the quality of the improvement achieved by the experimental treatment by focusing on the percentage point difference between its incidence rate and that of the placebo, once we merely agree upon a benchmark measure of what the maximum difference might be. Then simply divide the actual difference by that agreed-upon maximum. While the maximum would be established as an arbitrary agreement among experts, it would often produce more understandable data then the current relative risk measure. Here's why. If the incidence rate for the placebo is 10/18000 and that for the experimental treatment is 0/18000, we would announce with great fanfare that the treatment had 100% relative-risk effectiveness; which, IMO, would be sort of meaningless (except for derived computations about implied reductions in demand for hospital beds in a multi-million-size base population).
I have made this argument in my own Bulletin as such: Your odds of GETTING Covid have only been reduced by 1%. However among that 1%, your odds of a negative outcome have been reduced by the 95%.
Thanks for the explanation, and i understand the importance of publishing the relative risk number. But surely publishing the absolute risk reduction alongside would give people the knowledge to decide if they want the medication or not. Eg. Why would i take medication/ vaccine for a virus if the absolute risk reduction is .8%... most medications have side effects, they might be of benefit for one thing but cause other problems... that way people are empowered to calculate the risk for themselves.
As far as I understand, absolute risk reduction is mainly useful for calculating the Number Needed to Vaccinate. Even if an intervention is 100% effective and this intervention prevents a disease which is 0.1% fatal, it means that we have to vaccinate 1000 people to save one.
Wouldn’t a simple explanation be that result of a relative risk reduction calculation applies to the treatment of a general sample of people, and the absolute risk reduction is the difference in outcomes when comparing a treated group versus a placebo group?
A bit late to comment here... But I was just thinking.... by ignoring ARR (or not revealing it to the public)... it can effect the way people make decisions of whether to accept or reject the intervention. I'm no statistician... but just playing with hypothetical numbers. Say for discussion's sake... the numbers happen to be much higher... e.g. Instead of 8/18198, it's actually 800/18198 for the intervention. Then, for the placebo group it's16200/18325 (instead of 162/18325). This would ramp up the ARR:- ARR = 88.4 - 4.4 = 84% (A better number...) - instead of 0.84%. Reporting this alone would be enough (perhaps) for people's psychology. RR on the other hand, would still remain at 4.4/88.8 = 0.05, which leads to the same efficacy percentage (RRR) of 95%. A good number as well. So now that both numbers are high... it might be in the interest of the organization/corporation to report both numbers... In the case when the experiments were done, ARR was very low... So it might have been best not to report it , though I think the FDA asks to report both. I donno, am I making sense? :) Would appreciate any feedback... Trying to understand all this from various points of views :)
Drbeen The problem is definition and consistency of what they (vaccine manufacturers) are measuring: when they say the INT group or the CTR group got C19, how did they establish that anyone got C19? was it PCR test? was there variability in measurement? Also, the general public ought to be better informed in as many ways as possible so that we understand perfectly and it becomes common knowledge how to understand RRR in the context of ARR. I feel you have made inroads to this later request but not enough IMHO. I hope in future you can explain it in more different ways. 20,000 people in INT group 20,000 people in CTR group, how many got saved from C19?
Does there come a point where the number is so small it is insignificant or considered a statistical rounding error? and if so what is it? eg 8 Vs 162 out of 18,000 - they are both sub 1% and therefore it would be a reasonable expectation for the numbers to be different if run again on the same size group. For example 8 could be 80 and yet still sub 1% but giving a very different relative risk and the actual risk pretty much unchanged?
Thanks for refreshing my understanding of statistics but statistics are only helpful to me if I can see the conditions during trials and the data of persons chosen for the trials etc. ( Not especially for the vacc trials but in general) Because one side is the mathematical numbers and the other side and as important are environment and the people themselves when you measure something that is relevant for „real life“, with living beings. So I tend to look rather at statistics that are made very transparent concerning mentioned factors. In an ongoing pandemic with restrictions that change all the time that can make a huge difference to the outcome and the kind of criteria how to choose, who to choose for the trial groups. ( example: age might play a big role bc elderly people might take restrictions more seriously, behaving to the rules, due to their higher risk so I would like to see if the age groups are well balanced i.e. same nr of same age in both groups and the urgence for studies/ results doesn‘t make it easier to choose thoroughly) Also the timeline can be important. When did trials start? How are participants chosen? Are they chosen before trial start? Or could people drop in for participation over a longer period of time? Was there an interest that could push people to participate? These are just examples, I‘m not telling the vacc trials aren‘t reliable, it‘s just my general reluctance to look at stat‘s without knowing other factors during a study/trial that could have influenced the results.
This type of considerations should be done also including all adverse event: symptomatic disease, severe disease, ICU, death. The RRR and ARR and NNTT should be calculated also accordingly, wouldn't you agree? Separated by age and comorbidity classes this would be of real help in making a decision on vaccines, personally and at administration level. Do you think we can get these calculations by the data we have from most studies, or that we should get them?
If it was positive PCR plus at least one symptom, then wouldn’t the Study has missed out on positive asymptomatic cases, bearing in mind majority of Covid cases are asymptomatic?
This is from an INTERIM phase 3 trial report (completion 2023). Environment was not same as roll out during very infectious variants. Neither was it for serious disease or death but whether there was a +ve RT PCR test at what Ct amplification? Given the low prevalence of SARS-CoV2 in the trial it seems clear the RT PCR results included false positives. What was placebo? For AZ it was not normal one but another non C19 vaccine that lowered the immune response.
However, it is very important to know your absolute risk to begin with before calculating your relative risk reduction from a vaccine by seeing which risk groups you are in. If you do not have hypertension, diabetes, are old with a compromised immune system, or are obese, then you are not in with the 98%% group of those that have serious complications or death resulting from infection. If you are then in that healthy 2% group your absolute risk is 2%. Now we can add that, according to several international studies, there is a natural immunity for 30 to 50% of the general population probably resulting from previous exposure to coronavirus infections from the common cold. In any case this means that if you are in that 2% group we can further reduce your absolute risk by let’s say 40% if we split the naturally immunity range that was from 30 to 50%. This means that your absolute risk of having a serious outcome from a Covid-19 infection is approaching 1%. Now you can decide whether it is worth getting vaccinated to further reduce that 1% risk.
I would just like to add that if anyone has better numbers for my analysis above by all means plug them in as a way of correction. What I am really interested in and trying to get at is that you have to know where you are at the start to know where you may be at the end.
Very clear explanation, however if upwards of 15:/: of people with Covid end up hospitalized with examples in Canada of younger ages no comorbidities ... also need to consider potential hospitalization
It’s statistics but doesn’t paint a full picture and take into consideration long hauler group and some people that suffer organ damage that’s the problem with this disease it’s creating a lot of damage to what could well be a good chunk of those that survive.
excellent points. But it never was about the risk for healthy people with a good immune system. We already knew the risk was very small. It was about protecting the old and vulnerable. With that done the new variants might change the equation. And the issue of long Covid symptoms remains.India has presented some interesting variables.
@@MrBallynally2 I disagree healthy people and kids are getting impacted. There is still alot we don’t know and are learning therefore in my view assuming is just that assuming and somewhat magical thinking that if your young and healthy your immune. To much of this started early on assuming it was the old and vulnerable with an unknown it’s wise to make no assumptions and remain curious and open. Also the VOC might have less impact then we think currently a new study shows that on reinfection the virus uses 2 items that are a byproduct of damage that our bodies make one being billiroubin that this virus literally uses as cover/coating to evade antibodies. See below article on study that came out this week The SARS-CoV-2 coronavirus plays hide-and-seek with antibodies It’s a tactic that is sometimes used in horror movies. Pretending to be a zombie so that the real zombies won’t see you. Or pretending to be totally unresponsive, so that the aliens think you are one of them (Invasion of the Body Snatchers). It turns out that the coronavirus uses a very similar and clever strategy. Researchers at the Francis Crick Institute, Imperial College London, Kings College London and University College London have found that the SARS-CoV-2 coronavirus uses pigments in the body to hide from the immune system. This process involve the waste substances biliverdin and bilirubin. These cause the green and yellow patches on the skin when bruised. The substances - released mainly during the breakdown of old or broken red blood cells - are able to bind to the spike protein of the coronavirus. And when that happens, it becomes a lot harder for antibodies to attack the virus. The researchers say this also helps explain why some people get very sick and others don’t at all. The full study was published in the journal Science Advances. Green and yellow Biliverdin and bilirubin are compounds that occur naturally in the body. They are released during the breakdown of so-called heme compounds that attach to oxygen molecules in red blood cells. The compounds can be seen when bruises change color. This is because heme compounds are first broken down to biliverdin (green) and then to bilirubin (yellow). The toxic substance bilirubin normally leaves the body via the liver and gallbladder. What the researchers discovered with the help of blood serum and antibodies from people previously infected with COVID-19, is that biliverdin prevents antibodies from binding to the spike protein in 30% to 50% of cases. This significantly weakens the immune system. Acts as a booster For the study, the team used, among other things, X-ray images to look in detail at the interaction between spike proteins, antibodies and biliverdin. What they discovered was that biliverdin attaches itself to the so-called N-terminal domain (NTD) of the spike protein. This stabilizes it, preventing the spike from opening up and exposing parts of its structure. As a result, some antibodies are unable to access their target areas and therefore can’t bind to the virus and neutralize it. According to Annachiara Rosa of the Crick Institute, the interaction is especially strong as soon as SARS-CoV-2 reaches the lungs: “If a patient’s lungs are infected, the virus damages blood vessels and causes an increase in the number of immune cells. Both of these effects can contribute to an increase in biliverdin and bilirubin levels in the surrounding tissues. And with more of these molecules, the virus has more opportunities to hide from antibodies. That is a really remarkable process, because the virus may be benefiting from a side effect of the damage that it has already caused.” Also see our COVID-19 archive.
people also like 1 in 10 type numbers saw RR 1.3 obesity vs colon cancer how much greater is my risk if fat? need absolute numbers to say well 1 in 300 or what ever
So this RRR is significantly lower in the younger age groups I suppose? I mean the risk reduction would not be even close between a 15 year old and a 65 year old... right?
Oh I disagree. To me the 95% relative risk reduction is far more meaningless than the 0.84% absolute risk reduction. It is cheating by not advertising both numbers.
@@josephduncan4356 Both are useful metrics. the ARR will always be low since it depends on the event rate. You can still have a super, high-effective vaccine. the RRR gives a quicker, more meaningful idea of the vaccine's efficacy since it ignores the transmission setting and gives a good general overview
@@andreasmelkersson3592 Yeah but the point is that the effectiveness rate means very little when there are a small number of cases. If you were willing to accept a tiny absolute effectiveness rate then you could probably show that drinking orange juice is 99% effective and preventing getting hit by a train.
Do those figures include mortality "from" or "with" the device within 28 days of injection during the trial. The relative risk/actual risk to life or of significant illness must include all serious issues not just RT-PCR positive test result from a fragment of SARS-CoV2 RNA strand that is not known to be active or infectious - can get positives 6 weeks after infection or breath in inert strand in say a hospital waiting room.
You’re saying that relative risk is given more relevance because it means quality of the absolute risk? You mean that RR value is directly extrapolated into the whole population -despite its an experimental sample- in order to establish public health policy and not the absolute risk?
I think absolute reduction is important. Example: many big pharma companies claim their drug reduces in a 50% the chances to get disease X. But then the actual people getting the disease is 2 in 100000. So reducing it to 50% means now it's ONE person getting the disease, not TWO. That is a 50% reduction. See what they do there? Absolutes are important. Maybe even more. A 50% of a very small number is a small number.
I am afraid your presentation misses the main point in particular after minute 22'30. The FDA requires that :The reason is obvious. If I tell you that I can reduce your risk by 90%you cannot make a proper decision if you do not know what your absolute risk is without the treatment. Why do you think the Pharma does not comply with what the FDA guideline ? Looks a much better selling argument to say only efficacy (i.e., Relative Risk Reduction) is 90 % than adding that the absolute risk reduction is 0,7 % for Moderna or 1,1 % for Pfizer by way of illustration.
A key to understanding that mass vaccination (and vaccination mandates) won’t contribute to flattening the curve. Early treatment will have a much larger effect, but is unfortunately largely ignored.
I would not necessarily say that reporting RRR is cheating, but not reporting ARR is also not telling it all, which might not be a sign of adamant honesty on the side of those reporting on (their own) treatments and products.
@@cylon5741 On Dr Syed's side maybe just an intentionally neutral definition of the two concepts. On producers' side definitely an intentional omission that I do not condone for no reason, no matter what the good will of an explanation might be.
@@cylon5741 Both are needed to accurately understand the implications of an intervention. A low baseline risk to begin with will translate into a low decrease in absolute terms even with a high relative effect and conversely, if the baseline risk is very high to begin with, a moderate relative effect can translate into a pretty sharp decrease in absolute terms. .
@@andreasmelkersson3592 but when baseline difference is not much,it can be ignored..this is where it is being published wrong to show big numbers...Even with same factor you can have different baselines but difference should not be too high. Too many journals make it over the top just because they found a small difference in baseline which actually is common and makes no sense in describing it as risk.
In the case of a novel virus, where the entire population is at risk then surely the absolute efficacy is more relevant? And the. Set this against the known and unknown risks of the “novel” vaccines..!
Normal studies include other medical interventions that can reduce risk of death but our Gov has refused to let Drs use them. We also have the deadliest vaccine in vaccine history and the adverse events are just not happening to frail ppl. My question is: Moderna and Phifer EAU studies, there is no statistically significant differences between the control and vaccinated groups. That seems like it might be pretty important to disclose and discuss. Am I thinking of this incorrectly? It seems like you can’t discuss RR and ARR when technically there is no difference between control/vaccinated groups. Help to understand plz.
What is the the significance of absolute Baseline risk? For example, if the absolute risk of a particular negative outcome is 90% that would mean 90% of a population is going to get that negative outcome. An intervention with 50% efficacy would benefit 45% of the whole population. If the absolute risk for same negative outcome is only .09% then an intervention with 100% efficacy would only be benefiting.9% of the population. So absolute risk matters a lot. I don't think your talk it besides that very well. The problem with the covid vaccine is that so few people are at risk of serious outcomes from covid it really doesn't matter much whether we intervene or not. Worse than that by taking the vaccine a person moves from a population with a low risk of a negative outcome into another group with an even with a higher risk of the outcomes you're trying to avoid like death
How did they define/dx an actual covid case. We know the testing is pretty sketchy. If they used PCR to determine whether someone had covid or not, I think their data is completely contaminated by a test that is not diagnostic. Would be interesting also to compare the control population demographics with the study population. Lastly, as a former big pharma rep, I am highly skeptical of any data they put out. Yes we have the fda but when your former executives go back and forth between jobs in pharma and fda you have a serious credibility problem. Final final….there is so much money provided by big pharma to fda, cdc, nih etc through “grants”….it’s laughable to think that there’s no contamination. Pfizer has a deplorable record when it comes to honesty and integrity. We pretend everyone has integrity and I get the professional courtesy in the physician community but come on man; research and academia often completely fails to account for the reality of the world. Look at the real world failure of the vaccine in Massachusetts, Ireland and Israel to prevent transmission. These are some of the most vaccinated people on earth. The study numbers do not hold up in the real world yet the propaganda machine rolls on.
Let's say you make a vaccine for a disease that affects 1 out of a million people, and it is 100% effective. Is it worth it to vaccinate everybody in this case? This is the real question regarding the covid vaccines. In addition, the vaccines should be compared against treatment options like ivermectin.
Thanks. You always get at least one smart a** who comments about this thinking the person in the video is cheating w the numbers. I hope those posters have learned something now..
relative risk is just a way of making small improvement look more significant than it is
I think the numbers in the Pfizer trial were 170 out of 43,000 (168 from placebo and 8 from vaccine arm with positive PCR and ‘at least one symptom). Hence the famous 95%. This hurts my brain.
@@ritcha02 I see your point ... without the vaccine 168 people would have caught th virus ... with th vaccine only 8 people would have caught the virus. Now here's the hard part for me at what point do you shut down the economy mask all the citizens and keep children at home ... using the same numbers insisting on a vaccine disrupted 43000 lives ... potentially saved 160 lives ... still cost 8 lives though there's nothing to go home to ..
but that's ok we'll just keeping printing money and pretending it's worth something just remember ... you get what you pay for. Eventually the quality of the underlying asset matches the value of what it is being exchanged for
@@ronaldjohnson7449 exactly, same is true over and over for statins for example
This part was missing in my earlier comment : The FDA requires "Provide absolute risks, not just relative risks. Patients are unduly influenced when risk information is presented using a relative risk approach ; this can result in suboptimal decisions. Thus, an absolute risk format should be used"
Personally I agree with you... also relative risk reduction(which is vaccine efficacy) is just that- a relative risk reduction of contracting Covid. It is not the same as Protection against contracting Covid, which unfortunately many get confused by the figures given and their significance.
But ppl wouldn't understand it..and hunk less thacn 1% seems really bad. But I do agree, we are supposed to get all the info.
@@vanessap8717 it is really bad, it does not just seem so...
Absolute risk reduction is essential to doctors and patients alike to decide on risk-reward. Relative risk is normally used to compare between trials. But this being misrepresented in the media, and the wrong figures are shared to make an informed decision.
I would prefer both figures to be shared, which I believe is part of the recommended guideline for how trial results should be reported.
Please discuss efficacy of the sinovac. That’s the one receiving by our country the Philippines when will this approved
Terrible!!! Misleading. What an insult. Fooling the masses! Shame :`( How is it that such fallacy is rampant to push this experiment. Why,,why why
What is the significance of absolute risk? For example, if the of a particular negative outcome is 90% that would mean 90% of a population is going to get that negative outcome. An intervention with 50% efficacy would benefit 45% of the whole population. If the absolute risk for same negative outcome is only .09% then an intervention with 100% efficacy would only be benefiting.9% of the population. So absolute risk matters a lot. I don't think your talk it besides that very well.
The problem is the AR was not revealed to the public for any Covid vaccine even though the FDA requires full disclosure of both ARR and RRR so as to not mislead people.
These vaccines are not approved by the FDA. I understand Pfizer is pursuing approval and perhaps then they will release both numbers.
@@tibitaylor according to the data they're getting from the global campaign, they're NEVER gonna get an approval unless they manipulate heavily the numbers.
1-3 percent according to lancet - that’s the AR for them.
@@tibitaylor It's approved now. I still don't see them reporting the RRR.
0.7% arr for pfizer
Nice. Except if the vaccine had no side effects 95% relative risk reduction would be great. When the side effects of an intervention is significant the number needed to vaccinate minus 1 says how many people get the side effects with no benefit. 10000÷84 . More than 110 people vaccinated will experience the side effects without any benefit.
I understood that relative risk reduction was devised to help researchers in distinguishing between studies.Absolute risk reduction is the risk factor that should be told to the patients...
Thanks for clearing that up, doc! Wouldn't it be more transparent if they also included the risk of having a negative outcome of the treatment, medication or vaccine, like hosptalisation due to complications or even death?
"Wouldn't it be more transparent"
Yes, and this is why they do not include this information.
think absolute reduction is important. Example: many big pharma companies claim their drug reduces in a 50% the chances to get disease X. But then the actual people getting the disease is 2 in 100000. So reducing it to 50% means now it's ONE person getting the disease, not TWO. That is a 50% reduction. See what they do there? Absolutes are important. Maybe even more. A 50% of a very small number is a small number.
Excellent presentation on data with clear communication, making it easy to understand. Thank you, Dr Been, for all the work you’ve been doing so far!
Dr. Mo been - you did a phenomenal job with this explanation. I am still trying to understand why they only report one number - That being relative risk reduction - instead of following the FDA guideline which states they should report both numbers together. Thank you very much!!
To roll out vaccines for everybody and earn billions.
Thank you very much for your excellent review.
Engineers would consider such infinitesimal reductions in disease to be nothing more than statistical noise. We tend to work with larger deviations. 🙂
A slight bias in subject selection, for either group, would render such numbers meaningless. This is absolutely astounding.
My understanding is that typically the control group receives standard of care. By having corrupt medical leadership not treating outpatients with available interventions--vitamin D, HCQ, IVM, etc--the control group does not do as well as people should when being treated with the best combination of interventions available.
So the vaccines appear to be more efficacious than they really are in comparison to alternative interventions.
In my view, I would rather stick to absolute risk reduction as the data demonstrates from the get go (without intervention) the risk is very low 0.88%. Vaccine intervention is not required. Relative risk reduction is a red herring that distracts one from that fact.
Absolutely, I have also posted a similar comment, it is unbelievable that Drbeen despite explaining correctly the difference between RRS and ARR still makes the wrong conclusion
I'm 75. My risk is higher than 0.88%. In other words, other factors need to be taken into account when talking about individual risk. Beyond that, I also care about the plight of society as a whole. Whenever others don't care about getting sick (because their risk is relatively low), then the number of people who can infect the more vulnerable like me, either directly or through a younger member of the same family, shoots up. That's a concern in a number of ways. For example, in whole societies the risk percentages get applied to very large numbers. Hospital capacities can be overwhelmed, causing even more preventable deaths. Vaccines can reduce the extent to which hospitals are overwhelmed, and relative risk reduction would shed some light on that problem.
Finally, although the absolute risk reduction of 0.88% may seem low, every single infected person raises the possibility that a random mutation will lead to a truly nasty variant of the virus that is much more deadly and/or transmissible than the ones we already know about. That risk is too often neglected, in my opinion. I believe the situation is more complex than you imply.
@@RussCR5187 In that case you take the vaccine however let's not force the other 99% of us in that direction. Besides, have you looked into Ivermectin? You should.
@@michelm3624 👍👍👍
@@motumer But what about the mutation concern I brought up in connection with the 99% who don't get the vaccine and don't care whether they get infected or not?
PS -- Yes, for a number of months now I have had a supply of ivermectin large enough to cover my entire family, and I successfully used it myself as an early treatment after coming down with a fever and nasty headache. Don't actually know if the symptoms were from covid or influenza or common cold. But truth be told I don't really care what it was in this time of pandemic. Better safe than sorry.
If you ever wondered what a good physician looks like...listen to this gentleman's words.
I think you forgot an important part of the equation. The benefit of Absolute Risk Reduction is that it takes into account the size of the study’s participants. If you have a relatively small study group showing good efficacy vs a large group, this is very different. The good efficacy in the small group may be more greatly effected by the make up of the individuals. This effect is more balanced out in a large study. So ARR is important in fully evaluating the benefits of an intervention.
Plus. It can be easily manipulated. If the participants were excluded in the two groups in a proportion that ARR keeps the same but the CER becomes always lower, the RRR can be as high as they want, because every number divided by a denominator less than 1 becomes always higher, by has no relevance in clinical means, it only matters for scientists to evaluate experimental protocols. so my opinion is that the RRR shouldn’t even be cited because is affected by the experimental modeling , and the ARR is not, that why it’s called absolute.
@@leandromantovani4282 Great explanation! Thank you.
@@leandromantovani4282 sorry I didn't understand that bit .are you a professional, andin any case can you provide me a material to search this further
Bingo. The RRR is independent of the size of the cohort.
So if the ARR has a large number, the results are more reliable. In the case of Pfizer then, is 18000 participants a good size? Which in turn would provide a reliable / reasonable RRR? So how is the Pfizer efficacy (RRR - 95%) misleading?
So the selection of the experimental and control groups becomes critically important when numbers are so small. They must be carefully age matched, matched for comorbidities, matched for asymptomatic previous Covid disease/ potential immunity. What am I forgetting ?
Very good Dr. Been. I wonder why other ppl are making the absolute risk reduction look bad. Thank you for making this video, it makes sense. I'm still not getting the jab lol for other obvious reasons.
Thanks for this explanation and all the info and science you bring to this discussion.
Thank you for reviewing the data from Pfizer and calculating it in an easy to learn manner.
@Monroe Bothwell Do you really think that people are going to fall for your spam?
I was hoping you would cover this I was still unclear on a few points and you did a beautiful job explaining.Thank you ❤️
My pleasure.
@@DrBeenMedicalLectures Can you do a video to cover the role of the endocrine system in immune regulation. So patients can evaluate for themselves if endocrine disruptors "may" be playing a role?
www.hormone.org/your-health-and-hormones/endocrine-disrupting-chemicals-edcs
You could have 1 million people in a trial, if 2 get sick in the control group and only 1 in the intervention group, you have a 50% RRR. Contrary to what is said in this video, relative risk reduction does not show the quality of the data but it is a statistical shenanigan to amplify noise and make it appear as real!
Succinct!!! 👌🙌👏👏👏👏
Not really, pal.
Flights from DELHI to NEW YORK ??? 😯 Unbelievable? 😰
Thank you so much Dr Been for explaining what I know that most people aren't even aware of when it comes to statistical reporting let alone understand how it can be completely misunderstood and mis interpreted even by intellectuals alike. This however highlights my questioning of why we are trying to vaccinate everyone when only a fraction are at risk of getting seriously sick from the virus. To me it would be like giving everyone blood pressure medication when only those with chronic high blood pressure would need it. Why not just vaccinate the sick and elderly?
Thank you for your videos. I have learned important information with your videos. Great Job!
If we tell people that one out of 100 vaccinated people can be prevented to have the disease , I really doubt a lot of people will jump in to get vaccinated, not to mention the side effects. And the vaccine companies didn’t provide people the ARR for a good reason.
@Chloe M It's much bigger than that.
@@andreasmelkersson3592 It's actually less. 142 needed to be vaccinated by Pfizers vaccine to prevent 1 case. 88 for Modernas. However if it's being efficient for preventing severe disease as it logically should then it's still beneficial. However things like this from the vaccinecompanies aren't building much trust.
DrBeen is a really Cool person 👍🏼
God of explanation.
Something to consider: the number used to represent the risk of contracting the virus in the control group may be seriously flawed if they are using the PCR test. This value is easily manipulated by altering the cycle threshold. The question is: Are the researchers being transparent about the CT used for each group? Could there be any financial motivation to use different parameters in each arm of the study? The pharmaceutical industry does not have a good track record of integrity in self-reported data when it comes to justifying their bottom line.
Yeah but the point is that relative risk reduction can be a meaningless stat when your study has a tiny number of cases.
Correct, but the sample sizes should be reported too. That kind of study would be discredited right away.
Well done ! Best explanation ever ❤️
Delivered splendidly
Helped alot
God bless u
How is this compared with IVM /Mouthwash (Listerin) nd nasal spray
Very good question.
Thankyou for all the informative videos.
Kindly discuss whether Indian vaccines are safe for breastfeeding mothers.
Loved this discussion.. hoping to see more of these epi/biostat related videos...
Kindly make more videos on Epi/Biostat related videos esp/for USMLE ...Thank You
If you are a hiring manager...you look for people like Dr. Been.
Thank you- subscribed.
There's another factor that's not taken into account. If the treatment itself causes the same or similar harms as the disease in a significant size, that needs to be taken into account. It also confounds the analysis with regards to causation.
Please tell us about indian vaccine covaxin.. We are really very confused between.. Astrazeneca and covaxin
Nicely Explained sir
And what if you found out the major difference in the cases between the placebo group and the test group is a couple groups in the placebo group went to a party but no one in the placebo group did anything? When the cases are that small it is definitely possible.
It is good explanation but I don’t think still the vaccines are a good strategy for eliminating covid
it was more a strategy to give people a better outcome if they got covid. on average that seems to have worked out.
So In the Covid trials the did they measure who had positive PCR together with at least one symptom from each arm and that’s the relative risk? So the number takes into account that some would be expected to have got Covid anyway? Did they look at severity of symptoms?
Mitzvah. A good or righteous deed
Got it 🙏
Wouldn't it be if you gave it to 100 people that 20 %of those people would have gotten the events or but now 10 %. Also that is only tracking 2 weeks after second shot right?
What are the relative risk of the side effects?
What people talking about covid in the comments are misunderstanding is that the control group is affected by non-pharmaceutical interventions (NPIs). Therefore it’s not a “true” control group.
At the moment we cannot have a true control group, because govs have regulation in place to stop the spread.
Hence, when talking about ARR for covid vaccines it’s in relation to the measures. So this means that the vaccines are a little more effective than current measures, allowing countries to reduce their NPIs for more societal and economic benefits again. Which is why they’re being rolled out.
Its an even playing field as both the control and intervention group are exposed to the same government regulations to stop the spread.
I feel that we would have a perfectly good way of looking at the quality of the improvement achieved by the experimental treatment by focusing on the percentage point difference between its incidence rate and that of the placebo, once we merely agree upon a benchmark measure of what the maximum difference might be. Then simply divide the actual difference by that agreed-upon maximum.
While the maximum would be established as an arbitrary agreement among experts, it would often produce more understandable data then the current relative risk measure. Here's why.
If the incidence rate for the placebo is 10/18000 and that for the experimental treatment is 0/18000, we would announce with great fanfare that the treatment had 100% relative-risk effectiveness; which, IMO, would be sort of meaningless (except for derived computations about implied reductions in demand for hospital beds in a multi-million-size base population).
I have made this argument in my own Bulletin as such:
Your odds of GETTING Covid have only been reduced by 1%. However among that 1%, your odds of a negative outcome have been reduced by the 95%.
If someone has auto immune issues should they get vaccines
No
Thanks for the explanation, and i understand the importance of publishing the relative risk number. But surely publishing the absolute risk reduction alongside would give people the knowledge to decide if they want the medication or not. Eg. Why would i take medication/ vaccine for a virus if the absolute risk reduction is .8%... most medications have side effects, they might be of benefit for one thing but cause other problems... that way people are empowered to calculate the risk for themselves.
As far as I understand, absolute risk reduction is mainly useful for calculating the Number Needed to Vaccinate.
Even if an intervention is 100% effective and this intervention prevents a disease which is 0.1% fatal, it means that we have to vaccinate 1000 people to save one.
Wouldn’t a simple explanation be that result of a relative risk reduction calculation applies to the treatment of a general sample of people, and the absolute risk reduction is the difference in outcomes when comparing a treated group versus a placebo group?
A bit late to comment here... But I was just thinking.... by ignoring ARR (or not revealing it to the public)... it can effect the way people make decisions of whether to accept or reject the intervention. I'm no statistician... but just playing with hypothetical numbers.
Say for discussion's sake... the numbers happen to be much higher... e.g. Instead of 8/18198, it's actually 800/18198 for the intervention. Then, for the placebo group it's16200/18325 (instead of 162/18325). This would ramp up the ARR:-
ARR = 88.4 - 4.4 = 84% (A better number...) - instead of 0.84%. Reporting this alone would be enough (perhaps) for people's psychology.
RR on the other hand, would still remain at 4.4/88.8 = 0.05, which leads to the same efficacy percentage (RRR) of 95%. A good number as well.
So now that both numbers are high... it might be in the interest of the organization/corporation to report both numbers...
In the case when the experiments were done, ARR was very low... So it might have been best not to report it , though I think the FDA asks to report both.
I donno, am I making sense? :) Would appreciate any feedback... Trying to understand all this from various points of views :)
Relative risk makes sense only when there is difference at absolute level.
Drbeen The problem is definition and consistency of what they (vaccine manufacturers) are measuring: when they say the INT group or the CTR group got C19, how did they establish that anyone got C19? was it PCR test? was there variability in measurement? Also, the general public ought to be better informed in as many ways as possible so that we understand perfectly and it becomes common knowledge how to understand RRR in the context of ARR. I feel you have made inroads to this later request but not enough IMHO. I hope in future you can explain it in more different ways.
20,000 people in INT group 20,000 people in CTR group, how many got saved from C19?
what is practical signifcance of RR some say RR should be around 2 or .5 ?
Does there come a point where the number is so small it is insignificant or considered a statistical rounding error? and if so what is it? eg 8 Vs 162 out of 18,000 - they are both sub 1% and therefore it would be a reasonable expectation for the numbers to be different if run again on the same size group. For example 8 could be 80 and yet still sub 1% but giving a very different relative risk and the actual risk pretty much unchanged?
Thanks for refreshing my understanding of statistics but statistics are only helpful to me if I can see the conditions during trials and the data of persons chosen for the trials etc. ( Not especially for the vacc trials but in general)
Because one side is the mathematical numbers and the other side and as important are environment and the people themselves when you measure something that is relevant for „real life“, with living beings. So I tend to look rather at statistics that are made very transparent concerning mentioned factors. In an ongoing pandemic with restrictions that change all the time that can make a huge difference to the outcome and the kind of criteria how to choose, who to choose for the trial groups. ( example: age might play a big role bc elderly people might take restrictions more seriously, behaving to the rules, due to their higher risk so I would like to see if the age groups are well balanced i.e. same nr of same age in both groups and the urgence for studies/ results doesn‘t make it easier to choose thoroughly)
Also the timeline can be important. When did trials start? How are participants chosen? Are they chosen before trial start? Or could people drop in for participation over a longer period of time? Was there an interest that could push people to participate?
These are just examples, I‘m not telling the vacc trials aren‘t reliable, it‘s just my general reluctance to look at stat‘s without knowing other factors during a study/trial that could have influenced the results.
The unvaccinated have 0 chance of covid vaccine adverse effects. 162 versus 8 out of 18 thousand odd is almost no significant difference
This type of considerations should be done also including all adverse event: symptomatic disease, severe disease, ICU, death. The RRR and ARR and NNTT should be calculated also accordingly, wouldn't you agree? Separated by age and comorbidity classes this would be of real help in making a decision on vaccines, personally and at administration level. Do you think we can get these calculations by the data we have from most studies, or that we should get them?
Could you kindly explain how they defined the “Event” in the vaccine trials. Thank you
Positive PCR.
@@koliszemrany4951 plus one symptom.
If it was positive PCR plus at least one symptom, then wouldn’t the Study has missed out on positive asymptomatic cases, bearing in mind majority of Covid cases are asymptomatic?
This is from an INTERIM phase 3 trial report (completion 2023). Environment was not same as roll out during very infectious variants. Neither was it for serious disease or death but whether there was a +ve RT PCR test at what Ct amplification? Given the low prevalence of SARS-CoV2 in the trial it seems clear the RT PCR results included false positives. What was placebo? For AZ it was not normal one but another non C19 vaccine that lowered the immune response.
thanks for explaining so clearly. one question> does vaccine reduces the severity of infection in vaccinated people?
However, it is very important to know your absolute risk to begin with before calculating your relative risk reduction from a vaccine by seeing which risk groups you are in. If you do not have hypertension, diabetes, are old with a compromised immune system, or are obese, then you are not in with the 98%% group of those that have serious complications or death resulting from infection. If you are then in that healthy 2% group your absolute risk is 2%. Now we can add that, according to several international studies, there is a natural immunity for 30 to 50% of the general population probably resulting from previous exposure to coronavirus infections from the common cold. In any case this means that if you are in that 2% group we can further reduce your absolute risk by let’s say 40% if we split the naturally immunity range that was from 30 to 50%. This means that your absolute risk of having a serious outcome from a Covid-19 infection is approaching 1%. Now you can decide whether it is worth getting vaccinated to further reduce that 1% risk.
I would just like to add that if anyone has better numbers for my analysis above by all means plug them in as a way of correction. What I am really interested in and trying to get at is that you have to know where you are at the start to know where you may be at the end.
Very clear explanation, however if upwards of 15:/: of people with Covid end up hospitalized with examples in Canada of younger ages no comorbidities ... also need to consider potential hospitalization
It’s statistics but doesn’t paint a full picture and take into consideration long hauler group and some people that suffer organ damage that’s the problem with this disease it’s creating a lot of damage to what could well be a good chunk of those that survive.
excellent points. But it never was about the risk for healthy people with a good immune system.
We already knew the risk was very small. It was about protecting the old and vulnerable.
With that done the new variants might change the equation.
And the issue of long Covid symptoms remains.India has presented some interesting variables.
@@MrBallynally2 I disagree healthy people and kids are getting impacted. There is still alot we don’t know and are learning therefore in my view assuming is just that assuming and somewhat magical thinking that if your young and healthy your immune. To much of this started early on assuming it was the old and vulnerable with an unknown it’s wise to make no assumptions and remain curious and open. Also the VOC might have less impact then we think currently a new study shows that on reinfection the virus uses 2 items that are a byproduct of damage that our bodies make one being billiroubin that this virus literally uses as cover/coating to evade antibodies. See below article on study that came out this week
The SARS-CoV-2 coronavirus plays hide-and-seek with antibodies
It’s a tactic that is sometimes used in horror movies. Pretending to be a zombie so that the real zombies won’t see you. Or pretending to be totally unresponsive, so that the aliens think you are one of them (Invasion of the Body Snatchers). It turns out that the coronavirus uses a very similar and clever strategy. Researchers at the Francis Crick Institute, Imperial College London, Kings College London and University College London have found that the SARS-CoV-2 coronavirus uses pigments in the body to hide from the immune system.
This process involve the waste substances biliverdin and bilirubin. These cause the green and yellow patches on the skin when bruised. The substances - released mainly during the breakdown of old or broken red blood cells - are able to bind to the spike protein of the coronavirus. And when that happens, it becomes a lot harder for antibodies to attack the virus. The researchers say this also helps explain why some people get very sick and others don’t at all.
The full study was published in the journal Science Advances.
Green and yellow
Biliverdin and bilirubin are compounds that occur naturally in the body. They are released during the breakdown of so-called heme compounds that attach to oxygen molecules in red blood cells. The compounds can be seen when bruises change color. This is because heme compounds are first broken down to biliverdin (green) and then to bilirubin (yellow). The toxic substance bilirubin normally leaves the body via the liver and gallbladder.
What the researchers discovered with the help of blood serum and antibodies from people previously infected with COVID-19, is that biliverdin prevents antibodies from binding to the spike protein in 30% to 50% of cases. This significantly weakens the immune system.
Acts as a booster
For the study, the team used, among other things, X-ray images to look in detail at the interaction between spike proteins, antibodies and biliverdin. What they discovered was that biliverdin attaches itself to the so-called N-terminal domain (NTD) of the spike protein. This stabilizes it, preventing the spike from opening up and exposing parts of its structure. As a result, some antibodies are unable to access their target areas and therefore can’t bind to the virus and neutralize it.
According to Annachiara Rosa of the Crick Institute, the interaction is especially strong as soon as SARS-CoV-2 reaches the lungs: “If a patient’s lungs are infected, the virus damages blood vessels and causes an increase in the number of immune cells. Both of these effects can contribute to an increase in biliverdin and bilirubin levels in the surrounding tissues. And with more of these molecules, the virus has more opportunities to hide from antibodies. That is a really remarkable process, because the virus may be benefiting from a side effect of the damage that it has already caused.”
Also see our COVID-19 archive.
people also like 1 in 10 type numbers saw RR 1.3 obesity vs colon cancer how much greater is my risk if fat? need absolute numbers to say well 1 in 300 or what ever
So this RRR is significantly lower in the younger age groups I suppose? I mean the risk reduction would not be even close between a 15 year old and a 65 year old... right?
Oh I disagree. To me the 95% relative risk reduction is far more meaningless than the 0.84% absolute risk reduction. It is cheating by not advertising both numbers.
Wrong, wrong and wrong.
@@andreasmelkersson3592 what is wrong?
@@josephduncan4356 Both are useful metrics. the ARR will always be low since it depends on the event rate. You can still have a super, high-effective vaccine. the RRR gives a quicker, more meaningful idea of the vaccine's efficacy since it ignores the transmission setting and gives a good general overview
@@andreasmelkersson3592 Yeah but the point is that the effectiveness rate means very little when there are a small number of cases. If you were willing to accept a tiny absolute effectiveness rate then you could probably show that drinking orange juice is 99% effective and preventing getting hit by a train.
@@andreasmelkersson3592 ARR is not always low...see the difference between lung cancer in smokers vs non-smokers.
dr Been, your control is 20% not 100 %, please clear that, thank you so much
Efficacy = Absolute Risk Reduction. Why are they using RELATIVE RISK REDUCTION..MISLEADING.
reporting bias that unfortunately leads to authority bias ie 'expert halo' Dr Fauci effect
Efficacy is NOT equal to ARR.
Efficiency = Relative Risk Reduction.
Why are you deleting my reply.
@@tarmahomedjoosub5552 Wrong. Read more Relative risk reduction vs Absolute risk reduction.
Read Babywin Wea post
Do those figures include mortality "from" or "with" the device within 28 days of injection during the trial. The relative risk/actual risk to life or of significant illness must include all serious issues not just RT-PCR positive test result from a fragment of SARS-CoV2 RNA strand that is not known to be active or infectious - can get positives 6 weeks after infection or breath in inert strand in say a hospital waiting room.
You’re saying that relative risk is given more relevance because it means quality of the absolute risk? You mean that RR value is directly extrapolated into the whole population -despite its an experimental sample- in order to establish public health policy and not the absolute risk?
I would want to know both relative and absolute to make any meaningful decision. Providing only 1stat is misleading. It achieves a goal of who?
I think absolute reduction is important. Example: many big pharma companies claim their drug reduces in a 50% the chances to get disease X. But then the actual people getting the disease is 2 in 100000. So reducing it to 50% means now it's ONE person getting the disease, not TWO. That is a 50% reduction. See what they do there? Absolutes are important. Maybe even more. A 50% of a very small number is a small number.
link to efficacy data?
I am afraid your presentation misses the main point in particular after minute 22'30. The FDA requires that :The reason is obvious. If I tell you that I can reduce your risk by 90%you cannot make a proper decision if you do not know what your absolute risk is without the treatment. Why do you think the Pharma does not comply with what the FDA guideline ? Looks a much better selling argument to say only efficacy (i.e., Relative Risk Reduction) is 90 % than adding that the absolute risk reduction is 0,7 % for Moderna or 1,1 % for Pfizer by way of illustration.
09:55 I think your calculation of CER - (IER/CER) may be incorrect.
Any chance of an example with figures?
A key to understanding that mass vaccination (and vaccination mandates) won’t contribute to flattening the curve. Early treatment will have a much larger effect, but is unfortunately largely ignored.
I would not necessarily say that reporting RRR is cheating, but not reporting ARR is also not telling it all, which might not be a sign of adamant honesty on the side of those reporting on (their own) treatments and products.
@@cylon5741 On Dr Syed's side maybe just an intentionally neutral definition of the two concepts. On producers' side definitely an intentional omission that I do not condone for no reason, no matter what the good will of an explanation might be.
@@cylon5741 Both are needed to accurately understand the implications of an intervention. A low baseline risk to begin with will translate into a low decrease in absolute terms even with a high relative effect and conversely, if the baseline risk is very high to begin with, a moderate relative effect can translate into a pretty sharp decrease in absolute terms. .
@@andreasmelkersson3592 but when baseline difference is not much,it can be ignored..this is where it is being published wrong to show big numbers...Even with same factor you can have different baselines but difference should not be too high.
Too many journals make it over the top just because they found a small difference in baseline which actually is common and makes no sense in describing it as risk.
So RRR is basically a quality score for the intervention? Whilst ARR is the actual real world affect of the intervention? Is this right?
This video explains it poorly. Reply to this comment if you are interested in the truth.
@@andreasmelkersson3592 What is the truth?
Have they done this study on kids? If so what has that shown?
In the case of a novel virus, where the entire population is at risk then surely the absolute efficacy is more relevant? And the. Set this against the known and unknown risks of the “novel” vaccines..!
Normal studies include other medical interventions that can reduce risk of death but our Gov has refused to let Drs use them. We also have the deadliest vaccine in vaccine history and the adverse events are just not happening to frail ppl.
My question is: Moderna and Phifer EAU studies, there is no statistically significant differences between the control and vaccinated groups. That seems like it might be pretty important to disclose and discuss. Am I thinking of this incorrectly? It seems like you can’t discuss RR and ARR when technically there is no difference between control/vaccinated groups. Help to understand plz.
What is the the significance of absolute Baseline risk? For example, if the absolute risk of a particular negative outcome is 90% that would mean 90% of a population is going to get that negative outcome. An intervention with 50% efficacy would benefit 45% of the whole population. If the absolute risk for same negative outcome is only .09% then an intervention with 100% efficacy would only be benefiting.9% of the population. So absolute risk matters a lot. I don't think your talk it besides that very well.
The problem with the covid vaccine is that so few people are at risk of serious outcomes from covid it really doesn't matter much whether we intervene or not. Worse than that by taking the vaccine a person moves from a population with a low risk of a negative outcome into another group with an even with a higher risk of the outcomes you're trying to avoid like death
How did they define/dx an actual covid case. We know the testing is pretty sketchy. If they used PCR to determine whether someone had covid or not, I think their data is completely contaminated by a test that is not diagnostic. Would be interesting also to compare the control population demographics with the study population. Lastly, as a former big pharma rep, I am highly skeptical of any data they put out. Yes we have the fda but when your former executives go back and forth between jobs in pharma and fda you have a serious credibility problem. Final final….there is so much money provided by big pharma to fda, cdc, nih etc through “grants”….it’s laughable to think that there’s no contamination. Pfizer has a deplorable record when it comes to honesty and integrity. We pretend everyone has integrity and I get the professional courtesy in the physician community but come on man; research and academia often completely fails to account for the reality of the world. Look at the real world failure of the vaccine in Massachusetts, Ireland and Israel to prevent transmission. These are some of the most vaccinated people on earth. The study numbers do not hold up in the real world yet the propaganda machine rolls on.
100/.0084 is almost 12,000. This seems to say more about covid than about the vaccine.
Let's say you make a vaccine for a disease that affects 1 out of a million people, and it is 100% effective. Is it worth it to vaccinate everybody in this case? This is the real question regarding the covid vaccines. In addition, the vaccines should be compared against treatment options like ivermectin.
Thanks. You always get at least one smart a** who comments about this thinking the person in the video is cheating w the numbers. I hope those posters have learned something now..
for the algo
The explanation is good from a statistician's point of view but is absolutely rubbish from a physician's point of view.