“How Not To Get Fooled by the Medical Literature”
Despite everything that’s happened, Dr. Vinay Prasad still feels entitled to teach a 19-part online course titled How Not to Get Fooled by the Medical Literature, along with his fellow “medical conservatives“. Dr. Prasad also has many admirers who speak of his “lifelong commitment to scientific rigor” and even today describe him as “smart, driven by evidence, fearless, unafraid to take on the medical establishment“. Let’s examine his real-world contributions to evidence-based medicine (EBM) to see if his course is worth taking and this high praise is justified.
“RCTs are feasible “
In 2024, along with Dr. Mariana Barosa and EBM luminary Dr. John Ioannidis, Dr. Prasad published one of his typical papers, an opinion piece titled Evidence Base for Yearly Respiratory Virus Vaccines: Current Status and Proposed Improved Strategies. They “proposed” the following:
Well-conducted experimental studies, particularly randomized trials, are necessary to address persistent uncertainties about influenza and COVID-19 vaccines. We propose a new research framework which would render results relevant to the current or future respiratory viral seasons. We demonstrate that experimental studies are feasible by adopting a more pragmatic approach and provide strategies on how to do so. When it comes to implementing policies that seriously impact people’s lives, require substantial public resources and/or rely on widespread public acceptance, high evidence standards are desirable.
The key phrase here was “we demonstrate that experimental studies are feasible“. Their paper had an entire section titled CLINICAL TRIALS ARE NECESSARY AND FEASIBLE that said:
These cases demonstrate that RCTs are feasible and not prohibitively expensive if one incorporates pragmatic elements… what is most needed is robust effectiveness data on important clinical endpoints, such as severe illness, hospitalizations and death.
By boldly asserting that such randomized-controlled trials (RCTs) were feasible, Drs. Prasad and Ioannidis could then turn around and depict anyone who felt otherwise as lacking “high evidence standards” and indifferent to “robust effectiveness data on important clinical endpoints.”
Dr. Prasad explicitly stated this many times. In an article titled The Scientists Who Undermine Randomized Trials he said:
In the last few weeks there have been several examples of scientists trying to undermine randomized trials— arguing they are not possible, not practical, or not useful. All of these arguments are false, as I will detail.
Dr. Prasad repeatedly said his philosophy was “RCT or STFU for all” and claimed that anyone who disagreed must “hate EBM“. He felt his detractors, “left-wing public health” in this case, weren’t just wrong, they deliberately undermined RCTs.
In one of his many vulgar tirades, Dr. Prasad excoriated public health officials and rejected “forgiveness” for them. He wanted them to “admit they didn’t run any studies” even though he “told them to“. In yet another profane rant, Dr. Prasad spoke with contempt for doctors who corrected his anti-vaccine disinformation, blasting them as “fucking morons… who don’t know anything about evidence-based medicine.“
In Dr. Prasad’s telling, EBM was his domain, and the stakes couldn’t be higher. During a debate from before the election titled Which US Presidential Candidate is More Anti-Science?, Dr. Prasad argued that Kamala Harris was the larger risk. His top reason was:
Harris will continue approving annual COVID19 boosters, including for kids as young as 6 months old, without randomized trials.
As MAHA/MAGA loomed, Dr. Prasad claimed that the single greatest threat to the scientific enterprise was COVID boosters without RCTs and that Trump and Kennedy had to be put in charge to prevent this EBM catastrophe.
“I would conduct randomized experiments.”
While Dr. Prasad denigrated public health officials as incompetent buffoons who deliberately undermined RCTs, he claimed to be wiser and routinely boasted of his entirely hypothetical accompaniments regarding EBM. “I would conduct randomized experiments,” Dr. Prasad said about his imagined alter-ego.
I will disclose every time I use AI for an article here, and for this one I asked ChatGPT to:
Scour Dr. Vinay Prasad’s tweets, articles, blogs, YouTube videos and make a table of all the randomized trials (RCTs) he either “called for”, “proposed”, “argued for” or claims he “would have” done.
Here was the result:
| RCT / research question | Population / intervention | What Prasad said or did |
| School closure vs. reopening | Schools randomized to closure/reopening strategies | Explicitly said school closure/reopening could have been studied with randomized trials |
| Business/workplace closures | Businesses/areas randomized to closure vs. opening | Listed absence of randomized trials of business closures as a major evidence gap |
| Community mask mandates | Communities/groups randomized to mask mandate vs. no mandate | Repeatedly called for cluster RCTs of community masking |
| Masking children | Children randomized to masking vs. no masking | Said a pediatric masking RCT should have been performed |
| Masking in schools | Schools randomized to masking strategies | Specifically listed masking in schools |
| Masking in the United States | Large-scale community masking | Listed masking in USA among trials that should have been done |
| Masking after vaccination | Vaccinated populations randomized to masking/no masking | Specifically called for post-vaccine masking RCTs |
| Masking of health-care workers indefinitely | HCWs randomized to continued masking vs. stopping | Used as an example of an RCT that could be done |
| Perfect mask compliance | Individuals randomized to continue vs. discontinue masking | Asked whether compliant people randomized to continued masking have better outcomes later |
| Different mask strategies | Cluster RCT with multiple masking arms | Proposed testing different masking strategies |
| 3 feet vs. 6 feet distancing | Groups randomized to different physical-distancing requirements | Specifically listed 3 vs. 6 feet |
| Cohorting | Schools/workplaces randomized to cohorting strategies | Specifically listed cohorting |
| HEPA filtration in schools/homes | Clusters receiving HEPA filtration vs. usual conditions | Said this should have been studied in cluster RCTs |
| CR boxes / portable air cleaners | Classrooms/locations randomized to filtration | Listed CR boxes among absent RCTs |
| Plexiglass barriers | Locations randomized to plexiglass vs. no plexiglass | Explicitly listed plexiglass |
| Hand sanitizer | Locations/groups randomized to sanitizer availability/use | Proposed as a component of a factorial NPI trial |
| Face shields | Groups randomized to face shields | Proposed as a component of a factorial NPI trial |
| Masks + sanitizer + plexiglass + partitions + shields | Factorial design testing combinations of NPIs | Proposed a factorial design to determine proportional benefit of individual interventions |
| College-campus COVID measures | College populations randomized to different mitigation strategies | Listed college-campus interventions as lacking RCT evidence |
| Asymptomatic testing | Groups randomized to routine asymptomatic testing vs. not | Explicitly listed asymptomatic testing |
| Testing in preschool/schools | Schools randomized to testing strategies | Specifically listed preschool/school testing |
| Quarantine duration | Different quarantine durations randomized | Explicitly called for an RCT of quarantine duration |
| Restaurant masking rules | Mask only entering/leaving/bathroom vs. continuous masking | Used as an example of something that could have been randomized |
| Alternative COVID-vaccine dosing schedules | Different dose intervals/doses | Listed RCTs of alternative dosing strategies |
| Bivalent COVID booster vs. control | Randomized booster vs. appropriate control, severe-disease endpoints | Argued for clinical-endpoint RCTs |
| COVID boosters in low-risk/vaccinated people | Vaccinated/previously infected people randomized to booster vs. no booster | Repeatedly argued for clinical-endpoint RCTs |
| COVID vaccine in healthy children 5–11 | Children randomized to vaccination vs. control | Said a large RCT should have been done |
| Paxlovid in vaccinated/previously infected people | Vaccinated/lower-risk adults randomized to Paxlovid vs. control | Argued for RCT evidence before extending EPIC-HR findings |
| Paxlovid in low-risk vaccinated people | Low-risk vaccinated adults randomized to treatment/no treatment | Identified as an RCT that should have been compelled |
| Paxlovid and long COVID | People with long COVID randomized to Paxlovid/control | Discussed need for randomized evidence |
| COVID anticoagulation strategy | Full-dose vs. intermediate vs. prophylactic anticoagulation | Said the question required RCTs |
| COVID vaccine/booster clinical endpoints | Randomized vaccine vs. control with hospitalization/severe disease/death endpoints | Made clinical-endpoint RCTs a general regulatory requirement |
| Annual COVID vaccine updates | Seasonal vaccine vs. appropriate comparator | Advocated randomized studies to determine clinical benefit |
| Annual influenza-vaccine effectiveness | Flu vaccine vs. control, hard clinical endpoints | Said flu shots would benefit from more randomization |
| Multi-arm influenza-vaccine trials | Several vaccine formulations + control | Proposed repeatedly randomizing large numbers and advancing best-performing formulation |
| Influenza-vaccine deployment | Randomized strategies for how/when vaccines are deployed | Explicitly proposed randomized trials of deployment |
| Influenza challenge trials | Controlled influenza exposure after randomized vaccination | Mentioned challenge trials as randomized studies |
| Routine childhood vaccination schedule: U.S. vs. Denmark | Regions randomized to U.S. vs. Danish-style schedule | Explicitly proposed a cluster RCT |
| Childhood-vaccine schedule: hospitalization | Same schedule RCT with hospitalization endpoint | Specifically proposed hospitalization |
| Childhood-vaccine schedule: all-cause mortality | Same schedule RCT | Specifically proposed all-cause mortality |
| Childhood-vaccine schedule: disease-specific outcomes | Same schedule RCT | Proposed measuring diseases targeted by additional vaccines |
| Childhood-vaccine schedule: neurocognitive development | U.S. vs. lower-dose schedule | Proposed routine neurocognitive testing |
| Teclistamab in relapsed disease | Teclistamab vs. observation/appropriate control | Used as example of a trial people incorrectly claimed could not be done |
| Smoldering myeloma treatment vs. observation | Treatment vs. observation, survival endpoint | Said this could/should be randomized |
| New cancer drugs powered for mortality | New cancer treatment vs. control with overall survival | Argued new cancer drugs could be tested with mortality endpoints |
| Free COVID tests | Free testing vs. no free testing | Used randomization to free COVID tests as an example |
| Coronary-artery calcium testing | CAC testing vs. no CAC testing | Said definitive test would randomize people to CAC testing vs. no testing |
| Routine CBC/BMP/CMP testing in healthy adults | Different testing frequencies vs. none | Said randomized studies should be conducted |
| Healthy-person primary-care visits | Different frequencies of routine primary-care visits | Pointed to randomized evidence and argued questions should be revisited |
| Whole-body MRI screening | Whole-body MRI vs. usual care | Proposed a large RCT powered for all-cause mortality |
| Prostate-cancer screening policy | No screening vs. screening vs. patient-choice approach | Suggested a randomized study comparing approaches |
| Alcohol consumption | Multi-arm randomized alcohol-consumption strategies | Explicitly proposed a multi-arm randomized trial |
| Vaccine coadministration | COVID + influenza + RSV vaccines together vs. separate/no coadministration | Called for RCTs examining safety and efficacy of coadministration |
| Vaccines in pregnancy | Vaccination vs. appropriate control with clinical endpoints | FDA framework called for premarket RCT evidence |
| Pneumococcal vaccines – clinical outcomes | Vaccine vs. control measuring pneumonia | Proposed requiring evidence of actual pneumonia reduction rather than antibody titers |
| New vaccines/products – clinical endpoints | New products tested in randomized trials using patient-important outcomes | Proposed premarket RCTs assessing clinical endpoints for most new products |
| Expansion of vaccine indications | Existing vaccine in new population vs. control | Said populations should be included in premarket RCTs rather than extrapolated using immunogenicity |
| Multiple vaccines administered simultaneously | Different combinations/timing of vaccination | Called for better randomized evidence for benefits/harms |
If calling for RCTs were an Olympic sport, Dr. Prasad would win the gold every year, though a clear pattern emerges from this. Dr. Prasad only called for RCTs for things he didn’t like, especially COVID mitigations and vaccines. However, when these measures were studied via RCTs, such as boosters and the pediatric COVID vaccine, including for babies and toddlers, Dr. Prasad didn’t care at all. He just claimed these trials need to be repeated ad infinitum. I discussed the reason for this in my article Doctors Who Performatively Fetishized RCTs Aren’t Out to Advance Medical Research, But Rather to Sow Doubt & Mistrust.
Conversely, Dr. Prasad never called for RCTs for the policies and people he supported. He never called for an RCT of “focused protection”, for example. Advocates of herd immunity via mass infection never needed RCTs to back up their agenda. He further lauded any and all observational studies that purported to show vaccine harms, even claiming an awful VAERS dumpster dive was a “bombshell finding from dream team of authors“. Dr. Prasad made countless evidence-free claims about SARS-CoV-2 itself, claiming it was “natural and healthy” for children to contract it. While the vaccine was held to the highest standards of EBM, the virus was held to no standard at all.
Predictably, now that MAHA/MAGA is in power, Dr. Prasasd is all done calling for our medical establishment to do RCTs. I asked ChatGPT about this too – Has Dr. Vinay Prasad ever criticized Kennedy and MAHA for not doing more randomized controlled trials? Here was the answer:
I found evidence that Prasad repeatedly called for RCTs while addressing Kennedy’s agenda, but I have not found a clear instance in which Prasad publicly criticized Kennedy/MAHA by saying, in effect, “you promised RCTs and haven’t done enough of them.” I have not found a clear statement from Prasad saying something like:
“Kennedy/MAHA are failing because they aren’t conducting enough RCTs.”
Though Dr. Prasad said, “RCT or STFU for all”, what he meant was “RCT or STFU for you, but not for me”.
“We intend to stop the study due to slow enrollment and therefore the inability to generate relevant post-marketing data.”
My final question to ChatGPT was whether Dr. Prasad had ever worked on any of the RCTs he called for. The answer was “no.” Wiser souls than me figured this out long ago. In 2018 Dr. Eliezer Van Allen said to Dr. Prasad:
You are an oncologist at an academic medical center. You clearly have knowledge & feelings about trial design. You even did some power calculations already. So… write some grants, write some protocols, get some buy in, and do something to be part of the solution.
Just saying the same thing over and over again, whether in social media or in medical literature, ain’t gonna get it done.
He was right then, and except for Dr. Prasad’s career arc, nothing’s changed. While his competent predecessors at the FDA delivered Operation Warp Speed, Dr. Prasad repeatedly promised new RCTs on FDA podcasts (here, here, and here) but failed to deliver them. An RCT of the COVID vaccine failed to enroll enough participants while Dr. Prasad led the FDA’s vaccine division. “We intend to stop the study due to slow enrollment and therefore the inability to generate relevant post-marketing data,” the companies said. Oh well! This was the exact type of trial he and Dr. Ioannidis had previously deemed “feasible”. I guess this means Dr. Prasad undermines RCTs and hates EBM.
As such, the question isn’t whether Dr. Prasad was correct about the value of RCTs. The obvious truth is that no one tried to “undermine” them, and no one “hates EBM”. Dr. Prasad just made that up to dodge substantive critiques and discredit “left-wing” public health officials.
Rather, the question is whether Dr. Prasad was correct about the feasibility of RCTs and whether it was impressive for him to create long lists of RCTs for other people to do.
I don’t think so, especially considering these lists as well as his academic papers must now be considered in light of his dismal performance at the FDA. Not only did he fail to deliver the RCTs he promised, but he also singlehandedly rejected a successful Moderna flu vaccine RCT, a decision that wiser FDA regulators later overturned. Unsurprisingly, vaccine companies announced their intentions not to do further vaccine-RCTs thanks to MAHA.
Meanwhile, his former bosses Trump and Kennedy are warning about chemtrails and gallivanting with Andrew Wakefield. They are promoting quack treatments, and under their leadership the headlines read- NIH Grant Terminations Disrupt Hundreds of Clinical Trials, Affecting More Than 74,000 Participants. Some people “hate EBM” after all, and Dr. Prasad endorsed and empowered them. This is his real-world contribution to EBM, and it’s the only thing that should matter about him.
Dr. Prasad’s sycophants and enablers don’t feel that way, however. They act as if his failed tenure at the FDA was just a fever dream, and they excuse his support for MAHA/MAGA. No big deal! They still lavish him with praise all because he sent Tweets and made YouTube videos calling for RCTs. They claim that by doing this, Dr. Prasad was bravely upholding the highest standards of EBM, whereas his detractors either knew nothing of the subject, or were Philistines, indifferent to its principles.
It turns out, however, that writing the words “RCTs are feasible” is a lot easier than doing RCTs. Some of us knew this all along, and we were never impressed by all the amazing, wonderful RCTs Drs. Prasad “called for” and “would have” done. We recognized that he glorified RCTs not to advance EBM, but rather to weaponize it and advance his political agenda, something he didn’t hide at all.
So no, I won’t sign up for Dr. Prasad’s course or laud him as an EBM guru. I say all this as someone who values EBM so much, I volunteered to be in a COVID-vaccine RCT in 2020. My contribution was small, but at least it was real, and not everyone can say that.